Combined Notes for the Lecture Series

Lecture Notes for Course: ELC003 – Introduction to Renewable Energy Sources

1

ELC003: Renewable Energy Sources

Course Notes

Simon Watson, Ralph Gottschalg, Richard Blanchard

Department of Electronic and Electrical Engineering


2

Contents

1 Wind Power .................................................................................................................. 7

1.1 Wind Characteristics .............................................................................................. 7

1.1.1 Introduction ................................................................................................... 7

1.1.2 Atmospheric Winds........................................................................................ 7

1.1.3 Summary of Boundary Layer Equations ......................................................... 8

1.1.4 Log Law for Vertical Wind Profile ................................................................. 9

1.1.5 Simplified Power Law.................................................................................. 10

1.1.6 Wind Speed Variation and Averaging .......................................................... 10

1.1.7 The Weibull Distribution.............................................................................. 12

1.1.8 Estimation of Weibull Parameters ................................................................ 14

1.1.9 Calculation of Energy Yield ......................................................................... 15

1.2 Wind Resource .................................................................................................... 16

1.2.1 Introduction ................................................................................................. 16

1.2.2 Measuring Wind Speed at a Site ................................................................... 16

1.2.3 Estimation of Long Term Site Wind Speed .................................................. 18

1.2.4 Computational Models ................................................................................. 19

1.2.5 Wind Flow over Hills ................................................................................... 20

1.2.6 Initial Site Assessment ................................................................................. 22

1.3 Wind Turbine Aerodynamics ............................................................................... 25

1.3.1 Power Available in the Wind........................................................................ 25

1.3.2 Power Fluctuations ....................................................................................... 25

1.3.3 Extracting Energy From the Wind ................................................................ 26

1.3.4 Aerodynamic Lift and the Aerofoil............................................................... 32

1.3.5 The Tangential Induction Factor................................................................... 34

1.3.6 Relationship between Thrust and Torque, and the Lift and Drag forces ........ 35

1.3.7 Relationship between the Angle of Attack and the Lift and Drag Coefficients

37

1.3.8 Performance of a Wind Turbine ................................................................... 38

1.3.9 Wind turbine types ....................................................................................... 39

1.3.10 Electrical generator options .......................................................................... 40

1.4 Notation and Units ............................................................................................... 41

1.5 References ........................................................................................................... 41

2 Hydro Power ............................................................................................................... 42

2.1 Energy from Water .............................................................................................. 42

2.2 Developing a Hydropower Scheme ...................................................................... 42

2.2.1 Resource and Equipment Requirement ......................................................... 42

2.2.2 Choice of Turbine ........................................................................................ 44

2.2.3 Environmental Considerations ..................................................................... 45

2.2.4 Basic Calculation of Energy ......................................................................... 45

2.2.5 Calculation of Power .................................................................................... 46

2.2.6 A Typical Run of River Scheme ................................................................... 47

2.3 The Measurement of Head ................................................................................... 48

2.3.1 Why is the measurement of Head Important? ............................................... 48

2.3.2 The Variation of Head with Flow ................................................................. 49

2.3.3 Estimation of Net Head ................................................................................ 50

2.3.4 Turbine Setting Losses ................................................................................. 52


3

2.4 The Flow Duration Curve .................................................................................... 54

2.4.1 Variation of Flow Over a Year ..................................................................... 54

2.4.2 What is a Flow Duration Curve? .................................................................. 55

2.4.3 Q-Values ...................................................................................................... 56

2.5 Glossary .............................................................................................................. 56

2.6 Introduction to Water Turbines ............................................................................ 57

2.6.1 The Development of Water Turbines ............................................................ 57

2.6.2 Specific Speed ............................................................................................. 58

2.6.3 Shape Factor ................................................................................................ 59

2.6.4 Impulse Turbines ......................................................................................... 60

2.6.5 Reaction Turbines ........................................................................................ 65

2.6.6 Draft Tube ................................................................................................... 67

2.6.7 Cavitation .................................................................................................... 68

2.6.8 Selecting a Turbine for a Particular Site ....................................................... 68

3 Electrical System Aspects ............................................................................................ 69

3.1 Apparent, Real and Reactive Power ..................................................................... 69

3.2 The Balance of Active and Reactive Power .......................................................... 71

3.3 The Significance of Power Factor ........................................................................ 72

3.4 The Reactive Power of a Capacitor ...................................................................... 73

3.5 P and Q Transfer in Power Networks ................................................................... 74

3.6 The Generator ...................................................................................................... 75

3.7 Comparison of Synchronous and Induction Generators ........................................ 76

3.8 Connection to the Electricity Network ................................................................. 78

3.8.1 Power Factor Correction for an Induction Generator .................................... 78

3.9 Soft-Start Units .................................................................................................... 80

3.10 Power Quality ...................................................................................................... 81

4 Tidal Power ................................................................................................................. 82

4.1 Introduction ......................................................................................................... 82

4.2 Causes of Tides.................................................................................................... 83

4.2.1 Tides in the open ocean ................................................................................ 83

4.2.2 Mechanisms for Tidal Enhancement............................................................. 85

4.3 Tidal Barrages ..................................................................................................... 88

4.3.1 Turbines for Tidal Barrages.......................................................................... 88

4.3.2 Operational Strategies .................................................................................. 88

4.3.3 Choice of Barrage Site ................................................................................. 89

4.3.4 Case Study: La Rance .................................................................................. 90

4.3.5 Other Sites ................................................................................................... 93

4.4 Tidal Current Schemes ......................................................................................... 96

4.4.1 Turbines for Tidal Current Schemes ............................................................. 96

4.4.2 The Resource ................................................................................................... 97

4.5 Impacts of Tidal Schemes .................................................................................... 99

5 Wave Power .............................................................................................................. 100

5.1 Introduction ....................................................................................................... 100

5.2 Wave Climate .................................................................................................... 101

5.3 Shore Mounted Technology ............................................................................... 108

5.3.1 Oscillating Water Columns (OWC) ............................................................ 108

5.3.2 The OWC device on Islay .......................................................................... 109

5.3.3 Limpet ....................................................................................................... 111

5.3.4 Tapchan ..................................................................................................... 112

5.3.5 Locations for Shore Mounted Schemes ...................................................... 113


4

5.4 Near Shore Technology ..................................................................................... 114

5.4.1 Osprey ....................................................................................................... 114

5.4.2 The 'Mighty Whale' .................................................................................... 115

5.4.3 FWPV ........................................................................................................ 115

5.4.4 Point Absorber Float Systems .................................................................... 115

5.4.5 Pendulor .................................................................................................... 117

5.5 Offshore Technology ......................................................................................... 117

5.5.1 The Duck ................................................................................................... 117

5.5.2 The Clam ................................................................................................... 117

5.5.3 Pelamis ...................................................................................................... 118

6 Solar Power ............................................................................................................... 120

6.1 Solar Characteristics .......................................................................................... 120

6.1.1 The Solar Spectrum .................................................................................... 120

6.1.2 The interaction between radiation and matter ............................................. 121

6.1.3 The mechanism of absorption..................................................................... 122

6.1.4 Blackbody Radiation .................................................................................. 122

6.1.5 Radiation on earth´s surface ....................................................................... 123

6.2 Solar Resource ................................................................................................... 126

6.2.1 Geometry of the Sun, the Earth and the Collector Plane ............................. 126

6.2.2 Variation of Extraterrestrial Radiation with Season and Latitude ................ 128

6.2.3 Estimation of the Global Radiation on Tilted Planes ................................... 129

6.2.4 Devices for Measuring Global Radiation .................................................... 132

6.3 Basic principles of Photovoltaic (PV) Cells........................................................ 132

6.3.1 The crystalline silicon cell .......................................................................... 133

6.3.2 The basics of cell operation ........................................................................ 134

6.3.3 Cell spectral response ................................................................................. 138

6.3.4 Equivalent circuit for cell ........................................................................... 139

6.3.5 Maximum Power Point .............................................................................. 141

6.3.6 Effects of changes in irradiance and temperature ........................................ 142

6.3.7 Other types of cell ...................................................................................... 144

6.3.8 Applications ............................................................................................... 144

7 Biomass..................................................................................................................... 146

7.1 Introduction ....................................................................................................... 146

7.1.1 What is Biomass? ....................................................................................... 146

7.1.2 Developed Countries .................................................................................. 150

7.2 Conversion Routes for Biomass ......................................................................... 151

7.3 Biomass as a Fuel .............................................................................................. 152

7.3.1 Solar Store and the Carbon Cycle ............................................................... 152

7.3.2 Calorific Value and Moisture Content ........................................................ 154

7.4 Biogas ............................................................................................................... 156

7.4.1 Introduction ............................................................................................... 156

7.4.2 Control of Anaerobic Digestion .................................................................. 157

7.4.3 Biomass to methane ................................................................................... 159

7.4.4 Example Domestic Waste Treatment – Wanlip 5MW Biogas Plant ............ 159

7.4.5 Bio-hydrogen Production ........................................................................... 159

7.5 Introduction to Liquid Biomass Fuels, Biodiesel and Bioethanol ........................ 160

7.5.1 Biodiesel .................................................................................................... 160

7.5.2 Transesterification...................................................................................... 160

7.5.3 Bioethanol.................................................................................................. 162

7.6 Conversion of Biomass 1: Pre-treatment and Direct Combustion ....................... 164


5

7.6.1 Resource Base ............................................................................................ 164

7.6.2 Pre-treatment ............................................................................................. 165

7.6.3 Direct Combustion of Biomass ................................................................... 166

7.7 Conversion of Biomass 2: Gasification and Pyrolysis ........................................ 171

7.7.1 Gasification ................................................................................................ 171

7.7.2 Pyrolysis .................................................................................................... 174

7.8 Power Generation from Biomass Fuels .............................................................. 179

7.8.1 Steam Systems ........................................................................................... 179

7.8.2 Combined Heat and Power ......................................................................... 188

7.9 Conclusions ....................................................................................................... 190

7.10 References ......................................................................................................... 190

8 Integration of Renewables ......................................................................................... 191

8.1 Integrating renewables - the issues ..................................................................... 191

8.2 The operation of power systems ......................................................................... 192

8.2.1 Required Characteristics ............................................................................ 192

8.2.2 Power system hardware .............................................................................. 193

8.2.3 Demand forecasting ................................................................................... 194

8.2.4 Generation scheduling and spinning reserve ............................................... 195

8.2.5 Contingency analysis ................................................................................. 196

8.2.6 Optimum economic dispatch ...................................................................... 197

8.3 Plant generation costs and capabilities ............................................................... 199

8.4 Aggregation ....................................................................................................... 201

8.5 Penetration levels from variable sources ............................................................ 202

8.6 Cycling costs ..................................................................................................... 204

8.7 Reserve costs ..................................................................................................... 205

8.8 Discarded Energy .............................................................................................. 206

8.9 Penalties due to increasing penetration ............................................................... 206

8.10 100% Renewable Energy Generation ................................................................. 208

8.10.1 Combining different renewable energy sources .......................................... 208

8.10.2 Energy Storage........................................................................................... 209

9 Electricity Trading and Renewable Energy in the UK ................................................ 210

9.1 Introduction ....................................................................................................... 210

9.2 The State Owned ESI ......................................................................................... 210

9.3 The Electricity Pool ........................................................................................... 211

9.3.1 Overview ................................................................................................... 211

9.3.2 The Operation of the Pool and Pool Prices ................................................. 211

9.4 Hedging Your Bets ............................................................................................ 212

9.5 Deregulation ...................................................................................................... 213

9.6 The Non-Fossil Fuel Obligation and Renewable Obligation ............................... 213

9.7 The Climate Change Levy ................................................................................. 214

9.8 The New Electricity Trading Arrangements (NETA) ......................................... 215

9.8.1 Background ................................................................................................ 215

9.8.2 Buying and Selling Power under NETA ..................................................... 215

9.8.3 System Sell and System Buy Prices ............................................................ 215

9.8.4 The Balancing Market ................................................................................ 216

9.8.5 The British Electricity Transmission and Trading Arrangements ................ 217

9.9 The Impact on Renewable Energy Sources ........................................................ 217

9.9.1 Overview ................................................................................................... 217

9.9.2 Mitigating the Effects of NETA ................................................................. 217

10 Tutorial Questions ................................................................................................. 219


6

10.1 Wind Power ....................................................................................................... 219

10.1.1 Wind Characteristics and Resource ............................................................ 219

10.1.2 Wind Turbines ........................................................................................... 220

10.2 Hydro Power ..................................................................................................... 221

10.3 Tidal Power ....................................................................................................... 224

10.4 Wave Power ...................................................................................................... 225

10.5 Solar Power ....................................................................................................... 227

10.6 Biomass Power .................................................................................................. 229

10.7 Integration ......................................................................................................... 230

10.8 Electricity Trading ............................................................................................. 231


7

1 Wind Power

1.1 Wind Characteristics

1.1.1 Introduction

These lectures deal with the wind resource in general. In particular they will cover the nature

of atmospheric winds with emphasis on the lower boundary layer, the long-term probability

distribution of wind speeds, and techniques for wind resource assessment. The latter include

computational models and measurement based approaches.

1.1.2 Atmospheric Winds

Wind is the large-scale movement of air masses in the earth's atmosphere. It is caused

primarily by pressure gradients arising from differential heating of the earth's surface. In a

sense therefore it is another form of solar energy. Centripetal and Coriolis forces (see XFigure

1X below), arising from the earth's rotation also effect the air movement.

On a weather map, contours representing a common pressure are called isobars (given in

millibars where 1 mbar = 100 Pa and a standard atmosphere is 1013.25 mbar). Clearly, the

pressure gradient is perpendicular to the isobars. Because of Coriolis forces and centripetal

forces caused by strongly curved isobars, the wind will be deflected from a direction along

that of the pressure gradient. The resulting wind is called the gradient wind. For straight or

slightly curved isobars, it is also called the geostrophic wind. In the northern hemisphere,

winds rotate anti-clockwise into low pressure regions (cyclones), and clockwise out of high

pressure regions (anti-cyclones).

The discussion so far has ignored the influence of the earth's surface and the whole issue of

turbulent mixing which can be mechanically or thermally driven. The characteristics of this

lower layer of the atmosphere, known as the boundary layer, are critical to the exploitation of

wind energy.

The wind near the ground is significantly affected by such things as the topography of the

ground, including buildings, and the differential heating between land and sea.


8

Figure 1: The Coriolis effect: the two curved lines are of constant latitude, with point B

located directly south of point A. A parcel of air is moving south from A towards B. If

we ignore air friction, the speed of the air will remain constant with respect to the

ground. The direction will change, however, because the of the earth’s rotation under

the parcel.

1.1.3 Summary of Boundary Layer Equations

Wind speed increases steadily with height, and according to the 'no slip' condition, is zero at

ground level. This increase with height is known as wind shear. The wind shear profile

depicted in XFigure 2X is to some extent, dependent on atmospheric stability. Three stability

states are defined: stable, neutrally stable and unstable which are dependent on surface

heating and cooling.

If a volume of air is displaced vertically (and adiabatically) it will tend to return to its original

location if the atmosphere is stable normally caused by cooling of the underlying ground. If

it stays at its displaced location the conditions are said to be neutrally stable. In an unstable

atmosphere with a significant amount of surface heating, it will continue to move, due to

buoyancy forces, in the direction in which it was displaced.

The more unstable the conditions, the greater the mixing with the result that the velocity

gradients are lower.


9

Figure 2: The lower boundary layer.

1.1.4 Log Law for Vertical Wind Profile

A stability dependent form of the wind shear profile with height is given by a log law of the

form:

zz )L(z/+)z1n(z/k

U = U(z) 0ss0

*

(1)

where U* is the friction velocity (proportional to the square root of the turbulent shear stress,

which is assumed constant in the lower boundary layer); k is the von Karman constant (~0.4);

z is the elevation above the ground level; and zo is the surface roughness length. The

stability, s is a function of z/Ls where Ls is known as the Monin-Obukhov length.

For neutral stability which is usually taken to apply to the higher wind speeds associated with

wind turbine operationF

1F, this equation reduces to:

U(z) = U

k.1n

z

z

*

0 (2)

Since U* is difficult to evaluate, this formula is usually rewritten in terms of a reference wind

speed U(zr), at reference height, zr:

U(z) = U(z )1n(z / z )

1n(z / z )r

0

r 0 (3)

1 At relatively high wind speeds turbulent mixing of the air tends to „neutralise‟ any changes in vertical

temperature gradient created by surface heating or cooling.


10

Guidance in evaluating zo can be found in Reference 1. A useful approximation is:

zo = /30 (4)

where is the average height of roughness elements. Alternatively, reference can be made to

a look-up table such as XTable 1X.

Type of terrain Z0

Mud Flats, Ice 10-5

to 3x10-5

Calm Sea 2x10-4

to 3x10-4

Sand 2x10-4

to 10-3

0.01

Mown Grass 0.001 to 0.01

Low Grass 0.01 to 0.04 0.13

Fallow Field 0.02 to 0.03

High Grass 0.04 to 0.1 0.19

Forest and Woodland 0.1 to 1

Built up area, Suburb 1 to 2 0.32

City 1 to 4

Table 1: Typical values of surface roughness length and power law exponent (see

later), for various types of terrain

1.1.5 Simplified Power Law

For neutral conditions, present when wind speeds are high, a simple power law has been

found to provide a reasonable fit to the data. This expression is widely used by engineers, for

example when looking at peak wind loads on tall buildings.

U(z) = U(z )(z / z )r r (5)

where depends on the surface roughness.

Over the height range of 10m to 30m the following expression relating to zo has been

proposed.

= 1/ -1n(Zo / 15.25) (6)

For zo in the range 0.0001m and 1m, accuracy, in determining , of the order of a few percent

can be expected. Unlike the log law, there is no physical basis and in general it is not

recommended.

1.1.6 Wind Speed Variation and Averaging

Wind speed at a given location is always varying. There are changes in annual mean from

year to year; variations with season, with passing weather systems (synoptic), on a daily

z0


11

basis (diurnal) and from second to second (turbulent).

The Van der Hoven spectrum ( XFigure 3X) shows features corresponding to these different time

scales. The spectrum represents the strength of the wind speed variation at different

timescales. It can be seen that there is significant amount of variation at the 1-minute

timescale (turbulent) and at the 4-day timescale (synoptic). A spectral gap, separating

turbulent variation from slower variations, is apparent in the range of 10 minutes to 1 hour.

This is why the analysis of long term variations and wind statistics are based on 10 minute or

1 hour mean wind speed values.

Figure 3: The Van der Hoven Spectrum showing the amount of variation in wind speed

on a particular time-scale.

Time averaged wind speed values over a time interval T are defined by:

2/

2/

0

0

)(/1Tt

TtdttUTU (7)

where U(t) is the instantaneous wind speed at time t.

Despite inter-annual variations, it is conventional to work with annual statistics based on

hourly (or ten minute) averaged values. Frequency distributions are calculated based on

small wind speed intervals. In XFigure 4X, a typical wind speed distribution is presented; in this

instance 0.5m/s intervals have been used. Three features are obvious: the relative

frequencies are only positive for wind speeds greater than or equal to zero (by definition wind

speed cannot be negative); there is a peak in the distribution at somewhat less than the mean

of the distribution; and there are occasional occurrences of very high winds.


12

Figure 4: A typical wind speed distribution

1.1.7 The Weibull Distribution

It has been found that the frequency distribution of wind speeds at most sites can be well

represented by the two parameter Weibull probability density function. The probability of

the wind speed having a value U is given by

])[-(U/C )(k/C)(U/C = p(U)k1-k

exp (8)

where k is known as the shape parameter and C the scale parameter.

Figure 5: Example Weibull distribution

Frequency distribution of Wind Speed

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.2

5

1.2

5

2.2

5

3.2

5

4.2

5

5.2

5

6.2

5

7.2

5

8.2

5

9.2

5

10

.25

11

.25

12

.25

13

.25

14

.25

15

.25

16

.25

17

.25

18

.25

19

.25

20

.25

21

.25

22

.25

23

.25

24

.25

25

.25

26

.25

27

.25

28

.25

29

.25

Wind Speed (m/s)

Rel

ati

ve

Fre

qu

ency

Weibull Distribution with C = 9.3m/s and k=1.8

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.1

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30

Wind Speed (m/s)

Pro

bab

ilit

y


13

XFigure 5X shows an example distribution. The cumulative probability distribution associated

with this is obtained by integration of the function between zero and some value, V. This

gives the probability Q, that the wind speed is less than V as:

Q(U < V) = 1 - [-V / C ) ]kexp (9)

Since this equation is easily evaluated it should be used to calculate the probability of the

wind speed lying in a particular range by applying it twice at the two values defining the

range of interest and differencing the results. Sometimes (1-Q) is plotted; this can be viewed

as the reverse cumulative distribution. An example is given in XFigure 6X.

If the shape parameter, k, takes the value of 2, the Weibull distribution reduces to the well

known, one parameter, Rayleigh distribution. There is some physical basis to this simpler

form in that it can be derived by assuming wind to be isotropic and uniformly distributed

with no prevailing direction and that wind speed variations in orthogonal directions are

independently normally distributed.

In general sites do exhibit a prevailing direction and the more flexible, but not physically

based, Weibull distribution is applicable.

If the parameters k and C are known for a given site the annual mean wind speed aU can be

calculated from:

aU = C. (1+1

k) (10)

where is the gamma function, defined to be

(y) = e x dxo

-x y-1 (11)

Reverse Cumulative Weibull Distribution (C=9.26 and k=1.77)

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30

Wind Speed (m/s)

Pro

ba

bil

ity

Figure 6: Reverse cumulative distribution


14

1.1.8 Estimation of Weibull Parameters

From the cumulative distribution it is clear that the probability, Q (inverse cumulative) that

the wind speed exceeds V is given by:

Q(U > V) = [-(V / C ) ]kexp (12)

Care should be taken since this is sometimes also referred to as the cumulative Weibull

distribution. Taking logarithms twice on both sides gives

k1n(C)-k1n(V) = 1n(-1nQ) (13)

Hence if 1n(-1nQ) is plotted against 1n(V), where Q is calculated from the data for a wide

range of values of V, then the result should be approximately a straight line with k as the

gradient and an intercept at -k.1n(C).

Linear regression can be used to find the best straight line fit to the data. XFigure 7X shows the

result of such an analysis showing that the Weibull distribution is a reasonable fit to the data.

ln(-ln(Q)

ln(V)

Intercept = -k ln(C)Slope = k

Figure 7: Linear regression for Weibull parameter fitting

Although widely used, linear regression applied to logarithms and double logarithms of the

dependent and independent variables will not provide the best choice of parameters k and C.

A more rigorous approach to parameter estimation uses maximum likelihood techniques.

These are recommended if it is important to have great accuracy. In many situations however

rough estimates of k and C will suffice. In these circumstances the following approximate

relations are very useful.

From work by Bowden and others we have the results:


15

3.0)k(1.6for /U2C 1/2

(14)

and

2.3)k(1.8 for )U/(2C3k 1/23 3

(15)

Provided the ranges indicated are adhered to the accuracy of these formula is within 1%.

Another approximate relation, which does not depend on knowing the mean of U3, is:

Uk

-1.086

(16)

But note that the standard deviation here is the long-term (say annual) value, and not that

associated with turbulence intensity.

In special cases the wind speed at one point will be approximately proportional to the wind

speed at another point. For example if the points are two heights at a given location for

which the simple log law (neutral stability) applies, then if U1 = aU2, then k1 = k2 and C1 =

aC2.

1.1.9 Calculation of Energy Yield

It is straightforward to calculate the energy yield from a wind turbine to be placed on a given

site, using the Weibull parameters and the wind turbine power curve.

The average power from the wind turbine, assuming 100% reliability, is given by

P = p(U) P(U) dUu=0

(17)

where P(U) is the power output from the wind turbine at wind speed, U and p(u) is the

probability given by distribution function. This integral will in general have to be evaluated

numerically.

If the record of hourly mean wind speeds from the site is available this can be used directly

with the power curve to calculate the energy yield which would have occurred had the turbine

experienced that specific wind history.


16

1.2 Wind Resource

1.2.1 Introduction

Prediction of the wind resource at a given site is a crucial part of the development of

commercial wind energy generation. Without an accurate assessment of the available wind

resource no meaningful evaluation of a particular wind generation proposal can be attempted.

There are increasing demands for accuracy as project developers' plans will be scrutinised by

the banks and other organisations involved in project finance.

At different stages of project development different levels of assessment are appropriate. At

the early stages a superficial and low cost appraisal of a site is often undertaken to see if

sufficient potential exists to justify a time consuming and potentially costly measurement

campaign.

Smaller installations, such as those contemplated by individuals, cannot afford expensive site

assessment exercises. Greater reliance is therefore placed on predictions based on computer

modelling or other information such as local knowledge.

In all cases, what is being attempted is an estimate of the long-term wind resource; ideally the

wind resource over the expected lifetime of the wind turbine/s which could be up to 20 years.

Usually it is the long-term past wind resource that is being estimated at the site; the

assumption being that the long-term wind resource is not changing.

1.2.2 Measuring Wind Speed at a Site

1.2.2.1 Mast

If a site is being assessed for its long term wind power potential, it is common for a developer

to erect a mast on the site instrumented to measure wind speed and direction. Normally

anemometers are placed on at least two heights up the mast. This is to allow for an instrument

breaking down and also to give some information about the wind speed profile or wind shear

(see Section X1.1.3X). This allows a more accurate extrapolation of wind speed with height. It is

also preferable to ensure that at least one of the anemometers is placed at the hub height of

the wind turbines to be erected at the site. The axis of the wind turbine rotor is at the hub

height. You will learn more about wind turbines in Section X1.3X. Wind direction is normally

measured using a wind vane at at least one height on the mast. A guyed lattice or tubular mast

is usually used with the measuring instruments mounted sufficiently far from the mast itself

(on booms or otherwise) so as not to be influenced by the mast aerodynamically. A data

logger is then used to record measurements at regular intervals, typically 10 minute averages.

1.2.2.2 Instrumentation

For wind site assessment purposes the key parameters to be measured are wind speed and

direction. In some instances ambient air temperature and atmospheric pressure are also

measured.

The standard transducer for measuring wind speed in wind site assessments is the cup


17

anemometer. There are more sophisticated transducers for specialist wind speed

measurement, such as hot wire and doppler anemometers, but their particular characteristics

are not required for site assessment. Cup anemometers have the advantage of being robust

and relatively cheap. Rotation of the cup is sensed and the instrument can be configured to

produce either an analogue output (proportional to wind speed) or a pulse train with each

pulse representing a fixed amount of rotation, equivalent to a fixed run of wind. The rate of

pulsing gives the wind speed. Counting pulses is an attractive approach because the ideal

integrated values of the wind run are directly measured.

The time response of cup anemometers is specified in terms of a distance constant, d. This is

the length of wind run for the output, in response to a step change in wind, to reach within 1/e

of its final value. For a given change in wind speed, δv this gives a time constant equal to

d/δv seconds. Unfortunately, cup anemometers respond more rapidly to increases in wind

than decreases, as shown in XFigure 8X. This results in the so-called over-speeding effect which

means an overall overestimate of the wind speed. In order to limit this effect a sufficiently

fast response anemometer should be used. Generally a 5m distance constant is considered

acceptable. A further source of measurement error arises because the cup also responds to

some extent to vertical wind components.

Figure 8: Anemometer dynamic response

Because cup anemometers do not respond to changes in the direction of the horizontal

component, a separate instrument is required to measure wind direction. This is normally

done using a wind vane. A wind vane can either be a potentiometer device where a change in

continuous voltage is measured as the vane turns and its resistance changes or by a series of

reed switches to give discrete voltage measurements.

Ambient temperature has traditionally been measured using a platinum resistance

thermometer or a thermocouple, although simple semiconductor devices can now be used

which are sufficiently accurate. Suitable signal conditioning circuitry is supplied with

proprietary instruments. Compact commercial pressure transducers are available for the

measurement of atmospheric pressure. They come supplied with appropriate signal

conditioning. Nowadays a range of low cost commercial data loggers suitable for data

collection are available. These are battery powered and store data in ram or on storage cards

ready for downloading to a PC at regular intervals. Some loggers are available with modems

for data transmission.


18

There are many ways of presenting wind data. The wind rose indicates at a glance the

strength and directional dependence of long term wind. An example is shown in XFigure 9X.

This figure shows that the wind at this site blows most frequently from the 240-degree

direction of the compass. It is normal conventional to talk about the direction from which the

wind blows, so that a south-westerly wind is blowing FROM the south-west TOWARDS the

north-east.

Figure 9: A typical wind rose for the UK

1.2.3 Estimation of Long Term Site Wind Speed

Data collected at a site can only tells us about the wind over the period of data collection. In

order to make estimates of long-term resource some approach to extrapolation is required.

The standard approach used by the wind energy developers is known as Measure-Correlate-

Predict (MCP). Data collected at the candidate site is correlated with data over the same

time period available from a nearby site for which long term records exist. Usually this is an

official meteorological site. There are a number of varying approaches which fall under the

category of MCP, many of them involving linear regression. Once the relationships

(correlations) between the two sites have been calculated, these are used to appropriately

scale the long-term data from the met site. If possible 10 or more years of data is used. The

accuracy of the MCP approach is dependent on the quality of the data at the candidate site

and the met site, and the duration of the measurements at the candidate site. XFigure 10X shows

a schematic of the procedure.

0%

10%

20%

30%

40%

0

30

60

90

120

150

180

210

240

270

300

330

> 20 m/s

15 - 20 m/s

10 - 15m/s

5 - 10 m/s

0 - 5 m/s

Strong prevailing South West wind rose


19

Figure 10: The Measure-Correlate-Predict procedure.

It is generally agreed, due to seasonal variations in the correlations, that at least 8 months of

data collection is required for accurate predictions (see Reference 2). Often developers will

collect data over one year.

1.2.4 Computational Models

A number of computational models are available to the wind energy analyst to help in site

evaluation. These are now available as commercial software packages. These methods

perform some transformation of data available from reasonably nearby met sites so as to map

it onto the site of interest so as to account for variations in the topography and major surface

features of the two sites. A common approach is to use an MCP analysis to produce the long-

term prediction at the measuring mast on a site and then to use a computational model to

produce an area map of the wind speeds around the mast in order to find the best places to put

the wind turbines in a wind farm.

WAsP (Wind Atlas Analysis and Application Program) which was developed as part of the

European Wind Atlas, is perhaps the best known of the computational models. MS-Micro

developed by the Canadian Atmospheric Environment Office has a similar basis. Both

models solve the equations of conservation of mass and momentum with several simplifying

assumptions for the flow of air over terrain. As mentioned above, a significant amount of

energy can be stored in the turbulent part of the wind. The generation and dissipation of

turbulent kinetic energy is not trivial to model. The two models mentioned above use a

simple parameterisation of turbulence known as a first order turbulence closure model. Mass

consistent models such as NOABL have been used in the past, but as these contain no

turbulence model and only solve the equation for conservation of mast they are not as

accurate. NOABL has been used to produce a wind map of the UK on a 1km grid. In WAsP,

a flow model is used to calculate the effect of topography and ground level roughness,


20

including roughness transitions. It is applied to both the candidate site and the reference site.

For a range of directional sectors the solutions are used to extrapolate the reference data

upwards through the boundary layer to give the local geostrophic wind regime. This is then

transported to the candidate site and then transformed down to the candidate site using the

characteristics calculated from the flow solution. Account is also taken of local obstacles.

More complicated models for solving the flow of wind over terrain have been developed in

recent years such as CFX and FLUENT. Many of these models solve the full set of Navier-

Stokes equations for fluid flow and can provide fairly accurate results for specific sites.

However, to produce these results required specialist knowledge to set up the boundary

conditions. In addition, these types of models need a significant amount of input data and

computing time to produce the wind flow field. Models such as WAsP and MS-Micro,

though less accurate, nonetheless produce reasonable results in gentle terrain and are user-

friendly. These tend to be favoured by wind farm developers at the present time.

During the project based around the use of WindFarm, you will have the opportunity to see

MS-Micro used in conjunction with an MCP analysis of a potential wind farm site.

1.2.5 Wind Flow over Hills

1.2.5.1 Introduction

The computational models above can be used to calculate the effect of wind flowing over

non-flat topography. In theory such models can calculate the wind speed at a given point in

quite complex topography but need a lot of input data and a certain amount of computing

time.

Generally, wind farms are placed on the top of hills to take advantage of the speed-up of the

wind that occurs. This is basically due to conservation of mass flow. If you assume the upper

level wind is unperturbed by the flow, then the mass flow is restricted by a hill. Therefore the

„parcel‟ of wind must flow faster over the hill to conserve mass flow (see XFigure 11X). There

are rough „rules of the thumb‟ that can be used to calculate the wind speed on top of

geographical features (Reference 3).


21

Figure 11: CFX calculation of wind speed-up over a hill and wind slow-down in a valley.

1.2.5.2 Simple Guidelines for Wind Speed Changes Over Hills

Consider a hill as shown in XFigure 12X with height h and half-width measured at half-height of

L. An anemometer is placed on a mast at height Δzp above the ground a certain distance away

from a hill on flat ground. The anemometer measures a wind speed U0. An estimate is

required of the wind speed UT at the same height above ground at the top of the hill.

Figure 12: Hill-top for which an estimate of wind speed is required.

Ut can be considered as the sum of the undisturbed wind speed U0 plus a factor due to the

effect of the hill:

TppT UzUzU 0 (18)

Speed-up of wind on

top of hill Slow-down of wind

in valley


22

The factor due to the hill is given by:

pT zSUU 0 (19)

which is expressed as a speed-up factor ΔS multiplied by the undisturbed wind. The speedup

factor is then expressed as:

LzASS p /expmax (20)

and ΔSmax is given ny:

LBhS /max (21)

where A and B are empirical constants which depend on the type of topographic feature.

Approximate values are proposed in XTable 2X.

Terrain Type A B

2D hills (ridges) 3.0 2.0

3D hills 4.0 1.6

2D escarpments 2.5 0.8

2D rolling terrain 3.5 1.55

3D rolling terrain 4.4 1.1

Table 2: Approximate values for A and B constants used in guideline equations for wind

speed changes over topographic features.

This is only a simple guideline and assumes:

Gentle slopes (<0.35)

Moderate/strong winds (>6m/s)

Horizontal length scales ~1km

Uniform roughness

Neutral stability

1.2.6 Initial Site Assessment

Before a wind farm developer goes to the expense of erecting a mast to measure wind speed

or running a numerical model to get a detailed wind map for a specific area, s/he can turn to

regional wind speed maps to get an estimate of the expected wind speed over a wide area.

The developer can then concentrate efforts on the area with the largest potential wind power

resource. Large scale maps of wind speed have been produced for areas around the world. In

the UK, the NOABL model has been used to produce a wind map on a 1km grid using data

from a number of Met. Office stations (see XFigure 13X). The WAsP model has been used to

transform site specific wind speed measurements at meteorological stations to regional wind

speed maps known as the European Wind Atlas (see XFigure 14X).


23

Figure 13: A wind speed map of the UK at 10m agl on 1 1km grid.


24

Figure 14: The European Wind Atlas.


25

1.3 Wind Turbine Aerodynamics

1.3.1 Power Available in the Wind

The power PW in the wind flowing at right angles through an area A can be shown (see Box

1) to be proportional to the cube of the wind speed and is given by:

3

02

1UAPw (22)

where is the density of air, and U0 is the undisturbed wind speed.

This equation also shows that power is proportional to the density of the air, varies slightly

with height and temperature (for 15 C at sea level = 1.225kgm-3

), and is proportional to the

swept area A.

Box 1. The derivation of Power Contained in the Wind.

A parcel of air of mass m, moving with velocity U0 has kinetic energy given by ½ mU02. If

the density of the flowing air is , then the kinetic energy per unit volume of air is given by

½ U02 .

If we consider an area A perpendicular to the wind direction, as shown below, the

volumetric flow rate through A is U0 A.

The power contained in the wind is the kinetic energy of the air that flows per second

through A, ie volumetric flow rate x energy per unit volume:

3

02

1UAPw

1.3.2 Power Fluctuations

The cubic relationship between wind speed and power has important consequences for the

variability in the power output from a wind turbine. XFigure 15X shows the variation of wind

speed and corresponding power output from a 30kW wind turbine over a 5 minute period.

We can see that the variations in the wind speed due to turbulence result in extremely large

power fluctuations.

AU0


26

Figure 15: Wind speed over a 5-minute period (measured at the Rutherford Appleton

Laboratory)

1.3.3 Extracting Energy From the Wind

A wind turbine is a device for extracting kinetic energy from the wind. To a first

approximation, only that mass of air which passes through the rotor disc is affected. If we

assume that the affected mass of air remains separate from the air which does not pass

through the rotor disc, a boundary can be drawn between the two. This can be extended

upstream as well as downstream, XFigure 16X, forming a long stream-tube of circular cross-

section.

The stream-tube is an important idea and is used as a model in deriving many important

equations regarding the extraction of energy from the air. No air flows across this boundary

and so the mass flow rate (i.e. the mass of air flowing through an area per second) will be the

same all along the stream-tube. Therefore, as the air slows down the cross-sectional area of

the stream-tube must expand to accommodate the slower moving air.


27

Figure 16: Energy extracting stream-tube of a wind turbine

The turbine first causes the approaching air gradually to slow down which results in a rise in

the static pressure. Across the turbine swept surface there is a drop in static pressure such

that, on leaving, the air is below the atmospheric pressure level. As the air proceeds

downstream the pressure climbs back to the atmospheric value causing a further slowing

down of the wind. This is illustrated in XFigure 17X.

Thus, between the far upstream and far wake conditions, no change in static pressure exists

but there is a reduction in kinetic energy. The changes in velocity and pressure can be

explained by considering the energy within the system via the Bernoulli equation.


28

Distance

Press

ure

Turbine

Atmospheric

Pressure

Distance

Win

d S

peed

Figure 17: The pressure distribution and wind speed along the streamtube.

1.3.3.1 The Betz Limit

Wind turbines are capable of converting only a fraction of the kinetic energy contained in the

wind into mechanical energy. With a wind turbine there is a wind speed change between the

entry and exit of the stream tube of figure 2 for which the conversion efficiency is a

maximum. The calculation of this maximum point uses the idea of an actuator disc which is

a disc partially transparent to wind flow.

If there were no change in wind speed, the actuator disc is completely transparent and no

energy would be extracted and the power from the wind turbine would be zero.

If all the kinetic energy of the wind were abstracted by an opaque actuator disc, the wind

velocity downstream would be zero. This is clearly an impossibility since a constant mass

flow must be maintained within the stream tube. Thus, a wind turbine cannot completely

obstruct the flow of air, and so it can only extract a proportion of the kinetic energy from the

wind.


29

We define the Power Coefficient, CP, for a wind turbine as the power, P, extracted by the

wind turbine divided by the power in the wind, PW contained in an area swept by the rotor if

the turbine had not been there i.e.

w

pP

PC (23)

Substituting PW from Eq. X22X we obtain

3

021 UA

PCp (24)

The theoretical maximum fraction of wind energy that can be converted into useful energy by

a wind turbine is known as the Betz Limit. This turns out to be 59.3%. The Betz Limit, was

first formulated in 1919 by Albert Betz a German aerodynamicist. It applies to all types of

wind turbines and represents the maximum theoretical limit of power that can be extracted

from the wind. See Box 2 for its derivation.

Modern designs of wind turbines for electricity generation operate at CP values of about 0.4

although values approaching 0.5 have been reported. No wind turbine has been designed

which is capable of exceeding the Betz limit. Other losses in efficiency for a real turbine

arise from the aerodynamic drag on the blades, the swirl imparted to the airflow by the rotor,

and power losses in the transmission and electrical systems.

The Betz limit can be derived by considering changes in energy along the stream tube and

therefore uses the equation for energy conservation in fluids known as the Bernoulli equation.


30

Box 2. The Derivation of the Betz Limit using Momentum theory

The analysis of the energy extraction process begins by using simple momentum theory. In

this analysis, the wind turbine aerodynamics are modelled by an actuator disc. This is a disc

which extracts energy from the wind flux by providing a resistance but not fully obstructing

the flow of air. Hence a pressure drop is developed across it.

As air velocities are relatively low, changes in air density are negligible and so the air flow

can be regarded as incompressible. Since kinetic energy has been extracted from the wind by

the turbine, the velocity is reduced, and hence the stream-lines must diverge as they pass

through the rotor disc i.e. the flow through a stream-tube containing the disc has a cross-

sectional area smaller than that of the disc upstream and an area larger than the disc

downstream. This is because the mass flow rate must be constant within the stream tube far

upstream (0), at the rotor (1) and far downstream (2), hence

221100 UAUAUARateFlowMass (25)

This is known as the continuity equation and is one of the fundamental equations in

aerodynamics.

We now introduce a quantity called the axial flow induction factor (or inflow factor), a, which

is defined as the fractional decrease in wind velocity from far upstream to the rotor plane i.e.

the velocity at the rotor is reduced by a factor aU0. Therefore a is given by

01 )1( UaU or 0

10

U

UUa (26)

The air which passes through the disc undergoes an overall change in velocity, )( 20 UU and

the rate of change of momentum is equal to the overall change of velocity multiplied by the

mass flow rate i.e. The force along the streamline on the rotor disc is given by the rate of

change of momentum i.e.

01201120 )1()()( UaAUUUAUUmomentumofchangeofRate (27)

The force causing this change of momentum comes entirely from the pressure difference

across the actuator disc because the stream tube is otherwise entirely surrounded by air at

atmospheric pressure.


31

Therefore:

0120111 )1()()( UaAUUAPP (28)

Where 1P and 1P are the pressures on the two sides, at the disk.

We now introduce a second fundamental equation of aerodynamics, the Bernoulli Equation,

which is a restatement of the law of conservation of energy. It states that, under steady

conditions, the total energy in the flow of a fluid, comprising of kinetic energy (KE), static

pressure energy (PE) and gravitational potential energy(GE), remains constant provided no

work is done on or by the fluid. Thus for a unit volume of air:

KE + PE + GE = constant (29)

The Bernoulli Equation is applied separately to the upstream and downstream sections of the

streamtube. Separate equations are necessary because the total energy is different up-stream

and down stream. Applying the equation upstream we obtain

111

2

112

1000

2

002

1 zgPUzgPU (30)

where P0 is the upstream pressure (atmospheric pressure).

Assuming the flow to be incompressible ( )( 10 and horizontal )( 10 zz we obtain

1

2

121

0

2

021 PUPU (31)

Similarly downstream

1

2

121

2

2

221 PUPU (32)

where P2 is the downstream pressure (atmospheric pressure).

Subtracting Eqs. ( X31X) and (X32X) noting that P0=P2, we obtain

)( 2

2

2

02

111 UUPP (33)

Combining equation (X33X) with equation (X28X) we obtain

02 )21( UaU (34)

This shows that half the axial speed loss in the streamtube takes place upstream of the actuator

disc and half downstream. The force on the air becomes

)1(2)21()( 2

01

2

0

2

012

1111 aaUAUaUAAPPF (35)

As this force is concentrated at the actuator disc, the rate of work done by this force (power) is

given by force multiplied by the velocity at the disc, hence

23

011 )1(2 aaUAUFP (36)


32

The Power Coefficient is:

3

0121 UA

PC p (37)

where the denominator represents the power available in the wind, in the absence of the

actuator disc, hence

2)1(4 aaC p (38)

CP can be maximised by differentiating CP with respect to a and setting equal to zero:

0)31()1(4 aada

dC p (39)

The meaningful solution of equation ( X39X) gives a = 31 . Substituting this value of a into

equation (X38X) gives

593.027

16pC (40)

In practice good performance can be obtained for values of a between 0.2 and 0.5.

1.3.4 Aerodynamic Lift and the Aerofoil

When air flows over an aerofoil two forces are generated; lift and drag. Drag is measured

parallel to the airflow. The lift force is measured perpendicular to the air flow and it is the lift

force that gives rise to shaft torque in a wind turbine to generate power. The lift coefficient is

defined as:

cU

SpanUnitperForceLiftCL 2

2

1 (41)

The drag coefficient is defined as:

cUCD 2

21

SpanUnitperForce Drag

(42)

where c is the chord length (distance from the blade‟s leading edge to its trailing edge – see

XFigure 20X).

An important quantity defining how well a particular shape can extract energy from the wind

is the ratio of the lift force to the drag force. For good extraction efficiency we want to

maximise the lift force and minimise the drag force i.e. we require a high lift to drag ratio.

An aerofoil is a specially designed shape to generate lift forces when moving relative to a


33

fluid. An aerofoil cross-section is shown in XFigure 18X.

An important consequence of the Bernoulli Principle is that areas where the wind speed is

high are at low pressure, whereas areas where the wind speed is low are at high pressure.

The air hitting the aerofoil flows above and below the aerofoil and meets again at the trailing

edge (the sharp point at the back of the aerofoil). The air flowing over the top of the aerofoil

is accelerated and this increase in speed causes a decrease in pressure. It is this pressure

difference across the aerofoil which produces the lift force perpendicular to the aerofoil and is

related by the equation: Lift Force = Pressure difference x Surface Area.

Figure 18: The Lift and Drag forces on a stationary aerofoil

By convention the lift and drag forces are defined as being perpendicular and parallel to the

wind direction (not the aerofoil).

The air moving over the aerofoil also produces a drag force in the direction of the air flow.

This is an undesirable effect and is minimised as much as possible in high performance wind

turbines. Both the lift and drag forces are proportional to the air density, the surface area of

the aerofoil and the square of the incident wind speed.

Suppose now we allow the aerofoil to move. This motion will combine with the motion of

the air to produce a relative wind direction (i.e. relative to the aerofoil) shown in XFigure 19X.

The lift and drag are perpendicular to and along the relative wind direction.

The useful work that can be extracted from the wind and turned into rotational motion of the

wind turbine rotor is due to the force in the direction of the aerofoil motion, which must be

maximised. The force perpendicular (i.e. in the direction of the undisturbed wind) is a

loading force on the wind turbine and therefore makes no contribution to the power of the

turbine and needs to be minimised.

We calculate these forces by resolving the lift and drag components into the tangential force

F1 and the force F2 in the direction of the undisturbed wind. These forces and the overall

performance of the wind turbine depend on the construction and orientation of the blades.


34

Figure 19: Lift and drag forces on a moving aerofoil

We therefore need to look in greater detail at the blade aerofoil geometry and define two

angles:

a) The pitch angle is the angle between the chord of the blade and the plane of rotation as

shown in figure 6. The pitch angle is a static angle, depending only on the orientation of

the blade.

b) The angle of attack is the angle between the chord line of the blade and the relative wind

direction. It is a dynamic angle (i.e. prone to change), because it depends on both the wind

speed and the speed of the rotating blade. Any given point on a rotating blade, with

angular velocity , will have a speed of r, where r is the distance of that point from the

hub. It is clear therefore that the angle of attack for a straight blade will vary along its

length.

Figure 20: Definition of pitch angle , and angle of attack .

1.3.5 The Tangential Induction Factor


35

We have already seen that the wind turbine blades generate lift and drag forces. We now

want to know the relationship between these forces and the corresponding thrust (i.e. the axial

force on the wind turbine) and the torque (i.e. the useful force in the tangential direction

turning the rotor). Although the axial force has only nuisance value, it is important because it

defines the strength of the blades and the support structures

Upstream of the turbine there is no tangential velocity component of the air, since the air flow

is purely normal to the wind turbine. However, the forces generated cause the rotor to rotate

and hence, due to the conservation of angular momentum, an equal and opposite tangential

momentum in the air flow is induced. This appears as a helical wake shown in XFigure 21X.

Figure 21: Helical wake induced in the air flow.

The change in tangential velocity of the air is expressed in terms of a tangential flow

induction factor a . This has a relatively minor effect on the angle of attach and is ignored in

a simplified analysis.

1.3.6 Relationship between Thrust and Torque, and the Lift and Drag

forces

With information about how the aerofoil characteristic coefficients CL and CD vary with the

angle of incidence, it is now possible to determine the forces on the blades for a given values

of the axial a. The axial force on the blades is equal to the rate of change of axial momentum

of the wind

Consider a turbine with N blades of radius R each with chord c and set pitch angle

measured between the aerofoil zero lift line and the plane of the disc. Both the chord length

and the pitch angle may vary along the blade span.

Let the blades be rotating at angular velocity Ω and let the wind speed be U0. XFigure 22X

shows the velocities relative to the blade chord line at radius r.


36

Figure 22: Blade element velocities and forces.

The resultant relative velocity W at the blade (ignoring the term a‟) is:

222

0 )()1( raUW (43)

which acts at an angle to the plane of rotation, such that

)1()sin( 0 aW

U (44) and W

r)cos( (45)

The angle of incidence is then given by = - . The lift force per unit length on each

blade, L, normal to the direction of W is therefore LCcWL 2

21 and the drag force per unit

length on each blade, D, parallel to W is therefore DCcWD 2

21 , where c is the chord

length.

Resolving along and at right angles to the plane of rotation we obtain:

Tangential force F1 = L sin - D cos = L sin( + ) - D cos( + ) (46)

Loading force (thrust) F2 = L cos + D sin = L cos( + ) + D sin( + )

(47)

Theses are the two fundamental equations from which the loadings on the wind turbine and

the torque given to the turbine rotor can be calculated for a given wind speed, rotor speed and

pitch angle. F1 represents the force turning the rotor (this force acting at a given blade radius

produces a moment known as the torque) and F2 is the axial thrust. The torque produces

useful work, whereas the axial thrust will try and overturn the turbine and must be resisted by

0


37

the blades, tower and foundations.

1.3.7 Relationship between the Angle of Attack and the Lift and Drag

Coefficients

From XFigure 23X it can be seen that, for angles of attack less than about 13 , the coefficient of

lift is proportional to the angle of attack, . It can be shown theoretically that the relationship

between CL and is CL = 2 .

In practice CL = 0a , where 0a , called the lift-curve sloped

dCL , is about 0.1/degree, giving

a value slightly less than 2 .

The lift force depends on two parameters, the angle of attack and the relative velocity

between the wind and the blades

The variation of CL and CD with the angle of attack is shown in figure 9 for a typical aerofoil.

Such aerofoil characteristics are obtained from wind tunnel tests.

Figure 23: Experimental data of the variation of the lift and drag coefficients with

angle of attack for a particular aerofoil.

A high lift to drag ratio CL/CD is essential for a high efficiency rotor. The peak lift to drag

ratio indicates the optimum value of for maximum efficiency of a turbine blade. The peak

value generally occurs at values of around 4-6 .

The lift and drag will have optimum values for a single angle of attack only, for a given value

of W. It is for this reason that wind turbine blades are twisted so as to maintain a nearly

constant angle of attack from hub to tip. All modern blades of large wind turbines are twisted

although some smaller blades are not because straight blades are easier and cheaper to

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

-4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34

Angle of Attack (Degrees)

Coeff

icie

nt

CL

CD


38

produce and the cost reduction may more than offset the loss in performance.

It can be seen that for small values of , CL increases linearly with and CD is very small.

However, for this aerofoil, at an angle of attack of about 13 there is a sudden drop in the

value of CL and a sharp rise in the value of CD. This is due to sudden flow separation from

the suction side of the aerofoil; this is called stall.

If the angle of attack exceeds a certain critical value (10° to 16°, depending on the Reynolds

Number), separation of the boundary layer on the upper surface takes place. This causes a

wake to form above the aerofoil, reduces the lift and increases the drag. The flow past the

aerofoil has then stalled. A diagram of a stalled aerofoil is shown in XFigure 24X. A flat plate

will also develop lift but will stall at a very low angle of attack because of the sharp leading

edge. Arching the plate will improve the stalling behaviour but a much greater improvement

can be obtained by giving thickness to the aerofoil together with a well rounded leading edge.

Figure 24: Diagram showing stalled flow around an aerofoil

1.3.8 Performance of a Wind Turbine

The actual performance of a wind turbine is defined by its power characteristics which are

usually presented as the normalised power coefficient, CP plotted against tip speed ratio.

This has the advantage of reducing a family of power-rotational speed curves to a single

curve. The power coefficient will depend on the relationship between wind speed and speed

of rotation of the turbine rotor. The speed of rotation for a WT ranges from a few hundred

rpm for small sub-kW machines to a few tens of rpm for very large machines.

The tip speed ratio is defined as:

0U

R

SpeedWind

tipbladeofSpeed (48)

where Ω is the rotational speed in rads-1

, R is the rotor radius and 0U is the undisturbed wind

speed.


39

The usual method of presenting the performance of a wind turbine is through the CP - curve

as this gives a dimension-less graph i.e. the performance of a design will be approximately

independent of the size of machine.

The performance of a typical modern three bladed turbine is given in XFigure 25X, showing the

type of losses that arise and how they contribute to the departure from the ideal. We note that

the maximum value of CP is only 0.47, significantly less than the Betz limit of 0.59. The

difference is caused by drag and tip losses at higher values of and stall losses at lower

values . However, even if none of the above losses are included the analysis, the Betz limit

is not reached because the rotor design will not be act as a perfect actuator disc.

Figure 25: A Cp - curve for a modern wind turbine showing losses.

1.3.9 Wind turbine types

The simplest method of control is stall regulation. The rotor is kept at a nominally fixed

speed (through the connection to the grid) and as the wind speed increases the value of falls

towards zero, and so therefore does the efficiency of the rotor. In this way the output of the

turbine is limited. The pitch angle of the blades does not change and so no moving parts are

required. Stall regulation is also able to response to sudden changes in wind speed such as

gusts. It has the disadvantage that it leads to a high starting torque and the characteristics of

stall are not well understood so that at rated power there can be some degree of variation in

the power output.

Another common method of control for fixed speed turbines is to pitch the blades when the

maximum output of the turbine has been reached, so as to produce a less efficient

characteristic.

It should be clear from the Cp- curve that for a fixed speed rotor, wind speed variations will

0

0.1

0.2

0.3

0.4

0.5

0.6

0 2 4 6 8 10 12 14 16

Tip Speed Ratio

Pow

er C

oef

fici

ent

No Losses Included

Tip Losses Only

All Losses Included

Stall Tip Losses Drag Losses

Losses


40

give rise to variations in aerodynamic efficiency. In order to maximise the output from the

turbine when it is operating below rated power, variable speed operation can be considered,

aimed at keeping the operating point as near as possible to Cpmax.

In practice, the rotor speed will not be able to follow all the turbulent wind speed variations.

The tracking performance will depend on the nature and the rating of the electrical/power

electronic connection.

Variable speed operation can also have the advantage of reducing transient drive train loads

caused by turbulence. Instead of torque variations due to wind variation being felt directly by

the gear box, they are smoothed to some extent through the acceleration and deceleration of

the rotor inertia.

Above rated power, the most common control approach is to use blade pitch control to limit

the rotor speed. This has the advantage that a fairly slow blade pitch control rate is all that is

required.

Variable speed operation of stall regulated wind turbines is, in principal, possible. It is

however far from straightforward and so far is not available commercially.

1.3.10 Electrical generator options

For fixed speed operation the two obvious candidate electrical machines are the

synchronous generator and the asynchronous or induction generator. The advantages and

disadvantages of the two in the context of wind generation, are detailed below.

synchronous generator

advantages : reactive power can be controlled, high efficiency

disadvantages: synchronising equipment required, intolerant of torque fluctuations,

transient stability can be a problem

asynchronous (induction) generator

advantages : synchronising not required, small variations in rotor speed are

allowed, rugged cage rotor design

disadvantages: reactive power must be supplied by network

Overall the ability of the induction generator to allow, through variation of slip, small

changes of rotor speed which dramatically alleviate transient drive train loads means that this

is the favoured choice. Indeed, synchronous machines are never used for fixed speed

applications. Pole switching is sometimes employed to give two speed operation.

For variable speed operation the choice of electrical machines is basically the same,

although both the cage and the wound rotor variants of the asynchronous generator are

considered, and permanent magnet synchronous generators can be used in addition to the

more conventional brushless machines.

Many different arrangements are possible. Wound rotor induction generators have been used

with slip energy recovery (either Kramer or Scherbius arrangements). The advantage of these


41

approaches is that the rating of the power electronic path can be a small fraction of the overall

machine rating. This has not prevented straight application of power electronic variable

speed drives to cage machines and synchronous machines. Some large multi-pole

synchronous generators have been specially designed for wind turbines. When combined

with a variable speed drive, it has been possible to remove the gear box from the wind turbine

altogether. DC machines are not used in wind turbines due to their cost and maintenance

requirements.

Whether fixed or variable speed operation is preferable depends on a range of issues, but

noise generation and power quality (related to the strength of the grid connection) are two of

the most important.

1.4 Notation and Units

Symbol Quantity Units

U0

PW P A m CP a a´ Q

W p CL CD

F1 F2 µ

Re c

Undisturbed Wind Speed Density of air = 1.225 Kgm-3 Power contained in the wind Power Area Mass Power Coefficient Axial Flow Induction Factor Tangential Flow Induction Factor Torque Rotational Speed of Turbine Wind velocity relative to plane of rotation Pressure Lift Coefficient Drag Coefficient Angle of Attack Pitch Angle

Angle of relative wind speed ( + ) Rotational turning force on a turbine Axial Thrust on a turbine Viscosity Reynolds number Aerofoil Chord Length

ms-1

Kgm-3

Watts Watts m2 Kg None None None Nm Rads-1 ms-1

Nm-2 None None Degrees Degrees Degrees N N kgm-1s-1

None m

1.5 References

1. Wind Structure and statistics, by U. Hassan and D M Sykes, Chapter 2 in Wind

Energy Conversion Systems, ed L. L. Freris.

2. William Lloyd; Wind Resource Assessment Using Measure-Correlate-Predict

Techniques; CREST MSc thesis, 1995. (available in CREST library).

3. J L Walmsley, P A Taylor, J R Salmon, Simple Guidelines for Estimating Wind

Speed Variations due to Small-scale Topographic Features – An Update,

Climatological Bulletin 23(1), 1989, pp 3-14.


42

2 Hydro Power

2.1 Energy from Water

Approximately 70% of the Earth‟s surface is covered with water. Much of this is seawater,

which is part of the water cycle. Weather patterns driven by solar energy take up water from

lakes, rivers and the sea and deliver it as precipitation (rain, snow, hail etc.). A large

proportion of this precipitation falls on moderate to high ground and either enters the

watercourse or becomes runoff, which returns to the sea as rivers. Figure 26 shows this water

cycle.

2.2 Developing a Hydropower Scheme

2.2.1 Resource and Equipment Requirement

Hydropower is the extraction of energy from water by the use of turbines to capture the

downward flow of water from a high level to a low level and convert it to motive power.

Hydro electricity schemes which use the motive power to drive a generator are generally

categorised as either large-scale (>5MWe) or small-scale (<5MWe). XTable 3X illustrates the

typical scales with the associated head of water. The total installed capacity of large-scale

schemes in the UK amounts to l,360 MWe (1.36GW) with the majority of viable schemes

having already been developed.

Figure 26: The Water Cycle


43

Table 3: Typical values of schemes

Output Head

Large Scale >5MW 50-1000m

Small scale <5MW 10-1000m

Micro 10-100kW 5-30m

Low head 10-100kW 1-5m

A hydro scheme will consist of:

A suitable rainfall catchment area

A river with a suitable drop in level (vertical distance gives the HEAD)

A water intake above a weir or behind a dam

A pipe, channel or barrage to convey water to the turbine

A turbine, generator and electrical connection

A tailrace to return the exhaust water to the main flow.

Hydroelectric schemes extract energy from water in one of two ways, depending on the

geographical location of the water source.

These are described below:

1. Run-of-river (Figure 27) where the flow of water in the river is used directly and since

there is no storage, energy can only be extracted as the flow permits. In periods of low

flow, the available output is limited.

Figure 27: Run of the River Scheme


44

2. Using a dam (Figure 28) which captures and stores most of the water until the energy

output is needed, or until water is drawn off if the reservoir also serves as a water supply

reservoir.

2.2.2 Choice of Turbine

The choice of turbine machine depends upon the head and flow available:

In hilly or mountainous areas where there is a high hydraulic head of water (perhaps

up to 1000m), high-speed impulse turbines are used.

In river valleys and low-land areas where there is less head of water (medium heads

are usually less than 100m, but sometimes heads can be as low as 2-3m), high speed

reaction turbines are more suitable.

High-head schemes will involve the installation of a pipe or penstock either from a reservoir

or at high level above natural waterfalls. This will divert part of the river flow through an

impulse turbine installed in a plant room at the lower level before being returned to the river.

These schemes can incur high civil engineering costs and may require extra measures in

relation to land restoration, tree screening and the preservation of visual amenity of the

natural waterfall.

Medium-to-low head schemes will generally be developed near to existing weirs on rivers

whereby a part of the water flow is diverted by a channel to a turbine situated on one side of

the riverbank before being returned to the river at the lower level. Barrage schemes, however,

will involve the installation of a turbine or pipe within the wall of the weir with the majority

of the water (except in extreme flood conditions) being directed through the turbines.

Figure 28: Large Scale Hydro Scheme


45

The capacity of the generation plant installed will be determined by the net head of water

available, the turbine discharge rate, friction losses in any pipework and the efficiency of the

turbine and generator. The energy extracted will in turn reflect the variation in flow of water

throughout the year in relation to the turbine discharge capacity.

2.2.3 Environmental Considerations

Hydropower schemes need to be developed sensitively to avoid any detrimental

environmental effects and would be subject to environmental controls, primarily imposed by

the Environment Agency, which has responsibility for the environmental protection of

watercourses in England and Wales, covering water resource management, fisheries,

conservation, flood defence, navigation and recreation. In certain circumstances, other

organisations, such as the Countryside Commission, also have powers of regulation. These

relate specifically to the conservation of species or habitats. Non-water related aspects, such

as buildings and transmission line construction would be regulated by the local and regional

planning authorities where appropriate. Further allowance may need to be made for

environmental mitigation measures, such as replanting any ground, hedgerows and trees

disturbed during construction and possible screening of any pipes or water channels installed

above ground level.

It may be necessary to construct approved fish passes/ladders to allow fish migration and to

fit fish screens and/or fish guidance systems to ensure that fish are not inadvertently drawn

into the water intake of the turbines installed.

It is very unlikely that a hydropower scheme will be allowed to use the entire river flow for

generation. A portion of the flow, termed the compensation flow will need to by-pass the

turbines to maintain the ecology and aesthetic appearance of the river, to keep the river bed

wetted over the majority of its width, and to maintain a visible flow over any waterfalls or

weirs. Hydropower schemes can have a beneficial effect in that they aerate the water passing

through them and increase the quantity of air in the water downstream. It may be necessary to

make some provision to reduce noise levels if water channels or turbines are installed in near

proximity to occupied premises.

There are potential drawbacks to some schemes, particularly when the impounding of water

behind a dam leads to flooding of a valley. This will affect farming, industry, forestry,

wildlife etc as well as displace population. The latter often leads to social and cultural decline

of the population involved. There are many examples of population displacement but the

largest impact is that of the Three Gorges scheme in China which will involve over one

million people. One of the case studies examines the Three Gorges scheme in detail.

2.2.4 Basic Calculation of Energy

Imagine that we wish to raise one litre of water through a vertical height of one metre. The

energy or work required to do this is given by the equation

(49)

m = mass of water

g = acceleration due to gravity = 9.81 ms-2

Energy = mgH Joule


46

H = vertical lift in metres

V= volume of water

= density of water in kgm-3

Since the mass of one litre of water is given by V

for fresh water is about 1000 kgm-3

(or as 1m3 = 1000 litres, = 1.00 kgl

-1 )

and for sea water is about 1025 kgm-3

(1.025 kgl-1

)

Then we can say that the energy to raise our litre of water is either mgH or V gH

That is 1/1000*1000*9.81*1 = 9.81 Joules

This simple calculation is fundamental to calculated energy from, water flow, be it related to

hydro or tidal power generation. It enables us to calculate how much energy is available in a

reservoir, hydro scheme or tidal barrage.

For example, if we know that a reservoir holds 25,000 cubic metres of water at a height of 17

metres above a river then the amount of energy needed to lift the water into the reservoir is

calculated using:

V gH = 25,000 * 1000 * 9.81 * 17 = 4169250000 Joules or 4.169 GJ

This energy needed to fill the reservoir with water may have come from rainfall in the

catchment area filling the reservoir or it may have come from pumping water during times of

surplus energy to use the reservoir as an energy storage system.

Whatever the means of the water entering the reservoir, the exploitation of the potential

energy stored in the reservoir is the same. The water is allowed to flow from the reservoir

through pipes to a turbine. The passage of water causes the turbine to spin, and this in turn

drives an electrical generator at several hundred rpm to generate electricity. The overall

process is reasonably efficient provided the following: the pipes are internally smooth,

without too many bends and have large enough diameters, the turbine is correctly chosen as

the right type and specification to suit the head and flow, and the generator is of high

efficiency. If all of these conditions are satisfied, then net efficiencies of 80-90% are

achievable.

2.2.5 Calculation of Power

We talked about the energy needed to raise one litre of water by one metre. If we release this

water at a flow rate of Q m3s

-1, then the water power, P, is given by

(50)

So, for example, if we release all of our water in one second, then the flow would be 1ls-1

or

1000

1 m3s

-1 and the power would then be

Power = Q gH


47

1000

1 * 1000* 9.81*1 = 9.81 Watts

Of course, this power would only last for one second before we ran out of water.

On a more realistic scale, if we release the water from our reservoir at five cubic metres per

second (a cubic metre per second is often called a cumec), then the water power would be 5*

1000* 9.81* 17 = 833850 Watts

2.2.6 A Typical Run of River Scheme

A typical scheme is illustrated in XFigure 29X. The water supply to the turbine is first collected

at a weir and carried along a channel with little loss of head into a forebay or holding tank.

The penstock then delivers the water under pressure to the turbine. After passing through the

turbine, the water returns to the river course via the tailrace. Inside the penstock, the water

loses potential energy (mgH) and gains kinetic energy ( 25.0 mv ). At the end of the penstock,

we can equate potential and kinetic energies if we assume that there has been no loss of

energy in the pipe.

mgHmv25.0 (51)

and so we can calculate the velocity of the water leaving the pipe at its lower end as

gHv 2 (52)

Figure 29: A typical run-of-river hydro scheme.

A plan view of a scheme is shown in XFigure 30X, and this raises some important practical

issues. Note the inclusion of spillways and silt basins. The variability of rivers in

mountainous regions can be very great and in periods of extreme flood, the flow can be ten or

one hundred times the average flow. These high flows will carry boulders in addition to the

sand/grit that is normally contained in a rapidly flowing river. Obviously, the grit can damage


48

turbine blades and so silt basins are incorporated in the scheme to avoid it reaching the

turbine. The larger boulders and other debris are carried downstream at the weir and do not

enter the channel.

A trash rack is generally employed to prevent floating items such as logs from entering the

penstock.

Figure 30: Civil Works of a Run of a River Scheme

2.3 The Measurement of Head

2.3.1 Why is the measurement of Head Important?

The gross head is the vertical distance that the water falls through in generating power, i.e.

between the upper and lower water surface levels. This is shown in Figure 31.

The power generated from a hydropower scheme is given by

Power = g Q H (53)

Where:

is the „water to wire‟ efficiency –typically (70-80% for a small scheme)

is the density of water (1000kg/m3)

g is acceleration due to gravity (9.81m/s2)

Q is the flow through the turbine (m3/s)

H is the Net Head (m) i.e. including head losses

Therefore, we can see that the power is proportional to the head and the flow. In general, the

physical size, and hence to a large extent cost, of the turbine is governed by its design flow,

rather than the hydraulic head. For example, a 100kW scheme at 300m head will require

about 40litres/s whilst at 3m head will require a flow of 4,200litres/s (assuming 80%

efficiency).


49

Therefore, it is clear that a high head

scheme is preferable to a low head scheme

and head is, generally speaking, more

important than flow (although clearly flow

is also important!).

The revenue generated by a scheme is

proportional to the energy capture which is

proportional to the head. Therefore, an

accurate measurement of the head is

essential to predict the annual energy

capture from a scheme. This measurement

is particularly critical for low head

schemes.

2.3.2 The Variation of Head with Flow

It is found with many low head sites that the weir can literately disappear under flood

conditions. This gives rise to the paradoxical situation of generating no power at all at very

high flows since the head becomes low or disappears. There is generally a decrease in head

at high flows since the tailrace tends to rise at a faster rate that the headrace. The variation of

head with flow will depend on the actual geometry of the weir and river system.

This loss of head with increased flow, and hence loss of power at high flows, must be taken

into account when estimating the annual energy capture at a potential site, otherwise the

annual energy capture could be significantly lower than predicted. However, this problem

will generally only be significant on low head schemes .

XFigure 32X shows an actual measured head-flow data from the Dulverton site. We can see that

at very low flows, which are usual during the summer months the gross head is about 3.4m.

However, as the flow increases we can see that the level of the tailrace rises to a greater

extent than the headrace and for the majority of the time the head is between about 3.2 and

2.8m. At high flows (say greater than 15m3/s) the tailrace begins to rise very quickly and

under flood conditions the tailrace comes up almost to the weir crest and the weir itself is

hardly visible anymore.

Figure 31: The hydraulic head


50

Figure 32: Variation of head with flow for the low head site at Dulverton.

2.3.3 Estimation of Net Head

Having established the gross head available, it is necessary to allow for the losses arising

from trash racks, pipe friction, bends and valves. In addition to these, certain types of turbine

must be set to discharge to atmosphere above the flood level of the tail race and therefore all

of the head cannot be utilised.. The gross head minus all these losses is the net head, which is

the head available to drive the turbine.

2.3.3.1 Pipe Losses

The greater the water velocity in the pipe, the higher the frictional losses. In general,

assuming the pipe is flowing full, the head lost (hf) is given by:

Dg2

vLhf (54)

Where L = length of pipe (m)

v = mean velocity in the pipe

g = the gravitational constant = 9.81m2/s

D = pipe diameter (m)

= pipe friction factor, which depends on the roughness of the internal pipe

surface, the pipe diameter and the flow velocity.

It should be noted that the head losses due to pipe friction increases with increased velocity

and decreases with the pipe diameter. However, it should also be noted that as D increases v

will decrease for a given flow (since Q = ( D2/4)v).

0

0.5

1

1.5

2

2.5

3

3.5

4

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40

Gro

ss H

ead

(m

)

Flow (m3/s)

Head-Flow Curve for a Small-Scale Hydropower Scheme


51

Therefore, the frictional losses will be strongly dependent on the diameter of pipe used to

transport the water from intake to powerhouse.

2.3.3.2 Turbulence Losses

Water velocity is an important parameter in frictional losses; it is also critical in the

turbulence losses associated with trash racks, pipe bends, sudden enlargement or contraction

of pipes, valves etc. These losses are denoted by ht and are given by the general equation:

g2

vKh

2

t (55)

In this way we can quantify the value of K for each loss in the components which will impede

water flow (e.g. trash rack, bends etc.). We can then simply add all the losses together and

arrive at a total loss of head and hence calculate the net head.

Here we will only deal with turbulence losses by the trash rack. Other losses such as pipe

bends and valves which maybe placed in a system are also of great importance but are not

covered here.

2.3.3.3 Head Loss at a Trash Rack For the loss of head due to the flow of water through a trash rack is calculated from a formula

by Kirchmer.

sin)g2/v()b/t(Kh 2

0t3

4

(56)

Where ht = head loss (m)

t = bar thickness (mm)

b = width between bars (mm)

vo = approach velocity (m/s)

g = the gravitational constant = 9.81m2/s

=angle of inclination from the horizontal

XFigure 33X illustrates the parameters v0 and in Equation 4. It should be noted that as the

angle of inclination is decreased, i.e. the bars become more horizontal, the effective surface

area will increase and hence the water velocity perpendicular to the trash rack will decrease.

This will significantly reduce the overall frictional losses due to the trash rack.

XFigure 33X also shows a range of K values for different types of bars that can be used in

construction of the trash rack. It can be seen that a significant improvement in head loss can

be made by using round or profiled bars.


52

Figure 33: The head loss through a trash rack. Source: [ESHA 1994].

2.3.4 Turbine Setting Losses

As we shall see in the next unit, turbines are fundamentally of two kinds, impulse and

reaction. In the former case (e.g. Pelton and Turgo) the flow converts all of its kinetic energy

into power through jets striking buckets or blades. The spent water then flows away at

atmospheric pressure. Accordingly the turbine itself must be kept clear of the tailwater which

would otherwise exert a drag on the turbine wheel if a small part of the turbine was

submerged. It is therefore necessary to keep the turbine above the highest flood level.

However, this clearly means that some of the available head cannot be utilised. This loss of

head is referred to as the turbine setting loss and must be added to the other minor losses to

determine the true net head on a reaction turbine. This is shown schematically in Figure 34.

Note that this loss does not occur with reaction turbines which operate fully submerged.


53

To summarise, all losses with the exception of setting losses are strongly influenced by the

water velocity. A general principle is “slow moving water is good”. Keep water velocities

low. However, there will clearly be a compromise between reducing velocities and cost

since, for example, large diameter pipes are more expensive. In general the velocity of water

in pipes should be between 1 and 3m/s. However, an iterative process should yield an

optimum solution between minimising losses and the extra cost involved.

Figure 34: Setting-loss on an impulse turbine


54

2.4 The Flow Duration Curve

From Eq. X50X we have seen that the two critical parameters in determining the potential

power from a site are the hydraulic head and the flow in the river. However, the flow in the

river will tend to vary, and often substantially vary, on a day by day basis. It is of great

importance to know what kind of variation there is over a year period. This data is often

shown in the form of a flow duration curve. This can be determined:

By purchasing flow data from the Environment Agency

Using a software package such as HydrA

Measuring the flow yourself

2.4.1 Variation of Flow Over a Year

The flow in a given river will vary greatly throughout the year, generally having high values

during the winter months and low values during the summer months in Europe. In general, a

small scale hydropower scheme in the UK correctly sized will not be expected to have

sufficient water to run continuously throughout the summer months.

XFigure 35X shows the mean daily flow for the river Barle in Somerset for the particular year

1980 from January through to December. We can make the following observations:

1. For a couple of months in the summer the flow dropped to very low values, less than

1m3/s.

2. The river flooded (over 20m3/s for this river) on several occasions.

3. The river level tends to rise fairly quickly but this level reduces gradually i.e. there is

a sharp “leading edge” followed by a slow decay. This is a general characteristic of

all rivers.

Figure 35: The mean daily flow on the river Barle at Dulverton (Somerset) in 1980.

0

5

10

15

20

25

30

35

Flo

w m

3/s

Mean Daily Flow on River Barle (Somerset) Jan - Dec for 1980


55

Since the flow is continually varying we require some method of representing and

quantifying this variation. The usual form that this data is presented is in a flow duration

curve and is explained below.

2.4.2 What is a Flow Duration Curve?

A flow duration curve represents the flow as a function of percentage of time that a flow is

exceeded. This is best explained by looking at an example. The flow duration curve in

XFigure 36X is generated by the computer program HydrA.

The Y-axis is the flow in m3/s and the X-axis is „percentage exceedance probability‟. For

example, from this graph we can see that a flow of 0.5m3/s is exceeded about 95% of the

time, a flow of 2.41m3/s is exceeded 50% of the time, the mean flow is 4.83m

3/s which is

exceeded just under 30% of the time, and flows of more than 20m3/s are exceeded about 1%

of the time.

This particular FDC is for a rather “flashy river” in Somerset, England. By the term “flashy”

we mean that the river is prone to large variations in flow and the river level can rise very

quickly in a short period of time.

Figure 36: A typical flow duration curve calculated from HydrA


56

2.4.3 Q-Values

When we refer to the percentage of time that a flow is exceeded we called it a “Q-value”

For example the flow which is exceeded 95% of the time is referred to as the Q95 value. The

flow which is exceeded 40% of the time is called the Q40 value etc. Clearly the lower the Q-

value the higher the flow e.g. the Q10 flow will be much greater than the Q95 flow.

Historically in the UK several Q-values have had particular significance.

Q95

In general, a reserve flow imposed on a particular hydropower site was historically set at Q95

i.e. the licensing authority in the UK (previously the National River Authority, now the

Environment Agency) would insist that a flow equal to the Q95 value must remain in the river

at all times. That is to say this amount of water would not to available to the hydropower

scheme but must remain flowing over the weir (for example) at all times when natural flows

allow. This restriction was perfectly reasonable and ensured that a minimum level of water

remained in the river to allow fish passage and protected wildlife.

However, this simple rule appears not be relevant anymore and more stringent and more

complex restrictions are often imposed by the Environment Agency.

Q40

Since the flow in a river is continuously varying, the determination of the optimum size of

turbine for a particular site is not trivial. A large turbine will give a higher maximum power,

but will only operate at full power on a small number of days. Conversely a very small

turbine will operate throughout most of the year but will not be able to take advantage of the

larger potential power capture at higher flows.

Clearly there will be an optimum somewhere between the two. It was found that, in general,

taking a turbine design flow equal to the Q40 value would give a solution close to this

optimum (assuming a Q95 reserve flow). However, since more complex abstraction regimes

are being imposed on developers and the fact that the economics of small scale hydropower is

generally more difficult, this is no longer really satisfactory.

2.5 Glossary

Compensation flow: water flow that does not pass through the turbine, usually to maintain

some flow in the river or over a weir.

Control System: controls the speed of the turbine or the electrical output.

Draft tube: a gently increasing diameter tube returning water to the river but recovering

some energy from it.

Forebay: a tank or area where water enters the penstock.

Gearbox: changes speed between turbine and generator (not always needed).

Generator: produces electricity due to the rotation of the attached turbine.

Guide vanes: direct the flow of water onto turbine blades.

Leat: a channel that conveys water in open flow i.e. the water is not under pressure.

Penstock: a pipe that conveys water under pressure.


57

Runner: the rotating part of a turbine that includes the turbine blades.

Settling tank or area: allows sediment to sink under gravity rather than entering the

penstock and turbine.

Spillway: allows excess water to return to the river.

Tailwater/tailrace: water leaving the turbine to return to the river.

Trashrack: a crude filter to avoid rubbish entering the turbine.

Turbine: device-extracting energy from the flow of water.

Weir: an artificial waterfall.

2.6 Introduction to Water Turbines

2.6.1 The Development of Water Turbines

In 1832, a French Engineer, Benoit Fourneyron, patented a more efficient water-wheel: in

fact, this was the first successful water turbine. The novelty of Fourneyron's design was the

vertical axis, the use of guide vanes to direct the water radially outwards onto the runner

blades and the fact that the turbine functions completely submerged.

The turbine operates because of the reaction of the runner blades. Machines of this type are

termed Reaction Turbines. All these features lead to a large flow of water through the

turbine, through many guide vanes, and hence a large power capacity, with conversion

efficiencies of up to 80%. The runner rotates at high speed and so can be readily used for

electrical generation. Having given up its energy, the water falls away into the outflow. The

power is controlled by raising or lowering a ring between the guide vanes and the rotor

blades.

Another water wheel where it is possible to apply multiple streams or jets of water is the

Pelton Wheel (Pelton Wheel Turbine), named after the American Lester Pelton, who patented

the concept in 1880.

Figure 37 shows the typical

arrangement for a Pelton Wheel.

The water feeding the wheel is at

high pressure due to the head of

hundreds of metres, and it is

directed onto the buckets at high

velocity through a jet. In this

case, we are not concerned with

capturing energy from the fact

that the water loads one side of

the wheel and causes it to rotate

but we are able to

capture the kinetic energy from

the impulse of the water jet on

the blades. Hence the generic

term of Impulse Turbine for this

kind of turbine. The water will

give up most of its energy to

the buckets and

runner

deflecto

r Spear valve

reservoi

r

head

Figure 37: Typical arrangement for a Pelton Wheel.


58

therefore drive them at high speed. The difficulty of limited space for water flow can be

overcome in the Pelton by using two jets instead of one. This can be further extended to three

or four jets to accommodate greater flow.

Another American engineer, James Francis, experimented with inward flow radial machines

and this resulted in the modern machines known as Francis turbines which have supplanted

the Fourneyron design because of the advantage of collecting the tailwater in the centre and

allowing it to exit under gravity. Francis turbines generally operate in the mid pressure range,

that is, somewhat lower pressures or heads than the Pelton turbines.

Lower pressure reaction turbines have axial flow with large flow cross sections and resemble

a ship propeller.

2.6.2 Specific Speed

The speed of the runner blade relative to the water striking it is critical for the efficiency of

the turbine. Propeller type turbines run best when their blade tips move faster than the water.

In the case of the Kaplan, the blades should move at twice the water velocity, whereas the

skirted Francis is most efficient when the two speeds are roughly equal. The Pelton

theoretically performs best when its blades are moving at half the water velocity.

When selecting a turbine for a specific situation, factors like head, flow and power must all

be considered, but a valuable tool for selecting the most appropriate turbine type is specific

speed (Ns). This is related to the output power (P in kW), effective head (H in metres) and

rotational speed (n revolutions per minute).

2/5H

PnN s (57)

The turbine rotational speed is dictated by the generator speed, and would usually have to be

adequate to drive the generator at 1500 or 3000 rpm (with the appropriate number of pole

pairs 2 and 1 respectively in these cases) through a gearbox.

Equally, the specific speed can be calculated from the equation,

))((500W

Bs

v

v

R

rN (58)

where

r = radius of water jet or flow R = radius of runner

Bv = velocity of blade

Wv = velocity of water

Essentially, then, it is the ratios of water/blade ( r / R ) radii and blade/water ( Bv / Wv )

velocities that govern the specific speed and hence lead to the choice of turbine type. The size

of the turbine will be determined by the power requirements. Table 4 shows the specific

speed ranges for different types of turbine.


59

Type of Turbine Specific speed range

Pelton one jet 10-35

Pelton two jets 10-45

Turgo 20-80

Cross-flow 20-90

Francis 70-500

Kaplan 350-1000

Propellor 600-900

2.6.3 Shape Factor

Another, but equivalent, way of looking at specific speed is the shape factor. The shape factor of

a turbine is defined as

4/52/1

2/1

gH

P (59)

= density of water =1000 kg m-3

= angular velocity of the runner

P = Power in kW

g = acceleration due to gravity =9.81 ms-2

H = effective head m

Table 4: Specific speed range for

different types of turbine


60

2.6.4 Impulse Turbines

2.6.4.1 The Pelton Wheel

Impulse turbines are most suited to high pressure situations where the water head might be

hundreds of metres. The Pelton wheel is a wheel with a set of double cups or buckets

mounted around the rim (see Figure 38 and Figure 39). Water is projected onto the buckets at

high velocity. As it hits each pair of buckets in turn, the water jet splits into two, half passing

round each of the curved bowls or buckets. In doing so, the water gives up some of its kinetic

energy to drive the wheel. Under ideal conditions, almost all of the water kinetic energy is

converted to mechanical energy to the wheel: the theoretical optimum condition is when the

bucket velocity is half of the water jet velocity. This type of turbine is called an impulse

turbine because the energy is delivered to the wheel in a series of short impulses to a

sequence of buckets. Impulse turbines usually run at high speed in air at normal atmospheric

pressure.

Figure 39: Commercial

Pelton Runner

Figure 38: Small Pelton Runner


61

The power input to a Pelton wheel is determined by the usual product of effective head and

flow rate of the water together with the constants for water density and gravity.

gQHP (60)

But we know that the water velocity is governed by the head since

kinetic energy = potential energy.

mgHmv25.0

(61)

and so we can calculate the velocity of the water leaving the pipe at its lower end as

gHv 2

(62)

If the area of the water jet is A m2

Then Q = A v

So Q = A gH2

The power is then

P = Hg A gH2

Or approximately

P (kW) = 45A 3H

(63)

It is desirable for the water to reach each pair of buckets without interference from other

buckets. A small cutaway in each bucket helps with this but it is also important to space the

buckets to minimise such interference and this is done by ensuring that the wheel diameter is

ten times the diameter of the water jet. Greater power can be achieved by utilising more jets (

j= 2,3 or 4 jets)

P (kW) = 45 j A 3H (64)

2.6.4.2 Optimum Speed for a Pelton Wheel


62

The optimum speed of a Pelton wheel can be calculated from the change of momentum of the

water when it strikes the bucket. If the buckets are stationary, then no energy will be

transferred from the water to the turbine, but energy will be dissipated in splashing.

If the buckets move at the water velocity, then the water never catches up with the bucket and

so cannot give up any of its energy to the bucket.

It is possible to show using conservation of momentum principles that the theoretical

maximum energy transfer occurs when the buckets are moving at half the water velocity.

2.6.4.3 Performance of Pelton Wheels

Figure 41: Single Jet Pelton Turbine Efficiency Curve

1.1 1.2

1.3 Figure 5: Single Jet Pelton Turbine Efficiency Curve

Figure 40: Twin Jet Pelton Turbine Efficiency Curve

2.1 2.2

2.3 Figure 4: Twin Jet Pelton Turbine Efficiency Curve

20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

2 PERCENTAGE OF TURBINE FLOW

TURBINE EFFICIENCY

(%)

92

91

90

89

88

87

86

85

84

83

20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

3 PERCENTAGE OF TURBINE FLOW

TURBINE

EFFICIENCY (%)

92

91

90

89

88

87

86

85

84

83


63

Figure 40 and Figure 41 show typical efficiency curves for a Pelton wheel as a function of

water flow rate for one and two jet operations. Note that the efficiency is reasonably high and

does not fall particularly rapidly with flow.

Pelton Wheels have to withstand high velocity water impacting on the buckets and so the

buckets must be manufactured from a very durable material such as aluminium, steel or

phosphor bronze.

Although they are designed for very high head applications, Pelton wheels are also employed

in small-scale schemes with lower heads. The simplicity of the design and the fact that the

runner is in air and rotates at high speed makes them suitable for electrical generation.

Remote sites across the world are often fitted with Pelton wheels for these reasons. Other

reasons for using Pelton wheels include:

Greater tolerance to sand and other particles in water

Ease of fabrication

Allow easier access to working parts

Less subject to cavitation

Display flatter efficiency curves if a flow control device is fitted.

2.6.4.4 Other Impulse Turbines

The Turgo turbine was developed from the Pelton in the 1920s. Single shallow cups replace

the double ones so that water enters on one side where it strikes the cups in turn. As with the

Pelton, it is at its most efficient when the rotating blade velocity is about half of the water

velocity.

The shape of the cups accommodates larger water flows than the Pelton, giving the Turgo the

advantage of higher generating capacity.

Cross flow (also termed Mitchell,Banki or Ossberger) turbines (see XFigure 42X) consist of a

cylindrical shaped runner on a horizontal axis. The Pelton and Turgo axes can be horizontal

or vertical. A rectangular nozzle directs the jet onto the full length of the runner, the water

strikes the blades and gives up most of its kinetic energy. As the water falls through the

cylinder, it strikes the blades again on exit thereby imparting a small additional amount of

energy to the runner before leaving the turbine. The effective head can be increased by using

a draft tube to draw a partial vacuum inside the casing.

The cross flow machines can accept large flows of water, and in principle can be made with

long enough cylinder lengths to take larger and larger flows. Naturally, there will be a

practical limitation to this as the blades will be prone to flexing and have short lives if they

are too long. Cross flow machines are attractive because they are suitable for a wider range of

heads and flows. Control of the water flow is possible with a partition device, which cuts off

supply to either 1/3 or 2/3 of the runner.


64

Figure 42: A cross flow turbine.


65

2.6.5 Reaction Turbines

Reaction turbines are suited to lower heads than the impulse machines. A significant

distinction between impulse and reaction turbines is that impulse machines run in air whereas

reaction machines are submerged and in general will rotate faster than impulse machines. If

the rotational speed is high enough, then the electrical generator may be coupled directly onto

the shaft of the runner, removing the need for a speed-raising gearbox.

Reaction turbines are more geometrically complex and hence are likely to be more expensive,

but their high speed and efficiency often compensate for this.

2.6.5.1 The Francis Turbine

As discussed earlier, the Francis turbine was developed to permit a large flow of water onto

the runner blades, and this is done using a scrolled casing to deliver the water. Water passes

through guide vanes which turn the flow radially inwards and onto the runner blades. The

runner blades extract most of the energy from the water, which leaves along the axis via a

draft tube. The guide vanes are used to regulate the water flow as it enters the runner. They

can therefore be used as part of a control system to match the flow to the turbine loading.

Figure 43 shows a schematic view of a Francis turbine.

Figure 43: Cross-section of a Francis and its draft tube.

2.6.5.2 Propeller and Kaplan Turbines

For very low heads, propeller turbines are used. They are so called because they are similar to

a ship's propeller. However, they work by being driven by water rather than driving the water

past a ship. The runner consists of three blades in the case of very low heads or six blades for

Internal View External View


66

other conditions, and is located inside a tube in such a way that its shaft can be taken out

where the tube changes direction. The generator can thus be outside the water chamber. A set

of guide vanes, swivelling gates or wicket gates regulate the flow to the propeller. Propeller

turbines have fixed runner blade geometries, and tend to show poor performance in part flow.

Where it is possible to spend more money, such as large scale hydro sites, a variable pitch or

Kaplan version of the propeller can be used (see Figure 44 and Figure 45). This means that

the runner blades can be swivelled, which, together with the wicket gate adjustment, has the

benefit of maintaining high efficiency under part flow conditions.

Figure 45: Photograph of a

Kaplan Turbine

Figure 44: Kaplan Turbine


67

2.6.6 Draft Tube

The draft tube concept applies to all reaction turbines and to the partially evacuated crossflow

turbine. It is used to increase the performance by means of energy recovery. As the water has

to exit the turbine with some velocity in order to get away from the turbine, it still possesses a

significant amount of kinetic energy. To recover most of this energy, the water velocity must

be reduced gradually without incurring heavy friction losses. This recovery of velocity head

is usually achieved by using a draft tube with a gradually increasing cross sectional area so

that the final exit of water takes place through a area of twice that of the inlet. The angle of

the expansion of the walls is typically 7 degrees.

Consider area A1, the kinetic energy = 0.5mV12

And at A2 the kinetic energy is = 0.5mV22

Since the flow of water is the same through both areas

Q = A1 V1 = A2 V2

Therefore, if the area reduces by a factor of two in the draft tube.

V2 = A1/A2 V1 = 0.5 V1

V2 2 = 0.25 V1

2

And so the kinetic energy is reduced to a quarter of its value at inlet, thereby minimising the

loss of energy, without the draft tube being too long.


68

2.6.7 Cavitation

Cavitation arises in turbines during operation at low pressures. As water passes over the

runner, the pressure may decrease to the point where the water vaporises, forming bubbles.

When these bubbles are subjected to greater pressure elsewhere in the system, they can

collapse violently producing miniature explosions which can fatigue and erode nearby

surfaces. Cavitation leads to pitting of the runner blades, reducing turbine efficiency and

shortening their lives. It also leads to poor part load flow characteristics.

One design feature to reduce this problem is to ensure that the height of the turbine above the

tail water is not too great as this will lower the working pressure and encourage cavitation.

2.6.8 Selecting a Turbine for a Particular Site If we know the head and annual water flow, then we can suggest how much power we could

expect from our site. To decide on the best turbine, we can calculate the specific speed, which

will lead us to a small choice of possible turbines, and we can then use the look up chart

shown in Figure 46 as a final check.

Figure 46: Turbine selection charts for small to large scale turbines.


69

3 Electrical System Aspects

3.1 Apparent, Real and Reactive Power

Power systems are designed to transfer energy from a generator to a consumer with

maximum efficiency and to deliver this energy at virtually fixed voltage. Over the years

techniques have been developed to analyse power systems based on the concepts of active

and reactive power. A brief review of the techniques will be provided here/

Electrical networks consist of sources of energy, which can be considered as “active”

elements and “passive” elements such as resistors, capacitors and inductors. The vast

majority of electrical consumption equipment (whether industrial, domestic or commercial)

consist basically of inductive-resistive components. Overhead cables have some limited

capacitance, the higher the voltage the more pronounced the capacitive component.

Underground cables however, contain substantial capacitance.

The circuit in XFigure 47X, which consists of resistance R, and inductive reactance XL,

represents a typical consumer connected to the electrical grid. The grid voltage phasor V

impressed across the consumer terminals results in a current phasor I which when flowing

through R and XL results in voltage drops across these components of RV and LV

respectively.

P

I

Q

R

V

RV LV

a a‟

Energy Source

LX

Figure 47: A inductive-resistive network representing a typical consumer.


70

According to Kirchhoff‟s Voltage law the phasorial sum of voltage drops is equal to the

applied voltage i.e.

LR VVV (65)

The phasor diagram illustrating this equality is shown in Figure 48a where RV is in phase with

I and LV is leading I by 90 . The angle is the angle by which the current lags the voltage

in this circuit.

By multiplying the sides of the voltage triangle in Figure 48a by I , the result is the “power

triangle” shown in Figure 48b. The product IVR represents the power that is irreversibly

converted from electrical into other energy forms. e.g. heat or mechanical power. This is

known as “Active Power” and is denoted by P. It is measured in Watts (W), kilowatts

(kW = 103W) or megawatts (MW = 10

6W).

The product IVL does not represent irreversible energy conversion (because inductors are

storage devices) but is energy which flows backwards and forwards between the the energy

source and the inductor. This fluctuating power is denoted by Q and is known as “Reactive

Power”. It is measured in Reactive-Voltamperes (VAR‟s) or kVAR or MVAR.

The product VI is known as S, the Apparent Power with units of Volt-amperes (VA),

kilovolt-amperes (KVA) and Megavolt-amperes (MVA). The active and reactive power are

obtained from the volt-amperes through the following expressions:

cosVIP (66)

sinVIQ (67)

where cos is known as the circuit “Power Factor” and is known as the power factor angle.

VLV

RV

I

IVP R

IVQ L

VISa b

Figure 48: a) Voltage triangle – phasor diagram from figure 1and b) “power triangle”

showing apparent real and reactive power for an inductive-resistive network. S is the

Apparent Power, P is the Active Power and Q is the Reactive Power.


71

3.2 The Balance of Active and Reactive Power

In Figure 49 the consumer “imports” or is a “sink” of active power P given by equation X66X.

This concept can be extended to reactive power so that the consumer can also be considered

as “importing” or being a “sink” of reactive power Q given by equation X67X.

If at the interface aa , shown in XFigure 47X, we were to look into the terminals of the energy

source we may conclude that as P and Q are coming out of the terminals the source is

“exporting” or “generating” P and Q.

This way of looking at P and Q is of great benefit to power systems engineers who analyse

energy networks in terms of these quantities. As P and Q are scalar quantities they can be

added at any node on the network as shown in figure 3.

At the node of Figure 3 the incoming P and Q are equal to the outgoing P and Q respectively

i.e.

nodeaat

P 0 and nodeaat

Q 0

Q1 Q2

P2P1

P3

Q3 Q4

P4

ConsumerConsumer

GeneratorLarge Interconnect System

Figure 49. Active and Reactive Power balance.


72

3.3 The Significance of Power Factor

Figure 50 represents a basic circuit of energy transportation from a generator to a consumer

through a transmission line, which is simply represented by a series inductance and

resistance. The consumer is conveniently modelled by an inductive resistive impedance.

Following the reasoning in section X3.1X the consumer or load “absorbs” both PL and QL. The

generator or “the utility” has to supply both the P of the load and the loss in transmission

given by tRI 2 where tR is the resistance of the transmission line. The transmission loss is

strongly influenced by the load QL.

To appreciate this, assume that the load consists solely of an inductor, which absorbs Q but

not P. In this case the energy meter at the consumer premises which records kilowatt-hours

(i.e. energy purchased) indicates zero but the finite current in the transmission system results

in tRI 2 which has to be supplied by the utility. Clearly this is a most undesirable scenario

which the utility endeavours to discourage through special tariffs which penalise the

absorption of Q by large loads.

Additionally, the utility has the legal obligation to supply power to the consumer at a more or

less fixed voltage. It can be shown that for transmission lines, with tt RX , the voltage LV

at the consumer terminals is particularly sensitive to changes in Q rather than P. It is now

obvious that the utility is keen to encourage consumers to draw P at minimum Q i.e. with

cos close to unity.

VL

QL

Generator

QC

PL

RtXt

Transmission Line Consumer (load)

I

RL

XL

C

a

a

b

b

Figure 50: Single transmission system


73

3.4 The Reactive Power of a Capacitor

Redrawing Figure 48b) for a resistive capacitive circuit results in Figure 51. QC is now

negative therefore the reactive power associated with a capacitor is “exported” or

“generated”.

Capacitors are used extensively in power systems to generate or “inject” reactive power at

strategic points of the network. Their importance is illustrated in the transmission circuit of

Figure 50. If a capacitor C is connected across the load terminal and C is sized to generate

QC locally so that QC = QL, the reactive power absorbed by the load, then the consumer will

appear to the utility as having a unity power factor. The utility will be truly satisfied with

such an arrangement as this will minimise losses in the transmission line resistance Rt and

will ensure little variation in consumer voltage.

IVP R I

VIS IVQ CC

Figure 51: Apparent, Real and Reactive power for a resistive/capacitive circuit

Figure 5: Apparent, Real and Reactive power for a resistive/capacitive circuit


74

3.5 P and Q Transfer in Power Networks

In Figure 52 a typical transmission line is shown which consists of inductive reactance X and

resistance R. Typically the voltages at the two extreme ends of such a line are somewhat

different in magnitude and slightly out of phase. This is illustrated by the two sine waves in

Figure 52, which show, BA VV and that AV leads BV by an angle .

The ratio of inductive reactance X to resistance R varies from over 20 in very high voltage

lines to less than 2 in low voltage distribution lines. To simplify matters in this unit we will

consider only the extreme case of X >> R when the resistance can be legitimately neglected.

For this condition obtainable in the high voltage network and to a good approximation, the

active and reactive power can be shown to be given by:

sinX

VVP BA (68)

)cos( BAA

A VVX

VQ (69)

)cos( ABB

B VVX

VQ (70)

-

VA

A X R

0BB VVVB

AA VVTransmission Line

Figure 52: Transmission line characteristics


75

For reasons beyond the scope of this unit, in a well run power system the angle (known as

the load angle) is of the order of a few degrees. The effect of this on equations X68X to X70X is to

make the active power P very sensitive to changes in but much less so to changes in V. The

converse is true for reactive Power Q which is sensitive to changes in V but not to changes in

We can conclude that at the high voltage network, active power is mainly transported by

virtue of the angular displacement between voltages at adjacent nodes, but reactive power is

transported through voltage magnitude differences.

As the grid transmission line voltage is transformed down to the primary and secondary

distribution level the ratio of reactance to resistance decreases and equations 3 to 5 become

progressively less accurate.

3.6 The Generator

All grid connected wind turbines drive a three phase alternating current (AC) generator to

convert mechanical power to electrical power, usually at an output voltage of 690V.

Generators are divided onto two classes; Synchronous and asynchronous. Both have a non-

rotating part, the stator, which is similar for the two types of generator and consist of a three-

phase winding on a laminated iron core; the currents which result when the stator is

connected to the grid produce a magnetic field rotating at constant speed. Although the

stators are similar, the rotors (the rotating part of the generator) are quite different.

The rotor of a synchronous generator (also known as an alternator) has a field winding

carrying a direct current supplied through slip rings. The field winding creates a constant

magnetic field, which locks onto the rotating field created by the stator winding currents.

Thus the rotor always rotates at a constant speed in synchronism with the stator field and the

network frequency. In some designs the rotor magnetic field is produced by permanent

magnets, but this is not common for large machines.

The rotor of an induction machine consists of a “squirrel cage” of bars, short-circuited at each

end and embedded in an iron cylinder. There is no electrical connection onto the rotor, and

the rotor currents are induced by the relative motion of the rotor with respect to the rotating

field of the stator.

If the rotor speed is exactly equal to the speed of the rotating field produced by the stator,

there is no relative motion and so there are no induced rotor currents. Therefore an induction

generator always operates at a speed that is slightly higher than the speed of the rotating field

stator. The difference in speed is known as the slip and is between 1 to 3% when the WT is

producing its rated output.

Induction generators are not common compared to synchronous generators (for example all

utility power systems generators are of synchronous type) however there are many millions

of induction motors in service throughout the world. An induction generator is essentially an

induction motor with torque applied to the shaft, which endeavours to run the rotor above

synchronous speed. Conversely, in an induction motor the torque endeavours to slow down

the rotor below synchronous speed.


76

The performance of an induction machine can be described by the torque-slip curve of XFigure

53X where slip is defined by:

s

rss (71)

where s is the synchronous speed of the rotating magnetic field and r is the rotor speed.

Slip is the “normalised” speed difference and is positive for motoring and negative for

generation. The slip is plotted along the x-axis and the torque on the y-axis. A slip of zero

corresponds to the rotor speed being in synchronism with the stator field of the stator and so

no torque or power is generated.

The small variation of speed with increments of torque inherent in the induction generator of

a wind turbine are of considerable advantage in suppressing surges of torque in the drive-train

and of electrical power into the network, during gusts.

2 s

generatingmotoring

max

normal

generating

no

load

SM

0

min

-SM

s

slip,s

Figure 53: Operating Characteristic of an induction generator

3.7 Comparison of Synchronous and Induction

Generators

The majority of generators used throughout the world are of synchronous type. However, for


77

fixed speed wind turbines induction generators are normally used. Synchronous generators

are slightly more efficient than induction machines and have a major advantage in that their

reactive power flow can be controlled. In a synchronous generator the direct current flowing

in the field windings is independently controllable. By increasing it, reactive power may be

exported to the network. By reducing it, reactive power may be drawn from the network.

Adjustments in the flow of reactive power provides control over the voltage of the network at

the point of connection of the synchronous generator. The advantages and disadvantages of

each machine are given in XTable 5X.

Table 5: The advantages and disadvantages of synchronous and induction generators

Synchronous Generators Induction Generators

Advantages

Advantages

More efficient

Reactive power flow can be

Controlled

Suitable for variable speed

operation through an electronic

interface

Suitable for connection to very weak

networks

Less expensive

More rugged and robust

Responds to gusts in a non-oscillatory

way

Small change of speed with applied

torque reduces stress in drive-train and

spikes in power fed into network

Can be simply synchronised to the mains

Disadvantages

Disadvantages

More expensive

Responds to gusts in an oscillatory

manner

Requires precise synchronisation to the

mains

May lose synchronism under severe

network transients

Never used now for fixed speed

operation

Consumes Reactive Power

Requires power factor correction

Not suitable for connection to very weak

electrical networks


78

Overall the ability of the induction generator to allow, through variation of slip, small

changes of rotor speed which dramatically alleviate transient drive train loads means that this

is the favoured choice. Indeed, synchronous machines are never used for fixed speed

applications.

Pole switching or two separate induction machines is sometimes employed to give two-speed

operation. Additional compliance at above rated wind speed, when it is most required since

the magnitude of the turbulent variations increase with wind speed, can be achieved by

increasing the generator slip. The Vestas Opti Slip system increases the generator slip at, and

above, rated power by switching in additional resistance to the generator rotor winding

circuit.

For variable speed operation the choice of electrical machines is basically the same, although

both the cage and the wound rotor variants of the asynchronous generator are considered, and

permanent magnet synchronous generators can be used in addition to the more conventional

field winding excited machines. In all cases a power electronic converter has to interface the

variable frequency, variable voltage generator to the fixed frequency and fixed voltage grid.

Such converters consist now of power electronic transistors which are switched in such a way

(pulse width modulation switching) that only small amount of harmonics are inject into the

grid.

Some large multi-pole synchronous generators have been specially designed for wind

turbines. When combined with a variable speed drive, it has been possible to remove the

gear-box from the wind turbine altogether.

DC machines are not used in wind turbines due to their cost and maintenance requirements.

Whether fixed or variable speed operation is preferable depends on a range of issues, but

noise generation and power quality (related to the strength of the grid connection) are two of

the most important.

3.8 Connection to the Electricity Network

The voltage at the point of connection of a WT is usually 690V. A transformer at the base of

the each tower is used to increase the voltage to, say 11kV, for transmission to the utility

network. Induction generators require power correction capacitors and these are usually

located in a cabinet at the base of the tower together with protective switchgear and control

equipment.

3.8.1 Power Factor Correction for an Induction Generator

Even when the real power output from an induction generator is zero it will still draw

considerable reactive power to magnetise its iron core. As torque is applied and real power is

exported to the network more reactive power is absorbed. The relationship between active

and reactive power for an induction generator is shown in XFigure 54X.


79

Figure 54: The relationship between active and reactive power for an induction

generator.

The induction generator power factor will vary from zero at A to say 0.9 at B. In order to

improve the power factor it is convenient to fit local Power Factor Correction (PFC)

capacitors at the generator terminals. These have the effect of shifting the overall

characteristic downwards to A‟ B‟. Capacitors that generate a reactive power not exceeding

80% of the no-load induction generator import (0A) are usually fitted.

If connection to the mains is lost and the WT accelerates then a larger compensation may lead

to self-excitation and higher voltages from the induction generator, This is known as

“islanding”.

Q Import

(MVAR)

P export (MW)

B

B´

A

A´

Effect of

PFC

0


80

3.9 Soft-Start Units

When induction generators are connected to an electrical network, there is a large inrush of

current as the iron core is magnetised to establish normal operating conditions. This is

similar to the direct-on-line starting problems of induction motors. Wind turbines use the

same soft start equipment used to start large induction motors. This is shown schematically

in XFigure 55X.

A soft-start unit works by inserting two thyristors back-to-back in each phase of the supply.

When an induction machine is connected to the mains, the thyristors are used to control the

voltage applied to the stator and hence will limit the inrush current. They effectively allow

the iron core to be magnetised over a longer period of time.

Soft-start units absorb a small amount of electrical power which would reduce the output

from the turbine. For this reason they are usually bypassed by a contactor as soon as the

generator is fully magnetised.

Bypass Contactor

Generator Network

Figure 55: A wind turbine soft start unit (only one phase is shown).


81

3.10 Power Quality

Wind farms “embedded” into distribution networks will influence the “power quality”.

Power quality is defined in terms of the “steadiness” of the frequency and the “purity and

steadiness” of the sine wave voltage supplied to the consumer. Embedded generation such as

a wind farm will have no effect on the frequency of a large interconnected system but may

well have a detrimental effect on local system voltage.

The usual cause of concern are i. transient network voltage disturbances referred to as

“flicker” because of their effect on lighting and ii. harmonic distortion of the voltage

waveform.

Voltage flicker may be caused either by the connection and disconnection of generators onto

the grid or by transient torque pulsations from the wind turbine rotor being translated into

network voltage variations.

Standards are in place to control the connection of loads likely to degrade the power quality

of the distribution network and these are also applied to wind farms. The prediction of

voltage flicker from wind farms is difficult and either data is required from installations

already in service or very detailed simulation modelling of the local system is necessary.

Wind turbines can contribute to flicker because tower shadow and turbulent winds can cause

rapid variation in torque and hence in real and reactive power. Pitch regulated wind turbines

are generally worse than stall-regulated and variable speed generation produces less flicker

than fixed speed.

Switching wind turbines on and off can cause step changes in line voltage but this can be

minimised through the soft-start units described earlier.

With wind farms, it is normal to connect turbines one at a time. The severity of flicker will

depend on how frequently switching occurs, the gustiness of the site, the size, design and

operating mode of the wind turbine and the strength of the local grid.


82

4 Tidal Power

4.1 Introduction

Tidal power has a long history. Records indicate that tide mills were being worked on the

coasts of France, Spain and Britain prior to 1100 AD. These consisted simply of a storage

pond, which was filled during the flood tide through a sluice and emptied during the ebb tide

through a waterwheel.

The technology used in these mills and the applications of the mechanical power were much

the same as in hydropower water wheels. The tidal mills captured seawater at high tide and

released it to provide low speed high torque output to grind corn etc.

These mills remained in common use for many centuries: they were gradually displaced by

the cheaper and more convenient fuels and machines made available by the Industrial

Revolution.

XFigure 56X shows a good example of this in Saint Suliac in La Rance estuary in France.

Figure 56: Saint Suliac in La Rance estuary in France


83

4.2 Causes of Tides

4.2.1 Tides in the open ocean The Moon and the Sun‟s gravitational fields cause the natural rise and fall of coastal tidal

waters. Since the Moon is closer to Earth, albeit much less massive, it has a dominant effect

upon tides. The Moon is 2.2 times more influential than the Sun. We could consider tidal

energy to be mostly a form of lunar energy!

The Earth rotates on its axis once every 24 hours. In the Earth‟s frame of reference the Sun

appears to orbit the Earth once every 24 hours. The Moon orbits the Earth once every 29 days

approximately. In the Earth‟s frame of reference, the Moon appears to orbit the Earth once

every 24 hours and 50 minutes. This difference in periods between the apparent orbits of the

Sun and Moon leads to phase changes with larger spring tides (See XFigure 57X) during in

phase behaviour and smaller neap tides (see XFigure 58X) when the Sun and Moon are out of

phase. Spring tides can be twice as large as neap tides.

Figure 57: Spring tides when gravitational effects of Sun and Moon are in phase. This

occurs twice per month.

Figure 58: Neap tides when gravitational effects of Sun and Moon are out of phase. This

occurs twice per month.

The Moon and the Earth can be considered, due to their mutual gravitational attraction as two

objects orbiting a common centre of gravity, which is far closer to the centre of the Earth than

the centre of the Moon as the Earth is far more massive than the Moon. The pair orbit this

common centre of gravity once every lunar month (about 29 days). The effect of the Moon‟s


84

gravity is stronger on the Earth‟s surface facing the moon and weaker at the surface facing

away from the Moon than at the centre of the Earth. Because of this difference in

gravitational pull, the Earth‟s oceans will tend to bulge towards the Moon when facing the

Moon and bulge away when facing away from the Moon. This leads to two high tides per day

across the Earth‟s surface as the Earth rotates and the high tides have period of approximately

12h25m. When the bulge is at its greatest height at a point on the Earth‟s surface this gives

rise to the flood tide (high tide). When it is at its least, 6h12m later, this is known as the ebb

tide (low tide).

XFigure 59X shows this Earth-Moon system. The Earth of mass E and Moon of mass M, rotate

about the their common centre of mass with angular velocity Ω. The centre of mass is a

distance l from the centre of the Earth. The Earth and Moon are separated by a distance R.

The Earth rotates on its own axis with angular velocity ω. The radius of the Earth is r.

Figure 59: The Earth-Moon system and effect on tides

The radial force, F1, on a mass of water m on the surface of the Earth closest to the Moon is

given by:

rmlrmrR

GMm

r

GEmF 22

221 (72)

where G is the gravitational constant.

If the mass m is on the opposite side of the Earth the radial force, F2, is then:

rmlrmrR

GMm

r

GEmF 22

222 (73)

The centrifugal forces on the Earth and Moon must balance the gravitational forces between

them so:

lRMlER

GEM 22

2 (74)

As R r, Eq. X72X can be written to a good approximation as:


85

rmlrmR

r

R

GMm

r

GEmF 22

221 21 (75)

Similarly, Eq X73X can be written as:

rmlrmR

r

R

GMm

r

GEmF 22

222 21 (76)

Substituting Eq X74X into Eqs X75X and X76X gives:

rmR

GMmr

r

GEmFF 22

3221 2 (77)

Hence the radial force on the oceans facing towards and away from the Moon are the same.

This shows that the magnitude of the flood tide is the same on the opposite sides of the Earth.

By integration of Eq X77X and by making various assumptions it is possible to show the

maximum (spring) tidal range, h, is given to a good approximation, by:

3

4

ER

Mrh (78)

For the Moon this gives h=0.35m and for the Sun this gives h=0.16m. The combined height

of around 0.5m gives the expected tidal range for deep oceans away from land.

4.2.2 Mechanisms for Tidal Enhancement

Over much of the surface of the oceans, the tidal range (the vertical rise and fall) is rather

small, less than one metre, but in certain places, there is an enhancement of the range.

Enhancement may be due to:

1. Funnelling (as is the case of Figure 60, Severn Estuary) -

The tide is gradually constrained from the sides and so increases in height - or the reverse,

later in the cycle.


86

2. Resonance (as is also the case in the Severn) -

The estuary has a resonant period equivalent to the tidal period. The length and depth of

the estuary are very important for Resonance: they should approximately satisfy the

expression:

L/h 0.5

= 35000 m 0.5

Estuary Length, L

Resonance occurs when the length of the estuary

corresponds to a quarter of a wavelength.

Velocity of the tidal wave = (gh)0.5

Where h = water depth

T = period =12h25m =44700s

For L= quarter wavelength = (gh)0.5

44700/4

Then L/h 0.5

= 35000 m 0.5

3. Coriolis Effect (e.g. La Rance, see section X4.3.4X) -

The spinning of the Earth leads to the Coriolis effect (the effect that

causes weather systems). The tide is influenced by this and so tends to

increase in height at high tide on the northern French coast and be drawn

away from this coast at low tide, with the net effect of enhancing the tidal

range. For the equivalent coast in Southern England, the effects are in the

opposite sense creating a lower tidal range. Figure 61 is a photograph of

Mont St Michel in France. This indicates a tidal range of up to 14m.

Figure 60: Severn Estuary


87

4. Atmospheric Pressure –

Changes in atmospheric pressure can depress or elevate the sea level by tens of

centimetres.

5. Storms –

Storms can push water onto a coastline.

The first three of these enhancements are highly predictable, and so permit us to largely

determine the output from a tidal scheme years ahead. The last two are unpredictable except

in the short term.

Figure 61: Mont St Michel, France - tidal range of up to 14m


88

4.3 Tidal Barrages

Constructing a barrage across an estuary and allowing tidal waters alternately to fill the

estuary through sluice gates and then to empty it through turbines can generate energy.

In the modern version of the tidal mill, a barrage constructed across an estuary is equipped

with a series of gated sluices and the waterwheel is simply replaced by a bank of low-head

axial turbines. Where it is necessary to maintain navigation to the upper part of the estuary, a

ship-lock may be required.

4.3.1 Turbines for Tidal Barrages

Electricity is generated using large axial flow turbines of diameters up to 9m. In view of the

continuously varying head of water that drives them, it is necessary to regulate the blade

angle of the guide vanes or of the turbine runner, or both, for maximum efficiency. If the

turbine is also to be used in both directions for generation, or in reverse for pumping, variable

control of both the guide vanes and the turbine runner (i.e. double regulation) may be

advisable. The conventional „bulb‟ turbine containing the generator in a pod located in the

water passage directly behind the turbine runner is generally recommended for tidal schemes.

4.3.2 Operational Strategies Ebb generation is the most simple mode of operation for a tidal barrage scheme. The

operating cycle consists of four steps:

1. Sluicing on the flood tide, to fill the basin.

2. Holding the impounded water until the receding tide creates a suitable head.

3. Releasing the water from the basin to the sea via turbines, on the ebb tide, until the tide

turns and rises to reduce the head to the minimum operating point.

4. Holding until the tide rises sufficiently to repeat the first step.

Ebb generation with flood pumping is a modification of this mode which allows increased

energy output. By using the turbines in reverse as pumps, the basin level and hence the

generating head can be raised. The energy required for pumping must be imported but since

the pumping is carried out against a small head at high tide and the same water is released

later though the turbine at a greater head, this can produce a net energy gain with some

limited ability to re-time output. The energy gain through pumping could be small but useful

and typically in the range 3-13%.

Flood generation is the reverse of ebb generation and is rarely suggested alone, possibly

because it offers little storage opportunity and the basin is often being filled by a river which

would then reduce the total energy capacity.

Two-way generation (ebb and flood) is possible with reversible turbines and is used at La

Rance (together with flood pumping). The additional energy recovered may not justify the

extra cost and complexity of the turbines.

Tidal energy barrages are expected to have very long lifetimes. Their design life could be

about 120 years but with normal maintenance and replacement of turbine generators at 40-


89

year intervals, their lifetime could effectively be unlimited.

4.3.3 Choice of Barrage Site

Barrages are expensive and so a careful analysis of potential sites is needed. If the tidal range

is reasonably large, then to some extent, a simple view of the length of the barrage needed

will be that the basin should be as large as possible and the barrage as short as possible. The

parametric equation below was deduced by Baker from the data for about 20 prospective tidal

barrage site: it is a useful method of making a preliminary evaluation of a new site.

2

28.0

1

2loglog

RA

HLkU

(79)

where U = cost of electricity (p/kWh)

L = barrage length (m)

H = barrage height (m)

A = basin area (km2)

R = tidal range (m)

K = constant say 0.69


90

4.3.4 Case Study: La Rance

The tidal scheme at La Rance was constructed during the 1960s and fully operational by 1967

(see Figure 62). The northern French coast benefits from the Coriolis force, the spinning of

the Earth giving an enhanced tidal range to France compared with that of southern Britain.

The scheme consists of a barrage across the river Rance. Water from the high tide is trapped,

permitting generation once the tide has retreated. This is ebb generation, but an unusual

design at La Rance incorporates reversible turbines and so energy can also be derived from

the incoming (flood) tide. The characteristics of single action and double action systems are

shown in Figure 63 below. In the photograph in Figure 62, the sluice gates are open,

allowing seawater to flood into the basin. Note also the benefit of using the turbines to pump

water when the tide is fully in.

The major specifications in relation to the La Rance site are given in XTable 6X.

Figure 62: Tidal scheme at La Rance, France.

Figure 63: Operational strategies utilising the reversible turbines at La Rance. The

turbines can also be used in pumping mode.


91

Maximum tidal range 13.50m

Barrage length 750m

Barrage height 13m

Basin area 2,200Ha

Water volume 184,000,000 m3

24 Bulb type turbines Each is a 4 bladed Kaplan, see Figure 64,

5.35m diameter, running at 93.75rpm and

producing up to 10MW from a max flow of

275cumecs. A cross section of one the

turbines is shown in XFigure 65X.

Availability 96 to 97%.

Each turbine can also be used as a pump,

using 10MW to pump 225cumecs with a

head of 1m. This enables the operators to

effectively store energy at high tide if it

coincides with low demand on the grid, for

example, between 0100 - 0600 hrs.

Table 6: Barrage specification at La Rance, France

The total output of 600 GWh per year represents about 3% of the electricity consumed in

Brittany, at a cost of £0.02/kWh: this is the same cost as nuclear generated electricity.

Figure 64: A blade and a complete bulb turbine at La Rance


92

Figure 65: Cross section of a bulb turbine unit at La Rance. Note that the turbine is

placed as low as possible to minimise cavitation.

Contact details:

La Direction De L'Usine Maremotrice de La Rance

E.D.F.

France

02 99 16 37 14


93

Figure 66: A map of La Rance barrage.

The map in XFigure 66X shows the location of the scheme at La Rance, (marked as „EDF Usine

Maremortrice‟) and illustrates why it is such a good site: the barrage length is short compared

to the area of the basin. This gives a relatively low capital cost for large potential energy. A

road runs along the barrage, giving added value. The basin had a history of ancient tidal

malls, including one at Saint Suliac (see section X4.1X).

4.3.5 Other Sites

A very large potential resource exists in the UK due to the high tidal range along the west

coast of England and Wales where there are many estuaries and inlets available.

Potential schemes range widely in size, from 30MW (e.g. Conwy) to 8.6GW (the Severn

barrage). It has been estimated that exploitation of all practicable estuaries in the UK could

lead to electricity generation of up to around 20% of demand in England and Wales. A map

and list of the most important sites is given in Figure 67.

Figure 9: A map of La Rance Barrage


94

The barrage at La Rance in France, with an installed capacity of 240MW, is the only

significant tidal barrage power scheme in Europe and is - by a considerable margin - the

largest in the world.

Figure 67: Important UK sites for tidal potential.

barrage schemes.


95

A 1991 study commissioned by the EU estimated that the technically feasible energy

resources from tidal barrages across the EU could be as much as 105TWh/year (from 64GW

of installed capacity). This resource is unevenly distributed across Europe with the UK

(47.7%) and France (42.1%) sharing the bulk of the resource and Eire (7.6%) accounting for

most of the rest.

Worldwide, other examples of barrage schemes include:

Canada: 1984 - a 17.8MW plant at Annapolis. Tidal range 6.4m, basin area 6 km2

Russia: 1968 - a 0.4MW experimental plant at Kislaya Guba. Tidal range 2.4m basin

area 2 km2

China: 1980 - the 3.2MW Jiangxia station. Tidal range 7.1m, basin area 2 km2


96

4.4 Tidal Current Schemes

Useful energy has been obtained from the flow of water currents for centuries in the form of

waterwheels in rivers and estuaries. However, the production of electricity from tidal streams

is a relatively new concept. It relies on a different approach to conventional tidal barrage

schemes, and does not have the large capital commitments that are necessary for the barrage

schemes.

Tidal stream technology extracts energy from the flow of the currents, which are produced by

the rise and fall of the tides. These currents usually have a low velocity (1ms-1

) but this can

be modified by the local topography. In particular, the velocity can be magnified greatly in

straits between islands or between islands and the mainland. The tides can be predicted with

high accuracy; hence after measurements at a site, the energy available for conversion can be

forecast with confidence.

4.4.1 Turbines for Tidal Current Schemes

The technology is conceptually simple: a rotor (tidal mill) is placed in a suitable tidal flow,

which turns the rotor and (using a gearbox) a generator. It is similar to a submerged wind

turbine, except that the greater specific gravity of seawater results in much higher energy

densities in tidal streams than is found in winds of the same velocity. However, the water

velocities available in tidal streams (typically rated velocities of 2-3m/s on good sites) are

much less than the air velocities used by wind turbines. Since power output is proportional to

the cube of the velocity, tidal stream rotors do not generally produce significantly greater

output than wind turbines of the same size.

XFigure 68X shows the prototype 1.2 MW Seagen device developed by Marine Current

Turbines Ltd. This particular prototype is located in Strangford Lough, Northern Ireland and

consists of two rotors either side of a monopile. The advantage of this device is that the

turbine can be winched out of the water for maintenance.


97

Figure 68: The 1.2MW Seagen tidal current turbine at Strangford Lough, Northern

Island (courtesy of Marine Current Turbines Ltd).

4.4.2 The Resource

The UK has some of the best sites for the exploitation of tidal currents. The major factors that

determine the size of the resource are the velocity of the tidal current and the volume through

which it flows. This resource was evaluated at several locations around the UK. The power

contained in tidal currents (P) is given by: 2

3AVP

where A is the area swept by sea water of density at a velocity V. This cube law

relationship indicates that areas of high tidal current velocities are the most suitable for

development.

Tidal stream data is provided in Admiralty Tidal Stream Atlases. A survey of these identified

the promising areas for tidal stream exploitation as:

Area 1 - Pentland Firth, north east Scotland

Area 2 - Rathlin Island, Northern Ireland

Area 3 - Mull of Galloway, south west Scotland

Area 4 - Barry Island, Bristol Channel

Area 5 - Portland Bill, southern England

Area 6 - Alderney, Channel Islands.

These locations had maximum tidal stream velocities ranging from ~6m/s in the Pentland


98

Firth to ~2.5m/s in the Mull of Galloway and Rathlin Island. There were numerous small

sites which had high tidal current velocities (e.g. around islands in the Orkneys or to the west

of Scotland) but there was insufficient data to assess these sites accurately.

The annual electrical output was evaluated for each scheme assuming the characteristics of

current designs for horizontal axis turbines and using the following approach,

the separation of turbines was assumed to be twice the diameter of the blades for a row of

devices. For an array of devices, the separation of rows was taken to be ten times the rotor

diameter in order to avoid wake interaction effects. In addition, allowance had to be made for

access of repair ships and to maintain normal shipping routes, which resulted in a density of

37 turbines per square kilometre of seabed.

The rating of turbines was assumed to be 60% of the available power in a swept area

The efficiency of the turbine varies with the ratio of the tip speed to current speed and can

have a theoretical limit of 59% (the Betz limit). The assumed efficiency for this turbine

was based on efficiency performance curves with a maximum of 40% for the best designs

for current technology - higher efficiencies could result from further R&D or use of

different turbines

The availability was taken to be 90%.

The accessible tidal stream resource for the most suitable sites in the UK (including the

Channel Islands) is estimated to be approximately 36TWh/year. The actual resource is higher

but the current velocities in the remaining areas are so low that their exploitation would be

hopelessly uneconomic.


99

4.5 Impacts of Tidal Schemes

Tidal barrage schemes have a number of potentially detrimental impacts which have to be

balanced against their overall environmental benefit of helping to avoid greenhouse gas

emissions:

They have a visual impact as they significantly change the appearance of the estuary.

The shipping lanes are obviously blocked by the barrage and so provision has to be made

for locks to facilitate navigation of the barrage by vessels. This may have an economic

impact.

They may influence the build up of silt and sediment, but they are believed to raise the

barrage water level by half a metre or so. In the case of the Severn, this is likely to

permanently flood marshland where birds live and nest (see Figure 69). An interesting

aspect for the Severn is the possibility that the resonance effect that gives rise to the large

tidal range may be reduced by the construction of the barrage.

The structure will shorten the length of the estuary and so resonance may be reduced. At

La Rance, a road was incorporated into the design and it may be possible to gain some

benefit from roads in UK schemes.

The environmental disruption caused by tidal stream devices in the shoreline/nearshore areas

(e.g. through electrical transmission lines and changes to shipping lanes) would not be so

dramatic.

Figure 69: Wading birds in the Severn Estuary.


100

5 Wave Power

5.1 Introduction Following the OPEC oil crises of the early 1970s, the UK Department of Energy invested

millions of pounds in wave energy research and development. The UK wave energy resource

is amongst the greatest in the world, thanks to the country‟s geographical position, lying as it

does at the NE comer of the Atlantic Ocean. That is where it receives wind and thus waves

generated across several thousand kilometres of the Atlantic.

Wave energy research and development has been taking place for about 25 years, with some

progress towards exploiting the large energy potential of the world ocean wave climates.

Such an environmentally benign source of energy is highly desirable in that it could

contribute significantly to a sustainable energy future. Ocean waves are often powerful, but

with extremely low frequencies, of about 0.1 Hz (equivalent to 6rpm or to periods of around

10s), and the success in generating electricity demands that this frequency is raised to 500 -

1500rpm.


101

5.2 Wave Climate

The passage of wind over the surface of the sea will result in the gradual transfer of energy

into the water to produce waves. Solar insolation (with a power density of typically less than

one kWm-2

) produces the weather patterns and hence wind (wind power densities of 1.2 to

1.8 kWm-2

). As the wind goes on to produce waves (power density 50 kW per metre of

wavefront or crestlength), we can regard wave energy as a concentrated form of solar energy.

The distance over which this process occurs is called the "fetch” and longer fetches produce

larger, more powerful waves as do stronger winds and extended periods of wind. Sea waves

are characterised by their waveheight (H), period (T) and crest length (see Figure 70). A

particular wave climate will consist of many monochromatic waves, each with its own T and

hence wavelength, velocity and power per metre of crestlength ( , V and P).

In deep water:

2

gTV ms

-1 (80)

2

2gT m (81)

In shallow water, where the water depth is somewhat less than the wavelength,

gDV ms-1

(82)

For all depths, the power density is given by:

32

22 THgP kWm

-1 (83)

Figure 70: Wave Characteristics.

Wave Period T=9s

20 10

2

Wave Height (H = 3.8m)

-2


102

This is very nearly exactly THP 2 kWm

-1

Note here that the deep water waves are dispersive, that is to say that their velocity is

dependent on the period, whereas in shallow water, the velocity is only a function of depth, D

( gDV ). In deep water, usually taken to be at least half of the wavelength, the longer

period waves will arrive at a given location before their shorter and less energetic

counterparts.

In shallow water, the dependence on depth leads to refraction of waves which accounts for

the tendency of waves to approach beaches with crests almost parallel to the seashore. This

phenomenon can be used to identify the most energetic locations for shore or floor mounted

devices by ray tracing.

In the complex pattern of real waves, H and T are replaced by significant waveheight, Hs =

2Hrms, (where Hrms=H/ 2 for a sinusoidal wave) and energy period (Te). See XFigure 71X

below.


103

Figure 71: Surface level against time for waves with different Hs

The power density is then es THP2

2

1 kW per metre of crestlength.

e.g. Hs = 3.7m Te=9.5s, then P = 65 kWm-1


104

Another important feature of waves in water is the attenuation of power density as waves run

into shallow water. They gradually lose energy because of the interaction with the seabed and

so become less powerful. Figure 72 below shows the wave power density as a function of

depth for the Atlantic west of the Hebrides.

Figure 73 below shows a typical scatter diagram taken from an offshore Atlantic location.

Each number on the chart indicates 1/1000 part of a year and gives the average Hs and Ts for

that unit of time. Ts is the so-called „mean zero crossing time‟ defined as the duration of a

number of wave height readings divided by the number of upward crossings of the mean

water level. Typically, Te 1.12Ts. From the above equation we could calculate the power

density for each interval of time in Figure 4. Although not frequent, it can be seen that wave

heights over 10m, and periods above 13s, do occur and these represent severe storms with

power densities 600-1500 kWm-1

. Wave device structures have to be designed to withstand

such storms.

Figure 72: Wave power against water depth for the Atlantic - west of the Hebrides.


105

Energy is transmitted by the wave activity but there is no net movement of water. Instead, the

water particles stay almost in one place executing orbital paths. These orbits are large near to

the surface and smaller, declining exponentially beneath the surface.

The fraction of energy between the surface and depth d is given by:

Figure 73: Significant wave height Hs against zero crossing Ts


106

dF 4exp1 (84)

Figure 74 below illustrates this orbital behaviour in a time lapse photograph of a scale model

single Edinburgh Duck. The waves are travelling from the right and a dye marker indicates

the orbital activity. The time lapse is responsible for the blurred upper surface of the water at

the wave crests and troughs. The duck is also blurred but note that the water behind the duck

is not blurred because it is still: all of the incident wave energy has been absorbed by the

duck!

This image was instrumental in convincing the UK government that wave energy was worth

funding in the 1970s. It shows the elegance of the duck concept, with a shape that matched

the orbital characteristics of the waves. Technically, it shows good impedance matching.

There are some important features to pick out here:

1. The Duck pivots about a fixed axis which is attached to the sides of the tank. In a full-

scale version, the Duck will have dimensions of tens of metres and will be deployed in

deep water where it cannot be supported in this way. To permit floating devices to work

in deep water, a number of units are usually arranged on a common spine. In this way, the

spine provides a reasonably stable frame of reference for the units to move against. The

common spine is freely floating but because it spans a number of wave crests, it does

respond significantly to small waves. Figure 75 above shows a 20 duck test on Loch

Ness. The spine is 50m long, enough to span 5 to 20 crests. The model is at 1:10 scale, so

that a full-scale spine would be 500m long. Note how calm the water is behind the spine.

2. The power take off from the Duck has to convert seemingly random, low speed, high

torque movement into steady, high speed, low torque activity to drive a generator.

3. The photograph in Figure 75 shows virtually perfect wave energy capture. This is

possible if the device intercepts all of the wave flux, and is tuned to the particular wave

frequency (or equivalently, period). As real waves are complex and consist of a spectrum

Figure 74: Model Edinburgh Duck Figure 75: 20 Duck Tested on Loch Ness


107

of wave heights and frequencies, the device will respond well to frequencies near the

tuned frequency but will have lower efficiencies at other frequencies. Optimising the

design for a selected location requires that the tuned frequency is carefully chosen by

taking note of the annual distribution of wave frequency/period.


108

5.3 Shore Mounted Technology

5.3.1 Oscillating Water Columns (OWC)

Oscillating Water Columns are the most popular first generation devices. They are usually

designed as shore mounted structures having fixed frames of reference. The OWCs operate in

response to the incoming wave activity. They are essentially resonant devices.

Figure 76: Diagram of a Typical Oscillating Water Column

An incident wave crest will increase the pressure of the water inside the column, forcing the

internal water level to rise and in turn pushing air out from the top of the column (see XFigure

76X above). An air turbine extracts energy from this air flow. The air flow reverses to refill the

column when the wave trough appears at the mouth of the device. The natural frequency of

the OWC is determined by its physical size. For example, a path length l (which couples the

wave surface to the internal water surface) of 28m will have a resonant period of about 9s

using g

lT

22 and so will respond best to waves in the period range of 8 to 10s.

A critical feature of OWC has been the development of the Wells self-rectifying air

turbine, XFigure 77X below, which avoids the need for rectifying valves. Apart from being

able to extract energy from air flows both out of and into the column (hence the self


109

rectifying characteristic), the Wells has other valuable attributes. With its symmetrical

aerofoils arranged with a zero angle of attack, the Wells has a low drag and can be driven

at high rotational speeds of several hundreds of rpm and thus can be coupled directly onto

the shaft of an electrical generator without the need for a gearbox.

Figure 77: Wells Turbine

For example, a two-pole pair machine could be driven at 1500rpm in order to generate

electricity at 50Hz. A Wells rotor operates most effectively when the pressure drop across it

corresponds to 2 or 3 m of water, which is typical of ocean wave heights. As this pressure

drop is proportional to flow, the Wells has a linear power characteristic, making it ideally

matched to wave activity.

There have been a number of prototype OWCs built around the World, including Japan,

India, Portugal and Norway. A Norwegian device, the Multi Resonant Oscillating Water

Column (MOWC), was designed and manufactured in 1985 by Kvaerner Brug. The

oscillating water column chamber is set back into a cliff face which falls vertically to a water

depth of 60m.

Setting back the column produces two harbour walls which broaden the frequency response

curve by effectively providing a range of resonant frequencies, hence the term 'multi-

resonant'. The oscillating air flow is fed through a 2m diameter Wells turbine rotating within

the speed range 1000-1500 rpm. This turbine is directly coupled to a 600kVA generator, and

the output passed through a frequency converter before being fed to the grid. Performance

exceeded predictions and provided energy at about £0.04/kWh. Two successive severe

storms in December 1988 tore the column from the cliff and to date the scheme has not been

replaced, although future designs would be much more robust.

5.3.2 The OWC device on Islay

A prototype OWC was commissioned on Islay in Scotland in 1991 by the Queens University

Belfast team. This consisted of a concrete chamber with a footprint of 4m by 9m. It was fitted


110

with a 1.2m diameter biplane Wells turbine directly coupled to a 75kW wound rotor

induction motor operating as a generator above its synchronous speed of 1500rpm. This was

the first UK OWC and was built in a natural gully on Islay (Scotland) in 1991. The

photograph (Figure 78 below) show this prototype being subjected to storm waves.

Figure 78: The prototype OWC commissioned on Islay in

Scotland


111

5.3.3 Limpet

Following the extensive work on the first Islay OWC, a second OWC, termed LIMPET, was

commissioned in November 2000 (see schematic in XFigure 79X). Rated at 0.5MW, this also

benefits from significant improvements over the original design:

Purpose excavated gully rather than using a natural gully

Three chambers each with a Wells turbine

Shaped floor of the structure to turn water flow through 90 degrees.

Figure 79: The LIMPET oscillating water column


112

5.3.4 Tapchan

This is a shore-mounted device that contains some inherent energy storage. As waves

approach the mouth of the channel, they become compressed in width and so build up in

height. "TAPCHAN" stands for Tapered Channel, and it is the design of this tapered channel

which enables the scheme to harvest energy from the ocean waves which arrive at the mouth

of the channel, see Figure 80.

The 350kW prototype built in 1985 in Norway had a mouth which was a 40m wide horn

shaped collector capable of accepting energy from waves with a range of frequencies and

directions.

Waves entering the collector are fed into the wide end of the tapered channel where they

propagate towards the narrow end with increasing wave height. The channel walls on the

prototype are 10m high (from -7m to +3m) and 170m long. Because the waves are forced into

an ever narrowing channel, their height is amplified. However, their overall energy content is

slightly diminished because the 'width' of the wave crest is reduced, until the crests spill over

the walls into the reservoir at a level of 3.5m above the mean sea level. The wave energy has

been converted into potential energy and is subsequently converted into electricity by

allowing the water in the reservoir to return to the sea via a low head hydroelectric Kaplan

turbine system where a 350kW induction generator delivers electricity into the Norwegian

grid. The Tapchan concept is both elegant and simple. With very few moving parts, it has a

low maintenance cost, and high reliability. The storage reservoir also helps to 'smooth' the

electrical output. Tapchan 'collects' waves in the reservoir and so the output from the Kaplan

turbine is dependent on the relatively steady water level in the reservoir and this is only

slightly affected by the arrival of individual waves.

Tapchan therefore has an integral storage capacity which is generally not found in other wave

energy converters.

Figure 80: Typical Tapchan Scheme


113

Indonor AS is building a 1.1MW Tapchan on the Indonesian island of Java (see XFigure 81X).

This consists of a 7m wide, 7m deep channel, which narrows down to 25cm over its 60m

length. The bay has its own natural basin with an area of 7500m2

capable of holding water

4m above sea level. The total cost is expected to be about £6M and the predicted cost of

electricity exported from the plant is £0.05/kWh. Tapchan represents an ideal wave energy

converter where a suitable site can be found, as, for example, on Java, with a natural basin

and a low tidal range.

Figure 81: Proposed Tapchan for Indonesia.

5.3.5 Locations for Shore Mounted Schemes

The Tapchan concept has many beneficial attributes; it is relatively cheap, simple, reliable

and easily maintained and has a small inherent amount of storage which facilitates the

smoothing of electrical output from the wave energy scheme.

In view of these attributes, there would be merit in deploying many Tapchan schemes but the

consideration of site requirements leads us to the constraints on available locations. The

Kaplan being significantly reduced compromises the efficiency of the scheme. High tidal

ranges also compromise the performance of OWCs. The number of possible sites for shore

mounted Tapchan and OWC around the world is thus limited by tidal ranges and shoreline

constraints, and consideration of other wave conversion technologies is therefore necessary if

we are to harness a significant proportion of the large potential resource. XFigure 82X below

shows the potential UK sites for shore-mounted schemes.


114

Figure 82: Potential UK Sites for Shore Mounted Schemes

5.4 Near Shore Technology Near shore locations in water of depths from 10 to 25m can exploit the greater power density

of the waves that are found in shallow water (see Figure 72). There are many more potential

sites and hence a greater degree of resource accessibility.

5.4.1 Osprey

OSPREY could be classed in this category. A 2MW OWC wave energy plant equipped with

four Wells turbines, Osprey I, was launched in August 1995 by Applied Research and

Technology (ART) off the north coast of Scotland at a total cost of £4M. This was intended

to be the World's first commercial scale near shore wave energy device. The plant was

prefabricated and towed to its station. Unfortunately, it was destroyed by the tail end of


115

hurricane Felix before it could be secured at its chosen location. A replacement incorporating

a 1.5MW wind turbine on the top is planned.

5.4.2 The 'Mighty Whale'

Figure 83: The 'Mighty Whale'

The Mighty Whale (see XFigure 83X), a Japanese floating OWC 50m long, 30m wide, with

three chambers fitted with air turbines rated at 50,30 and 30 kW and has been deployed 1.5

km from the mouth of Gokasho Bay in 40m of water.

5.4.3 FWPV

A floating Swedish version of TAPCHAN has already been tested as a prototype in the form

of a floating vessel FWPV - Floating Wave Power Vessel. It is equipped with a sloping front

ramp which collects the waves and stores the water temporarily on board before returning it

to the sea via a turbine. A larger model will be deployed in Scottish waters under the Scottish

Renewables Obligation (SRO).

5.4.4 Point Absorber Float Systems

Float systems, with a body floating on the surface but slack or tight moored to the sea bed via

a pump, are attracting some attention. These can act as point absorbers which draw in energy

from a greater width of water than their own physical diameter (i.e. have capture width

ratios greater than one). Capture width is the ratio of effective width to actual device width,

and is often used in place of efficiency.


116

The Hose pump wave energy converter (see Figure 84) has been developed over 15 years by

Technocean in Sweden and is intended to pump sea water from an array of hose pumps fixed

to the sea bed. The hoses consist of rubber tubes reinforced with spirally wound cording. By

carefully choosing the spiral angle, it is possible to develop significant volume

reduction/pressure increase by stretching the tube along its axis. The reinforcing cords retain

their length by reducing the diameter of the tube, and hence the contained volume. A Pelton

wheel extracts energy from the water as it is released from an upper reservoir back to sea. A

hose-pump lightbuoy is undergoing pre-production tests, and evaluation of wave power

plants in Ireland, Spain, Sweden and the USA have been carried out. The Swedish

Government has halted research funding because it does not envisage wave energy as a major

contributor to Sweden's energy system despite the low cost/kWh predicted for such schemes.

The potential along the Swedish coast is about 5-10TWh/year (at an average power of 0.57-

1.1GW, representing 3-7% of demand) but the potential along the Norwegian coast is

estimated at around 3.0-3.5GW, which could contribute 12-15% of Sweden's electricity

demand via the Nordic grid.

A five phase project to build a semi-commercial hybrid wave/hydro power plant at Amorgos

in Greece is in preparation and the subject of bids for funding to the EU.

The first phases would have a mean rating of 6MW from an average wave climate of 8-

9kW/m, with the ultimate goal of a 50-100MW plant.

A small lkW float scheme was installed in 25m of water 2.5km offshore of Hanstolm,

Denmark in January 1996. The converter demonstrated a reliable performance over a nine-

month period, surviving a storm with waves of 9.6m. A 300MW scheme in the Danish North

Sea is estimated to produce electricity at a cost of £0.11-0.18/kWh, with possible reduction to

£0.08-0.13/kWh, from a climate averaging 15 to 23kW/m. Transmission over 100km would

represent approximately 20% of the cost.

The Interproject Service (IPS) converter was developed in the early 1980s and tested at 1:10

scale in a lake and at full scale prototype at sea. It consists of a tall buoy with a tube open at

both ends attached underneath. A piston inside the tube is linked to the buoy and power is

Figure 84: Hose Pump Wave Energy Converter


117

extracted by the interaction of the buoy and the water in the tube. A new configuration, the

sloped IFS, embodies a number of attractive features, and offers reliable power for small

electricity networks on small islands.

5.4.5 Pendulor A shore-mounted concept, the Pendulor, has been tested over 15 years in Japan. A concrete

chamber with a length corresponding to a quarter of a typical wavelength is open at the sea

end and is fitted with a hinged plate.

The wave activity in a normal sea state thus stimulates maximum or resonant response and

causes the plate to execute maximum amplitude oscillations which are damped by the

hydraulic power take-off mechanism. Costs are estimated at £0.07/kWh. Figure 85 shows a

proposed four chamber, 250kW, Pendulor planned for the southern coast of Sri Lanka.

5.5 Offshore Technology

5.5.1 The Duck The Edinburgh Duck was described earlier (see section X5.2X); it may ultimately be capable of

delivering electricity at around £0.10/kWh.

5.5.2 The Clam The SEA-Coventry Clam (see Figure 86) is an 80m diameter rigid toroid, rated at 3MW,

delivering an average of 1MW, and floating in deep water. Twelve air cells are arranged

around the circumference of the toroid and these cells are all coupled together by an air

ducting which contains twelve Wells turbines. Thus the air forced from one cell will pass

Figure 85: Pendulor


118

through at least one turbine on route to other cells.

Each cell is sealed against the water by a flexible rubber membrane. Performance

measurements, together with mathematical modelling, full scale component testing and

outline design and costing, led to a team estimated cost of delivered electricity of about

£0.05/kWh. A full scale experimental prototype is needed to prove the concept, to refine the

design and optimise costs. A major consideration for all design teams is then to obtain the

best performance from a range of wave heights and periods, and to design structures to

survive the worst storms.

Arrays of many floating devices in water 50m or so deep will be a few kilometres offshore

and will constitute wave farms, capable of delivering MW or GW to the main land via subsea

cables. As they will be moored to the sea bed, they will be able to swing to face the principal

wave direction in order to maximise the total energy capture in a particular sea state. Both

device teams have made considerable progress in design and technology, further

developments of the Clam and Duck are necessary before the wave farm concept can be

realised.

Large-scale offshore devices will be deployed in deep water to exploit the higher wave power

density to be found there. Since these systems will have to float, they are designed to be large

enough to span several wave crests. In this way, the wave forces on the structure are averaged

at an acceptably low value, permitting the structure to survive and moreover to provide a

reasonable stable frame of reference. Arrays of such devices could form wave farms and

transmit substantial amounts of electricity to the Irish, UK and Norwegian grids, with the

long term prospect of feeding into a European supergrid. The collective output from an array

will be smoother than individual devices. Offshore devices would transmit electricity via

subsea cables. Modular construction methods should keep permitted the minimisation of

capital costs and ensure rapid deployment and early operation.

5.5.3 Pelamis

Pelamis Wave Power Ltd have developed the Pelamis wave power generation device which

is based on a spin off of the duck design, exploiting the hinges used to connect the ducks

together (see Figure 75). XFigure 87X shows a picture of a Pelamis device. The device consists

of four large floating sections connected by hinges. As the sections move with the waves, the

movement is resisted by hydraulic rams which pump hydraulic fluid through hydraulic

Figure 86: The Clam


119

turbines via smoothing accumulators. These turbines are connected to generators which

produce power which is fed via a cable to a point on the sea bed which is in turn connected to

a transmission line which runs to shore. Each device can produce 750kW and a farm

consisting of the three of these devices has been commissioned off the northern Portuguese

coast at Aguçadoura (see XFigure 88X).

Figure 87: The Pelamis wave power generator (courtesy of Pelamis Wave Power Ltd).

Figure 88: The three Pelamis device wave farm near Aguçadoura, northern Portugal

(courtesy of Pelamis Wave Power Ltd).


120

6 Solar Power

6.1 Solar Characteristics

6.1.1 The Solar Spectrum

The energy of an individual photon corresponds to the wavelength of the photon which is

related to the frequency . The shorter the photon wavelength, the higher is its energy. This

relation is described by:

chhE (85)

where h is Planck´s constant (6.625 10-34

Js) and c the speed of light (3 108m/s).

Solar radiation consists of photons of differing energy. The distribution of the photons

according to their wavelength or energy is called a spectrum; the height of the curve indicates

the relative contribution by photons of a given wavelength. In passing the earth’s atmosphere

solar radiations interacts with atmospheric particles (gas, molecules, dust etc.) so that the

spectrum as well as the energy of radiation reaching earths surface differs from that outside

atmosphere. XFigure 89X shows the spectrum of solar radiation outside atmosphere and on

earth´s surface. The differences between the two curves are due to the interaction of the

radiation with the atmosphere. The entire energy Etot contained in radiation can be obtained

by adding up the contributions of the individual wave ranges of the spectrum:

Figure 89: Spectral irradiance outside atmosphere and on earth's surface


121

dEEtot0

(86)

XFigure 90X shows how colours in the visible part of the spectrum relate to wavelength.

The energy range beyond the visible part of the solar spectrum is called ultraviolet (UV).

Radiation with photons of lower energy, is called infrared radiation (IR). XTable 7X gives the

relative proportions of the ultraviolet, visible, and infrared ranges of the solar spectrum.

Table 7: Contributions of different spectral ranges to the total energy of the solar spectrum

6.1.2 The interaction between radiation and matter

When a photon interacts with the atoms of a material, either the photon transfers its energy

completely to the atom or it does not transfer any energy at all. If the energy is transferred,

the photon ceases to exist. There are various forms of interaction (see XFigure 91X).

Figure 90: Colours in the visible range of the solar spectrum

Spectral range wavelength portion [%]

ultraviolet < 380 nm 8

visible 380 780 nm 46

infrared > 780 nm 46

absorption

transmission

reflection

scattering

annihilated

conserved

photon interaction

Figure 91: Types of interaction between radiation and matter


122

The reflection of radiation at a surface of the material can be ”specular” or ”diffuse”. When

the angle of incidence is equal to the angle of reflection, the reflection is called specular,

whereas when the reflected radiation is uniformly distributed in all directions, it is called

diffuse.

When radiation strikes a body, a part of it is reflected, a part is absorbed and, if the material is

transparent, a part is transmitted. The fraction of the incident radiation reflected is defined as

the reflectance , the fraction absorbed as the absorptance , and the fraction transmitted as

the transmittance . In general, the proportions depend on the wavelength and the incidence

angle . These three components must sum to unity:

+ + = 1 (87)

6.1.3 The mechanism of absorption

The energy transmission from a photon to the absorbing matter can happen in two ways,

which have different effects on the condition of the matter and which underlie the thermal or

non-thermal utilisation of the radiation. These are illustrated in XFigure 92X.

On the one hand, the energy of the photon can lead to an increase in the kinetic energy of the

atoms or molecules, which becomes evident through an increase in temperature at

macroscopic level. In the second process, the internal energy of the atom changes in that an

electron assumes a higher energy level called an excited state through energy transmission, or

may even no longer be bound to the atom, so that it exists, for example, in a solid as a free

charge carrier. This process changes the chemical or electrical properties of a material and

forms the basis of the non-thermal utilisation of radiation, e.g. in the generation of electrical

energy with solar cells or in the conversion into chemical energy through photosynthesis in

plants.

6.1.4 Blackbody Radiation

Materials which absorb all radiation are described as black bodies or blackbody. By

definition a blackbody is a perfect absorber and emitter of radiation.

Figure 92: Atomic mechanism responsible for the absorption of photons


123

A remarkable property of the blackbody radiator is that the spectrum of its radiation, and thus

the energy contained in the radiation, depends on its temperature alone. The dependence of

the spectral irradiance E that is the energy at a certain wavelength radiated by a black body

having absolute temperature T, is described by Planck´s equation:

1

12

/5

2

kThcehcE (88)

where k is Boltzmann‟s constant (1.38 10-23 J/K).

XFigure 93X shows the spectrum of a black body at different temperatures. The higher the

temperature, the greater the energy contained in the radiation field; this energy is given by the

Stefan-Boltzmann law, obtained by integration over all wavelengths:

4TE (89)

where is the Stefan-Boltzmann constant (5.67 108Wm

2K

4).

A further characteristic is that the maximum of the spectrum (obtained by differentiating

Planck´s equation and equating to zero) shifts towards shorter wavelengths as temperature

rises.

mT

2898max (90)

6.1.5 Radiation on earth´s surface

The amount (and the spectrum) of radiation incident at a certain location and its daily and

seasonal variation depend on several factors. The geometry of sun and earth and the earth‟s

orbit around the sun cause the seasonal differences of insolation, the length of day and the

highest position of the sun above horizon. The rotation of the earth around its axis results in

Figure 93: Spectral irradiance of a black body at various temperatures


124

the daily course of the sun. The extraterrestrial radiation (radiation measured outside

atmosphere) is more or less constant. Passing through the atmosphere, the radiation interacts

with components of the atmosphere. The major factors are scattering from molecules and dust

particles, absorption by the atmosphere, and refraction. The resulting intensity and the

spectrum is strongly influenced by the local climate and weather conditions.

6.1.5.1 Terms and Quantities

That portion of the incident solar radiation which comes directly from the apparent solar disc,

without without having been scattered or reflected, is called direct or beam radiation. Diffuse

radiation is the result of scattering by the atmosphere. The sum of the direct and the diffuse

solar radiation on a surface is called the total solar radiation. The term global radiation is

used for the total solar radiation on a horizontal plane on the earth´s surface.

The instantaneous radiation intensity incident on a surface per unit area is called irradiance

and is measured in W/m2. The irradiation or, as usually used with solar energy, insolation is

the incident energy per unit area, found by integration of irradiance over a specified time,

usually an hour or a day. XTable 8X gives the symbols used for various quantities.

6.1.5.2 Extraterrestrial Radiation

The characteristics of the radiation reaching the earth are determined by the physical

processes occurring in the sun and the distance from the earth. The sun is a sphere of

intensively hot gaseous matter with a diameter of 1.39 106 km. The energy produced in the

interior of the solar sphere is transferred out to the surface and then radiated into space. While

the temperatures in the centre reaches several million degrees, the surface of the sun has a

temperature of about 5780 K.

Because of the thermal origin of solar radiation, the sun can be approximated as a blackbody

at 5780 K.

Table 8: Symbols for various irradiation and insolation quantities

instantaneo

us

(irradiance)

(W/m2)

hourly

average

(irradiation

(W/m2)

daily sum

(kWh/m2

)

monthly

average

of daily sums

(kWh/m2)

radiation outside the

atmosphere

(extraterrestrial)

referred to a horizontal

surface

Go

Io

Ho

OH

global radiation

(horizontal) on the

earth's surface

G I H H

direct beam radiation on

a horizontal surface

Gb Ib Hb bH

diffuse radiation on a

horizontal surface

Gd Id Hd dH


125

The radiation reaching the earth´s atmosphere depends on the diameter of the sun, the

distance between the earth and the sun and the sun’s total emission of radiation. The last

quantity varies slightly and changes with the occurrence of sun spots in cycles of roughly 9 to

11 years but in general, these small variations are not relevant for the utilisation of solar

energy.

As the earth rotates round the sun in a elliptical orbit, the distance r between sun and earth is

changing rather periodically. The deviation from the average distance r0 of 1.496 108km is

described by the eccentrity E0

2

00

r

rE (91)

whose variation in the course of one year is approximated by

365

360cos033.010

nE (92)

where n = day number of the year {1,...,365}. The extraterrestrial radiation Gon measured on

the plane normal to the radiation varies around 3% as:

scon GEG 0 (93)

In December/January the distance is at its minimum and therefore the extraterrestrial

radiation reaches its maximum of 1412 W/m2 while in June/July the earth is at the point of

maximum distance to the sun and a minimum of 1322 W/m2 may be received from the sun

outside the earth‟s atmosphere ( XFigure 95X).

The average intensity Gsc of extraterrestrial solar radiation is called the ”solar constant”.

Attempts to determine the value of the solar constant have been made regularly since 1900.

The actual value based on satellite measurements is 1367 W/m2 ( 4 W/m

2 measurement

uncertainty). The variation of Gsc due to variable sun activity (sunspots etc.) is less than 1%.

Figure 95: Monthly variation in extraterrestial radiation

6.1.5.3 The influence of the atmosphere

TFigureT 94: Annual variation of the normal extraterrestrial solar irradiance Gon [Duffie 91]


126

The interaction between the radiation and the atmosphere leads to changes in the quantity and

quality of the radiation. The absorption and scattering processes vary with wavelength. Parts

of the ultraviolet radiation are absorbed in the stratosphere, the outer layer of the atmosphere.

In the upper troposphere, water molecules and carbon dioxide absorb parts of the infrared

radiation. Aerosols lead to the scattering of visible light, mainly in the blue visible part of the

solar spectrum. This diffuse radiation can be seen on sunny days as the blue sky. In the lower

and upper troposphere, clouds reflect part of the radiation back into space. The remainder is

transmitted or reflected down to the earth‟s surface.

In addition to these processes, the solar spectrum at the earth‟s surface is determined by the

amount of air that the radiation has passed through. With lower sun positions, the direct

fraction decreases because more and more photons may be absorbed or scattered away from

their direct path. The distance that radiation covers while passing the atmosphere is measured

in the unit of air mass (AM).

The conditions outside atmosphere correspond to air mass zero (AM0). Solar radiation that

has passed the atmosphere vertically has an AM1 spectrum. AM 1.5 corresponds to a sun

shining from an elevation angle of 42° above the horizon and is internationally chosen as a

reference spectrum for solar cell characterisation. XFigure 96X shows the spectral distribution at

various values of the air mass.

6.2 Solar Resource

6.2.1 Geometry of the Sun, the Earth and the Collector Plane

A general task with the utilisation of solar energy is the calculation of the available radiation

falling on a given receiver plane. The calculations of the direct and the diffuse fraction must

be carried out separately, as beam radiation comes from a specific direction, changing with

time of day and season, whereas the diffuse radiation is incident more of less uniformly from

all directions. XTable 9X illustrates the variability in global radiation for different weather

Figure 96: Spectra of solar radiation for various air mass values, definition of air mass


127

conditions and what proportion of the global radiation is diffuse.

The following terms and quantities are used to determine the sun‟s position in the sky and the

position of the receiver plane ( XFigure 97X):

the location of the receiver on earth‟s surface is given by the geographical latitude

(north positive, south negative) and the geographical longitude (0 Greenwich)

the declination is the angular position of the sun at solar noon with respect to the plane

of the equator, i.e. the zenith angle at solar noon at the equator (north positive, south

negative). As the earth‟s axis of rotation is not parallel to the axis of the earth‟s orbit

around the sun, but tilted by 23.45 , varies in the course of the year between 23.45 and

+23.45 .

365

284360sin45.23

n

the hour angle is the angular displacement of the sun east or west of the local meridian

due to rotation of the earth around its axis. It is equal zero, when the sun reaches its

highest position above the horizon; so, is a measure of the time of day. A displacement

of 15 corresponds to a period of one hour (negative in the morning, positive in the

afternoon).

18025.0 sT

where time is given in minutes (0 < TS < 1440). The true solar time Ts is obtained from the

standard zone time Tzone as

tloczonezones ETT 4

where zone is the longitude of the relevant time zone. The additional correction Et (time

equation) considers that the speed of the earth on its elliptical orbit is not constant:

364

81360with )sin(5.1)cos(53.7)2(sin87.9

nBBBBEt

the zenith is the direction in the sky vertically above

the zenith angle z is the angle between the zenith and the sun vector equal to the

direction of the beam radiation

the elevation angle (90°- z) is the angle between the earth-sun vector and the horizon

the tilt angle is the angle between the horizontal and the receiver plane respectively the

angle between the surface normal to the receiver plane and the zenith ( =0° for horizontal

planes, =90° for vertical planes)

Table 9: Radiation intensity for various weather conditions

weather clear blue sky hazy/cloudy,

sun visible as

whitish yellow

disk

overcast sky,

dull day

global radiation 600 1000

W/m2

200 400

W/m2

50 150 W/m2

diffuse fraction 10 20% 20 80% 80 100%


128

the azimuth angle describes the orientation of tilted planes (normally South=0°, East

negative, West positive)

solar azimuth angle s is the angular displacement from south of the projection of beam

radiation on the horizontal plane

the incidence angle is the angle between the normal of the receiver plane and the direct

radiation.

the basis for computations of the radiation incident on a receiver is the calculation of the

incidence angle that describes the position of the sun relative to the receiver plane. The

position of the sun at a certain time, and so the incidence angle, depends on the day within the

course of the year, the time of day and the location of the receiver‟s site on earth‟s surface.

The zenith angle z is expressed by the following relation:

coscoscossinsincosz

and the incidence angle:

cos = (sin Ф cos - cos Ф sin cos ) sin

+ (cos Ф cos + sin Ф sin cos ) cos cos

+ cos sin sin sin

Some special cases frequently occurring lead to simplified expressions:

With horizontal surfaces ( = 0 ) the incidence angle equals the zenith angle:

cos = cos z

South inclined surfaces with tilt angle equal to geographical latitude

( =Ф, = 0 ):

cos = cos cos

6.2.2 Variation of Extraterrestrial Radiation with Season and Latitude The extraterrestrial radiation Got on an arbitrarily oriented surface is given by


129

z

oonot GGGcos

coscos (94)

For the radiation falling on a horizontal surface, this equation reduces to:

zono GG cos (95)

or

coscoscossinsin365

360cos033.01

nGG sco (96)

The highest annual average of extraterrestrial solar radiation occurs at the equator. Seasonal

changes at the equator are only small because the day length is almost the same throughout

the year, only the angle of incidence varies from season to season. At the poles, the daily

extraterrestrial solar input varies between winter and summer between zero and more than 12

kWh/m2 which is the highest value of daily extraterrestrial irradiation. At mid-latitudes

(between the polar circles and the tropics) this effect is less pronounced. However, strong

seasonal variations complicate the effective utilisation of solar energy.

6.2.3 Estimation of the Global Radiation on Tilted Planes

The radiation incident on an arbitrarily orientated receiver consists of the direct radiation Gbt,

the diffuse radiation from the sky Gdt and another diffuse fraction Grt reflected by the ground

(albedo). In order to estimate the radiation on tilted planes (Gt), information about these three

different components is required.

Figure 97: Geometry of the sun, the earth and the collector plane


130

rtdtbtt GGGG (97)

The direct radiation incident on the receiver Gbt is calculated as

z

bbnbt GGGcos

coscos , (98)

where Gbn is the normal beam radiation intensity.

The fraction of the diffuse radiation Gdt reaching the tilted surface from the sky can be

estimated as

2

cos1ddt GG (99)

if the diffuse fraction is considered to illuminate the receiver homogeneously from all

directions of the sky dome This equation is based in an isotropic model of the diffuse fraction

which assumes the diffuse radiation to be equally intensive from all parts of the sky. In

reality, however, the area of a small disk of approximately 10° radius around the sun is much

brighter than the average sky (circumsolar radiation) and the horizon also supplies a higher

than average intensity (horizon brightening).

The third fraction of solar energy reaching a tilted plane (Grt) is mainly determined by the

ground reflectance (albedo). Grt may be expressed as

2

cos1GGrt (100)

where the ground is assumed to be homogenous and diffusely reflecting.

Considering all three components the global radiation intensity on tilted planes is expressed

as

2

cos1

2

cos1

cos

cosGGGG

GGGG

d

z

d

rtdtbtt

(101)

In general, an orientation towards south is favourable (in the northern hemisphere), as then

the radiation can be received equally well during the morning and the afternoon. Tilting the

receiver has different effects on the collecting of direct and diffuse radiation. The usable

diffuse fraction is smaller for a tilted receiver than for a horizontal one, as it then „sees‟ only

part of the sky; the larger the tilt angle, the smaller is the diffuse fraction ( XFigure 98X).

Figure 98: Diffuse radiation on a tilted receiver


131

Optimum use of direct radiation is only achieved when the receiver surface is always

perpendicular to the incident radiation. The more oblique the incident angle, the smaller is the

amount of useful energy. At higher latitudes, as the sun is low in the sky in winter even at

noon, a large tilt angle is advantageous, whereas in summer, a smaller tilt angle is better.

XFigure 99X shows a receiver tilted for optimum capture during winter and what this translates

to in summer.

Effect of tilting the receiver plane

4.00

4.50

5.00

5.50

6.00

6.50

7.00

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

sola

r in

sola

tion [kW

h/m

²day]

global 25° tilt tracking

Figure 100: Radiation gains due to tracking

If for constructional reasons the receiver must face eastward or westward, then in this case it

should be tilted less than it were directed south.

Figure 99: Direct radiation on a tilted receiver.


132

Even higher amounts of solar energy may be gained if the receiver plane is not installed at a

fixed angle but mounted such that it always faces towards the sun. Tracking of the receiver

enables one to receive 20% to 50% more energy compared with a fixed installation. In most

cases, however, it is cheaper to increase the converter size by this amount than to install the

whole system on a tracking device. In XFigure 100X the long-term monthly averages (at San

Juan) for global radiation on horizontal plane, fixed plane at 25° tilt and a plane which always

faces the sun normally (tracking) are given.

6.2.4 Devices for Measuring Global Radiation

The basic device for measuring global radiation is called a pyranometer. A typical

pyranometer (see XFigure 101X) is a thermopile where the difference between the heating of a

black surface (which absorbs most radiation) and a white surface (which reflects most

radiation) produces an output in mV which when calibrated will give a reading of the global

radiation. Both surfaces are normally shielded from wind and temperature The device is

normally mounted on a level surface free from obstructions to diffuse and direct radiation. It

should also be mounted to avoid any significant reflections which are not representative of

the surroundings.

Figure 101: A Middleton EP08 pyranometer.

6.3 Basic principles of Photovoltaic (PV) Cells

The photovoltaic effect was first observed in 1839 with developments resulting in the current

crystalline silicon cells starting in the 1950's. More recently there have been developments in

amorphous silicon technology, and other thin films which hold the promise of perhaps the

cheapest cells in the future.


133

The high performance end of the market is currently dominated by the crystalline silicon cell

and so these notes will concentrate on this particular device.

6.3.1 The crystalline silicon cell

A typical n-on-p silicon solar cell ( XFigure 102X) is made from a wafer of pure silicon crystal

doped with a minute quantity of boron to make it what is call "p-type". Phosphorus is

diffused into the front surface to form a highly-doped n-layer, with the p-n junction a fraction

of a micron below the surface. The negative contact is a fine metallic grid on the front, while

the positive contact usually covers the whole of the back. An antireflective coating is applied

to the front surface. Most modern cells are 10cm x 10cm square but they can be longer or

smaller and of any shape.

Figure 102: Section through a typical crystalline silicon solar cell


134

6.3.2 The basics of cell operation

The principle is similar to the photo-electric effect in that photons incident on the cell

material excite electrons thereby dislodging them from the valence band to the conduction

band. The subtlety of operation of the semiconductor cells such as the crystalline silicon cell

relate to the fact that the photon energies are often less than the energy gap, or bandgap,

between the valence and conducting bands of the natural material (1.12 eV for silicon).

Crystal lattice structures are attractive as the bandgaps tend to be low and the electrons and

holes can move fairly freely. (Note that 1 eV is the energy of a photon of wavelength

approximately 1.24 m).

The cell consists of a junction of n-type and p-type silicon, ie a typical semi-conductor

junction. The doping of the n-type silicon is with phosphorus, which having five valence

electrons compared to silicon's four creates and extra electron to appear in the conduction

band. The p-type silicon is doped with the acceptor boron (three valence electrons), creating

a deficit of one electron (hole) per atom. XFigure 103X show the effects of doping.

n-type semiconductor with phosphorus atoms

p-type semiconductor with boron atoms

Figure 103: the Effect of doping silicon with phosphorous and boron

Silicon Atom

Phosphorus Atom

Boron Atom

Moving Electron

Key

Electron


135

The concentrations of electrons, n, and holes, p, depend on temperature and are given by:

(102)

where n = p = ni and Eg is the bandgap energy and the Boltzmann constant, k = 1.38x10-23

JK-

1. Nc and Nv are the density of states in the conduction and valance bands respectively and ni

is known as the intrinsic carrier density.

These can be written in terms of the Fermi energy, EF, as:

n =Nc exp-(Ec-EF/kT) (103)

and

p = Nv exp-(EF-Ev/kT) . (104)

Note that for n-type material the Fermi level (the average energy of the charge carriers) is

very close to the conduction band, whilst for a p-type material it is close to the valence band.

Figure 3 shows what happens to the energy levels at the junction.

In practice, a shallow junction is formed by diffusing phosphorus onto a boron doped silicon

base. At the junction, conducting electrons from the n-region diffuse into the p-region where

they combine with the holes at the acceptor atoms to give the impurity atom an overall

negative charge. Correspondingly holes migrate from the p-region to the n-region creating a

layer of positively charged impurity atoms. The result is the creation of a reverse electric

field at the junction (positive on the n-side and negative on the p-side. The area in which this

field is established is known as the depletion area or depletion zone (since no free electrons

and holes exist there) or the barrier layer (because of the reverse field) as shown in XFigure

104X. It is the existence of this build-in field which helps the electrons to reach the conduction

band when excited by a photon. Each electron liberated to the conduction band leaves a

corresponding hole so the photons are said to create electron-hole pairs. The more energetic

photons (shortest wavelengths) are absorbed close to the surface of the cell, longer

wavelength photons penetrate further, most being absorbed within 10-4

m.

np = ni2 = NcNvexp(-Eg/kT)


136

Figure 104: Charge depletion and band bending in a p-n junction.

+

+

-

-

-

p-type n-type

Fermi Level

Conduction Band

Valence Band

+


137

The built-in voltage is given approximately in terms of the doping concentrations NA and ND:

qVbi = kT ln(NAND/ni2) (105)

If an external bias voltage is applied to the cell, this will act to reduce or increase the built in

potential/field, and current will flow as a result.

e-

-ve

+ve e+

Forward bias

e-

+ve

-ve

e+

Reverse bias

In the n-region, just below the surface, the average energy of the charge carriers (the Fermi

level) is near the top of the bandgap where many electrons are in the conduction band and

only a few hole in the valence band. Conversely, in the p-region, the Fermi level is near the

bottom of the bandgap.

The electrons and holes diffuse through the crystal, driven to produce an even distribution.

Some end up recombining after a very short period of time (of the order of 1 ms). This

neutralisation results in heat generation. Others manage to reach the junction, where they are

separated by the reverse field, the electrons being accelerated towards the negative contact,

and the holes towards the positive contact. When the cell is connected across a load, the

electrons will be forced from the negative through the load (conductor) to the positive contact

where they recombine with the holes there.

Detailed analysis shows the current to follow the diode equation:

n type

qVbi p type

W

EF


138

Idark = Io [exp(qV/kT)-1] (106)

where Io is the reverse bias saturation current. There is a strong rectifying potential due to the

built-in potential. However, under sufficiently high reverse bias, junction breakdown will

occur. The photo-current works in the opposite direction to the dark current.

I forward bias

V

reverse bias

6.3.3 Cell spectral response

Not all photons have sufficient energy to liberate electrons to the conduction band. The limit

on wavelengths is given by

max = hc/Eg = 1.24/Eg (eV) m (107)

Semiconductors are transparent to radiation with > max. Losses associated with these

photons are called non-absorption losses. The absorption coefficient ( ) is clearly very

dependent on wavelength. Even for wavelengths of sufficient energy, the coefficient is finite

and some photons will pass right through the device without being absorbed. The losses

associated with these are called transmission losses. So far we have considered the radiation

that has penetrated the surface of the cell, but a certain proportion of the incident radiation

will be reflected at the surface. These losses are called reflection losses. The optical

absorption coefficient of the device takes into account these three sources of loss. Good

design will minimise reflection losses through use of anti-reflection surfaces finishes and

minimise transmission losses by providing sufficient depth of material and sometimes by the

use of photo trapping techniques. For a particular junction, non-absorption losses are fixed.

Not all liberated electrons contribute to useful current generation. This is because, even if no

external voltage exists across the devise, some of the charge carries are created at a distance

from the junction (where they can be separated by the built-in field) and so must travel

towards the junction before they can be separated. As already mentioned, some of them will

re-combine before they reach the junction. The performance of a cell in this regard is given

by its quantum efficiency. This is the average number of electrons extracted from a cell for

each incident photon under short circuit conditions. The terms internal and external quantum

efficiency are distinguished by whether the radiation intensity already takes into account


139

reflection losses or not. Quantum efficiency represents the combined effect of photo-

generation and the collection of carriers.

Overall it is clear that the generated current is made up of contributions produced by photons

of various wavelengths. If the current generated at a particular wavelength by a unit of

irradiance is plotted as a function of wavelength, the resulting curve is called the absolute

spectral responsivity or, more commonly, the spectral response of the cell. In terms of the

quantum efficiency this is:

s( ) = q /hc x Quantum Efficiency

The current generated by a solar cell in radiation of known spectral composition can be

computed thus:-

where: IG is the generated current density (A.m-2

)

s( ) is the absolute spectral response at wavelength (A.W-1

)

E( ) is the absolute spectral irradiance of the incident radiation at

wavelength (W.m-2

. m-1

).

As both the spectral response and the solar spectral irradiance curves are sharply peaked, a

slight change in either can significantly affect the generated current. A typical spectral

response for silicon is shown in XFigure 105X.

Figure 105: Absolute spectral response in Amps per Watt

6.3.4 Equivalent circuit for cell

The solar cell can be depicted as a constant current generator shunted by the p-n junction

(XFigure 106X), which, as we have seen, acts like a positively-biased diode. RS is the lumped

).d).E( s( = I G (109)


140

internal series resistance and capacitance can usually be neglected.

Figure 106: Cell equivalent circuit

The load current IL is the difference between the generated current (photo-current) IG and the

junction (dark) current IJ:

where: IO is the dark reverse saturation current of the diode, proportional to exp(-

Eg/kT)

q is the charge on an electron (1.6 x 10-19

C)

k is the Boltzmann Constant (1.38 x 10-23

J.K-1

)

T is the absolute temperature of the cell (K)

A is the constant between 1 and 2 (varies with type of cell)

V is the terminal voltage (V).

0

0.5

1

1.5

2

2.5

3

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Voltage (Volts)

Cu

rren

t (A

mp

eres

)

Figure 107: IV characteristic of a typical cell

I - I = I JGL

1 -)R . I + (VAkT

q I - I = SLOG exp


141

This gives the characteristics I-V (current against voltage) curve for a PV cell as shown in

XFigure 107X. In the short circuit condition, when there is no voltage across the terminals of the

cell, IJ is very small, so practically all of the generated current passes through the external

link. The short-circuit current ISC is therefore a useful measure of the generated current.

From the expression for IL, we can deduce the following expression for the open-circuit

voltage

VOC at IL = 0:

6.3.5 Maximum Power Point

XFigure 108X hows the current-voltage characteristic of a typical modern crystalline silicon

solar cell under Standard Test Conditions (STC), which are:-

Irradiance 1000 W.m-2

Spectral The reference distribution for AM1.5 total sun-light, as defined

in

distribution IEC 904-3.

Cell temp. 25 2 C.

The relationship with IG and IJ is indicated by broken lines. Points A and B indicate ISC and

VOC respectively.

Maximum power PMAX is represented by the area of the largest rectangle that can be fitted

under the curve. In this case, it is 1.2 W at 0.48 V. The power output at the rated voltage,

which is usually at or near the maximum power point, is commonly referred to as the rated or

"peak" power. It is expressed in "peak" watts (Wp).

In the case illustrated in XFigure 108X, the conversion efficiency at STC, based on the total

surface area of the cell (including front grid and contact) is 12.0%. Most commercial

crystalline silicon cells lie between 12 and 18% at present.

1) + I/I( . q

AkT = V OGOC ln (110)

100% x 2)m(W. Irradiance

2)m(W. P = efficiency Conversion

-

-MAX


142

0

0.5

1

1.5

2

2.5

3

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Voltage (Volts)

Cu

rren

t (A

mp

eres

)

A

B

Figure 108: I-V characteristics of a typical silicon PV cell under standard test

conditions (10cm by 10cm square) showing maximum power point.

For modules (encapsulated assemblies of solar cells), conversion efficiencies are lower (10 -

17%), because of the non-productive areas between and around the cells.

Most modern crystalline silicon cells have fill factors exceeding 0.72. In this case, it is

2.5 x 0.48/(2.75 x 0.6) = 0.73.

6.3.6 Effects of changes in irradiance and temperature

Changes of irradiance (XFigure 109X), such as one would experience in the course of a day,

affect the short-circuit current proportionally but have little effect on VOC, because of the

logarithmic relationship.

In concentrated sunlight, ISC remains proportional to the irradiance up to extremely high

values, provided the temperature is controlled at a constant value. As the VOC also increases,

albeit only slightly, one would expect the conversion efficiency to rise as the concentration

ratio is increased. So it does, up to a point, but the effect of series resistance progressively

reduces the fill factor, offsetting the gain in efficiency and limiting the improvement that can

be achieved. In practice it has been found that concentration ratios (by using a lens, for

V . I

P =factor Fill

OCSC

MAX

maximum power point point

Lecture Notes for Course ELC003 – Renewable Energy Sources

143

example) of about 20 are optimal from the point of view of efficiency.

Crystalline silicon cells respond very rapidly to sudden changes in irradiance, their time

constant being about 20 μs.

Figure 109: The effect of temperature and irradiance

To understand the effect of temperature we can for simplicity assume the constant A in the

equation for the cell current to be unity. This then gives VOC as:

The, photo-current IG increases slightly with temperature, partly because of increased

diffusion lengths for the minority charge carriers, and partly because of a small narrowing of

the bandgap which allows photons of lower energy to be absorbed. However these effects are

small and can be initially ignored.

This leaves two effects, the explicit dependence on T, and also the fact that IO is a function of

T. The reverse bias current can be expressed as:

IO = B x T3 exp(-Eg/kT) (112)

1) + I/I( . q

kT = V OGOC ln (111)


144

where B is a constant which is independent of temperature.

Combining (and simplifying) these equations gives the following approximate relation:

VOC = Eg/q – (kT/q) ln(B x T3/IG) (113)

The rate of change of open circuit voltage with temperature is approximately given by:

dVOC/dT = -(1/T)[Eg/q – VOC(T)] (114)

which has a value of –2.3 mV/oC for silicon cells at ambient temperature.

This decrease in VOC (plus a slight reduction in fill factor) easily outweighs the slight

increase in IG. and there is a reduction of cell performance with temperature. This is also

shown in XFigure 109X and can be approximated as a 0.4% reduction in efficiency per degree C

temperature rise.

6.3.7 Other types of cell

Although crystalline silicon makes up most of the cells used for terrestrial applications, a

large number of different cells are currently under investigation which could result in lower

costs. The most popular alternative at present to mono-silicon and multi-crystalline cells are

the amorphous silicon cell and the polycrystalline thin film cells such as cadmium telluride

and copper indium diselenide. BP Solarex and Siemens have recently launched thin film

modules onto the market. Very recently, work has started to commercially develop dye-

sensitised devices for electricity generation

6.3.8 Applications

PV cells are normally connected together to form units called modules. PV modules are

widely used for battery charging. Electronic interfaces provide charge control, protection,

and sometimes maximum power point tracking. More recently there has been interest in the

grid connection of PV.

There are three basic types of PV plant

Stand alone

Hybrid

Grid-connected

Stand-alone systems are generally used in remote locations which have no access to a public

utility grid. A PV plant has one or more inverters if it supplies AC power or operates with an


145

auxiliary power source such as a utility grid or a diesel/wind generator. Stand alone PV

systems are often called „solar home systems‟, especially in developing countries where they

are most common. XFigure 110X shows such a system where a PV panel is charging a battery

via a control unit. The control unit regulates the PV array and charging/discharging of the

battery. Electricity from the battery is drawn off via the inverter which converts DC to AC for

powering domestic appliances.

A hybrid system includes a PV array, one or more auxiliary power sources such as wind or

diesel generator, and often batteries. Although it requires a more complex controller than the

stand-alone or the grid-connected systems, its overall reliability is superior to the other two

systems.

Control Unit Control Unit

Battery Battery Invertor Inverter

PV Module PV Module

Figure 110: A solar home system

In a grid-connected system, the inverter must be capable of accepting the full range of solar

array voltage and power excursions, and must be capable of operating at the array peak-

power point instantaneously. In this case, the utility network acts as an infinite energy sink

and accepts all available power from the PV system. It also provides any local system load at

times when the PV output is insufficient. Two-way metering is usually required and this can

add to the system costs.

Regulations for the grid connection of PV are starting to emerge. The working standard in

the UK is G77, although this is yet to be included in the Distribution Code which controls the

connection of all generators to the electricity distribution network.


146

0

10

20

30

40

50

60

70

80

90

100

1860 1880 1900 1920 1940 1960 1980 2000 2020 2040 2060

% S

ha

re

Biomass

Coal

Oil

Gas

Water & Wind

Other Renewables

7 Biomass

7.1 Introduction

7.1.1 What is Biomass?

Biomass energy is: “recent organic matter originally derived from plants as a result of the photosynthetic conversion process or from animals, and which is destined to be utilized as a store of chemical energy to provide heat, electricity , or transport.” (Sims 2002 p1-2).

Biomass was the first fuel used for energy by humanity. Wood was burnt for warmth and cooking. Evidence of the first use of controlled fire has been found at the „Cradle of Humankind‟ World Heritage Site near Johannesburg, South Africa, dating from 1 million years before present (BP). At 40 000 BP animal oils were used in oil lamps in France. Biomass predominated throughout history as our chief energy resource In 19th Century USA 90% of energy was derived from wood. In the early 20th Century some 40% of agricultural land in the UK was devoted to producing grass and oats to feed horses. This was an energy crop as horses were the power of the economy (cars are still rated in horse power). XFigure 111X shows changes in patterns of energy use with suggestions on future patterns.

Figure 111: changing patterns in energy consumption


147

But the rise in the global consumption of fossil fuels since the industrial revolution has seen a decline in the use of biomass as a primary resource in the developed world. Fossil fuels have higher energy values per kilogramme. Only 3% of energy use in USA comes from biomass at the beginning of the 21st Century. In rural areas outside industrialised nations traditional wood fuel is still one of the most important fuel resources around the world. It is often the main source of energy in developing countries. In the Sub-Sahara this can be over 90% of the fuel available for heating and cooking. An estimated 2.4 billion people rely on non-commercial biomass for fuel.

XFigure 112X shows the percentage contribution to world energy resources. It can be seen that biomass, or the Combustible and Renewable Wastes contribution has remained approximately the same percentage but has increased in absolute amounts.

i

Figure 112: World Primary Energy 1973-2001, Source IEA 2002.

What reasons can you think of for the dramatic increase in energy use since 1973?

In terms of the contribution of biomass renewable energy to global energy, XFigure 113X illustrates the relative percentages.


148

Figure 113: Global contribution of energy resources.

However, there are differences between the contributions in developed countries and developing countries as shown in XFigure 114X.


149

(a) (b)

Figure 114: Contribution to primary energy in (a) Developed and (b) Less

Solid Fuels

25%

Natural Gas

24%

Nuclear

6%

Hydro

6%

Oil

36%

Bioenergy

3%

Nuclear

1%Hydro

6%Natural Gas

7%

Solid Fuels

28%

Oil

23%

Bioenergy

35%


150

7.1.2 Developed Countries

Can you suggest reasons why there are differences between the amounts of different energy resources used in developed and less developed parts of the world?

This is further illustrated with the relationship between income and biomass use as shown in XFigure 115X.

Figure 115: The relationship between income ($2 per day) and Domestic use of Biomass.

Biomass can be illustrated to be of major importance as an energy resource. It is estimated that perhaps as many as 2.4 billion people rely on biomass for their primary energy resources. With future population growth predictions, this number of people is likely to grow.

0

10

20

30

40

50

60

70

80

90

100

0 10 20 30 40 50 60 70 80 90 100

Percentage of Domestic Biomass Use

Perc

en

tag

e o

f P

op

ula

tio

n b

elo

w $

2 p

er

day

Mexico

Ecuador

Russia

China

Indonesia

TunisiaMorocco Uruguay

Algeria

Poland

Costa Rica

Romania

BrazilTurkey

Columbia

Bolivia

Chile Panama

Thailand South Africa

Peru Paraguay

El Salvador

Sri

Lanka

Pakistan

Tanzania

Mozambique

Zambia

Nigeria

India

Bangladesh

Nepal

ZimbabweSenegal

Guatemala Kenya


151

7.2 Conversion Routes for Biomass There are three main groupings of the methods of converting biomass to energy namely

Thermo-chemical conversion (Combustion Gasification and Pyrolysis)

Enzymatic (Anaerobic Digestion, Fermentation and Bio-Hydrogen)

Chemical (Esterification) Biomass is a fuel in the same sense as coal, oil or gas. Indeed, biomass conversion technologies mean that biomass can be used as a solid, liquid or gaseous fuel. XFigure 116X shows examples of the range of biomass fuels and their conversion to end products.

Figure 116: Energy from biomass, fuels to end-uses

Sources: Biomass is Forever, article in Energy Wise News April/May 2002: Power Plants - Biofuels made simple, Centre for Alternative Technology 1996: Renewable Energy Resources, Twidell and Weir, E&FN Spon 1986:

Poultry litter/straw bales, Wood

energy crops, forest and wood wastes, chips/pellets

Animal wastes/sewage sludge/biodegradable wastes

High energy sugar/ starch crops

High energy oil

crops, animal fats

Esterification Fermentation Anaerobic Digestion Pyrolysis Combustion Gasification

Bio diesel Ethanol Methane Bio oils/

fuel gas

Producer gas

Heat Electricity

Wood

stove/

boiler

Gas turbine Compression

ignition engine

Spark ignition

engine

Motive power

for transport

Methanol

Steam boiler/

CHP plant

Pri

nc

ipal F

ue

ls

Inte

rme

dia

ry

Pro

du

cts

En

d U

se

Tec

hn

olo

gie

s

En

d

Pro

du

cts

Co

nv

ers

ion

Pro

ces

se

s


152

7.3 Biomass as a Fuel

7.3.1 Solar Store and the Carbon Cycle Biomass is the only source of renewable fixed carbon. It has low sulphur content and for every 2 tonnes of biomass, 1 tonne of oil and 1.5 tonnes of coal can be left in the ground. While it is growing biomass produces oxygen and absorbs the same amount of carbon dioxide as it releases when it is burnt. The energy contained in biomass comes directly from the sun, by the process of photosynthesis, and therefore, if properly managed, will always be available unlike coal, oil and natural gas, which will eventually run out Biomass grows by using solar energy to convert carbon dioxide and water into carbohydrates (sugars, starches, cellulose etc) and oxygen represented as follows:

CO2 + H2O + (solar) energy [CH2O] + O2

When the biomass decays or, more importantly, is burnt then the energy is released as heat:

[CH2O] + O2 CO2 + H2O + (heat) energy

Therefore it can be seen that biomass acts as an energy store in a similar way to other fuels such oil, gas and coal. It is also important to know that the production of energy from biomass is almost a closed cycle, see below.

Based on El Bassam 1998.


153

Carbon dioxide (CO2) is taken from the atmosphere and used by biomass to grow. Approximately 200 billion tonnes of carbon is fixed by plants annually. This is 10x the energy equivalent used by people every year. About 800 million tonnes or 0.4% are used to feed the human population. When plants die and decay or, more importantly, are burnt then CO2 is released back into the atmosphere. As long as there is sufficient biomass left growing to absorb all the CO2 released from the decaying or burning biomass then the cycle can be said to be CO2 neutral. If biomass fuels are grown in a sustainable way, for example in sustainable managed woodlands or plantations, then the CO2 released by its conversion as a fuel will never be more that the CO2 taken up by new growth. In considering the conversion of biomass to energy it is important to consider the scale of the plant and volumes of biomass required. This is shown in XTable 10X.

Table 10: Relative sizes of energy conversion plants with volumes of biomass fuels consumed in oven dry tonnes (odt). Source Sims, R. (2002).

Plant capacity

Properties served

Annual fuel demand

Vehicle movements

Conversion technologies

Physical size

Investment costs

Domestic heating (15 kW)

Family dwelling

3-5 odt firewood

2-3 trailer loads/yr

Wood burner or boiler

Large suitcase

£100s

Small business heating (350 kW)

School or a small factory

80-120 odt wood or straw

30-40 small truck loads/yr

Wood / straw burner / boiler and fans

Garage for one car

£10,000s

Small electricity generating plant (250kW)

200-300 houses or a small industry

1500-2000 odt wood, straw

5-6 medium lorries/week

Gasifier or boiler and gas engine or steam engine

Small house and garden

£100,000s

Medium electricity generating plant (5MW)

4000-6000 houses or a small industrial estate

20,000- 30,000 odt of a range of biofuels

40-50 large lorries/week

Gasifier or boiler and gas engine or steam engine

Petrol service station and forecourt

£1,000,000s

Large electricity generating plant (30MW)

25,000- 35,000 houses or industrial estate

120,000- 140,000 odt using dry

120-150 large lorries/week

Steam turbine or gas turbine or small combined cycle

Large church and graveyard

£10s of millions


154

biomass fuels

Combined cycle gas turbine or pulverised coal fired station (500MW)

>500,000 houses or a large industrial site

800 Mm3 natural gas or 1Mt coal

Pipeline or 400-500 large lorries/week

Gas turbine and or steam turbine

Tower of London or Sydney Opera house

£100s of million

7.3.2 Calorific Value and Moisture Content

There are two critical elements when describing biomass usefulness as a fuel: calorific value (energy content) and moisture content. A fuels calorific value is the measure of its available (i.e. stored) energy, usually given in Joules (J) per unit mass (kg) or volume (m3). XTable 11X shows typical caloric values for different biomass fuels and conventional fuels.

Table 11: Average calorific values for different biomass and conventional fuels

However the useful energy in a fuel will be affected by the amount of water or

moisture present in the fuel. Moisture contributes nothing to the stored energy of a fuel and will reduce its useful energy. For example Table 2 shows that dry wood has an average calorific value of 20 MJ/kg but only 15 MJ/kg when containing 20% moisture

Fuel Calorific value

MJ/kg GJ/m3

Wood (oven dried, 0% moisture content) 20 13

Wood (air-dried, 20% moisture content) 15 10

Paper 17 9

Straw 14 1.4

Rice husk 15 -

Bagasse (sugar cane fibre, air-dried) 14 10

Ground nut shells 20 -

Natural gas (supply pressure) 55 0.04

Oil (petroleum) 42 34

Coal (UK average) 28 50


155

In addition if the biomass fuel is sold by weight then the amount of moisture present can affect the economics. It can be seen from table 2 that 1 tonne of dry wood has a higher calorific value than 1 tonne of wood containing 20% moisture (i.e. 200kg). However, transporting 1 tonne of dry wood will cost the same as transporting 1 tonne of wet wood and so there will be a cost associated with transporting water that reduces the fuels useful fuel. There are two methods used to calculate the moisture content, the “Wet Basis” and “Dry Basis”. The moisture content of a single piece or sample of biomass will be different using the two methods and so care must be taken to be clear which method is being used. Generally speaking the Wet Basis is used as a norm and so all figures quoted in these notes refer to Wet Basis. To determine the moisture content, a sample or a number of samples of the biomass is/are weighed wet then oven dried to expel all the moisture. The oven-dried weight is subtracted from the wet wood weight to determine the amount (mass) of water that was present. In the wet basis, the mass of the water is divided by the mass of wet biomass. In the dry basis the moisture content is then calculated by dividing the mass of the water by the mass of the oven dried biomass. The following calculation demonstrates the difference between the two methods: Example: A quantity of wood has a mass of 10kg. It is dried to an oven-dried

condition, and then it has a mass of 8kg. What is its moisture content on a) a Wet basis and b) a dry basis? WET BASIS

Mass of wet wood (10kg) – mass of oven dried wood (8kg) = mass of water (2kg)

)(..20.0)10(

)2(basiswetCM

kgWoodWetofMass

kgWaterofMass i.e. 20%

DRY BASIS

)(..25.0)8(

)2(basisdryCM

kgWoodDryofMass

kgWaterofMass i.e. 25%

In many respects the „wet basis‟ is more natural as values under the dry basis can go over 100% (if more water is present than solid matter).


156

7.4 Biogas

7.4.1 Introduction

Biogas production from waste is very widely used. There are about 1,000 plants in the UK generating methane from sewage and industrial effluents. There are probably another 250 generating biogas from refuse in landfills. In Europe as a whole there are well over 7,000, including some in Denmark taking mixtures of wastes and dedicated to energy production. The advantages over other biomass sources are that the raw material has negative value so gasification attracts an environmental tax premium. The disadvantage, like liquification, are the side streams, leachate, effluents and solids which need further treatment and can often be up to 50% of the original mass. Disposal or recycle routes for these wastes are difficult and becoming more expensive. Recent problems with the food chain in the UK for example have undermined confidence in processed waste as a soil conditioner. The process by which biogas is produced is known as “anaerobic digestion”. Anaerobic digestion has been used for waste treatment for over 100 years. There is considerable experience and most large sewage treatment works have anaerobic digesters. Energy recovery has always been part of the process. Anaerobic digestion can be defined as “Organic matter broken down by bacteria in the absence of air, producing a gas (methane) and solid (digestate). The by-products can be useful, for example biogas can be used in a furnace, gas engine, turbine or gas-powered vehicles, and digestates can be re-used on farms as a fertiliser.” Hwww.planningportal.gov.uk/england/professionals/en/1115310681665.html Anaerobic digestion is unusual biotechnology because it uses some of the first organisms to have evolved 4 billion years ago when the emerging biosphere was a completely different place. Much of what is known about the microbiology and biochemistry of anaerobic digestion has been unravelled in the last ten years. It has evolved in parallel with the rapid improvements in the understanding of molecular biology and genetics. A new group of bacteria have been identified and characterised as responsible for the generation of methane. They are known as the Archaebacteria; they have very special biochemical characteristics and were some of the first organisms to evolve. They played a key role in shaping the present environment. There are differences between aerobic and anaerobic processes that are summarised below in XTable 12X.

http://www.planningportal.gov.uk/england/professionals/en/1115310681665.html


157

Table 12: Simple contrast of the features of aerobic and anaerobic process engineering

(process intensitivity).

Aerobic Anaerobic

Capital cost Low High

Energy or running costs High Low

By-products Large amounts and need further treatment

Small well stabilised

Control Difficult but well known Simple at low rates understanding still developing

Environmental Odour can be a problem

at low energy. Can be noisy, can be fly problems

Some ddour a problem. More corrosive.

Seasonal Can be difficult but understood

Can be completely switched off out of season with rapid restart.

Treated effluent Suitable for discharge to

environment Not possible in reduced

state

Nutrients Need to be added for industrial effluent

Rarely necessary except start up

7.4.2 Control of Anaerobic Digestion

Unlike aerobic treatment anaerobic digestion is a sequential process which relies on a synergistic consortium of different micro-organisms for each stage. These activities need to be closely coordinated for the successful breakdown of biomass/sludge. The digestion of sludge and other solid organic polymers are the most difficult and take place in four successive steps: hydrolysis, acidogenesis, acetogenesis and methanogenesis ( XFigure 117X). The operation of anaerobic digestion is further complicated by these different microorganisms involved having different preferred environments and rates of growth. The stages can therefore

become unbalanced with each other. The rate limiting reaction is usually the hydrolysis step although under stress methane generation can be slowed down by excessive quantities of undesirable intermediates. The digestion of solid organic matter involves 4 sequential steps involving 4 different


158

types of bacteria : 1. Hydrolysis which involves the enzymatic conversion of solid organic matter to

soluble organic matter. 2. Acidogenesis which is a fermentation stage. 3. Acetogenesis 4. Methanogenesis, in which methane is formed.

Figure 117: Stages in Anaerobic Digestion

methane and carbon dioxide

Amino acids, sugars

Complex organic material

carbohydrate lipid

LARGE CARBOXYLIC ACIDS AND ALCOHOLS

Hydrogen acetate

protein

INTERMEDIATES, HIGHER ORGANIC ACIDS, BUTYRATE, LACTATE ETC.

biogas

4.METHANOGENESIS

3. ACETOGENESIS

2. ACIDOGENESIS

1. HYDROLYSIS


159

7.4.3 Biomass to methane The conversion of organic materials including biomass to methane is achieved by the bacterial decomposition of organic matter in the absence of air or oxygen to produce a mixture (biogas) of methane and carbon dioxide in a roughly 2:1 volumetric ratio (biogas). The carbon dioxide can be removed to leave Substitute Natural Gas (SNG) of pipeline quality, while the residual sludge retains its nitrogen to yield a fertiliser. It is a versatile process which can convert a wide range of raw material inputs including sewage sludge, farm wastes, Municipal Solid Waste (MSW), algae, water weeds and crop residues; its reaction temperature can be

maintained in the mesosphilic region of 35-37 C or at 55 C for faster thermophilic rates; the type of fermentation selected may be batch, semi-continuous or continuous. It includes digesters for large urban sewage works with reactor tanks up to 10,000 m3 capacity down to the 3-4 m3 Gobar vessels of rural India.

7.4.4 Example Domestic Waste Treatment – Wanlip 5MW Biogas

Plant Wanlip Sewage Treatment Works treats domestic and industrial waste water from Leicester and a number of satellite villages. Industrial waste includes Walkers Crisps, Everards Brewery and Nestlé. The volume of effluent treated is equivalent to 600,000 people about 185 million litres per day. The sewage and industrial effluent goes through settling to remove some water so that the total solids content of the waste stream up to 5-7% dry solids. The sludge is then mesophilically anaerobically digested for between 12-15 days. This converts approximately half the solids to biogas this is most of the easily biodegradable carbon and the residue is able to be recycled as a soil conditioner. The biogas is used in a CHP plant with 5 1MW engines.

7.4.5 Bio-hydrogen Production

It has been discovered that certain types of algae such as Chlamydomonas reinhardtii can produce hydrogen instead of oxygen under certain conditions such as a sulphur free environment. Current research has produced small quantities of hydrogen, 2% conversion of sunlight, but this is a rapidly evolving a subject. The possibilities of algae pond farms in unproductive desert areas yielding hydrogen are not beyond the realms of fantasy.


160

There is a requirement for 5% of road transport fuel to be derived from biomass by 2012

7.5 Introduction to Liquid Biomass Fuels, Biodiesel and

Bioethanol

7.5.1 Biodiesel

Since Rudolf Diesel‟s demonstration of groundnut oil as a diesel fuel at the Paris exposition of 1900, many others have emerged with various other alternatives. Today, there are a number of vegetable oils that have been tried as diesel alternatives but, with the exception of a few, none of them can be classed as satisfactory. They include sunflower, soybean, peanut, palm oil, linseed, rapeseed, winter rape, cottonseed, and canola oil.

All vegetable oil and animal fats consist primarily of triglyceride molecules as shown below in XFigure 118X.

Figure 118: Molecule of Triglycerol.

R1, R2, and R3 in XFigure 118X above represent the hydrocarbon chain of the fatty acid elements of the triglyceride. There is a three-carbon chain called the glycerol backbone that runs along the left side of the molecule. Extending away from this backbone are the three long chain fatty acids.

7.5.2 Transesterification To make biodiesel with similar chemical and physical properties to fossil diesel requires the separation of the glycerol backbone from the long chain fatty acids. This can be achieved by a number of processes. However, the process known as transesterification where the vegetable oil is reacted with methanol and a catalyst salt such as sodium hydroxide is a viable method. The chemical bonds between the glycerol and long chain fatty acids are broken and the methanol combines the long chain fatty acids to form what is called a fatty acid methyl ester or biodiesel. This is shown in XFigure 119X

CH - O - C - R2 3

CH - O - C - R2

CH - O - C - R2 1

O

O

O


161

Oil + Alcohol Glycerol + Fatty Acid Methyl Ester (Bio Diesel)

H

H C

COOR/

H C

COOR

H C COOR//

H

+ 3CH OH3 + RCOOCH/

3

RCOOCH3

R COOCH//

3

H

H

H

H

H

C

C

C OH

OH

OH

Figure 119: The Chemistry of Biodiesel Transesterification

The glycerol bi-product can be extracted by settling as it is denser than the biodiesel, as shown in XFigure 120X. Glycerol has many uses in the chemical industry.


162

Figure 120: Biodiesel separation from Glycerol

Biodiesel can be blended with fossil diesel or even used as pure biodiesel, as it has very similar chemical and physical properties. Car and engine manufacturers are certifying that biodiesel can be used for compression ignition engines.

7.5.3 Bioethanol The production of ethanol (C2H5OH) is nothing new, its being going of for six millenia. The microbiology of the brewing process is relatively uncomplicated - at the heart of the process is yeast which converts soluble sugars to ethanol and carbon dioxide. By and large the yeast belong to a small number of closely related genera, principal amongst these is the genus Saccharomyces. As far as ethanol-producing yeast are concerned, ethanol is a waste product of anaerobic energy metabolism. The yeasts can only breakdown a small range of short-chain soluable sugars such as glucose and fructose also known as six carbon atom-containing mono-saccharides (C6H12O6) and also a small number of di-saccharides such as sucrose. Starch is the only polymeric material hydrolysed to provide fermentable sugars in conventional ethanol fermentations; this requires an enzyme called amylase or by

Biodiesel

Glycerol


163

heat treatment. Following fermentation the following values of ethanol are typically produced.

Theoretical yield: 1.00g(C6H12O6) —> 0.51g (C2H5OH) + 0.49g (CO2)

Practical yield: 1.00g (C6H12O6) —> 0.46g (C2H5OH) + 0.44g (CO2) + 0.10g (Biomass)

The fermentation process, depending of the type of yeast used will yield a solution containing between 4-12% ethanol to water. To make use of the ethanol in engines requires dehydration known as distillation. Certain countries, most notably Brazil, have embarked on a programme of setting up what are in effect energy plantations. There are areas of land in which crops are grown solely for conversion to fuels. In Brazil the principal fuel crops are sugar cane, sorghum and cassava. Brazil has a favourable combination of land resources and climate to permit a sizeable proportion of its liquid fuel requirements to be met by this strategy. Most countries do not; in most countries land given over to fuel plantations would adversely affect food production. Although there are numerous programmes throughout the world in which various crops are under development for cultivation in marginal or poor quality lands, there have been no truly dramatic breakthroughs. The US has embarked on a programme of ethanol production using corn as the substrate. The use of conventional substrates has certain implications for the cost of the ethanol produced by the fermentation of such substrates. Typically, up to 80% of the cost of the ethanol product is attributable to the cost of the substrate in such cases. This reduces greatly the impact which technological innovations might have on the process. Ethanol can be blended with petrol in a ration of 20:80. This requires no modification of the engine. Ethanol has a lower energy content than petrol, but oxygen in ethanol improves the combustion process in the combustion chamber.


164

7.6 Conversion of Biomass 1: Pre-treatment and Direct

Combustion

7.6.1 Resource Base Wood is the main solid biomass resource in the UK that is suitable for thermo-chemical conversion for electricity generation. Wood is available from a variety of sources including energy crops, forestry residues and sawmill wastes such as shavings and sawdust. Energy crops are plants, which are grown specifically for energy purposes, see XTable 13X As well as the environmental advantages (i.e. CO2 neutral) already referred to, an important driver in European countries is the need to find new agricultural enterprises, since the problems of overproduction of food and surplus agricultural land increase. In northern European countries, Short Rotation Coppice (SRC), the production of fast growing trees, which are planted, maintained and harvested in a manner, which is closer to agricultural crop production than conventional forestry, has received much attention recently. This technique has given rise to a number of pilot projects in Britain but is much more commercially developed in countries like Finland and Austria.

Table 13: Comparative content of some energy crops

Species Heating value

(MJ/kg) Annual production rate ODT/ha

Poplar 18.75 12-15 (UK) Eucalyptus 19.35 Up to 50(E.grandis in Brazil) Miscanthus 18.0 16 (UK) 30 (S Europe) Giant Reed (Arundo

donax) 18.27 25-35 (S Europe)

The wood produced by SRC may be treated as a solid fuel and simply burned in a combustor to produce heat or steam. To produce a fuel, which is more easily stored, transported and used, wood may be briquetted or pelletised. Alternatively, it may be fed to a gasifier or pyrolyser to transform it into a gaseous fuel, the active constituents of which are hydrogen and carbon monoxide. This may in turn be used to drive a reciprocating engine or turbine which, if attached to a generator, will produce electricity. (In fact, electricity systems like this usually make simultaneous use of the heat produced as well – this is called Combined Heat and Power or CHP). The same options are available for forestry and wood industry wastes. Forestry residues are the branches and tops of trees which are removed on felling –


165

the so-called „lop and top‟ which is normally left on the forest floor. This amounts to about one third of the tree and constitutes a large, free and generally neglected resource. Problems of harvesting, transport and utilisation are exacerbated by the low bulk density of the material; and to address these difficulties, the residues could be converted to wood chips or pellets. It has been estimated that a further one third of the wood is transformed to waste in sawmills and other wood processing operations and this material may be similarly treated.

7.6.2 Pre-treatment

7.6.2.1 Drying and Grading The drying requirement depends on the combustor feed specification. Wet biomass at typically 50 Wt % (wet basis) is generally considered too wet. It would affect

combustion temperatures, gas quality, and giving rise to condensation problems and lower efficiencies. This, however, depends on the conversion technology used as 50 Wt % may be suitable for some combustors (i.e. space heating plant) and some gasifiers and 5 Wt % for a pyrolysis plant. However, some gasifiers will not work unless you dry the feedstock. Drying may be carried out in the field and in the storage pile, but this is slow, unreliable, causes loss of material from biological degradation and can cause fires. Rotary kilns are widely specified as dryers using waste heat and/or combustion of biomass feed. The feed used for the latter may be that rejected as screenings or fines, again depending on the gasifier feed specification.

Fluid bed, silo and steam dryers have all been used successfully for biomass. None is very efficient, however, and the energy and economic costs are high. These are outweighed, however, by the higher downstream gas cleaning requirements consequent on not drying. The operation is well established with extensive experience to draw on. Different combustors and reactors (gasifiers and pyrolysis) have different feed stock requirements. In most cases the burner will be designed for a particular type and grade of feedstock (i.e. logs, straw bales, wood chips, wood pellets etc) and changing the feedstock specification can result in poor performance or damage to the combustion or reactor plant. Automatic fuel handling and feeding systems can also be sensitive to the grade of feedstock. Screw feeders for instance tend to deliver a pre-determined volume of fuel, which is a function of the screw pitch and diameter, rather than a mass of fuel. Small regular shaped particles will pack more tightly than larger irregular shaped particles effectively delivering a greater mass of fuel per revolution of the screw feeder. This could lead to more or less fuel being delivered than the combustor or reactor is designed. Over feeding could result in damage to the combustor or reactor plant (e.g. refractory lining, heat exchangers etc.) while underfeeding could result in


166

reduced efficiency.

7.6.2.2 Densification Many biomass materials (i.e. saw dust, straw, rice husk) are available as the by product of agricultural, industrial, or food processing activities. Many of these materials can, after drying to appropriate moisture content, be used directly as fuel in combustors and reactors. However, these materials are light and have a low bulk density making them difficult to burn in a controlled manner and expensive to transport. Therefore they need undergo some form of densification to convert them into a more useable form. There are a number of proven densification technologies available including baling, briquetting and pelletising.

7.6.3 Direct Combustion of Biomass The different types of conversion technologies used in the direct combustion of biomass are shown in XFigure 121X.

7.6.3.1 Combustion Process The combustion process for different biomass fuels is basically the same consisting of three main stages; drying, pyrolysis and oxidation (see XFigure 122X). The effect of each stage on the overall efficiency of the combustion process will depend on the biomasses:

Moisture content

Calorific value of dry matter

Content of volatile matter

Ash content

Ash melting point

Particle size and size variation


167

Figure 121: Conversion technologies

Figure 122: Process of biomass combustion


168

The first stage of the process is the evaporation of moisture from the fuel. The moisture content and particles size of the biomass are important during this stage. High moisture content and large particle size requires a long drying time due to the relatively low surface to volume ration of the fuel. The heat for drying is supplied by radiation from the flames and from the refractory materials used for the furnace walls. In addition heat can be supplied from preheated primary air, of up to 400oC, obtained via a heat exchanger located in the furnace flue. As biomass gets close to being fully dry the temperature increases until at around 200oC pyrolysis starts and the volatile matter in the form of mixed vapours or vaporised tars and oils starts to be released. With further heating above 400oC, the fuel begins to oxidise and combustion of solid material starts. From now on the heat is supplied from the combustion itself. Combustion of the solid material will raise the temperature above the fuel bed igniting the volatiles, which are often seen as the yellow flame burning above the fuel. The final stages of combustion involve particles of char arising from the disintegration of the biomass. As surface area to volume ratio of the char is much higher than the original biomass, high combustion intensity and temperatures are achieved. Char is composed mainly of carbon and burns to produce CO2. The inert matter is made up of materials that are non-combustible and becomes clinker, slag or ashes. Three quarters or more of the energy of biomass fuels is in the volatile matter compared to less than half for coal. Any stove, furnace or boiler should be designed to ensure that these vapours burn. Air must reach all of the char for its complete combustion and so the fuel should be in as small particles as possible. This has the disadvantage of producing finer particulates that need to be removed from the flue gases. Airflow needs to be carefully controlled to ensure that there is sufficient oxygen for complete combustion of the fuel. Too little air will lead to in incomplete combustion resulting in low efficiency and high emissions of pollutants (i.e. carbon monoxide, smoke, particulates etc). Too much air can have a cooling or quenching effect reducing combustion temperatures resulting in incomplete combustion and high emissions of pollutants. Therefore the amount of air for a given amount of fuel is critical to efficient combustion. The relationship between the amount of air and the amount of fuel is known as the air to fuel ratio and is expressed as the mass (kg) of air to the mass (kg) of fuel. For example an air to fuel ratio of 35:1 means that 35 kg of air is required for every 1 kg of fuel. The air to fuel ratio will vary according to the type and grade of fuel being used.

Efficient combustion also depends on the achievement of the so-called Three Ts: Temperature, Time and Turbulence. Obtaining full oxidation of the fuel especially


169

the volatiles requires the fuel to be exposed to a minimum temperature for sufficient time to ensure that all necessary reactions are completed. Finally, the gases in the combustion chamber must be turbulent so that are thoroughly mixed in with and exposed to combustion air to ensure that they are fully oxidised. High turbulence enables this to be achieved without the need for excess amounts of air to be introduced into the combustion chamber and thereby cooling the process and reducing its efficiency

7.6.3.2 Types of Combustors Biomass can be burnt in a furnace or combustor to provide power usually by raising steam (see XFigure 123X). Although shown separately in XFigure 123X, the furnace and boiler unit are usually built into a single unit known as the boiler. Furnaces can vary from simple log burners to much more complex fluidised bed technology, which was developed for coal.

Figure 123: Power generation from direct combustion

Furnaces can be categorised into three main types: grate burners, suspensions burners and fluidised bed. Grate burners have two main sub-categories; pile burners and stoker burners. Suspension burners and fluidised bed are more efficient than great burners but tend to have much higher energy requirements to operate ancillary equipment. Grate burners

Pile burners are the most simple with the fuel lying in a pile on a fixed bed or grate through which combustion air flows and ash falls. This type of grate is simple but restricts the airflow resulting in low combustion efficiency, low burning rates and relatively low temperatures being achieved. However, pile burners are able to burn wet fuels, require little operator attention and can cope with minor interruptions in fuel supply, fluctuations in a moisture content and particle size.


170

Stoker burners facilitate the movement of the fuel on the grate and the removal of ash resulting in increased combustion efficiencies (around 90%), high burning rates and relatively high temperatures being achieved. There are number of proven designs available including the underfed cone grate, moving sloping grate, travelling gate and the spreader stoker. Suspension burners Suspension or cyclone burners are suitable for small chips and fines. The combustion chamber is cylindrical with primary air entering from below and secondary air entering tangentially at very high velocities. This creates a cyclonic flow of air over the fuel bed and most of the fuel becomes airborne. Therefore most of the combustion process takes place in suspension. This gives very good fuel to air mixing resulting in high combustion efficiencies and high heat release rates (750+ MJ/m3 of furnace volume). However, compared with grate burners suspension burners are expensive, are sensitive to particle size and moisture content and have high energy demands to operate ancillary equipment and are therefore unsuitable for small scale applications. Fluidised beds Fluidised beds were originally designed for burning pulverised coal but can be used with any solid fuel. The solid fuel particles are burnt in a bed of inert material such as sand or ash through which combustion air is forced. As the gas passes upwards through the packed bed of particles a pressure drop is formed across the bed. This pressure drop increases as the gas velocity increases until the bed of solid particles expands slightly. At this point the individual particles become supported in the gas stream with freedom of movement relative to one another. The bed is then said to be fluidised and has the appearance of a boiling liquid with a well-defined free surface. Higher gas velocities result in entrainment of the solid particles and the surface loses its well-defined interface. This „fluidisation‟ of the biomass allows for a high degree of mixing and high rates of heat transfer compared to a grate or suspension burner.


171

7.7 Conversion of Biomass 2: Gasification and Pyrolysis

7.7.1 Gasification

7.7.1.1 Gasification Processes

Thermochemical gasification is the conversion, by partial oxidation at elevated temperature of a carbonaceous feedstock such as biomass or coal into a gaseous energy carrier. Partial oxidation takes place when there isn‟t enough oxygen present for full oxidation to occur i.e. when less than the stoichiometric amounts of oxygen needed for complete combustion are present. As a result of this, partially oxidised products are formed. If the temperature is sufficient, then the primary products from the gasification of biomass are gases. These gases are:

carbon monoxide (CO)

carbon dioxide (CO2)

hydrogen

methane

trace amounts of higher hydrocarbons such as ethane and ethene

water

nitrogen (if air is used as the oxidising agent) Various contaminants such as small char particles, ash, tars and oils are also present with the gases. At high temperatures, charcoal and liquids are either minor products or not present in the product mixture. The partial oxidation can be carried out using air, oxygen, steam or a mixture of these. There are three basic gasification techniques: air gasification, oxygen gasification and pyrolytic gasification. Air gasification involves direct contact of biomass feedstock and air which produces

a gas with a low calorific value of 4-12 MJ/m3. (low compared to methane) This is due to the dilution of the product gases with nitrogen from air during the gasification process. The gas is suitable for boiler, engine, and turbine operation. Low energy value means not economic to store, so used immediately. Oxygen gasification results in the production of a medium calorific value gas of

between 12-27 MJ/m3 and is suitable for pipeline distribution and as synthesis gas


172

for conversion, for example, to methanol and gasoline. Pyrolitic gasification (from indirectly heated biomass gasifiers when air is used and heat transfer occurs via an inert solid medium) or steam gasification will also produce a gas with a similar calorific value as oxygen gasification. Gasification with air is a much more widely used technology. This is because oxygen production and usage incurs an extra cost and hazard. It also involves the complexity of having multiple reactors. The feedstock will contain a certain amount of moisture and so steam will always be present during the gasification even if it isn‟t added. This causes a type of reforming to take place. For biomass gasification, reforming refers to gasification in the presence of another reactant e.g. steam, steam-oxygen or steam-air. Steam reforming processes involve reactions of biomass and steam and of the secondary products formed from biomass and steam.

Steam-oxygen or steam-air gasification of biomass often includes combustion of a residual char from the gasifier, of a portion of the product gas, or of a portion of the biomass feedstock to supply the heat. The process can be carried out with or with out the use of catalysts. Catalysts are substances added to a reaction in order to speed it up. They themselves do not become chemically involved in the reaction.

The addition of steam increases the amount of hydrogen in the product gas.

Gasification occurs in a number of sequential steps: 1. Drying to evaporate moisture 2. Pyrolysis to give gas, vaporised tars or oils and a solid char residue 3. Gasification or partial oxidation of the solid char, pyrolysis tars and pyrolysis

gases.

When a solid fuel is heated to 300-500 C in the absence of an oxidising agent, it pyrolyses (physically or chemically breaks down by heating in the absence of air) to solid char, condensable hydrocarbons or tar, and gases. The relative yields of gas, liquid and char depend mostly on the rate of heating and the final temperature. The gas, liquid and solid products of pyrolysis then react with the oxidising agent – usually air – to give permanent gases of CO, CO2, H2 and lesser quantities of hydrocarbon gases. Char gasification is the interactive combination of several gas-solid and gas-gas reactions in which solid carbon is oxidised to carbon monoxide and carbon dioxide, and hydrogen is generated through the steam reforming reaction (mentioned above). The gas-solid reactions of char oxidation are the slowest and limit the overall rate of


173

the gasification process. Many of the reactions are catalysed by the alkali metals present in wood ash, but still do not reach equilibrium. The final gas composition, as mentioned above, is influenced by many factors such as:

feed composition

water content

reaction temperature

the extent of oxidation of the pyrolysis products Not all the liquid products from the pyrolysis step are completely converted due to the physical or geometrical limitations of the reactions involved, and these give rise to contaminant tars in the final product gas. Due to the higher temperatures involved in gasification compared to pyrolysis, these tars tend to be refractory and are difficult to remove by thermal, catalytic or physical processes. It is important to note that gasification is a series of reactions occurring simultaneously. Gases are not only produced directly from the feedstock but also from the products of the feedstock reactions (e.g. the char produced during pyrolysis will further react to form gases and carbon dioxide will react with oxygen to produce carbon monoxide). A much more general description of the overall process of gasification is:

Biomass + O2 Heat +CO2 + H2O

Heat + Biomass + H2O CO + H2

Gasification is used in heat and power. For heat it has advantages over combustion as it is easier to control the gas burner. Very little excess air is used so there is less heat loss in exhaust gases and it produces no particulates

7.7.1.2 Types of Gasifiers A number of different types of gasifier have been developed with varying operating conditions, temperatures and pressures. The most common reactor configurations the fixed bed gasifiers, which are the simplest type, and the fluidised bed. Fixed bed tend to be small,. Why, the ability to mix the fuel. Two types. Downdraft gasifier: biomass fuel enters at top air is drawn down through the fuel and gas collected near the base. Updraft gasifier: biomass fuel enters at top. Air enters at bottom and passes up against direction of fuel. Gas is drawn off at the top. Scale up to 10 MW. Fluidised bed: fuel and substrate enter at base with air pumped through. Gas leave at top. Scale limited by demand and supply. Scale 100MW.


174

7.7.2 Pyrolysis Pyrolysis is the slow, irreversible, thermal degradation of carbonaceous materials

between 400 C and 800 C either in the complete absence of an oxidising agent, or with such a limited supply that gasification does not occur to an appreciable extent. This latter case is sometimes termed partial gasification and is used to provide the thermal energy required for the pyrolysis process at the expense of char and liquid yields. The thermal energy for pyrolysis may be provided in a variety of ways either directly or indirectly. Indirect methods include external heating such as firing a rotary kiln with product gas and direct methods include hot gas transfer or molten metals. The products of pyrolysis are:

gas

liquid (bio-oil)

char (charcoal). The relative proportions of each depend on the method of pyrolysis and the reaction parameters. The slower the pyrolysis process, the more gases and char that are formed. Control over the temperature, heating rate and time of reaction will result in different products.

7.7.2.1 Pyrolysis Processes Pyrolysis is almost certainly the oldest method of processing one fuel in order to produce a better one. Conventional pyrolysis involves heating the original material in

the near-absence of air, typically at 300-500 C, until the volatile matter has been driven off. The residue is then the char, a fuel that has about twice the energy density of the original and burns at a much higher temperature. For many centuries, charcoal was produced by the pyrolysis of wood. This method of production still takes place in much of the world today. Depending on the moisture content and the efficiency of the process, 4-10 tonnes of wood are required to produce one tonne of charcoal, and if no attempt is made to collect the volatile matter, the charcoal is obtained at the cost of perhaps two-thirds of the original energy content. With more sophisticated pyrolysis techniques, the volatiles can be collected, and careful choice of the temperature at which the process takes place allows control of their composition. The liquid product has potential as fuel oil but is contaminated with acids and must be treated before use.


175

Altering the pyrolysis conditions can substantially alter the proportions of solid, liquid and gas product. An example of the variation in products with reaction temperature for various feedstocks is shown in XFigure 124X below.


176

Figure 124: Variation in product with various reaction temperatures and feedstocks.

Each of the feedstocks shown in the graphs above was pyrolysed in a 0.5m fluid bed

500C

0

5

10

15

20

25

30

35

40

45

50

Bio

solids

Corn

cobs

Man

ure

MS

W

Paper

Wood

chip

s

Feedstock

wt

%

Low-energy gas

Pyrolytic oil

Charcoal

700C

0

10

20

30

40

50

60

70

Bio

solids

Corn

cobs

Man

ure

MS

W

Paper

Wood

chip

s

Feedstock

wt

%

Low-energy gas

Pyrolytic oil

Charcoal

900C

0

10

20

30

40

50

60

70

80

Bio

solids

Corn

cobs

Man

ure

MS

W

Paper

Wood

chip

s

Feedstock

wt

%

Low-energy gas

Pyrolytic oil

Charcoal


177

reactor containing sand and an inert gas from compressed air-natural gas combustion with a slight excess of air (about 0.2 to 0.6%). The fluidising velocities were 0.3 to 1m/s. The products were low energy gas (5.89-11.78 MJ/m3) and charcoal. Feed rates were 50-200kg/h. The moisture contents of the feedstocks were not specified. The balance of the yield for each feedstock is water. The modes of pyrolysis are summarised: Slow pyrolysis at low temperatures and long reaction times maximises charcoal yields at about 30 wt %, comprising about 50% of the energy. This is usually known as carbonisation. The gas and liquid products are by-products, which may be used in the process for energy or exported. Fast pyrolysis maximises liquid yields at up to 80 wt % at relatively low

temperatures of typically 500 C but less than 650 C, and at very high reaction rates and short residence times of typically less than 1 second. Heating rate K/s 500-100

000.

Fast pyrolysis at higher temperatures of above 650 C, and at very high reaction rates and short residence times maximises gas yields at up to 80% wt.

„Conventional’ pyrolysis at moderate temperatures of less than about 600 C and moderate reaction rates gives approximately equal proportions of gas, liquid and solid products. This is less efficient for liquids production than fast pyrolysis in giving a multiplicity of products that are difficult to handle and market. In addition the liquid is significantly different chemically.

7.7.2.2 The Products of Pyrolysis Solid Products

When pyrolysis is optimised for charcoal production, yields of up to 30 or 40 wt % on dry feed are obtained. This occurs in slow pyrolysis with reaction times of hours or even days. Traditional beehive kilns are used in this process. Partial carbonisation at lower temperatures gives higher solid yields as the product contains a high level of volatiles. The lower temperatures help prevent their formation. Solids produced during this type of process are sometimes referred to as torridified wood. Liquid Products Of particular interest is the direct production of liquids because of their much higher energy density which reduces transport and handling costs, and because of their


178

potential as an alternative for conventional fossil fuels in many static applications such as boilers and for process heat. One such liquid product is „bio-oil‟. Bio-oil has a number of special features and characteristics which have caused some concern over direct utilisation. However, the number of successful applications is increasing and considerable progress is being made in understanding the properties and controlling those that are less attractive. Density, viscosity, surface tension and heating value are known to be key properties for combustion applications in boilers, furnaces and engines; but other special characteristics such as char level and particle size and ash content will have a major effect. These properties have been claimed to make it relatively unstable in both chemical and physical terms, causing problems with storage and utilisation. However, a number of successful applications have been demonstrated and considerable progress is being made in understanding the phenomena and controlling the less attractive properties. The key characteristics are:

Energy value 16-18 MJ/Kg (1/2 that of fossil diesel)

Can be stored and transported

Suitable for stationary engines and turbines

As a biofuel can decompose over time.

7.7.2.3 Applications of the Liquid Products of Pyrolysis Combustion

Liquid products are easier to handle in combustion than solids and this is important in retrofitting existing equipment. Existing oil-fired burners cannot be fuelled directly with solid biomass such as wood chips or chopped straw without major reconstruction of the unit which is not attractive in uncertain fuel markets. However bio-oils require only relatively minor modifications of the equipment such as the burners or even none in some cases. Powdered coal fired furnaces can relatively easily accept charcoal as a partial fuel replacement, as long as the volatile content is compatible with the furnace design. Turbines can be fired with pyrolytic oils or upgraded pyrolytic oils and perhaps even char/water slurries but there is little recent practical experience. The use of bio-oil and char in engines is another target. In most countries there is a small market for charcoal lumps and briquettes for leisure and industrial applications and small regional markets for firewood, usually as logs.


179

Power Generation from Liquids

A key advantage of production of liquids is that the fuel production can be de-coupled from the power generation. Peak power provision is thus possible from a much smaller pyrolysis plant or liquids can be readily transported to a central power plant of engines or turbine. There are additional benefits from potentially higher plant availability from the intermediate fuel storage. Bio-oil has been successfully fired in several diesel test engines where it behaves in a very similar way to diesel in terms of engine parameters, performance and emissions. Continuous runs of 10 hours have been achieved on raw bio-oil without dilution or processing. A diesel pilot fuel is needed, typically 5% in larger engines and no significant problems are foreseen in power generation up to 15MWe per engine. Gas turbine applications are also considered to be feasible and an international project has recently started to thoroughly evaluate the possibilities. This may be more suitable for larger scale applications.

7.8 Power Generation from Biomass Fuels

7.8.1 Steam Systems Heat obtained from the direct combustion of biomass, producer gas or from pyrolysis gas or oil can be use directly, to provide heat for steam raising which in turn can be used to drive steam engines or turbines for electricity generation. Most types of steam plant can be modified for use with any fuel type (solid, liquid or gas) with the main changes being made to the combustion equipment and the fuel handling system. Combustion equipment for solid fuels, such as biomass and coal, tend to be larger, more expensive and require greater operator attention than those for use with liquid or gaseous fuel. There are two basic steam systems; open cycle and closed cycle ( XFigure 125X). The open cycle is the most simple where a continuous supply of fresh water is pumped into the boiler and the resulting steam is then exhausted to atmosphere once it has been used by the engine or turbine. In the closed cycle system water is continually circulating through the system. The water enters the boiler is heated and converted to steam which then drives the engine or turbine. The steam is then condensed and returned to boiler as water to be used again.


180

Figure 125: Open and closed steam system

The simplest steam power plant would have an open cycle system operating at relatively low pressure of around 10 bar with an overall efficiency of around 5%. Improved efficiencies of up to 10% can be achieved by increasing the steam pressure to around 15 bar and then superheating it to around 350oC. This type of system is really only suitable for small power requirements using steam engines of say 10kW or less. Its main advantage is low cost and minimal water treatment requirements compared with closed cycle systems. More advanced closed cycle steam systems operate with boiler pressure of 40 bar or more, superheat the steam to 3800C and condensate is returned to the boiler at around 600C. This type of system can achieve efficiencies of around 20% and is suitable for use with high-pressure steam engines and steam turbines. Therefore it can be seen that efficiencies are a function of the steam pressure and temperature entering the engine or turbine and the pressure and temperature of the steam leaving the turbine. The higher the pressure and temperature the higher the energy content of the steam and conversely the lower the pressure and heat of the exhaust steam the lower its energy content. Therefore the greater the difference between the condition of the steam entering and leaving the engine or turbine the


181

higher the efficiency. Unfortunately it costs more to produce super-heated high-pressure steam than low pressure steam. Also the equipment needed to ensure that steam leaving the engine or turbine at low pressure and temperatures (i.e. vacuum equipment, condenser, cooling water etc) are more expensive than that needed for high exhaust temperature and pressures. Therefore efficiency gains have to be weighed against the cost of achievement and in most cases will determined by the size of plant being considered. The thermal efficiency of a power station is the useful energy (electricity (or in the case of combined heat and power heat and electricity)) leaving the station divided by the fuel energy entering the station. It is essentially a measure of the overall fuel conversion efficiency for the electricity generation process.

7.8.1.1 Steam Boilers There are many different designs of steam boilers, which can be classified two ways. First of all they can be classified as ether externally fired or internally fired boiler. Externally fired boilers are those where the furnace is situated outside the boiler proper. Internally fired boiler are those where the furnaces (or furnaces) are within the boiler proper. The most suitable type of boiler for power generation purposes is the water tube boiler. In such boilers the water is evaporated in tubes, which are arranged inside a heated chamber in which they are exposed to the radiant heat of the flames and the hot flue gases (see XFigure 126X)


182

Figure 126: Water tube boiler

The tubes are relatively small in diameter giving high surface area contact between hot gases and the water. The tubes are constructed as boilers with steeply inclined water tubes or with tubes set at a relatively low inclination. The tubes are interconnected at their ends by so-called headers, which are usually set at right angles to the tubes. The boiler feed-water preheated by the flue gases, enters the upper steam drum from where it flows through unheated or only slightly heated tubes to the lower headers or drums. From here it ascends into the water tubes, in which it is evaporated and is returned in the form of a water and steam mixture to the upper drum. For power generation purposes the steam also requires super-heating (raising is temperature to above the saturation temperature), which is achieved by heating the steam after it has left the top drum. The steam is separated from the entrained water and flows through the superheater tubes, which are heated by flue gases of sufficiently high temperature. The steam then flows to the consumer equipment. The separated water, together with additional water, flows back to the lower or drums, and the cycle is repeated. These boilers are very efficient and respond very quickly to variations in load making them ideal for power generation requirements.

7.8.1.2 Steam Engines A steam engine is a reciprocating engine that utilises the energy released when


183

steam expands to drive a piston, which in turn drives a crank to provide rotary motion. A flywheel is usually attached to the crank to smooth out the jerkiness usually associated with reciprocating motion. There are two man types of steam engine the double acting and the compound engine. In the double acting engine live steam from the boiler enters at one end of a cylinder and expands pushing a piston to the other end of the cylinder. When the piston has travelled to the end of the cylinder it has completed its stroke and the now expanded steam is allowed to escape from the cylinder. Live steam from the boiler then enters the cylinder from the other side pushing the piston back to the other end of the cylinder completing cycle. In the compound engine steam is expanded in several cylinders in successive stages. It is usual to have three cylinders (triple expansion) although some only use two. Steam from the boiler enters the first cylinder (high pressure stage) where it is expanded to provide motive power. The exhaust steam from this stage will be at a lower pressure but still contains useful energy so it is passed into a second cylinder where it expands further providing additional motive power and so on until it passes through all the cylinders. Because all the cylinders are attached to the same crank it is necessary that they all have the same stroke. However, the volume of the steam will increase at each subsequent stage as its pressure reduces due to its expansion. Therefore the lower pressure stages of a compound engine have larger diameter cylinders and pistons. Double-acting steam engines are usually low pressure with low-speeds, which makes then unsuitable for electricity generation applications. However, modern high speed compound engines are available with an output in range 10 to 1800 kW with engine speeds of 750, 1000 or 1500 r.p.m. This type of unit require steam temperatures of around 380oC and can operate at pressure up to 150 bar. This type of engine is particularly suited to combined heat and power (CHP) applications. The main limitation of the steam engine is the need for lubrication, which limits the superheating temperatures to between 350 and 380OC. Above theses temperatures the lubricating oil viscosity will be low and may evaporate, reducing its ability to lubricate effectively. Oil vapour entrained in the steam will need to be removed before the condensate can be returned to the boiler. There is also a flow limitation at the exhaust, which in practice limits exhaust pressure to between 0.3 to 1 bar compared to about 0.2 bar for turbines. Therefore there will be the potential for lower cycle efficiencies than can be obtained with steam turbine. For small energy outputs up to 500kW the efficiency of modern high pressure steam engines can, however, be better than that of a steam turbine and with operating life times in excess of 200,000 hours are worth considering.

7.8.1.3 Steam Turbines Until recently (mid 1980s) steam turbines were the main form of motive power for


184

medium to large-scale electricity generation (both grid and embedded). A steam turbine is a prime mover in which a rotary motion is obtained by the centrifugal force brought into action by changing the direction of a jet of a fluid (steam) escaping from a nozzle at high velocity (see XFigure 127X). There are two basic types of steam turbine classified by the action of the steam on the rotor blades; the impulse and the reaction turbine. The working principle of the impulse turbine is that a jet of steam from a fixed nozzle push against the rotor blades and impels them forward. The velocity of the steam is about twice as fast as the velocity of the blades. Only turbines with fixed nozzles are classified as impulse turbines. In reaction turbines the steam flows from a nozzle on the rotor. The steam is directed into the moving blades by fixed blades designed to expand the steam resulting in a small increase in velocity over that of the moving blades. These blades form a wall of moving nozzles that further expand the steam. The steam flow is partially reversed by the mobbing blades producing a reaction at the blades. Since the pressure drop across the blade (nozzle) is small the speed is comparatively slow. Therefore, more rows of moving blades are needed than in an impulse turbine. There are also a number of steam turbine designs on the market, which combine both the impulse and reaction principle. Steam turbines can be either single-stage or multi stage devices. Single stage turbines extract the energy from the steam as it expands through a single high-pressure turbine. Multi-stage turbines progressively expanded through a series of turbines (often three) from high pressure to low pressure. Steam from the high-pressure turbine, passes into an intermediate and then into a low-pressure turbine thereby maximising the energy extraction from the steam. Multi-stage turbines are more efficient than single-stage units but are more expensive. Steam turbines range in output from around 200kW to over 50MW.


185

Figure 127: Conventional steam turbine

7.8.1.4 Gas-fuelled Systems

Gas produced by gasification or pyrolysis can be used as a fuel for internal combustion (i.c.) engines (petrol or diesel) or gas turbine to provide motive power for electricity generation. Ordinary off-the-shelf automotive engines can be used although they will require some modification, especially diesel engines, to operate effectively on gas. Before it is introduced in to the engine or turbine the gas must be cleaned to remove tars and particulates and, in the case of internal combustion engine, cooled to below 30oC to help prevent overheating occurring. Various types of gas clean up systems have been developed including, cyclones, wet scrubbers (bubbling the gas through water), dry scrubbers (saw dust, wood wool, coir fibre etc), fibreglass filters and electrostatic filters. Gas cooling is usually achieved using a condenser or other type of heat exchanger. Wet and dry scrubbers are the simplest and cheapest method for cleaning gas but fibreglass and electrostatic filters are the most effective. Cyclones can be effective for removing particulates but very fine particles often stay entrained with the gas. Therefore cyclones are often used in conjunction with a filter or scrubber.

7.8.1.5 Internal Combustion Engines Producer gas can be used directly in internal combustion engines; especially spark ignition (petrol) engines but also compression ignition (diesel) engines. However,


186

before it can be used it the gas needs to be cooled to below 300C and cleaned. Both spark and compression ignition engines will require modifications to the valves, valve seats and carburettor before they can be used with producer gas. The dry nature of producer gas means that it does not have lubricating properties and therefore valve wear is more likely to occur. Therefore, the materials used for the valves and valve seats will need to be modified (i.e. hardened) or changed to improve wear resistance. Both types of engines need gas carburettors to introduce the gas to the engine. Spark ignition engines are cheap, if based on mass-produced automotive engines, and are simple to operate but are sensitive to changes in gas quality. Little modification is required to an automotive engine other than changing the carburettor for one suitable for use with gas and hardening the valves and valve seats. These engines can be operated on 100% producer gas achieving efficiencies of around 25% at full load but this can fall off rapidly when operating at part loads. Compression ignition engines are more expensive but are less sensitive to changes in gas quality and have better efficiencies (30 and 35%), which can be maintained through out regardless of load. This type of engine will not work on producer gas alone, as it will not ignite when compressed. Therefore to bring about ignition a certain amount of supplementary fuel (usually diesel) will be required. The amount can vary between 7% to 20% (by volume) but can go as high as 60% for some types of automobile engines. Compression ignition engines can operate on all ratios of producer gas/diesel oil, which can be desirable when producer gas production is subject to fluctuations. To operate effectively on producer gas, compression ignition engines require several major modifications including lower compression ratios than conventional diesels, provision of extra cooling to injectors, fitting of gas carburettors and hardening of valve and value seats. Internal combustion engines have a large number of moving parts, which will eventually fail due to wear or fatigue. While some failures can be eliminated through proper maintenance others will eventually cause the engine to be dismantled for repair and maintenance. Engines used in conjunction with generators for electricity production tend to be used for long continuous periods. Many larger engines, particularly compression ignition engines, are designed for continuous operation and have low maintenance requirements. However, many smaller engines in the 10 to 100kW range are often modified automotive engines, for which the design life is often below 10,000 hours. Considering that there are only 8760 hours per year then small engines being used 24 hours a day will only last for around one year before they need to be replaced. Generally spark ignition engines are less robust than compression ignition engines but don‟t need supplementary fuel. Therefore it is common to convert compression


187

ignition engines to spark ignition as a compromise between robust and reliable operation and to alleviate the need for supplementary fuels. Internal combustion engines suitable for use with gas range in size from around 20kW to 3 MWe for spark ignition (the larger sizes being modified compression ignition engines) and 500kW to 20 Mw for compression ignition engines. The power output of a reciprocating engine depends on the calorific value of the fuel and the efficiency of the engine. The calorific value of producer gas is below that of diesel or petrol and so the output of the engine will be lower. Typically a internal combustion engine will be derated by between 40% and 50% when using producer gas instead of petrol or diesel. For example an engine with a rated output of 100kW engine will only operate at between 50kW and 60kW when using producer gas as the fuel. To some extent by increasing the engine‟s compression ratio and reducing pressure looses in the manifold can limit derating to around 30%.

7.8.1.6 Gas Turbines Gas turbines work in a similar manner to steam turbine in that a fluid with a high energy content, in this case the hot combustion gases, produces rotary motion as result of being deflected by rings of blades mounted on a rotor.

Figure 128: Simple open-cycle gas turbine arrangement

In an open cycle gas turbine, air is continuously drawn from the atmosphere and the exhaust gases are emitted to atmosphere often via a heat exchanger (see XFigure 128X). A compressor draws in atmospheric air and forces it into the combustor where it is mixed with the gases and then flows under high velocity into the turbine to drive it. The compressor is coupled to the turbine and the complete unit is started by an


188

electric motor. A closed cycle gas turbine uses the same working fluid repeatedly. This means that another gas with superior physical properties to air, such as helium, can be used. The working fluid is heated via a heat exchanger, which means that any fuel can be used, as the products of combustion will not contaminate the working fluid. Open cycle are much more common and cheaper than closed cycle but should be used with very clean fuels to avoid problems with corrosion. The thermal efficiency of this type of power plant is about 35%. The development of aero-derived gas turbines and improvements in gasification technology for clean coal combustion has open the way for efficient and flexible biomass to electricity systems of around 20 to 100 MW capacity. Combined cycle gas turbine (CCGT) systems are another development that is finding favour in large-scale gas fired power stations. A CCGT system comprises of one or more gas turbines whose exhaust gases are utilised in a steam boiler. The steam is then used, wholly or in part, to drive one or more steam turbines. Modern CCGT plant can achieve efficiencies of 50% (see XFigure 129X).

Figure 129: Combined Cycle Gas Turbine Plant

7.8.2 Combined Heat and Power


189

Combined heat and power (CHP), or cogeneration, is the use of both the electricity and heat that are simultaneously produced from the same plant in the generation of electricity. The generation of electricity using conventional methods is relatively energy inefficient, with typically only 35-50% of the primary energy being converted to electricity and the remainder being rejected as waste heat. CHP recovers this rejected heat and converts it into a useful form, such as steam or hot water, which can then be used for a wide variety of processes, heating, and even cooling applications. Recovering the heat produced during electricity generation means that the overall fuel conversion efficiency is increased substantially, typically to over 70% and sometimes even 90%. Being more fuel efficient, CHP delivers both cost and environmental benefits. The UK Government guidance of CHP states that „good‟ CHP has thermal efficiencies of at least 75%. Hhttp://www.defra.gov.uk/environment/energy/chp/index.htm

A CHP plant is typically described by its electrical output, not its thermal output, which is expressed in either kilowatts electrical (kWe) or megawatts electrical (MWe). For example, 500 kWe plant means that it has the capacity to generate 500 kW of electricity. However, when sizing a plant for a particular situation it is its thermal capacity and not its electrical capacity that is the deciding factor. While electricity can be exported or imported to or from the grid the heat has to be used on site because current economics in the energy market dictate that it cannot be piped over long distances. CHP plant can use steam engines, steam turbines, internal combustion engines, gas turbine or CCGT. The main source of heat from steam engine is obtained from the condensing operation, which produces low-grade heat (i.e. below 100oC) that can be used for process heating or space heating requirements. Steam turbine CHP use either backpressure or pass-out condensing turbines. Heat is rejected from the steam turbine by heat transfer from the condensing operation and from the lubricating oil cooling system. This heat can then be used for site process or space heat requirements. Although size for size steam turbines have lower capital costs than gas turbines, they are less efficient and have higher operating costs, and are therefore no longer considered for most new CHP applications except as the secondary turbine for CCGT systems. Reciprocating engine systems are based on spark ignition or compression ignition engines. Two grades of waste heat are produced from reciprocating engine systems: high grade from the engine exhaust and low grade from the engine cooling system. These systems tend to be used for small to medium applications with the capacity of gas engines ranging from 20 kWe to 3 MWe and compression engines from 500 kWe to 20 MWe.

http://www.defra.gov.uk/environment/energy/chp/index.htm


190

In open-cycle gas turbines the high-grade heat in the exhaust gases is normally used in boilers to produce usable heat for process or space heating applications. It is also possible to use the heat rejected from the lubrication cooling systems for low-grade heat applications. The exhaust steam from CCGT plants is used to provide process or space heating. Typically open-cycle gas turbine CHP plant range from 1 to 40MWe and CCGT CHP plant 6 to 40 MWe.

7.9 Conclusions Biomass is a diverse resource with a considerable range of energy applications. Biomass can be converted to heat, electricity and fuel for transport. Generally it produces less carbon than fossil fuels, but needs to be managed sustainably to ensure long-term availability. Biomass is an important global resource that supports the basic energy needs of 2.4 billion people. But it produces CO2 so requires sustainable management.

7.10 References El Bassam N. (1998). Energy Plant Species. Their Use and Impact on Environment and Development. London. James & James. Sims, R. (2002). The Brilliance of Bioenergy. In Business and in Practice. London, James & James.


191

8 Integration of Renewables

8.1 Integrating renewables - the issues

Renewable energy sources possess characteristics that are different to those of conventional

sources. Specifically:

Wind power follows the natural fluctuations of weather patterns and is unpredictable

over the very short term but could be fairly predictable over the medium term and not

at all over the longer term

Solar power also follows weather patterns and is predictable over the short and

medium term on clear cloudless days but not so on days with cloud cover. In the long

term it would be predictable in the absence of cloud due to the predictable nature of

the rotation of the Earth. However, cloud clover is difficult to forecast so this always

compromises the prediction of solar power on all timescales

Tidal changes follow natural patterns but are highly predictable at all time scales,

though enhancements due to wind and pressure systems means that medium and long

term prediction accuracies are slightly reduced

Wave power is highly predictable over the short term, less so over the medium term

and not at all over the long term

Small hydro is predictable over the short and medium term but not so over the long

term

Biomass is one of the renewable energy sources that closely resembles traditional

fossil fuels in that it can be stored and used when required, though it is dependent on a

sustainable fuel source that can be produced reliably.

These somewhat diverse characteristics make renewable energy (RE) sources more difficult

to integrate in traditional electricity networks where generation from coal, oil, gas and

nuclear is under the control of the dispatching engineers.

These characteristics lead to the following important questions relating to the integration of

RE sources:

How would the integration of renewables affect the optimal plant mix and system

operation of the traditional generation units?

What penalties do variations and limited predictability impose on the rest of the

system?

How much extra back-up capacity is required to maintain the same system reliability?

To what extent does the variability of some renewables affects the value of their

output?

What benefit is there in greater geographical and source diversity?

What are the additional transmission requirements and how do these affect the cost of

energy from renewables?


192

How would the particular characteristics of variability and unpredictability of some

renewables constrain their maximum long-term contribution?

To what extent and at what cost would extra energy storage capacity benefit the

integration of renewables?

How should „external‟ or „social‟ costs of generating electricity through conventional

and renewable sources be incorporated in the pricing mechanisms?

These questions and only recently been posed and addressed.

In the sections that follow, an attempt will be made to summarise the up-to-date

understanding of the integration of RE and to present some answers to the questions posed

earlier. We will only examine the global or „macro‟ effects of integrating renewables.

We begin by reviewing the way in which power systems are operated at present.

8.2 The operation of power systems

8.2.1 Required Characteristics

Traditionally power systems have been operated so as to provide consumers high quality

electricity (fixed frequency and voltage, high reliability and purity of waveform) at the lowest

possible cost. This has been admirably achieved through procedures that have gradually

evolved since the end of the 19th century when electricity was first supplied to consumers

from centralised generation by way of transmission lines. From our own experience,

electricity provided to consumers in developed countries is indeed of high quality. This

quality is maintained through actions (behind the scenes) that are being taken continuously by

the electricity supply authorities. The study of these actions will give us an understanding of

how power systems operate and how the introduction of renewables may require a modified

approach if the quality of electricity is not to suffer.

XFigure 130X is a block diagram that describes diagrammatically the data flow related to the

operation tasks carried out by electricity industry in order to maintain the required level of

quality of supply.

The diagram illustrates a unified system in which one authority generates, transmits and

distributes the energy, much as in the case of the former Central Electricity Generating Board

(CEGB) in the UK.


193

Figure 130: Power system operation activities

As the electricity supply industry has become increasing deregulated the tasks in XFigure 130X

are now carried out by several different players.

Until recently, the electricity supply in the UK was run on the basis of a „pool‟ to which

independent generators would contribute energy and from which various suppliers would

purchase energy for resale to consumers via the transmission and distribution networks. This

economic model was superseded by NETA (New Electricity Trading Arrangement) and

BETTA (British Transmission and Trading Arrangements) which was set up to correct some

undesirable characteristics of the pooling arrangement. This will be covered briefly in the

lectures on NETA/BETTA.

In this lecture we will limit our discussion to the technical aspects of operation. XFigure 130X

then describes the underlying activities and information flows within a power system without

reference to individual generators or suppliers.

8.2.2 Power system hardware

The box entitled "Power System" in XFigure 130X contains the totality of the power system

hardware i.e. the generating stations, transformers, transmission lines, circuit breakers,


194

protection system and all consumers. This system is heavily instrumented with transducers

that measure voltage, current, active and reactive power at various nodes of the network and

the position of circuit breakers and isolators. Knowledge of all these quantities provides

operators with all the information needed to review the topology of the network and

determine the voltage profile, as well as the active and reactive power flows in the system.

These ten of thousands of bits of data are transmitted to control centres, usually over hired

telephone lines, and more recently through fibre optics lines. It is, of course, statistically

inevitable that owing to the numbers of transducers and lines involved, unreliable data will be

present. This data is therefore processed by a „state estimator‟ (a mathematical algorithm that

provides a reliable database out of an unreliable set of information). This apparent „miracle‟

is only possible because there is redundancy in the data by, for example, monitoring

quantities at both ends of a transmission line. Subsequently, this reliable data-set is stored in

the database of a computer.

8.2.3 Demand forecasting

Electrical power systems possess the following characteristics:

Electrical energy cannot yet be stored economically. Attempts to store it in

superconducting inductors or capacitors have met with limited success. An alternative

is to store energy in other forms which can be converted into the electrical form

efficiently and with little time delay. Hydro systems and pumped storage schemes

store energy in potential form and are used wherever topography permits. However

hydro schemes represent a fraction of the installed capacity of most industrialised

nations. Batteries and fuel cells store energy in chemical form but have not made a

serious impact due to cost or technical problems.

By far the largest proportion of generating plant is thermal in nature irrespective of

the fuel (coal, oil, gas, nuclear). A feature of this type of plant is the time involved in

preparing it from cold for synchronisation (several hours) and the restrictions in the

rate at which a steam driven turbogenerator can be loaded after synchronisation.

These operational delays are dictated by the thermal/mechanical safety requirements

of massive boilers and turbogenerator sets.

Thermal generators using steam turbines have an upper limit of active power

generation equal to the nameplate rating of the unit but also a lower limit dictated by

cavitation problems in the turbine blades at low throughputs of steam.

Consequently, when a turbogenerator is connected to the network bars it should be

loaded to a level at least equal to the minimum recommended by the manufacturers

(around 30% to 50% of rated power).

XFigure 131X illustrates the variability of demand over 24 hours. There are periods

during the day (e.g. 6 to 7 am) when the rate of demand growth is considerable.

Because of the response times of thermal plant, following the demand cannot be done

unless preparative action is taken some time before the event.

It may be concluded that there is an absolute necessity to carry out demand forecasting in

order to prepare and progressively load plant as required. Utilities have invested considerable


195

effort in forecasting the daily pattern of demand. Through years of experience they have

evolved sophisticated mathematical techniques to forecast demand dependent on consumer

behaviour and other factors such as weather. All methods are essentially based on the fact

that demand exhibits regular patterns. Forecasting techniques adjust past demand to present

weather and other conditions. Meteorological data on temperature, wind speed, humidity,

cloud cover and visibility are used as variables because such factors have an important

bearing on heating and lighting demand.

The art of load forecasting has been refined to such an extent that nationally estimates are

rarely in error by more than +3% and on average they are accurate to within +1.3%. However

even this degree of accuracy represents an error of 500MW in estimating demand when the

load is 40GW.

The box named "Load Forecasting" in XFigure 130X represents the mathematical activity by

which information is provided for the database one day or a few hours in advance of the

expected demand. This activity can be run off-line. The next activity of „generator

scheduling‟ is based on this forecast.

8.2.4 Generation scheduling and spinning reserve

Demand forecasting gives a fairly accurate picture of the expected load over the following 24

hours. As shown in XFigure 131X the most economic generators supply the base load, a topic to

be discussed at length in the next subsection. In anticipation of the increases in demand

above the base level a choice has to be made about which uncommitted generating sets from

a number of available generating units (some sets are not available as they are undergoing

repairs or maintenance) have to be prepared (heated, synchronised and partly loaded) and

eventually shut down when the demand declines. This is a complicated economic and

technical choice. One of the complexities is that start-up and shut-down costs are incurred

when a unit is committed or de-committed. Therefore it may, in the long run, be more

economical to maintain a lightly loaded unit during a trough in demand, even though the

instantaneous operating costs would be less if the unit were stopped.

Large interconnected electricity systems have a number of very robust defences or reserves

against unexpected changes in the balance between demand and generation:-

The mechanical and thermal inertia in the boilers and turbo-alternators of thermal

stations provide a buffer towards transient increases of demand

Fast response reserve capacity is provided by partly loaded hydro or pumped storage

(when available). Water driven plant can respond in a few minutes and be started up

automatically when the frequency falls below a critical value.

The total generating capacity on-line is always larger than the anticipated demand. In

other words, some of the generating units are partly loaded. The difference between

these two quantities is called spinning reserve. This strategy is of vital importance in

the maintenance of system reliability. The total available effective spinning reserve is

at least equal to the largest generating unit or major interconnection to the system. In

the case of the UK there is a 1000MW cross-channel DC link to France. The spinning


196

reserve ensures that if this interconnection were to be lost (and assuming the pumped

storage schemes in the UK network have no water in the upper reservoir), the output

of the partly loaded plant can be rapidly increased to compensate for the loss.

Within limits, the larger the available spinning reserve the greater the reliability of the power

system; however the presence of these additional part-loaded generating sets significantly

increases the system fuel costs. Any part-loaded machine has a poor efficiency because of the

standing losses. An extreme example is a ticking engine of a stationary car when the

efficiency is zero.

In addition to the fast reserve available from hydro, pumped storage and partly loaded

thermal, extra back up reserves known as standing reserve and standby reserve may be

carried by highly developed power systems. These reserves consisting of generators not

connected to the network, and provide two extra lines of defence that can be relied upon to

come into action if something truly disastrous befalls the network, such as simultaneous loss

of transmission lines or of large generating units.

The levels of reserve required at any given time depend partly on uncertainties in the

predictions of demand, but more usually on the need to be ready to deal with the sudden loss

of substantial amounts of generation, either due to power station faults or the loss of

transmission circuits.

In deregulated power systems all these activities are still carried out through the complex

financial instruments that define the contributions and obligations of generators, suppliers,

distribution network operators, transmission network operators and system operator.

8.2.5 Contingency analysis

The operating state of a power system can be classified as either secure or insecure. The

system is secure if it can ride a disturbance or „contingency‟ without shifting into a state in

which a piece of equipment, such as a transmission line, is overloaded. The system is

insecure if the disturbance could bring about such an overload.

Owing to economic and technical limitations, no power system is secure from all possible

disturbances. In practice, system security is determined with reference to a selected list of

contingencies. The accumulated experience of power systems engineers dictates the

contingencies contained in the list. Typical contingencies are the loss of any generating set or

transmission line. The more disturbances are added in the list (such as the loss of two

geographically close lines), the more tight the security becomes, the higher the redundancy

required in the system and the higher the cost of electricity to the consumer.

Contingency analysis consists of the simulation of the occurrence of each of the

contingencies in the given list to determine whether the operating constraints of the power

system component parts are violated. The necessary parameters for this analysis are

contained in the "Database" component in XFigure 130X. This provides information on the

system topology and the generators used (selected from the „generation scheduling‟) to


197

supply the demand. This analysis is nothing more than a repetitive solution of the load flow

problem executed as many times as the number of contingencies in the list. The analysis will

indicate whether the system is in a secure or insecure state. If in the latter, sophisticated

programmes are capable of providing the control engineers with actions necessary to bring

the system into a secure state. A typical action may require re-routing of power flows from

generators to consumers or a change in system topology.

8.2.6 Optimum economic dispatch

8.2.6.1 The merit order dispatch

Having solved the generator scheduling problem and having ensured through security

analysis that the system is in a secure state, the "Economic dispatch" box of fig.4.1 illustrates

the task of trimming the load on individual generators to achieve minimum production cost

on a short term basis.

Before dealing with this task, we will have a closer look at the operational characteristics of

traditional generation as depicted in XTable 14X.


198

Type of plant Physical characteristics

Economic characteristics

Role within system

Nuclear Inflexible. Best operated at steady load

Very high capital cost. Low running cost

Steady base load. 75% capacity factor.

Gas CCGT Can be flexible But poor load following response

Low capital cost, low running costs. Fuel supply contracts dictate running on high load

Steady base load. 80% capacity factor.

Coal/oil Flexible with good load following resonse

Medium capital cost, medium-high running costs

Demand following, units partly loaded to provide spinning reserve. Capacity factor from 20 to 70%

Pumped-storage Hydro

Extremely flexible

High capital cost, high marginal cost

Rapid response used for „peak shaving‟

Hydro with reservoir

Flexible High capital cost low running cost

Base load and demand following if required

Open cycle gas turbine

Flexible Very low capital cost very high marginal cost

„peak shaving‟. Capacity factor <<5%

Table 14: Operational characteristics of traditional generation

A crude way of applying economic dispatch is to arrange the plants in order of average

operating cost, the so called merit order with the lowest cost plant at the top of the list.

Starting from the top of the list, the plants are then loaded (each up to the limit of its capacity)

before the next plant on the list is brought into action. This loading philosophy is illustrated

in XFigure 131X.


199

Figure 131: Merit order dispatch

At the top of the merit order is nuclear power, characterised by low fuel costs and poor

flexibility. Next is Combined Cycle Gas Turbine (CCGT) plant, which is highly efficient and

also relatively inflexible. This is followed by coal plant with progressively higher running

costs and then by the band labelled „other‟ which represents a mixture of high fuel cost

flexible coal and oil plant. Finally the „shaving‟ of the peaks is done by pumped hydro which

is arranged to absorb energy during the low demand periods between 22.00 and 06.00 hours.

Open cycle gas turbines may be brought in very occassionally to meet unexpected peaks in

demand.

To summarise, in the merit order approach, units with low operating costs run preferentially

and therefore attract high load fasctors; they generate a disproportionately large share of

electricity relative to their capacity. They are called base load plants, or high merit plant.

Units with high operating costs are only run as a last resort; they attract low load factors and

generate a disproportionately small share of the total electricity; they are called peaking

plants, or low merit plant. Plant in between these two extremes is called load following plants

or middle merit plant.

8.3 Plant generation costs and capabilities

The cost of a unit of generated electricity „at power station gate‟ (which neglects transmission

and distribution costs) consists of two components. The first relates to the capital expenditure

of setting up the plant and other fixed and semi-fixed costs such as insurance, operation and

maintenance etc. The second relates to the cost of fuel and other fuel dependent factors such

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Time (hours)

Ge

ne

rati

on

(M

W)

Nuclear

Gas

Other

Coal

Pumped Hydro Generation

Pumped Hydro

Pumping


200

as plant efficiency etc.

Plant for the generation of electricity is characterised by a K number where:

K = Plant capital cost/ Plant operating cost

Plant with large K are typically nuclear power stations for which the capital costs are very

large but the fuel and marginal costs are small. Solar, wind, hydro, wave, tidal and some

biomass plant also fall in this category, possessing large values of K as the „fuel‟ is freely

available. As an exception, some biomass systems based on wood harvesting or the gathering

and sorting of waste are likely to possess medium values of K.

It follows that plant with high K values are suitable for base load operation, renewables being

at the top of the merit order list as they possess very low marginal costs. Energy from

renewables, therefore, should be used whenever available. Plant with low K (such as open

cycle gas turbines), are suitable for peaking load.

XTable 15X shows current and projected generating costs (these costs include both capital and

fuel components) for various generating options in the UK. These figures are taken from a

draft report of the Cabinet Office Performance and Innovation Unit (PIU). This study,

undertaken during the second half of 2001, was designed to review the strategic issues

surrounding energy policy for Great Britain, within the context of meeting the challenge of

global warming, while ensuring secure, diverse and reliable energy supplies at competitive

prices. These figures, provide ball park values of the present and future competitiveness of

renewables worldwide.

Technology Price now (p/kWh) Source 2020 price PIU

(p/kWh)

Coal 2.6-3.2 1998 White Paper 3-3.6

Gas 1.8-2.2 ditto; now ~2.4 due

to gas price rises

1.9-2

Nuclear 4.7-7.8 1995 White Paper

updated using RPI

3-4.5

Wind, onshore 2.4-3.1 NFFO5 contract

prices (1998)

1.5-2.5

Wind, offshore ~5 ditto 2-3

Energy crops ~5 ditto 2.5-4

Wave ~5 ditto 4-8

Table 15: Electricity generation prices

Plant may be classified into three categories. All conventionally fuelled plant and some very

large hydro, are capacity limited in the sense that their output is constrained by their

nameplate rating at which level they can be operated indefinitely, fault occurrence and

maintenance schedules permitting. This is because fuel supplies of coal, oil, gas and

fissionable uranium as well as water in a very large reservoir are available on tap.


201

Plant such as most hydro power and biomass, are energy limited as they are usually unable to

generate continuously at their maximum capacity as the energy resource might be limited e.g.

a wind farm can only generate at maximum output when the wind speed is above a certain

level. Finally, most renewables are variable as their power output is dependent on natural

cycles.

8.4 Aggregation

The efficiency of integrated electricity systems depends critically on the aggregation of both

demand and generation. The greater the aggregation, the smaller (in proportion) are the

variations in demand and the easier it is to predict them. At one end of the spectrum, the

minimum demand from a domestic dwelling is a few watts, the average is about half a kW

and the maximum is in the range 5 to 10 kW, i.e. 10 to 20 times the average. If each UK

household had to meet its own maximum demand - 5 kW, say, 100 GW of plant would be

needed for this sector alone. Aggregation smooths these variations in demand from all sectors

- domestic, commercial and industrial - so, nationally, the maximum demand in winter is 58

GW, about 1.5 times the year-round average demand of about 40 GW. Compare this to the

summer demand profile shown in Fig 4.2. As demands are added and smoothed, prediction

becomes easier; as mentioned earlier, the National Grid Company in the UK can predict the

match between supply and demand at each half-hour with an average error of about 1.3%.

Aggregation is vital to an electricity utility and can be illustrated, in principle, using random

number strings to simulate consumer demands. When added together, 10 random consumer

demands - between zero and 10 kW - combine to produce fluctuations between 40 and 70

kW. The demand from one and from 10 consumers is illustrated in XFigure 132X.

Figure 132 Smoothing effects of load aggregation

Taking wind power as an example, as capacity increases, it also benefits from aggregation,

5 10 15 200

2

4

6

8

10

0

20

40

60

80

100

Hours

One consumer (kW) Ten consumers (kW)

One

consumer

All ten


202

and this is illustrated in XFigure 133X. This shows how dispersing wind power around the

country reduces the power excursions. Although this comes from a simulationD

1D, it illustrates

the point effectively.

Figure 133: Smoothing effects of wind power agreggstion

8.5 Penetration levels from variable sources

A number of extensive simulation studies carried out by the Central Electricity Generating

Board (CEGB) in the UK before privatisation illustrate the impact of large inputs from

renewables. To illustrate the problems of integration, an example of extreme penetration is

depicted.

XFigure 134X shows the output which would have been expected from wind (dispersed 25GW)

and tidal (10GW) alongside the demand over a period of one month. In this example 30% and

13% of the consumer energy would be supplied from wind and tidal respectively. The figure

serves to emphasise that the source variability needs to be considered in the context of the

varying demand which it helps to meet. Dispersed wind varies less rapidly than demand, tidal

more rapidly.

0 5 10 15 20 250

200

400

600

800

1,000

Time, hours

Wind output, MW

Single

farm

Distributed

farms


203

Figure 134: CEGB simulation study comparing demand with simulated distributed

wind power and tidal power.

XFigure 135X (below) shows the effect of subtracting the output from wind and tidal power

from the demand, leaving a reduced load to be met by conventional thermal plant.

Figure 135: Residual demand to be met by conventional generation


204

Ideally, the financial benefit of wind and tidal power in the extreme penetration level of this

simulation would have been calculated. This is achieved by determining the cost of

supplying the total demand in XFigure 134X from traditional generation and then subtracting the

cost of supplying the residual demand in XFigure 135X. These would represent the „ideal‟ fuel

savings obtained by considering the reduced operating time required of thermal units.

Unfortunately in real life systems the following operating penalties have to be considered:

cycling costs, due to increased start-up and shut-down of thermal plants;

reserve costs, arising from the need to ensure that the system can respond adequately to

unpredicted changes; and

discarded energy, when the available variable input exceeds the amount which can be

safely absorbed while maintaining adequate reserves and dynamic control of the sysetm.

8.6 Cycling costs

In general, cycling costs are roughly proportional to the average variability (in MW/hour) of

the load on thermal plants in a given period. It can be shown that if variations in a variable

source and in demand occur roughly independently, the total resulting variation in the net

load to be met by the thermal plans is approximately a „sum-of-squares‟ addition of the

components:

Thus, for example, when the average power variation of the added source equals that of the

demand itself, the startup loads are not doubled but increased by 40%. This has some

important implications. The impact of fluctuations in variable sources at low penetrations is

practically zero; they are lost as noise among demand flactuations. This is visually illustrated

in XFigure 136X which shows a simulation study on the UK grid with a more modest wind

power capacity of 10GW compared to the simulations shown in XFigure 134X and XFigure 135X.

XFigure 136X indicates that the variability of the residual demand is very similar to that of the

gross demand.

(total variability of

load on thermal

units) 2


electricity demand)2


variable source)2

= +


205

Figure 136: Simulation of 10GW of wind in the UK system

Studies on renewables indicate that the variability in ascending order of severity is wave,

dispersed wind and tidal. Tidal power is the only source which is likely to incur substantial

cycling penalties. For the UK‟s Severn Barrage (which could supply 5-6% of UK demand

with an average 5.5GW cycle every 12.5 hours), simplified statistical calculations indicate

that penalties in the region of 7% of the „ideal‟ fuel savings will have to be taken into

account. This is illustrated visually in fig 4.7 where the substantial variability of the tidal

contribution is obvious.

In the case of dispersed photovoltaics, the available solar resource and load variations are

often closely related so that simple comparisons assuming rough independence may be

invalid.

8.7 Reserve costs

The problems arising from the possible unpredictability of variable sources may incur more

serious costs than those relating to cycling of thermal plant. The topic of system reserve

which includes spinning, standing and standby where discussed in section X8.2.4X. System

reserve is required not only to protect from the loss of the largest generator on the nework but

also from the errors in predicting demand. When errors in predicting the output from variable

sources occurs independently of those in predicting demand, the combined error is again a

sum-of-squares addition.

(average error in

predicting net load

on thermal units)2

(average error in

predicting

electricity demand)2

(average error in

predicting variable

input)2

= +


206

Again, for small penetrations of variable sources the prediction errors are lost among load

fluctuations. However, since demand is fairly predictable, forecasting errors from substantial

penetration of renewables may incur some penalty which is unlikely to exceed 5% of the

„ideal‟ fuel savings.

For relatively little expenditure the predictability of most variable sources could be greatly

improved. This would be accomplished partly through the installation of extra weather data

monitoring stations (e.g. anemometry towers a few tens of kilometers from major wind

farms) and partly through sophisticated computational techniques. Such a programme of

enhanced predictability has been used and is being further developed in Denmark.

8.8 Discarded Energy

As the capacity of the varible sources injected into a system increases, there might be

occassions when the available power from such sources cannot be used. This is clearly

illustrated in XFigure 135X, which shows a simulation where there are periods when the availble

power exceeds demand. However, even before this stage is reached, energy from variable

sources will have to be shed because the power system would need to keep a minimum level

of thermal plant generating to maintain adequate operating reserve and system control

capabilities.

Discarding energy from variable sources poses no operational difficulties. Output from wind

turbines can be controlled through blade pitch variation, from photovoltaics through inverter

control and through hydro, wave and tidal schemes through similar controlling techniques.

This discarding of energy, however, results in an economic penalty on variable sources which

becomes increasingly important at high penetrations. This penalty is difficult to assess as it

depends heavily on the flexibility of the base load units i.e. the extent to which they could be

operated stably at low power and upon how rapidly their output could be increased if

required. For the level of penetration expected over the next decade or so, the penalties due to

discarded energy are unlikely to be of any significance.

8.9 Penalties due to increasing penetration

In the previous sections the operational penalties incurred on thermal plants due to increases

in variable source penetration were reviewed.

A number of studies have been carried out aimed at providing some estimates of penalties

incurred as penetration of renewables increases. The majority of these studies relate to wind

power as this is the variable source with the largest installed world capacity to date. One such

EU-funded study has simulated the operation of the England and Wales National Grid

accommodating increasing amounts of wind power. XFigure 137X illustrates typical results

from this study. The plot shows how annual fossil fuel costs from conventional generation are

saved as increasing amounts of electricity are generated by wind power. If there were no

penalties due to penetration the ideal savings would results. It can seen that at increasing


207

penetrations the savings (where wind power is forecast up to a day ahead) reduce below ideal

for the reasons explained above. To give an idea of scale, the England and Wales electricity

system has a conventional generating capacity of around 70GW. XTable 16X gives recent

figures on penetration penalties based on EU-funded studies and on Danish and UK

thresholds linked to operational experience of wind farms. The additional costs are graduated

and depend on the level of penetration.

Figure 137: Fossil fuel savings for the England and Wales grid from increasing penetration

of wind power.

0

500

1000

1500

2000

2500

3000

3500

4000

0 10 20 30 40 50

£M

GW

Ideal

Forecasting


208

Wind power

penetration

Measures required

Cost penalties

Up to 5% None Negligible

5-10% Occasional instances when some

energy from the wind is dicarded

and more part loading of thermal

plant required

~0.1 p/kWh

10-20% As above, plus more use of pumped

storage or hydro to balance wind

power

~0.2p/kWh

20-50% May be necessary to build more

storage, or peaking plant, or retain

old coal plant,depending on

relative costs. (Extra storage will

benefit system as a whole)

~0.5p/kWh

Table 16: Implications of increasing wind energy supply on the UK network

8.10 100% Renewable Energy Generation

At the moment it is difficult to envisage a power system where all the demand is met by

renewable energy due to it intermittency and lack of total predictability. Renewable energy

needs to be implemented alongside conventional forms of generation to guarantee security of

supply. However there are possibilities for the future where this could happen:

8.10.1 Combining different renewable energy sources

There is evidence to suggest that combining wind and tidal power may be beneficial from the

point of view of reducing their combined variability. In certain hotter climates, there is also

evidence that wind and solar (PV) power may benefit from being in combination where the

wind is driven by thermal effects reaching a maximum in the evening. To foresee a truly

100% renewable energy supply would require the use of biomass generation. Wind, solar,

tidal, hydo and wave power would be used to supply the bulk of the energy requirements and

biomass generation with its flexibility could be used to balance the residual demand.


209

8.10.2 Energy Storage

Research is underway to provide cost effective energy storage to absorb the variation in RE

generation. Fuel cells, batteries, flywheels and superconducting magnetic energy storage are

all being investigated to this end. At present the cost of energy storage is too high and it will

be many years yet before it can be considered a viable option for a 100% renewable energy

supply.


210

9 Electricity Trading and Renewable

Energy in the UK

9.1 Introduction

The lectures on integration gave an overview of the physical operation of the National Grid

and considered some of the implications for renewable energy sources. The generation and

supply of electricity in a reliable system depends on the efficient scheduling and dispatch of

the generation plants to meet the system demand.

The efficient running of the system implies value for money to the consumer for the

electricity supplied. To this end the economic operation of the UK National Grid has evolved

during the last 12 years. In this lecture we will consider how the electricity supply industry

(ESI) has changed from the state owned Central Electricity Generating Board (CEGB),

through to the privatised Electricity Pool, deregulation and finally to the present day British

Electricity Transmission and Trading Arrangements. These changes will be considered in the

light of their impact on renewable energy generation.

9.2 The State Owned ESI

Before 1990, the entire UK ESI was in the state sector and was centrally planned and

operated. Since the Second World War this was carried out by the Central Electricity

Generating Board (CEGB). The remit of the CEGB was to provide a secure supply of

electricity to its consumers. All generating plant, transmission lines, distribution lines,

substations and metering was owned and operated by the CEGB as a monopoly. The

generating plant was operated in a merit order (see notes on Integration) with some degree of

modification to allow for transmission constraints. The merit order broadly reflected the

efficiency and flexibility of the generating sets. Electricity was supplied to consumers

through the Regional Electricity Companies which were also responsible for the operation of

the distribution network in their area (132kV and below).

Although this system had worked well for many years, it was a monopoly and severely

restricted opportunities for small scale privately owned generation.

In line with the then Conservative government‟s programme of privatisation to promote

competition and drive down costs, the CEGB was broken up into various sectors and sold off

to the private sector.


211

9.3 The Electricity Pool

9.3.1 Overview

In 1990, the Pooling and Settlement Agreement (PSA) was introduced and the ESI privatised.

The generators were split up between two privatised companies: National Power and

Powergen, the former being the slightly larger of the two in terms of generating capacity. The

nuclear power stations remained in the state owned company known as Nuclear Electric, due

to problems with decommissioning liabilities. The Regional Electricity Companies (RECs)

were individually privatised and given sole licences to supply electricity in their areas still

retaining responsibility for the distribution network in their areas. The RECs jointly owned

the National Grid Company (NGC) which was responsible for the operation of the

transmission network (275kV and 400kV) and the pump storage facilities in Wales. NGC

also served as the network operator responsible for the scheduling and dispatch of the

generation plant.

9.3.2 The Operation of the Pool and Pool Prices

Instead of generation plant being scheduled in a merit order reflecting efficiency, operating

cost and flexibility, generators now „bid‟ into a pool offering to generate a given output at a

given price for the day ahead. NGC would examine the bids, put them in order of increasing

cost and then from this list select the generation plants it would need to meet the demand in a

given half hour period, allowing for transmission constraints and spinning reserve. The most

expensive plant required in each half hour period would set the so-called System Marginal

Price (SMP) for that half-hour. To this price would be added a capacity credit in order to

ensure that sufficient capacity over and above maximum demand would always be available

for reasons of security of supply. This price was called the Pool Purchase Price (PPP) and

was paid to all generators selected within the half hour regardless of whether they had bid

less.

Inflexible plant such as nuclear would commonly bid in at a very low price or even zero to

ensure being selected. Note that they would still be paid PPP even though they would bid

zero. Combined Cycle Gas Turbines (CCGTs) with fixed gas contracts would also bid low to

ensure being scheduled. The next lowest prices would be operated by the coal-fired

generators which were more flexible but with higher operating costs. Finally, open cycle gas

turbines and pump storage would bid higher due to higher operating costs and greater

flexibility. It can be seen that in general the order of the bids would be in line with the old

merit order system but now prices were meant to be driven down due to the competitive

bidding nature of the Pool. XFigure 139X shows a comparison of the average daily half-hourly

demand and the PPP for November 1999. It can be seen that the trend in prices throughout

the day broadly follows the demand trend during the day. The peaks in prices reflect the more

expensive plant that need to be chosen by NGC to meet the peaks in demand.

The PPP was then „uplifted‟ slightly to cover the difference in the total cost incurred by NGC

in balancing generation and demand and what was paid out to generators through the PPP.


212

This uplifted price was known as the Pool Selling Price (PSP). The PSP is then charged

through to suppliers for the amount of electricity supplied to their customers.

Figure 138: NGC average daily half-hourly demand and Pool Purchase Price for

November 1999.

9.4 Hedging Your Bets

It can be seen from XFigure 139X that there is a wide variation in prices during the day. The

prices shown are merely averages prices during the month. It was quite possible to get price

fluctuations at times of peak demand (around 5:30pm) well over £100/MWh. In certain

severe weather conditions this could go over £250/MWh. For a supplier with a peak „tea-

time‟ demand of 2GW (2000MW) this could mean half-hourly cost fluctuations approaching

£0.25 million! Most suppliers operate to quite tight margins and charge a fixed price to their

customers so a sudden increase in Pool prices over a cold period could potentially bankrupt a

supplier. On the other hand, generators are paying a fixed price for their fuel and similarly a

dip in prices during a prolonged warm period could also spell disaster for their cash flow.

Soon after the introduction of the PSA, so-called „Contracts for Differences‟ came into

existence. These „financial instruments‟ gave generators and suppliers the ability to fix the

price of their generation or demand for a given period of time, typically a month or more. The

contracts did this by „hedging‟ against changes in the Pool prices.

Typically a supplier would forecast its demand profile by half-hour throughout a month

taking into account changes due the seasons, changes due to expected customer losses/gains,

etc. This would be the „hedged volume‟. The supplier would present this profile to a third

party broker which may could be a financial institution. This broker would then offer a fixed

price for the contract reflecting how volatile the broker expected the prices to be in the next

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

0

10

20

30

40

50

60

70

80

90

00:30 02:30 04:30 06:30 08:30 10:30 12:30 14:30 16:30 18:30 20:30 22:30

PPP (£/MWh)

Demand (MW)


213

month. The supplier would still have to pay the half-hourly PSP for demand but at the same

time if the PSP went above the fixed price, the broker would pay the supplier for the

difference between the PSP and fixed price multiplied by the hedged volume. On the other

hand, if the PSP went below the fixed price for any half-hour, the supplier would pay the

broker for the difference between the fixed price and the PSP multiplied by the hedged

volume. In this way the supplier was effectively paying a fixed price for its hedged volume.

Obviously, if the hedged volume was an under-forecast, the supplier would be exposed for

the difference between the forecast and actual demand. If the forecast was too high, then the

supplier might be paying over the odds for the contract, though it was possible to make a

profit with some speculation…

Generator could fix the price they were paid for a hedged volume of generation in exactly the

same way.

Eventually, bespoke contracts became „off-the-shelf‟ contracts with standard terms and

conditions and standard contract types, for example for base load or peaking. These were

known as Electricity Forward Agreements (EFAs). These could be bought and sold on Power

Exchanges.

The importance of such contracts became apparent when the independent supplier

Independent Energy loss its credit rating and could not obtain an EFA finally becoming

bankrupt due to excessively high generation prices in September 2000.

9.5 Deregulation

Immediately following privatisation, consumers could still only buy their electricity from

their local REC. However, in 1992, consumers with a average demand of greater than 1MW

could choose a supplier outside their local area. In 1994, the demand threshold was reduced

to 100kW. Eventually, in 1999, all consumers were able to choose their supplier.

At the same time, small independent generators and „Second Tier‟ suppliers entered the

market, breaking the monopoly of National Power, Powergen, Nuclear Electric and the RECs

(which latterly became known as Public Electricity Suppliers or PESs). This gave an

opportunity for renewable energy generators and specialist renewable energy suppliers (such

as Good Energy and Ecotricity) to enter the market. In addition, the PESs have been

„unbundled‟ so that supply, distribution, metering and meter reading have all become separate

businesses. Over time the larger suppliers like National Power and Powergen have become

fragmented and taken over so that they bear little resemblance to the companies they were at

privatisation. In addition, the National Grid Company, still System Operator and transmission

line operator, became an independent company.

9.6 The Non-Fossil Fuel Obligation and Renewable

Obligation

In order to cover the future decommissioning costs of the nuclear power stations still in the


214

public sector at privatisation, the fossil fuel levy was introduced which was a levy on all

electricity sold. The government of the time also saw this as a convenient mechanism to

support the fledgling renewable energy market in the UK. The vast majority of the levy went

towards the future decommissioning, but a small amount was ring-fenced for the Non-Fossil

Fuel Obligation (NFFO). Developers who wished to build renewable energy generation plant

(wind, small-scale hydro, biomass, etc) could „bid‟ into successive NFFO rounds or

„tranches‟ to in order to build a scheme which would generate for a fixed price. The Non-

Fossil Purchasing Agency (NFPA) ordered the bids and agreed contracts with up to and

including the most expensive scheme in order to fill the order in terms of contracted

generation. There have been 5 NFFO rounds with the prices shown in XTable 17X. It can be

seen that average bid prices have slowly reduced over time as the different renewables

technologies have become more competitive.

NFFO Price (p/kWh)

NFFO 1 (1990) 7.51

NFFO 2 (1991) 8.78

NFFO 3 (1994) 4.84

NFFO 4 (1997) 3.59

NFFO 5 (1998) 2.71

Table 17: Average NFFO contract prices (taken from the NFPA website).

The supplier in which the renewable energy generation project was built was obliged to take

the power from the generator. The fossil fuel levy made up the difference between what the

supplier paid the generator (which was effectively the PSP due to the reduction in demand)

and the bid price for the contract.

The latest NFFO contracts are for 15 years and will continue to run. NFFO has now been

replaced by the Renewables Obligation (RO). This will be the scheme which benefits new

renewable energy generation or existing renewable energy generation whose NFFO contracts

have now expired. The RO places an obligation on all suppliers to source a certain fraction of

their demand from renewable energy sources starting at 3% for 2002/2003 rising to 10% by

2010/2011. If a supplier cannot or will not comply then it can „buy-out‟ its obligation by

paying £30/MWh (2002 prices) for the required generation shortfall. This effectively fixes

the price for the premium paid to renewable energy generators through a system of

certificates known as Renewable Obligation Certificates (ROCs). The proceeds from the buy-

out „pot‟ are given back to compliant suppliers in proportion to their degree of compliance

i.e. how many ROCs they have bought. This can add extra value to the ROCs.

9.7 The Climate Change Levy

In April 2001, the government introduced the Climate Change Levy. This is a levy on all

energy sold to commercial consumers. Electricity from renewable energy sources is exempt

from this levy. The exemption scheme is run using a Levy Exemption Certificate (LEC)

scheme. Each certificate has a value of £4.30/MWh (2001 prices) which can be passed on in

part or full from the supplier to the generator for the electricity that the supplier supplies to its

commercial customers.


215

9.8 The New Electricity Trading Arrangements (NETA)

9.8.1 Background The electricity pool system was felt to have played its part in the successful deregulation and

opening up of the ESI. However, criticisms were made of the way in which pool prices could

be manipulated by a few key generators bidding into the pool. In fact, it was considered that

this „rigging‟ of pool prices was keeping wholesale electricity prices artificially high. In

addition, it was felt that generators and suppliers should be free to enter into bilateral

agreements instead of also buying from the pool, albeit with a bilateral hedging contract

sitting on top. Furthermore, it was seen as beneficial if third parties could enter the market to

trade in physical electricity much as they had been doing in CFDs and EFAs. For this reason,

the New Electricity Trading Arrangements were introduced on the 27th

March 2001.

9.8.2 Buying and Selling Power under NETA Generators and suppliers (above a certain size) are obliged to notify their position for each

half hour in terms of generation and demand for the day ahead to National Grid as system

operator. This allows NGC to carry out its balancing of supply and demand much as before.

Generators now „self-dispatch‟ rather than wait to be dispatched by NGC, though can alter

their output when requested to do so if they participate in the Balancing Market (see below).

Generators still self dispatch in a very similar order to the merit order under the CEGB

regime, but this time the order of dispatch is dictated solely by price. In addition, suppliers

and generators notify what contracts they have struck to the NETA central systems. These

contracts must be notified at least 1 hour1 ahead of actual time, known as „gate closure‟.

Instead of suppliers dealing in CFDs and EFAs to hedge an agreed volume against pool

prices they now trade in physical volumes at a fixed price. Any difference between the

physical volume contracted and actual volume generated or supplied is „cashed-out‟ at prices

set within the Balancing Market.

Suppliers and generators typically strike contracts on different timescales. A month or more

ahead of actual „delivery‟ they will strike „bilateral‟ contracts which is a straight contract

between two parties for a given volume of electricity at a given price. This accounts for 90%

of the volume of electricity bought and sold. Suppliers and generators may not exactly be

able to predict their demand or output exactly months ahead so around a day ahead, they buy

and sell chunks of power on a Power Exchange. This is much like a stock exchange for power

where chunks or power are bough and sold anonymously. This can be done up to gate closure

as mentioned above. Around 5% of electricity is traded in this way. The remaining 5% is

traded through t he cash-out market at the SSP and SBP.

9.8.3 System Sell and System Buy Prices If a supplier‟s actual demand in a half hour is higher than it has contracted for ahead of gate

closure, then the supplier must pay to the system operator the System Buy Price (SBP) for

this deficit. If the supplier‟s actual demand is lower than it has contracted for then it is paid

System Sell Price (SSP) for the excess. The principle is identical for a generator whereby it

will need to pay the System Buy Price for any deficit in contracted generation output and will


216

be paid System Sell Price for any generation excess above contracted output. The System Sell

Price is in general lower than the average bilateral contract price between the supplier and

generator. The System Buy Price is in general higher than the average bilateral contract price

between supplier and generator.

Figure 139: Average daily half-hour System Sell and System Buy Prices for March

2002.

XFigure 139X shows SSP and SBP for each half hour average over the month March 2002. The

SSP is around £10.70/MWh for this month, and the SBP is £23.70, though the SBP shows

rather more variation during the day crudely reflecting the shape of the national demand

curve. The typical contract price for this month would around £15/MWh. It can be seen that

there is an incentive for suppliers and generators to get their forecast and close as possible to

their actual position due to these non-symmetrical prices.

In the early days of NETA, the system or Imbalance prices were far more volatile with SSPs

becoming frequently negative (a supplier or generator would have to pay for any contract

surplus!) and the SBP reached values into the £1000s per MWh on occasion! Prices have now

settled down due to people learning how to play the market and changes to the way SSP and

SBP are calculated in the Balancing Market.

9.8.4 The Balancing Market

The system operator National Grid needs to give orders to generation plant to change their

output in order to match supply and demand just as in the days of the CEGB and the Pool.

This is done using another market known as the Balancing Market. Ahead of delivery,

flexible generators (and to a lesser extent flexible suppliers) can place bids to reduce power

output or offers to generate extra output. In the case of a supplier it would make a bid to

0

5

10

15

20

25

30

35

40

45

50

0 6 12 18 24

£/M

Wh

SSP

SBP


217

increase demand or an offer to reduce demand. After gate closure, NGC examines the bids

and offers and accepts those it needs to balance the system. Suppliers and generators pay their

accepted bids and are paid their accepted offers. Simplistically, the value of the bids sets the

System Sell Price and the value of the offers sets the System Buy Price, though there are

other costs and adjustments which feed into these prices. The total cost of balancing the

system in general is slightly less than the balance of the proceeds for the SSP and SBP. This

is due to the fact that two generators may be out of balance separately and pay for this, but

together their imbalances cancel out. The excess is reallocated to suppliers and generators in

proportion to their total generation output or total demand as appropriate.

9.8.5 The British Electricity Transmission and Trading Arrangements

Up until 2005, the privatised electricity supply industry in Scotland was run separately from

that in England and Wales with different transmission network operators. The Scottish

system had operated a „NETA like‟ trading system even whilst the Pool was still operating in

England and Wales. In April, 2005, NETA was rolled out across Great Britain and the

operation of the whole transmission network came under National Grid. This became known

as the British Electricity Transmission and Trading Arrangements (BETTA).

9.9 The Impact on Renewable Energy Sources

9.9.1 Overview

It was stressed above that there is an incentive to ensure that a generator or supplier gets its

forecast of its position (generation output or demand) as close as possible to the actual

position due to the penal nature of the SSP and SBP. In the case of a generator whose output

is not wholly predictable, this can present a problem. It can be seen that this will be the case

for many renewable energy sources in particular wind power, PV power, wave power and

hydro power. Tidal power though very variable is still highly predictable. Biomass generation

behaves in much the same way as conventional power plant and is thus predictable.

Interesting Combined Heat and Power plant is also at a disadvantage. As it is run to supply

the heat requirement primarily, there is less control over the electricity output.

For a wind farm with an energy contract price of £15/MWh, the cost of not being wholly

predictable under NETA and the effect of the imbalance prices, reduces the value of the

contract by around 20%. Instead of striking a contract, the generator could accept the SSP for

all output spilt, however this would reduce its income to around £10/MWh. It can be seen that

intermittent and unpredictable renewable energy sources are disadvantaged as the cost of

imbalance is charged directly to those that cause it rather than averaged over the whole

market as it would have been under the Pool.

9.9.2 Mitigating the Effects of NETA


218

9.9.2.1 Forecasting

One way in which an intermittent generator can hope to reduce the penal effect of the

imbalance prices is by accurate forecasting. In theory, a generator only needs to forecast his

power output 1h ahead at present. In practice, a small generator such a wind farm or a small

hydro plant will not have the resources to do this every half-hour and strike the required

contract using a Power Exchange. Also it takes time to prepare a forecast so the generator

may have to forecast ~4.5h ahead. There are both infrastructure and transaction costs

associated with doing this which are very high. In practice, many such generators will sell

their output to a supplier for a fixed price and the supplier will take the risk. The supplier

might strike a contract once a day or possible three times a day to balance the output of the

renewable generator. In this case a forecast needs to be available for several hours head. The

simplest forecast used is persistence, where it is assumed that the power output at a given

time ahead is the same as at the present time. This works well for short periods. For wind

power, above 3h a better method is to use a Met. Office forecast or a hybrid of a Met. Office

forecast and a statistical forecast based on the latest historic data.

9.9.2.2 Aggregation

The variability and to some extent the unpredictability can be reduced by adding together or

aggregating the output of several geographically spread generators, e.g. wind farms or hydro

generators. This was discussed in Section X8.4X. Typically, a supplier might contract with

several such generators and pass on the benefits to each generator from aggregating the

output.

9.9.2.3 Storage

The ultimate solution for an intermittent generator would be the use of storage. The storage

medium (a fuel cell or a battery, for example) could buffer the unpredictable variations in

output from the renewable generator to give a constant output. Unfortunately, at present, such

storage is uneconomic. Another option would be to use a controllable generation source

alongside the intermittent renewable generator. An example of this might be the use of a

biomass generator alongside a wind farm whose output was controlled to match (in anti-

phase) the output of the wind farm. Once again, the economics of this are not yet favourable.


219

10 Tutorial Questions

10.1 Wind Power

10.1.1 Wind Characteristics and Resource

1. A site has an average mean wind speed of 8 m/s and an annual standard deviation of

5m/s. Calculate rough estimates of the Weibull parameters k and C. Explain which

formulae were used and why.

2. Take a very simple power curve for a wind turbine specified as follows:

0 - 5 m/s: 0kW

5 -8 : 20

8 - 12 : 35

12 - 14 : 40

14 - 25 : 45

25 - : 0

Calculate the annual energy yield for the above site in Q1 (and also average power) from

the wind turbine if its reliability is 95%.

3. Average wind speed is measured on a mast at 12m above ground level (agl) in

reasonably flat terrain and is found to be 8 m/s. The surrounding ground coverage is

high grass (roughness length 0.08m). Use two expressions to estimate the average

wind speed at 25m agl. Quote any assumptions.

4. An average wind speed of 7 m/s is measured 10m agl on a mast situated in flat terrain

with a uniform covering of short grass (roughness length=0.03m). An estimate is

required of the mean wind speed at 10m agl on top of a nearby isolated ridge also

covered with short grass. The ridge is 10 m high, relative to the surrounding flat

terrain with a half-width at half-height of 20m. Give an estimate of the required mean

wind speed on top of the ridge and quote your assumptions. How would you calculate

an estimate of the wind speed at 20m agl on top of the ridge? What would be your

estimate?


220

10.1.2 Wind Turbines

XFigure 140X show a Cp-λ curve for the fixed-speed stall-regulated „Wind Champion‟ wind

turbine located at a site with average long term wind speed of 7.8 m/s and long term standard

deviation (based on 10 minute averages) of 4.2m/s. The turbine produces rated output at wind

speed U=13m/s and its Cpmax (the peak of the Cp-λ curve) is at 9m/s. The radius of the rotor is

17m. Assume air density is 1.22kg/m3.

1) Calculate the rotational speed of the turbine.

2) Calculate the rated power output of the turbine.

3) Estimate the annual energy production in MWh expected from the turbine assuming

no losses.

4) Describe how a stall-regulated machine is aerodynamically controlled.

Figure 140: Cp- λ curve for the ‘Wind Champion’ wind turbine.

Cp -Lambda curve

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0.22

0.24

0.26

0.28

0.3

0.32

0.34

0.36

0.38

0.4

0.42

0.44

0.46

4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Lambda

Cp


221

10.2 Hydro Power

1. Outline the main requirements of a typical hydro generation project.

2. What are the two basic types of hydro generation and how do they differ?

3. A hydro generation scheme is rated at 100 kW and has a flow rate of 2 m3/s.

Assuming that the density of water is 1000 kg m-3

and acceleration due to gravity is

9.81 ms-2

, calculate the required effective head (neglecting losses). What is the

velocity of the water at the end of the penstock?

4. The hydro scheme in Q3 has a penstock which is 20m long and has a pipe diameter of

1m. The losses in the pipe are 10% of the gross head. Calculate the pipe friction

factor, assuming the mean velocity of water in the penstock is 75% of the velocity at

the end of the penstock.

5. Head losses due to a trash rack are given by sin)g2/v()b/t(Kh 2

0t3

4

. Explain

each of the terms in this equation. Explain how turbulent losses in a trash rack can be

reduced.

6. XFigure 141X shows a flow duration curve at the Dulverton hydro generation site.

XFigure 142X shows the variation of flow with head for this scheme. XTable 18X gives

specific speed ranges for different turbine types. If this scheme had a 100 kW turbine

connected to a generator with a rotational speed of 1500 rpm via a gearbox with ratio

1:10 (one rotation of turbine converted to 10 rotations of generator) decide what type

of turbine would be most suitable for this site justifying your answer (neglect any

losses).

7. What are the main differences between an impulse and a reaction turbine? When is it

appropriate to use each type?

8. What is the principle of a draft tube?


222

Figure 141: Flow duration curve for Dulverton site.

Figure 142: Head-flow curve for the Dulverton site.

0

0.5

1

1.5

2

2.5

3

3.5

4

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40

Gro

ss H

ead

(m

)

Flow (m3/s)

Head-Flow Curve for a Small-Scale Hydropower Scheme


223

Type of Turbine Specific speed range

Pelton one jet 10-35

Pelton two jets 10-45

Turgo 20-80

Cross-flow 20-90

Francis 70-500

Kaplan 350-1000

Propellor 600-900

Table 18: Specific speed ranges for different types of turbine.


224

10.3 Tidal Power

1. With the aid of a diagram, describe the terms „spring‟ and „neap‟ tides and how they

arise. Why are there two high tides per day?

2. Calculate the maximum (spring) tidal range in the deep ocean given that the mass of

the Earth is 5.98 1024

kg, the mass of the Moon is 7.34 1022

kg, the mass of the

Sun is 1.99 1030

kg, the mean radius of the Earth is 6.37 106 m, the mean distance

between the centre of the Earth and the centre of the Moon is 3.84 108 m and the

mean distance between the centre of the Earth and the centre of the Sun is 1.49 1011

m.

3. Describe what phenomena can enhance the tidal range close to a land mass.

4. If a river estuary is 100km long, what is the average depth required for tidal

resonance?

5. With the aid of a diagram, describe the mechanism of two-way generation with flood

pumping when applied to a tidal barrage. What are the benefits and drawbacks of this

mode of operation?

6. An approximate equation for the cost of electricity from a tidal barrage scheme is

given by: cRA

HLkU

2

28.0

1

2loglog . Calculate what this cost will be for the La

Rance barrage, assuming k=0.46 and c=-0.447. For La Rance, the barrage length is

750m, the barrage height is 13m, the basin area is 22km2 and the mean tidal range is

8m. [NOTE: This equation has been slightly varied from the notes to give a better fit

to the data from the 24 sites from which the equation was derived].

7. If the theoretical maximum power output from a tidal current turbine can be

calculated in the same way as for a wind turbine using the equation: 2

3AVP ,

explain why a tidal current turbine of a given rotor size will produce roughly the same

amount of power as a wind turbine of the same rotor size even though water is much

denser than air.


225

10.4 Wave Power

1. Calculate the velocity of a wave in deep water with a wavelength 2m. What is the

speed of this wave in water of depth 0.5m? What is meant by a dispersive wave and

what consequence does this have for deep water waves of different wavelengths as

they race towards a shoreline?

2. Using the equation gDV , explain why wave crests tend to align themselves

parallel to a shoreline, regardless of the shape of the shoreline.

3. XFigure 143X shows a time trace of a series of waves passing a given measuring point

over 175s. From this trace, calculate the power density (in kW/m) of these waves

given that the significant wave height (Hs) is 13.7m.

4. Using the equation d

F 4exp1 , explain how the energy in a wave varies

with depth. Using a sketch illustrate this showing how a molecule of water behaves

with increasing distance below the water surface.

5. For a Salter‟s Duck explain at least two important design characteristics. What does

the term „impedance matched‟ mean for the Duck?

6. An Oscillating Water Column (OWC) contains a Well‟s turbine. Explain the principle

of an OWC and why a Well‟s turbine is used in this device. An OWC is designed

with a water path length of 25m. Calculate the resonant oscillation period for this

device.

7. What advantages does a Tapchan wave energy converter have over an OWC?


226

Figure 143: Time trace of a series of waves passing a given measuring point during a

period of 175s.


227

10.5 Solar Power

1) What energy is associated with a photon of yellow light (wavelength 600 nm)?

2) Explain the role of reflection, transmission, and absorption when radiation is incident on

an encapsulated PV cell.

3) If the sun can be represented as a black body at 5780K, calculate the intensity of radiation

emitted from its surface.

4) The eccentricity of the sun‟s orbit is 365

360cos033.010

nE . Use this equation to

show that the variation in the extraterrestrial radiation is approximately 3%.

5) Describe the meaning of Air Mass, and give some key example values.

6) At a particular time of day, the beam radiation component is 500 W/m2. A PV module of

0.4 m2 is installed at the site tilted at 40

o, and at this time the incidence angle is 25

o.

i) If the efficiency of the module under these conditions is 17%, calculate the

electrical output if the other radiation components are insignificant.

ii) On a different, cloudy day, there is no beam component. If the global radiation (on

a horizontal surface) is 250 W/m2

calculate the output from the module if the

efficiency under these conditions is 10% and the albedo is 0.1.

7) Explain why the region around the cell junction is sometimes referred to as the depletion

zone.

8) What is the Fermi energy? Where does it lie in the bandgap for the p and n regions of the

cell?

9) What is the dark current of a PV cell and what equation is used to model it?

10) The output current for a PV cell can be expressed as:

1 -)R . I + (VAkT

q I - I = SLOG exp

Explain all the terms in this equation and draw the associated equivalent circuit. Derive

from this equation an expression for the open circuit voltage, VOC.

11) Draw a typical I-V curve for a crystalline silicon cell. Mark the maximum operating

point. What is the effect on efficiency of increasing temperature on cell output? What is

the effect of reducing radiation? Draw appropriate I-V curves to illustrate the effects.


228

12) What are Standard Test Conditions ? What is the definition of fill factor?


229

10.6 Biomass Power

1. Write down the chemical equations by which solar energy is stored by plant material

and how it is released as heat. What is meant by the term „CO2 neutral?

2. A quantity of wet wood has a moisture content on a wet basis of 25%. What is its

moisture content on a dry basis? Which measure of moisture content is preferred and

why?

3. How is willow from short rotation coppice pre-treated before direct combustion and

why?

4. What are the three stages in direct combustion of biomass and what temperature

ranges do they correspond to?

5. Explain the importance of air-flow in direct combustion.

6. Draw a diagram to explain the principle of a suspension burner. Why is it more

efficient than a grate burner for direct combustion?

7. What are the three basic methods used in the gasification of biomass material? What

is the effect of steam when used in the gasification process? What is the advantage of

gasification over direct combustion?

8. Explain the principle of pyrolysis and describe what products it produces when

applied to a biomass material such as wood. What is the effect of temperature on

pyrolysis?

9. What are the limitations of a steam engine for power generation? What are the two

basic types of steam turbine?

10. Draw a schematic diagram of a combined cycle gas turbine (CCGT) plant and explain

the basic principle of its operation. What typical efficiency can be achieved by such a

plant?

11. What is a combined heat and power plant and what advantage does it have over a

CCGT? What disadvantage does it have?


230

10.7 Integration

1. What is the „predictability‟ of wind, solar, wave and tidal power in the short

(minutes), medium (hours) and long (days or weeks) term?

2. Why is demand forecasting necessary in a power system with a large fraction of

thermal power generation? What is the minimum level that a coal-fired power station

can be loaded?

3. What is spinning reserve and why is it important in a power system?

4. In what order would the following types of power plant be dispatched if they were

following a merit order: coal-fired, CCGT, wind, open cycle gas turbines, hydro pump

storage, nuclear? Justify your answer. When are the two major demand peaks in the

day?

5. Assuming statistical independence and normal statistics show that the fractional

variability of two consumers each with mean consumption μ1 and standard deviation

of consumption σ1 is reduced when the two are added together. This phenomenon is

known as aggregation. Why is aggregation important for wind farms connected to a

large power system and how might it be achieved?

6. Explain the concept of „penetration penalty‟ when applied to wind power. What are

the major causes of this penalty?


231

10.8 Electricity Trading

1. What is the principle of an electricity Pool system such as existed in England and

Wales between 1990 and 2001?

2. Explain the concept of a Contract for Differences. How have such contracts changed

under the New Electricity Trading Arrangements (NETA)?

3. What are the System Sell and System Buy prices under NETA?

4. What mechanism does National Grid (the System Operator) use to balance supply and

demand under NETA?

5. Explain how suppliers and generators contract for electricity and why there is an

incentive to forecast output/demand as accurately as possible under NETA.

6. Why are wind farms and combined heat and power plants disadvantaged under

NETA? How can these disadvantages be mitigated?

7. Name two financial support mechanisms in the UK for renewable energy generation.

Combined Notes for the Lecture Series

Documents

mont st michel

renewable

european wind

wound rotor

compression

wave energy

wind speed

operate ancillary