Probabilistic Economic Analysis Wind EnergyKenHumphreyBlairWalter

7/30/2019 Probabilistic Economic Analysis Wind EnergyKenHumphreyBlairWalter

http://slidepdf.com/reader/full/probabilistic-economic-analysis-wind-energykenhumphreyblairwalter 1/23

Mesoscale Modeling and ProbabilisticEconomic Analysis



2

Contents

• Project Principles

• Alternative Wind Assessment Technologies

- mesoscale modeling

- quantifying uncertainty

• Economic decision making under uncertainty

- limitations of traditional economics

- probability based decision making

• Conclusions



3

Projects aim to reduce risk / increase certainty

through development to financial close

Wide distributions of outcomes

when project first considered

Narrow distribution of outcomes

required for financial close

Project management to cost effectively move from;

•Concept to feasibility / consenting

•Feasibility / consenting to detail design

•Detail design to financial close / construction



4

Contents








• Conclusions



5

The initial assessment of wind resources

only justifies the feasibility study

Yes

No

Is

there enough

wind over

the site

Feasibility

study

Is thesite worth

consenting



6

There are a number of techniques to

assess wind resource

Financialclosure

+/- 3%$200k12 monthsHub Height WindMast

InitialAssessment

+/- 10% for80m wind

$50k12 monthsProspecting 30mwind mast

All stages+/- 5%$10kWeeksMesoscalemodelling

Best UseAccuracy for

80massessment

CostTimeType of

Measurement



7

Mesoscale modeling is recommended

for initial site economic evaluation

WRF Validations

5.0

6.0

7.0

8.0

9.0

10.0

11.0

12.0

13.0

5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0

Measured mean wind speed (m/s)

M

o d e l m e a n w i n d

s p e e d ( m / s )

Onshore

Offshore

Trendline



8

Model period can be selected to

represent long-term wind climate

3

4

5

6

7

8

9

10

11

- D Q

- D Q

- D Q

- D Q

- D Q

- D Q

- D Q

- D Q

- D Q

- D Q

- D Q

- D Q

- D Q

- D Q

- D Q

- D Q

0RQWK

0 H D Q : L Q G 6 S H H G P V

Monthly mean wind speed and 12-month moving average for long-term NCEP data



9

Mesoscale modeling gives a good overall view

of sites for conceptual turbine placement



10

High resolution modeling provides

insight to site dynamics

6.2

13.7

17.5 17.3

15.7

12.3

7.9

9.4

0.0

2.5

5.0

7.5

10.0

12.5

15.0

17.5

20.0

%

Wind Class Frequency Distribution

Wind Class (m/s)

Calms 0.1 - 2.0 2.0 - 4.0 4.0 - 6.0 6.0 - 8.0 8.0 - 10.0 10.0 - 12.0 12.0 - 14.0 >= 14.0

NORTH

SOUTH

WEST EAST

4%

8%

12%

16%

20%

WIND SPEED

(m/s)

>= 14.0

12.0 - 14.0

10.0 - 12.0

8.0 - 10.0

6.0 - 8.0

4.0 - 6.0

2.0 - 4.0

0.1 - 2.0

Calms: 0.00%



11

Wind assessment (mesoscale or alternatives)

contains uncertainties

Maximum

Wind SpeedMinimum

Wind Speed

+ One

Std Dev-One

Std. Dev

Mean

Within

One

std dev

66% of

time

m/sec @ 80m

8.1 9.5 10.9

Uncertainties•Measurement accuracy

•Vertical extrapolation

•Long-term correction

•Flow modeling

Together, these

amount to 15%uncertainty

eg



12

…. which can be expressed in terms of

probabilities

Probabilities

•P50 of 9.5 m/sec

•P75 of 8.5 m/sec

W ind Speed a t 8 0 m

0

0.1

0.2

0.30.4

0.5

0.6

0.7

0.8

0.9

1

4

. 4 5

4

. 9 4

5

. 4 3

5

. 9 1

6

. 4 0

6

. 8 9

7

. 3 8

7

. 8 7

8

. 3 5

8

. 8 4

9

. 3 3

9

. 8 2

1 0

. 3 1

1 0

. 7 9

1 1

. 2 8

1 1

. 7 7

1 2

. 2 6

1 2

. 7 5

1 3

. 2 3

1 3

. 7 2

1 4

. 2 1

m / se c

C u m

u l a

t i v e P r o b a b i l i t y

P75 means that there is a 75% probability that the annual average wind

speed will be greater than 8.5 m/sec.



13

Contents








• Conclusions



14

Feasibility assessment involves a number of

assumptions and uncertainties.

Assumptions Typical Uncertainty level

Wind Speed +/- 15%

Turbine Costs +/- 25%

Euro / NZ$ +/- 25%

O&M costs +/- 25%

Electricity Price +/- 20%

Wind speed

Accuracy of

+/- 15% appearsadequate for an

initial economic

assessment



15

Lets consider a site which could take up to

25 large 3 MW turbines

The question we are generally faced with is whetherthere is sufficient wind resources to justify negotiating aland agreement, measuring wind speed at hub heightand undertaking feasibility engineering ahead ofproceeding with a resource consent application.

Cost to

Find out

Value of

Project



16

Traditionally, we would make estimates of

economic variables and calculate value

Assumptions Source

• Wind Speed Mesoscale model

• Turbine Cost Manufacturer

• Euro / NZ$ Treasury

• O&M costs Manufacturer

• Electricity Price Consultant

In this example, the project looks OK with a 8.3 % IRR

Project Summary

Number of Turbines 25 V90

Installed Capacity 75 MW

Gross Annual Generation 305 GWh

Capital Cost 230,104,875 NZ$

Installed Cost 3,068 NZ$ / kW

Internal Rate of Return post tax 8.34 %

Net Present Value post tax* 5,112,576 NZ$

* Nominal After Tax WACC 8 %

NZWEA Mesoscale



17

… but we really need to recognize and quantify

the uncertainty in all assumptions

0.25n/a2.0Euro/NZ$

Min 1.0

Max 2.0

Min 2.25

Max 3.75

Min 8

Max 12

n/a

Min / Max

n/a1.37O & M Costs

(c/kWhr)

n/a3.0Turbine Cost

(Euro m))

n/a10Electricity Price

(c/kWhr)

1.429.5Wind Speed(m/sec)

St DevMeanDistributionFactor



18

Firstly, we can get a feel for the gross

energy yield from the development

Probabilities

•P50 of 305 GWhrs

•P75of 258 GWhrs

Cumula tive Probability for GW h

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

5 8 .

4 4

7 7 .

8 6

9 7 .

2 7

1 1 6 .

6 9

1 3 6 .

1 0

1 5 5 .

5 2

1 7 4 .

9 3

1 9 4 .

3 5

2 1 3 .

7 6

2 3 3 .

1 8

2 5 2 .

5 9

2 7 2 .

0 1

2 9 1 .

4 2

3 1 0 .

8 4

3 3 0 .

2 6

3 4 9 .

6 7

3 6 9 .

0 9

3 8 8 .

5 0

4 0 7 .

9 2

4 2 7 .

3 3

4 4 6 .

7 5

GW h



19

Then, we can describe project economics in

terms of probabilities

Probabilities

•P50 8.3% IRR

5 NPV(8%)

•P75 6.5% IRR.

(22) NPV(8%)

Cumula tive Probability f or I RR

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

- 1 .

0 0

- 0 .

1 0

0 .

8 0

1 .

7 0

2 .

6 0

3 .

5 0

4 .

4 0

5 .

3 0

6 .

2 0

7 .

1 0

8 .

0 0

8 .

9 0

9 .

8 0

1 0 .

7 0

1 1 .

6 0

1 2 .

5 0

1 3 .

4 0

1 4 .

3 0

1 5 .

2 0

1 6 .

1 0

1 7 .

0 0

I RR



20

Now our question can now be answered in a

somewhat more complete manner

Cost to

Find out

$1 million estimate

Value of

Project

P50 of NPV $ 5m

P75 of NPV $ (22)m

In this example, there is only a 55 % probability that the projectvalue will exceed $1 million (the cost of feasibility engineering)



21

Contents








• Conclusions



22

In practical terms, these techniques have

been used successfully in New Zealand

• At least 12 small wind farm sites have been evaluated with mesoscalemodeling; with

- two projects proceeding without further investment in windassessment (HorseShoe Bend and Weldcone)

- one project about to install a 30m wind mast to verify particular localtopographic effects

- a number of other projects under consideration by developers

- two projects rejected outright without any further investigation.

• Costs and timescales are modest - average cost for mesoscalemodeling with feasibility study is $10,000 to $15,000 in 6-8 weeks.



23

In summary, mesoscale modeling and decision

making under uncertainty has its place

• Mesoscale modeling is good for initial assessment andscreening studies.

• Physical measurement appropriate when more detailedunderstanding of site features and energy yield is needed

• Hub height measurement is needed for financing

• Probabilistic economic analysis has its place throughout

project development to highlight uncertainties of inputsand outcomes; to refine assumptions and to engage withthe financial community.

Probabilistic Economic Analysis Wind EnergyKenHumphreyBlairWalter

Documents