7/31/2019 ECEN 5007 - Lecture 10
1/50
ECEN 5007 SOLAR THERMAL POWER PLANTS
Lecture 10: Design and Optimization of CSTP PlantsJuly 31, 2012
Manuel A. Silva, Dr.Ing. - Manuel J. Blanco, Ph.D., Dr.Ing.
TWTH 17:00-19:30 - Class Room: ECCR 1B55
Office Hours: TWTH 15:30-16:30
7/31/2019 ECEN 5007 - Lecture 10
2/50
5 Optimization of plant configuration
THE DESIGN PROCESS
Optimization ofCSTP plant
The Need ofSimulations
The Design Process
Advanced DesignMethods
Classic DesignCriteria
Advanced DesignCriteria
7/31/2019 ECEN 5007 - Lecture 10
3/50
High number of parameters
q Optical, thermal, hydraulic properties of each component,q Efficiency curves, parasitic consumptions, etcq Size and cost of each componentComplex physical models
q Probabilistic (Monte-Carlo) or convolute optic models,q Thermo-hydraulic models,q High variability of input parameters (meteorological data).Operational strategies
q Defocusing strategy in the solar field,q Backup boiler usage,q Thermal storage system usage,q Start-Up and Cool-down strategies,q Production optimization with respect to economic criteria
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
4/50
Benefits / Uses of simulation:
q Estimating the final cost of energy from technological data.q Comparing amongst different technologies without the need to build everything.q Setting the next steps in R&D in order to reach a competitive technology.
Where shall we put the effort?q Translating a technological improvement into the final impact in the energy price.q Finding technological niches.Drawbacks / Dangers of simulation:
q Self-fulfilling prophecies.q Shape reality according to the model restrictions.q
Ignoring the accuracy of the model (hypothesis).
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
5/50
Key decisions:
q Solar technologyq Nominal electric powerq Thermal storageq HybridizationDesign parameters :
q Design Point: Instant of time and conditions for which the plant is designed to produceits nominal electric power.
q Solar multiple: Ratio of actual solar field size to the minimum size required to run aturbine at full capacity at design point.
q Capacity factor: Ratio of the actual output of a power plant over a period of time andits output if it had operated at full nameplate capacity the entire time
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
6/50
Classic design criteria
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
7/50
Design point
q Definition Instant of time (day of the year, and time of the
day) and external conditions for which the CSTP
plant is assumed to operate at nominal conditions.
q Information Needed DNI (850 950 W/m2) Date and time - Sun position - Optical efficiency
(cosine factor, shadowing, and blocking)
Ambient temperature (25 30 C) Relative humidity
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
8/50
Solar multiple
Available Solar Power
Solar Field Thermal Power
Gross Electric Power
Defocused Power
SM< 1
SM= 1
SM> 1
~ 1.2
SM> 1
~ 1.5
SM> 1
~ 2
PowerBlock
SolarField
Q
Q
Nominalthermal,
Nominalthermal,
SM =
q Definition Ratio of actual solar field size to the minimum size required to run a turbine at
full capacity at design point.
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
9/50
q Definition Ratio of the actual output of a power plant over a period of time and its output
if it had operated at full nameplate capacity the entire time.
Capacity factor
h8760ctorCapacityFa
Nominal
=
werElectricPo
ergyElectricEnYear
Online: 37%
Offline: 63%
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
10/50
LEC (Levelized Electricity Cost)
q Definition 1 Ratio of the total cost associated with
building and operating a CSTP plant and
the energy produced by that plant overits lifetime.
q Definition 2 Present price at which the electricity
produced by a CSTP plant must be sell sothat the Net Present Value associated to theCSTP investment project is zero.
UsefulLifeAnnualElectric
UsefulLifeAnnual
YearsE
YearsOPEXCAPEXLEC
+
= 0%
10%
20%
30%
40%
50%60%
70%
80%
90%
100%
2 01 0 2 01 1 2 01 2 2 01 3 2 01 4 2 01 5 2 01 6 2 01 7 2 01 8 2 01 9 2 02 0
LCOE
Year
EvolutionoftheLCOEforthedifferentCSTPtechnologies
ParabolicTrough(Max.) ParabolicTrough(Min.) Fresnel(Max.)
Fr esne l(Min.) Po we rTowe r( Max .) Po we rTowe r( Min .)
Dish-Engine(Max.) Dish-Engine(Min.)
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
11/50
CONCEPTDESIGN
TECHNOLOGY HTF
TESHYBRIDIZATION
POWERBLOCK
BOUNDARYCONDITIONS
DESIGN POINT
q Location and MDYq CSTP Technology
PT/ Tower / Working fluid
q Thermal Storage System (TES)q Hybridizationq Nominal Powerq Solar Multiple
Main characteristics
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
12/50
OPTIMIZATION
DESIGN POINT
RECEIVERSOLAR FIELD
SIZE & LAYOUTCONCENTRATING
TECHNOLOGY
OPTIMIZED DESIGN
PERFORMANCEEFFICIENCY
MATRIX
CRITERIA:CAPACITY FACTOR
LEC
q Central Receiver Heliostats Type Number Layout Receiver Type Tilt angle Tower
q Parabolic Trough PT Type Loop Number of Loops Layout (sub-fields) Heat Collection Element
Solar field
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
13/50
IS ITOK?
EFFICIENCY MATRIXEFF= F (AZ,EL)
DETAILEDMETEOROLOGICAL
DATA (TMY)
THERMAL STORAGE
TURBINE
DETAILED ANNUALEFFICIENCIES
LEC CALCULATION
FINAL DESIGNCONCEPT
RE-CONSIDERATION
YES NO
qAnnual Simulation Very Simple Models
(Efficiency Matrix)
Not a design parameter Checking the adecuation of the
design
Example: Luzergy, Solergy :
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
14/50
Advanced Criteria
q Four main subsystems: Solar radiation collection and concentration Thermal conversion of the solar energy Electric conversion of the thermal energy Thermal storage
q Desirable technical goals: Minimize optical losses Minimize thermal losses Maximize power block efficiency Minimize TES thermal losses Optimize plant operation
q Main goal: optimize economic performance Trade-off between efficiency and cost
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
15/50
mxCET CET,mx
q The design of the different sub-systems Is made separately for each one. Is optimized just for a unique design point. Is checked or adapted slightly for a few off-design points.
q The CSTP plant actually Works as a whole (the different systems are interrelated). Works very often in off-design conditions.
q The maximum efficiency at design point of each of the sub-systems does not guarantee the maximum efficiency of thewhole plant.
Holistic Approach
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
16/50
q Which is the actual goal of a commercial CSTP plant? Producing the maximum possible energy (Capacity Factor). Producing the energy at the lowest price (LCOE). Earn the maximum money...
qActual Goal: Maximum economic efficiency/profitability (cost effective). Usually working in off-design conditions.
q The optimum energy efficiency of the CSTP plant does notguarantee the optimum economic efficiency of the project.
The main goal is the maximum economic efficiency.Depends not only on plant cost and performance, but alsoon external factors..
Holistic Approach
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
17/50
q The profitability of a CSTP plant strongly depends on itsenvironment:
Legal Framework Market system (premium, PPAs) Financing (Bankability)
q Design should be adapted to the feasible and availablebusiness models in each situation.
New design criteria
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
18/50
California, 2000
Dispatchability:
q Thermal Energy Storageq Hybridization
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
19/50
INDICATORS USED FOR TECHNOLOGY ASSESSMENTCriteria Indicator Comments
Production costs LCOE ($kWhe) Levelized cost of electricity.
Capital costs CAPEX ($MWe) Investment costs of the project.
O&M costs OPEX ($MWhe) Operation and maintenance cost.
Financial return Modified IRR (%), Specific NPVModified Internal Rate of Return,Specific Net Present Value.
Financial risk MIRR
Standard deviation of the Modified IRR.
Cost reductionpotential LCOE potential reduction (%) Expected LCOE reduction
Plant efficiency Solar-to-electric net efficiency (%) Mean annual net efficiency of the plant.
MaturityInstalled capacity worldwide (GW), No. ofequivalent operating hours
Commercial plants in operationworldwide.
Developmentperspectives Project pipeline worldwide (GW) Commercial plants projects worldwide.
Local marketcompetitiveness
Share of local manufacturing in totalinvestment costs (%) From local suppliers information
Land use Land needed per energy produced (km2/MWh) Land occupied by the whole installation
Water demandWater consumption per energy produced(m3/MWh)
Consumption for cycle cooling andmirror cleaning.
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
20/50
q The way a CSTP plant is operated.q Routine operation definition:
Start-up and cool-down procedures DNI, Temperatures, mass flows(solar field, TES, PB) Recirculation while Start-Up / Cool-Down
Control Systems Maintenance/cleaning planning, etc
q This allow to: Reach a stable production on varying conditions. Produce energy at optimum conditions (highest possible efficiency).
At full load At partial loads
Schedule the production, up to some limits
Optimization of CSTP plants
Operationalstrategies
7/31/2019 ECEN 5007 - Lecture 10
21/50
q They severely affect the performance of the CSTP plant in the
long term.
q They affect the whole system:Energetic efficiency of each directly related sub-system.Efficiency/performance of the rest of the sub-systems.
E.g. The operating temperature of the solar fielddetermines the efficiency of the power block.
Dispatchability.Economic profitability.
q Main fields:Solar field operation.Thermal Energy Storage usage.Hybridization usage.
Operationalstrategies
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
22/50
Advanced Methods Probabilistic approach:
q Uncertainties in simulation parametersq Deterministic Vs. Stochastic simulationsq Probabilistic distributions of the results
Optimization of CSTP plants
7/31/2019 ECEN 5007 - Lecture 10
23/50
q The design of a CSTP plant is a complex exercise.Careful consideration should be given to the solar
resource assessment, investment costs estimates, andtechnology performance estimates.
Software tools are needed to assist in different stagesof the design process, from the definition of the designmeteorological year, to the energy yield estimates of theCSTP plant.
In most cases trade-off are needed between energyefficiency and cost.
Optimization of CSTP plants
Conclusions
7/31/2019 ECEN 5007 - Lecture 10
24/50
Overview of available computer tools
7/31/2019 ECEN 5007 - Lecture 10
25/50
SolTrace:
Monte Carlo ray tracer Optical simulation program of a great variety of solar concentrating systems Proprietary, non-flexible, non-expandable
1 Optical analysis: SOLTRACE
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
26/50
Tonatiuh:
Monte Carlo ray tracer Optical simulation program of a great variety of solar concentrating systems Open source, easy to use, expand, adapt, and maintain
1 Optical analysis: TONATIUH
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
27/50
C++ object oriented Monte CarloRay Tracer
Plug-in architecture. Operating system independent State-of-the-art GUI Open source
1 Optical analysis: TONATIUH
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
28/50
Tonatiuh:
Available at: http://code.google.com/p/tonatiuh/ Web infrastructure support for developers and users
Developers blog
Users groupMain web site
Moderator
Video channel
1 Optical analysis: TONATIUH
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
29/50
Power at the target
DifferencefromT
onatiuhs
estimate(%)
Parabolic trough
TonatiuhSolTrace
OVERVIEW OF AVAILABLE COMPUTER TOOLS
1 Optical analysis: comparison between SOLTRACE and TONATIUH
7/31/2019 ECEN 5007 - Lecture 10
30/50
Frequency distribution of photons
TonatiuhSolTrace
Parabolic trough
1 Optical analysis: comparison between SOLTRACE and TONATIUH
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
31/50
10 20 50 100 200 500 1000
0
50
100
Thousand rays
Maximum flux density
DifferencefromT
onatiuh
sestimate(%)
TonatiuhSolTrace
OVERVIEW OF AVAILABLE COMPUTER TOOLS
Parabolic trough
1 Optical analysis: comparison between SOLTRACE and TONATIUH
7/31/2019 ECEN 5007 - Lecture 10
32/50
Parabolic Trough Incident Angle Modifier
Sun Elevation: 5
1 Optical analysis: usage example with TONATIUH
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
33/50
Parabolic Trough Incident Angle Modifier
Sun Elevation: 10
1 Optical analysis: usage example with TONATIUH
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
34/50
Parabolic Trough Incident Angle Modifier
Sun Elevation: 20
1 Optical analysis: usage example with TONATIUH
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
35/50
Parabolic Trough Incident Angle Modifier
Sun Elevation: 30
1 Optical analysis: usage example with TONATIUH
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
36/50
Parabolic Trough Incident Angle Modifier
Sun Elevation: 40
1 Optical analysis: usage example with TONATIUH
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
37/50
Parabolic Trough Incident Angle Modifier
Sun Elevation: 50
1 Optical analysis: usage example with TONATIUH
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
38/50
Parabolic Trough Incident Angle Modifier
Sun Elevation: 60
1 Optical analysis: usage example with TONATIUH
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
39/50
Parabolic Trough Incident Angle Modifier
Sun Elevation: 70
1 Optical analysis: usage example with TONATIUH
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
40/50
Parabolic Trough Incident Angle Modifier
Sun Elevation: 80
1 Optical analysis: usage example with TONATIUH
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
41/50
Sun Elevation: 90
Parabolic Trough Incident Angle Modifier
1 Optical analysis: usage example with TONATIUH
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
42/50
DELSOL / WINDELSOL:
CRS field design Cone-optics and convolution approach, simplified economic optimization
2 Solar field optimization and analysis: DelSol/WinDelSol
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
43/50
New Solar Plant Optimization Code (NSPOC):
CRS field design and analysis Combination of optimization algorithms Cone-optics and convolution approach, www.nspoc.com
2 Solar field optimization and analysis: NSPOC
OVERVIEW OF AVAILABLE COMPUTER TOOLS
Fig. 3. Radiation flux maps for external and cavity receivers
7/31/2019 ECEN 5007 - Lecture 10
44/50
HOpS: the Heliostat Optical Simulation
Open-source tool just released by Google to analyze the optical behavior of heliostat fields overthe course of a year with ten-minute granularity.
2 Solar field analysis: HOpS
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
45/50
System Advisor Model:
PT CRS LFR Parabolic dishes PV & others
3 Plant performance: SAM
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
46/50
System Advisor Model:
Sensitivity analysis of operational strategies User friendly
3 Plant performance: SAM
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
47/50
System Advisor Model:
Integrated financial analysis Integrated probabilistic modeling Specialized in the USA regulative framework
3 Plant performance: SAM
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
48/50
EOS (goddess of the dawn):
Developed by GTER Parabolic troughs Thermal oil, DSG, (molten salts) Physical model of the solar field Empirical model of the power cycle Quasi-stationary, 10-min step by default Detailed calculation of mass flows and
temperatures
3 Plant performance: EOS
OVERVIEW OF AVAILABLE COMPUTER TOOLS
0
50
100
150
200
250
300
350
400
450
TemperaturadelHTF[C]
TLE(TiempoLocalEstandar)
En tra da la zos Sa lid a lazos Retorn ocamp o Impu lsin ca mpo
-600
-400
-200
0
200
400
600
800
Caudalde
HTF[Kg/s]
TLE(TiempoLocalEst andar)
CampoSolar CalderaGN Sistemadealmacenamiento Generadordevapor
7/31/2019 ECEN 5007 - Lecture 10
49/50
Optical analysis Helios Mirval
Optimization andanalysis HFLCAL
Plant performance SimulCET Solergy TRNSYS based
codes
3 Other codes
OVERVIEW OF AVAILABLE COMPUTER TOOLS
7/31/2019 ECEN 5007 - Lecture 10
50/50