Thermal Resistance Model
for CSP Central Receivers
by Oelof de Meyer
Email: dmeyeroe@eskom.co.za
Supervisors
Prof. Frank Dinter
Email: Frankdinter@sun.ac.za
Prof. Saneshan Govender
Email: GovendS@eskom.co.za
14 July 2015
Presenter Oelof de Meyer
Qualifications 2010, B.Eng. (Mechatronics), Stellenbosch University
2012, M.Sc. (Electrical Eng.), University of Cape Town
Work Experience 2011, Eskom : Control & Instrumentation
• Matimba C&I Refurbishment,
2011/11 - 2012/06
• Lead Discipline Engineer for:
• Eskom 100MW CSP Plant
2011/11 - present
• Solar Augmentation
2013/10 - present
• SERE Wind 100MW
2014/03 – present
Further Studies 2014, Eskom Power Plant Engineering Institute (EPPEI)
• PhD Program – Stellenbosch University
2
173 500 heliostats
140m tower
Ivanpah, 337 MW
Introduction : Molten Salt Central Receiver
10MW Solar Two
Molten Salt Central Receivers
2015/08/04 Gemasolar
20MW
110MW Crescent Dunes
5
PhD Research Topic
Operating Strategy and Philosophy
optimisation for a 100 MW CSP Plant
Operating Philosophy
(Process Plant)
Operating Strategy
(SA Grid)
Plant Model
System Operator
Solar Field &
Weather Data
Receiver
Model
Grid
2015/08/04
Thermal
Energy
Storage
Power
Block
6
PhD Research Topic (In Short)
• Power Purchase Agreement (PPA) initials “Operating Strategy”
• Agreement between Independent Power Producer (IPP)
and Utility / System Operator
• IPP design, build and optimise plant to adhere to PPA
• IPP business is money,
• thus maximise power production
• Minimise energy losses
• Minimise O&M costs
• = Max Revenue
For Example:
PPA = 75 MW between 4-10 pm
Thus storage of about 6 hours
Start/Stop turbine each day
Etc...
“Design Plant accordingly”
• What happens if you have a varying operating strategy?
• “Flexible Plant required”
• Optimisation of plant operation
Power Purchase
Agreement (PPA)
Weather Data
Plant Status
Operating Strategy
PhD Research Topic
Operating Strategy and Philosophy
optimisation for a 100 MW CSP Plant
Operating Philosophy
(Process Plant)
relationship
Plant Model
Solar Field &
Weather Data
Receiver
Model
*Done: Sep 2014
2015/08/04
Thermal
Energy
Storage
Operating Strategy
(SA Grid)
System Operator
*Done: Jan 2015
*Done:
June 2015
Power
Block
Grid
8
Model Capabilities
Solar Field &
Weather Data
Evaluate:
- Aiming Strategy used
- Receiver design/materials
- Tube strain per panel, O&M
12x10 Receiver Flux Map
Temperatures:
- Inner tube
- Receiver surface
- HTF outlet temperature
TMY Data
- Ambient Temperature
- Wind speed
- DNI etc.
Receiver Efficiency
- Receiver heat losses
- HTF thermal energy gained
- HTF mass flow rate
DELSOL3
Evaluate:
- Receiver design/materials
- Pressure drop across receiver
How does the model work?
Receiver design & configuration:
Height
19.24 m
Diameter 16.32 m
Panels
16
Tube diameter
50 mm
Tube thickness 1.5 mm
Flow regime:
- 2x flows with cross over halfway
- HTF enters from South panel
- HTF exit through North panel
Step 1:
Determine HTF mass flow rate
in flow regimes by determining:
- Heat loss per panel
- Heat gained per panel
*Initial surface temperature guess
values required
How does the model work?
Step 3:
Use bulk fluid temperature to determine:
- Inner tube temperature
- Outer tube temperature (surface)
- Corresponding heat loss
- Radiation, convection
Step 2:
Use HTF mass flow rate to determine
temperature rise in HTF
*Steady state model requires iterations due to
initial surface temperature guess values
Results
Morning, Mid-day and Afternoon – Case Studies
2015/08/04
14
Conclusion
You may think... Yes, results are nice
and pretty, but surely someone
developed this model already?!
YES!
Similar models are being
developed to analyse
- Receiver design
- Receiver material
- Pressure drop calculations
- Tube-strain of panels
- Etc...
- My overall research requires a model with the level of
similar to results on a basic design.
- The model will be assigned some intelligence during the
operating philosophy optimisation phase
- My model is not a “black box”
• Know what is going in and out – verified!
15
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Private Use
Free Lisence
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
-
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Commercial Lisence
X
System Plant Simulation
Performance
Heat Transfer Fluid
Analysis
X
Power Block
Finite Modelling
Thermal
X
Optical Analysis
Ray-Tracing
Finite Modelling
Mechanical
X
Lisence
Heliostat Field Design
Optimisation
Feasibility
Technology
Feasibility
Economic Impacts
X
X
X
X
Heliostat Field
Optimisation
X
Simulation
Finite Elements
X
Capabilities
Ray-Tracing
Feasibility
X
Software Platform
ANSYS
ASAP
ASPOC
COMSOL Multiphysics
DELSOL3
EnerTracer
EPSILON
ESEMflex
ESEMpro
ESOM
Fiat Lux
GateCycle
Greenius
Helios
HFLCAL
IPSEpro
NowCasting
NSPOC
OPTEC
OptiCAD
SAM
SCT
SenRec
SENSOL
SOLERGY
SolTrace
SOLUGAS
SolVer
ThermoFlow
TieSol
Tonatiuh
TracePro
TRNSYS (STEC)
2015/08/04
Visual HFLCAL
WINDELSOL
Heat Balance
Computational
Fluid Dynamics
Purpose
X
X
X
X
X
Source
ANSYS
Breault
Nevada Software
COMSOL
SANDIA
CIEMAT PSA
STEAG, DLR
Sun to Market
Sun to Market
Sun to Market
CIEMAT PSA
GE
DLR
SANDIA
DLR
SimTech
Sun to Market
Nevada Software
OptiCAD
NREL
CIEMAT PSA
SENER
SENER
SANDIA
NREL
DLR
Solucar / Abengoa
ThermoFlow
Tietronix
Google-Code
Lambda
DLR
DLR
SANDIA
References
ANSYS, 2014
Breault, 2014
ASPOC, 2014
COMSOL, 2014
Kistler, 1986
Blanco et al., 2000
EPSILON, 2014
Sun to Market, 2014
Sun to Market, 2014
Sun to Market, 2014
Tellez, 2013
GateCycle, 2014
Buck, 2013
Ho, 2008
Bode et al., 2012
ISPEpro, 2014
Sun to Market, 2014
NSPOC, 2014
Schoffel et al., 1991
OptiCAD, 2014
SAM, 2014
Tellez, 2013
Martin, 2007
Martin, 2007
Alpert et al., 1988
SolTrace, 2014
Buck, 2013
Garcia, 2007
ThermoFlow, 2014
Izygon et al., 2011
Tonatiuth, 2014
Lambda, 2014
Schwarzbozl, 2006
Schwarzbolz16
et al, 2009
Tellez, 2013
Summary
• Model provide methodology used to obtain
 HTF mass flow rate through receiver
 Surface temperature distribution
 Inner tube temperature distribution
 Heat losses
 Receiver Efficiency
 Result of Heliostat field aiming strategy
 Corresponding receiver pressure drop
 Tube-strain per panel can be obtained
2015/08/04
17
Tracking Error in OPEN-LOOP drive systems
Foundation motion
Mirror waviness
Panel alignment error
Tracking Error in CLOSED-LOOP drive systems
Tower Sway
SIGEL= 0
SIGAZ= 0
SIGSX= 0
SIGSY= 0
SIGTX= 0
SIGTY= 0
#_Helio= 8965
9.62
7.48
5.34
3.21
1.07
-1.07
-3.21
-5.34
-7.48
-9.62
69
271
365
314
277
278
319
365
256
60
67
280
394
362
328
330
369
394
263
56
Heliostat Aiming Strategy
117
441
597
558
481
485
565
592
419
102
175
622
826
819
750
755
829
820
593
156
239
790
1010
1010
949
956
1020
1000
756
216
273
886
1120
1110
1060
1060
1120
1110
847
246
282
913
1150
1140
1080
1090
1150
1140
872
254
273
886
1120
1110
1060
1060
1120
1110
847
246
239
790
1010
1010
949
956
1020
1000
756
216
175
622
826
819
750
755
829
820
593
156
117
441
597
558
481
485
565
592
419
102
67
280
394
362
328
330
369
394
263
56
GrossP=
1200
1000
800
600
400
200
0
SIGSX= 0.003
SIGSY= 0
SIGTX= 0
SIGTY= 0
#_Helio= 10491
9.62
7.48
5.34
3.21
1.07
-1.07
-3.21
-5.34
-7.48
-9.62
263
586
678
595
548
593
649
550
298
69
227
576
732
625
537
575
667
632
349
72
181
540
776
699
593
614
722
725
423
92
169
537
815
797
707
710
802
818
504
110
174
590
871
879
801
795
876
896
611
140
181
653
920
932
859
850
923
943
697
177
184
682
936
948
880
869
938
957
727
197
181
653
920
932
859
850
923
943
697
177
174
590
871
879
801
795
876
896
611
140
169
537
815
797
707
710
802
818
504
110
181
540
776
699
593
614
722
725
423
92
227
576
732
625
537
575
667
632
349
72
SIGSX= 0.003
SIGSY= 0.002
SIGTX= 0
SIGTY= 0
#_Helio= 11027
9.62
7.48
5.34
3.21
1.07
-1.07
-3.21
-5.34
-7.48
-9.62
229
473
688
760
753
734
674
513
279
103
209
449
684
777
768
751
704
544
296
111
186
426
679
799
810
806
771
611
341
131
189
450
723
841
854
860
849
705
408
157
200
493
787
898
905
914
918
791
474
185
211
528
839
946
950
959
969
851
520
206
216
540
857
963
966
976
987
870
536
212
211
528
839
946
950
959
969
851
520
206
200
493
787
898
905
914
918
791
474
185
189
450
723
841
854
860
849
705
408
157
186
426
679
799
810
806
771
611
341
131
209
449
684
777
768
751
704
544
296
111
SIGSX= 0.003
SIGSY= 0.002
SIGTX= 0
SIGTY= 0.003
#_Helio= 11252
9.62
7.48
5.34
3.21
1.07
-1.07
-3.21
-5.34
-7.48
-9.62
234
459
672
758
753
729
663
501
280
115
218
440
666
769
771
753
697
531
297
123
199
424
667
793
810
805
764
599
344
145
202
446
714
847
864
866
843
689
408
175
213
476
766
914
944
949
922
763
461
201
226
508
814
967
1000
1010
983
821
501
222
231
520
832
987
1020
1030
1000
840
515
229
226
508
814
967
1000
1010
983
821
501
222
213
476
766
914
944
949
922
763
461
201
202
446
714
847
864
866
843
689
408
175
199
424
667
793
810
805
764
599
344
145
218
440
666
769
771
753
697
531
297
123
HOUR
0.00
COSINE SHADOW
0.83
1.00
NetP=
BLOCK
0.98
AIR ATT SPILLAGE
0.92
0.99
COSINE SHADOW
0.80
1.00
Area=
GrossPM=
AIR ATT SPILLAGE
0.92
0.89
GrossP=
Area=
GrossPM=
NetP=
BLOCK
0.98
AIR ATT SPILLAGE
0.92
0.86
GrossP=
Area=
GrossPM=
COSINE SHADOW
0.79
1.00
NetP=
BLOCK
0.98
660.75 MWt
88.54 Mwe
659.40 MWt
9.13 m2
668.62 MWt
88.68 Mwe
659.33 MWt
9.13 m2
670.75 MWt
-1.73%
1 2 3 4 5 6 7 8 9 10 11 12
HOUR
0.00
9.13 m2
TOTAL
0.49
1200
1000
800
600
400
200
0
DAY
81.00
658.42 MWt
-1.40%
1 2 3 4 5 6 7 8 9 10 11 12
COSINE SHADOW
0.79
1.00
87.59 Mwe
TOTAL
0.52
1200
1000
800
600
400
200
0
HOUR
0.00
660.35 MWt
-0.35%
NetP=
BLOCK
0.98
9.13 m2
TOTAL
0.60
1 2 3 4 5 6 7 8 9 10 11 12
HOUR
0.00
651.72 MWt
-1.32%
GrossP=
DAY
81.00
SIGEL= 0
SIGAZ= 0
GrossPM=
1200
1000
800
600
400
200
0
DAY
81.00
SIGEL= 0
SIGAZ= 0
Area=
1 2 3 4 5 6 7 8 9 10 11 12
DAY
81.00
SIGEL= 0
SIGAZ= 0
Heliostat Error
"Wind"
SIGEL=
SIGAZ=
SIGSX=
SIGSY=
SIGTX=
SIGTY=
No Heliostat Error
(Ideal)
Heliostat Field & Flux Maps – DONE!!
AIR ATT SPILLAGE
0.91
0.82
TOTAL
0.47
88.71 Mwe
19
Receiver Model – DONE!!
20
Power Block Model – DONE!!!
2015/08/04
21
Start with PhD 
2015/08/04
22