Techno-economic Analysis of enhanced
Dry Cooling for CSP
SolarPACES Conference, Las Vegas, 19.09.2013
Massimo Moser
German Aerospace Center (DLR)
Institute of Technical Thermodynamics
Department of System Analysis and Technology Assessment
www.DLR.de • Chart 2
Motivation
• Which factors are important for the site selection of a steam power plant?
• Proximity to load centers
• Accessibility and presence of transmission corridors
• Availability of low-cost fuel
• Presence of water for cooling purposes
• Low-cost fuel is often available in water scarce regions. Examples:
• Coal plants located near to coal mines in South Africa
• CSP! (even if currently evaporation cooling is mostly used)
 In these cases dry cooling is the only viable option!
www.DLR.de • Chart 3
Dry Cooling - State of the Art and Improvements
• The heat exchange is governed by the dry bulb
temperature (TDB)
• Strong impact of TDB on cooling efficiency
• No water consumption/withdrawal
• Direct or indirect layout (Heller)
Tawney 2003
• Different approaches for the improvement of dry
cooling:
• Hybrid-wet cooling
• Deluge cooling
• ACC optimized design
• Optimized dispatch
www.DLR.de • Chart 4
Methodology
• Simplified model  sensitivity analysis of LEC on key parameters:
• ACC cooling design (Initial temperature difference)
• Solar multiple
• Solar field specific investment cost
• REMix model  Optimal plant dispatch of a dry cooled CSP plant:
• Standard dispatch (100 % till complete TES discharge)
• Optimized dispatch I (constant price)
• Optimized dispatch II (time-variable price, demand proportional)
www.DLR.de • Chart 5
Technical Model
ITD
= Initial Temperature Difference
TTD
= Terminal Temperature Difference
ITD
Steam Tcond
Air Outlet Temp.
TDB
Qcond
ITD − TTD
= A ⋅ U ⋅ LMTD = A ⋅ U ⋅
 ITD 
log

 TTD 
Turbine Efficiency = f(Tcond)
Gross Turb. Efficiency [%]
= TBD + ITD
TTD
Tcond
45%
40%
35%
30%
25%
20%
15%
10%
5%
0%
Gross Turb. Efficiency [%]
20
30
40
50
60
70
Condensation Temperature [°C]
• Assumption: constant ITD, TTD
80
www.DLR.de • Chart 6
Design Point Specifications and Investment Cost
Temp. Frequency [%]
Dry Bulb Temperature Frequency
10%
8%
6%
Temp Frequency [-]
TDB_DESIGN
4%
2%
0%
-04 00 04 08 12 16 20 24 28 32 36 40 44 48
Dry Bulb Temperature [°C]
• Assumption: design point specifications
have been set for the 20 % percentile of
the annual temperature
 Inability to maintain design output during
the hottest hours of the year
• A trade-off exists between CAPEX,
power block efficiency and annual
yield!
www.DLR.de • Chart 7
Sensitivity of LEC on key Parameters
CSP solar only - Dry Cooling Design =f(spec. inv. SF)
200 [€/m²]
150 [€/m²]
CSP solar only - Dry Cooling Design =f(SM)
SM1.3
250 [€/m²]
SM3
SM2
16.00
15.00
LEC [€cent/kWh]
LEC [€cent/kWh]
14.00
13.00
12.00
11.00
10.00
14.00
13.00
12.00
11.00
9.00
8.00
10.00
15
20
25
30
35
40
45
Initial Temperature Difference (ITD) [°C]
• In the case of elevate investment cost,
a highly efficient cooling has to be
preferred
• The same applies to conventional
power plants with high fuel costs!
15
20
25
30
35
40
45
Initial Temperature Difference (ITD) [°C]
• High SM (i.e. high plant operating
hours) requires high efficiency for more
quick amortization of elevate investment
cost
Assumptions: SM1.3 w/o TES / SM2 7.5 h TES / SM3 12 h TES
www.DLR.de • Chart 8
REMix – Analyzed cases
• 100 MW el_net CSP “Andasol-like”, solar-only operation, dry cooling
• SM 1.4; 7.5 h TES (not optimized; high dispatch flexibility)
• 3 sites in Jordan
Parameter
Unit
Kerak
Irbid
Aqaba
Latitude
°
31.18
32.55
29.52
Longitude
°
35.70
35.85
35.00
kWh/m²/y
2,545
2,537
2,371
TDB_AVG
°C
17.0
17.6
24.9
TDB_80_%
°C
25.1
24.7
32.5
DNI
• Hourly simulation with the optimizing tool REMix
• 3 operation strategies:
• Standard dispatch  used as reference
• Optimized dispatch I (constant price)
• Optimized dispatch II (time-variable price, demand proportional)
www.DLR.de • Chart 9
REMix - Results
Parameter
Unit
Kerak
Irbid
Standard
Operation
Optimized
Dispatch +
Fixed Price
Optimized
Dispatch +
Variable
Price
Standard
Operation
Optimized
Dispatch +
Fixed Price
Optimized
Dispatch +
Variable
Price
1254.4
1254.4
1254.4
1192.4
1192.4
1192.4
337
358
342
301
459.3
451.7
409.3
418.0
402.0
407.0
420.5
411.9
373.3
381.1
367.3
37.6
37.4
38.9
39.8
35.9
36.9
37.4
36.8%
36.3%
36.6%
36.9%
35.8%
35.3%
35.4%
35.2%
33.5%
33.8%
33.1%
33.5%
33.9%
32.3%
32.3%
32.4%
32.0%
0.0%
-2.0%
3.6%
0.0%
-3.2%
-1.1%
0.0%
-2.0%
7.1%
-
Standard
Operation
Optimized
Dispatch +
Fixed Price
Optimized
Dispatch +
Variable
Price
[GWhth]
1180.4
1180.4
1180.4
[-]
348
303
303
PEL_GROSS
[GWh/y]
419.0
427.8
421.4
444.5
PEL_NET
[GWh/y]
383.5
391.6
385.9
Plant Parasitics
[GWh/y]
35.4
36.2
ηGROSS - PB
[%]
36.5%
ηNET - PB
[%]
∆ LEC
[%]
Operation Strategy
QSF
n start-up - PB
Aqaba
Minimization of PB
347
302
start-up
procedures
www.DLR.de • Chart 10
REMix – Results Analysis
Exemplary Summer Electricity Generation - Constant Price
2500
100
2000
80
1500
60
1000
40
500
20
96
• In the case of constant feed-in price:
• Minimization of the number of
start-ups
• Max. production in the early
morning hours (lower TDB)
101
91
86
81
76
71
66
61
56
51
46
41
31
36
26
21
16
11
0
6
0
TES SoC [MWhth]; DNI [W/m²]
Pparasitics [MW]
TES SoC [MWh]
120
1
Electricity Generation [MW]
CSP P_el. Solar [MW]
Pgross [MW]
DNI [W/m²]
Hour [-]
Exemplary Summer Electricity Generation - Variable Price
2500
100
2000
80
1500
60
1000
40
500
20
Hour [-]
101
96
91
86
81
76
71
66
61
56
51
46
41
36
31
26
21
16
11
0
6
0
TES SoC [MWhth]; DNI [W/m²]
Pparasitics [MW]
TES SoC [MWh]
120
1
Electricity Generation [MW]
CSP P_el. Solar [MW]
Pgross [MW]
DNI [W/m²]
• In the case of time-variable price,
the impact of the temperature seems
to play a secondary role in the
optimization strategy
www.DLR.de • Chart 11
Conclusions
• Air cooled condensers will be the preferred option for large-scale
introduction of CSP in water-scarce and DNI-rich regions
• The optimal ACC design results from a technical-economic trade-off
between turbine efficiency and investment cost; In CSP plants, highly
efficient ACC are required for high SM/TES and high specific CAPEX
• Ca. 2 % LEC reduction can be reached in dry cooled CSP plant by partial
plant commitment shifting towards night hours
• In the case of demand-driven plant commitment, slightly higher feed-in
tariffs should be introduced. However, in this case CSP would be able to
displace the most expensive plants during peak periods.
Thank you
for your attention!
Contact
Massimo Moser
DLR German Aerospace Center
Institute of Technical Thermodynamics
 massimo.moser@dlr.de
 +49 711 6862 779