A Comparative Study of Heat Rejection Systems for sco 2 Power Cycles

Transcription

1 A Comparative Study of Heat Rejection Systems for sco 2 Power Cycles Timothy J. Held, Jason Miller and David J. Buckmaster 1

2 Heat rejection systems Common to all closed-loop cycles, sco 2 systems need to reject residual heat Heat addition Non-condensing cycle Condensing cycle Heat rejection 2

3 Heat rejection technology Air cooled Fin-fan cooler Low air-side dp (< 2 wc) Aux loads = fan drives Water-cooled Cooling tower (forced draft) Intermediate water-loop required for PCHE CO 2 -water HX Aux loads = fan drives + pumps Cooling Tower PFHE Accumulator/ pressurizer P-14 PCHE CO 2 Inlet CO 2 Outlet Circ pump Circ pump Hybrid solutions exist (WSAC, ACC with fogging) not considered for this paper 3

4 Condenser UA (kw/ C) CT water conductivity (S/m) Why you should not use direct CT water feed to your PCHE UA Cond a) The manufacturer advises against it b) Very small change in water conductivity resulted in significant PCHE performance loss Date 4

5 Temperature ( C) Condenser performance & nomenclature 70 CO 2 inlet Coolant CO Range Coolant inlet Approach Q tot dq UA = න 0 T CO2 T coolant 0 0% 20% 40% 60% 80% 100% Q/Q tot 5

6 Gross efficiency Cycle performance PHX RCBC baseline CO 2 54% High T recup Turbine 53% 52% 51% RC Brayton RC Condensation 50% 49% 48% Low T recup Cmpr 2 47% 46% 45% Cmpr 1 Cooler 44% T coolant ( C) Notes: Gross efficiency does not include coolant aux load Cycle performance plotted as function of coolant T 6

7 Power (kwe) Heat exchanger sizing Condenser/Cooler UA (kw/k) Cost generally proportional to UA Design value = 1500 kw/ C (both water & air) 7

8 Cooling tower cost (K$) Cooling tower sizing T approach ( C) Cost function of design approach temperature 3 C regarded as a practical minimum by CT suppliers, used for this study 8

9 CDF Local climate impacts Houston, TX Tdb Twb Temperature ( C) Hourly observations from 2014 converted to CDF format T wb affects water (evaporative) cooling, T db affects air cooling 9

10 T coolant ( C) Cooling tower performance as f(t ambient ) Akron Houston Phoenix T wb ( C) 20 C selected as design point Demin water loop adds another 3 C to approach temperature As Twb approaches 0 C, water T increases to avoid freezing 10

11 T coolant ( C) Coolant temperature distribution Houston, TX Water Air Twbi -5 0% 20% 40% 60% 80% 100% % hours Coolant temperature CDF calculated as function of wet/dry bulb temperature CDFs 11

12 Net power (kwe) Cycle performance over 1-year period Non-condensing cycle example, Houston, TX ACC WCC 0 0% 20% 40% 60% 80% 100% % hours Coolant CDFs convolved with Power = f(t coolant ) to calculated CDF of net power Results integrated to yield annualized average power 12

13 Net Power (kwe) Net Power (kwe) Net power output (non-condensing cycle) Annualized average 99% hot day 11,000 10,500 10,000 11,000 10,500 10,000 9,500 9,500 9,000 9,000 8,500 8,500 8,000 8,000 7,500 7,500 7,000 Akron Houston Phoenix 7,000 Akron Houston Phoenix WCC ACC WCC ACC ACC outperforms on average Hot-day performance falloff ~ 2-4% for non-arid climates 13

14 Auxiliary load comparison Load WCC ACC Fans CW pump 99 PFHE pump 112 Total Lower cooling tower fan loads largely offset by increased pump loads 14

15 Footprint comparison WCC 97m 2 5.7x17m ACC 240 m x 15.6m Cooling tower footprint smaller than ACC, does not include space for pumps, water treatment, etc. 15

16 Capital cost comparison WCC ACC Cooling tower/acc $ 143,000 $ 695,000 Water treatment $ 20,000 CT Pumps $ 44,000 CT/ACC Piping $ 41,000 $ 108,000 PFHE $ 83,000 PFHE Pumps $ 46,000 PFHE Piping $ 16,000 Filters $ 5,000 WCC PCHE $ 300,000 Total eqpt $ 698,000 $ 803,000 Installation $ 419,000 $ 482,000 Total $ 1,117,000 $ 1,285,000 Footprint (m²) ACC slightly (15%) higher 16

17 O&M cost comparison (annual) WCC ACC Water consumption $ 156,000 Water disposal $ 57,000 Water treatment $ 25,000 Chemical replacement $ 65,000 Pump maintenance $ 24,000 Fan maintenance $ 7,500 $ 30,000 Maintenance cost/yr $ 334,500 $ 30,000 $/kwh $ $ Significantly lower O&M costs, largely driven by elimination of water supply, treatment and disposal costs Reduced O&M pays back increased capex of ACC in << 1 year 17

18 Other reasons to prefer ACC Cooling water filter cleaning is unpleasant 18

19 Other reasons to prefer ACC Cold-weather operation is even less pleasant 19

20 Other reasons to prefer ACC Mimivirus world s largest virus, discovered in a UK cooling tower Legionella bacterium often found in cooling towers. Infects 10,000-18,000 people per year in the US alone Cooling towers are excellent industrial-scale Petri dishes 20

21 Conclusions Water cooling advantages Better performance on hot days, especially in arid climates Smaller footprint Air cooling advantages Better annualized average power output except in arid climates (where water cooling is impractical anyway) Lower O&M costs Simpler operation Near zero water consumption for sco 2 power cycles Costs and auxiliary loads are similar Air cooling nearly always the better choice 21

22 Backup slides 22

23 Cycle / cost modeling Optimization process defines cycle with minimum cost for given power output Multiple solutions generates curves of cost vs power Allows selection of optimal cycle architecture Held, T.J., Supercritical CO 2 Cycles for Gas Turbine Combined Cycle Power Plants. Power Gen International. 23

24 sco 2 vs steam Power optimized Cost optimized Normalized to steam power & cost from GT-Pro, power-optimized solutions ( cost-optimized point shown for reference) Same exhaust and boundary conditions used for sco % lower cost for same power 7-14% higher power for same cost 24