orc and kalina analysis and experience -...
TRANSCRIPT
1
ORC and KalinaAnalysis and experience
Páll Valdimarssonprofessor of mechanical
engineering
Sabbatical December 2003Lecture III
2
Energy and energy use
• Energy is utilized in two forms, as heat and as work
• Work moves, but heat changes temperature (moves the molecules faster)
• These are two totally different products for a power station
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
2
3
Convertability
• Work can always be changed into heat (by friction since ste stone age)
• Conversion of heat into work is difficult and is limited by the laws of thermodynamics. A part of the heat used has always to be rejected to the surroundings
4
Heat and work again
• Work is the high quality, high priced product
• Heat is second class quality, a low priced byproduct
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
3
5
A power plant
Source
Heat
Fuel
Rejected heatLosses
Electricity
Sellable heat
Power plant
6
Single flash - condensing
Production well Injection well
Separator
TurbineCooling tower
Pump
Condenser
Pump
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
4
7
T-s diagram
8
ORC with regenerator
Production well
Injection well
Turbine
Condenser
Pump
Regenerator
Boiler
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
5
9
T-s diagram
-2,0 -1,5 -1,0 -0,5 0,0 0,5 1,0-50
0
50
100
150
200
250
s [kJ/kg-K]
T [°
C]
2300 kPa
1000 kPa
350 kPa
90 kPa 0,2 0,4 0,6 0,8
0,00
45
0,02
6 0
,063
0
,15
0,37
0,
89 m
3/kg
Isopentane
10
ORC with paralell single flash
Production well Injection well
Separator
Turbine Cooling tower
Condenser
Pump
Condenser
Cooling towerTurbine
Boiler
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
6
11
ORC with serial single flash
Production well Injection well
Separator Turbine
Condenser
Cooling towerTurbine
Boiler
Throttling valve
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
7
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
8
15
Rogner, Bad Blumau
• The 250 kW air-cooled geothermal CHP plant generates electrical power as well as district heating using a low temperature geothermal resource.
• One standard containerized ORMAT CHP module, generating 250 kW electricity and 2,500 kW heat.
• The power plant is in commercial operation since July 2001.
16
Mokai, New Zealand
• The 60 MW Geothermal Power Plant is comprised of:» one 50 MW module operating on geothermal
steam» two 5 MW units operating on geothermal brine
• The power plant uses air-cooled condensers and achieves 100% geothermal fluid reinjection to produce electrical power with virtually no environmental impact.
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
9
18
Kalina
Production well
Injection well
Separator
Turbine
Condenser
Boiler
Throttling valve
Cooling tower
Regenerator
Pump
Brine hx
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
10
19
Geothermal experience
• Rough surroundings• Aggressive chemistry• Simple and reliable solutions• Geothermal energy - a mayor economic
factor
20
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
11
21
Conversion of heat to electricity
• The Carnot efficiency applies for infinite heat sources
• The maximum efficiency is lower than the Carnot efficiency for a source stream with finite heat capacity
• Kalina reduces entropy generation in the heat exchange process
22
Carnot efficiency
Hot reservoir
Cold reservoir
Work output
Heat in
Heat out
1
01TT
Carnot −=η
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
12
23
Maximum efficiency for a liquid source
−−=
21
2
1
0)(
ln1*
TTTT
THE hx
1T
2T
0T 0T
24
What is a Kalina process?
• A modified Rankine cycle, or rather:• a reversed absorption cycle• Ammonia - water working fluid• Patented by Exergy Inc and A. Kalina
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
13
25
The characteristics
• Heat is added in a combined boiling and separation process» at a variable temperature
• Heat is rejected in a combined condensation and absorption process» as well at a variable temperature
260 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
60
80
100
120
140
160
180
200
220
240
Tem
pera
ture
[°C
]
Ammonia mass fraction [-]
Mixture boiling at 30 bar
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
14
27
Heat addition - boiler
Source
Kalina
ORC
28
The working fluid
• Ammonia has a molar mass of 17, so steam turbines can be used
• The mixture properties are more complex, usually three independent properties are needed for the calculation of the fourth
• Therefore the cycle is more flexible, and can be closely optimized to the source
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
15
29
The benefit
• The working fluid is almost at the temperature of the source fluid when leaving the boiler
• The variable heat rejection temperature makes regeneration possible
30
Kalina vs. ORC
• Kalina is better when the heat source stream has a finite heat capacity
• ORC and Kalina are similar when the source is condensing steam
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
16
31
Boiling curves for Húsavík
0 200 400 600 800 1000 120020
30
40
50
60
70
80
90
100
110
120
Tem
pera
ture
[°C
]
Enthalpy [kJ/kg]
Water KalinaORC
32
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
17
33
34
Kalina again
Production well
Injection well
Separator
Turbine
Condenser
Boiler
Throttling valve
Cooling tower
Regenerator
Pump
Brine hx
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
18
35
36
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
19
37
38
0 200 400 600 800 1000 1200 1400 1600
0
50
100
150
200
Enthalpy [kJ/kg]
Tem
pera
ture
[°C
]
T - h diagram
6.5
6.56.5
6.5
31
3131
31
6.5
6.5
6.5
6.5
31
31
31
31
6.5
6.5
6.5
6.5
31
31
31
31
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
20
39
Comparison
Power[kW]
1. law 2. law Maxliquideff.
Volumeto
turbine[m3/s]
Kalina 2000 0,13 0,45 0,56 0,57ORC 1589 0,10 0,36 0,45 2,06Flash cycle 1589 0,10 0,36 0,45 22,4Themoelectricity 720 0,05 0,16 0,20 0
40
Comparison of Kalina and ORC
• Heat (10MW) is available down to 80°C• Cooling water (120kg/s) is available at
20°C• Heat exchangers have a pinch of 3°C• Condensers have a pinch of 10°C
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
21
90 95 100 105 110 115 120 125 130 135 1400
0.05
0.1
0.15
0.2
0.25
Temperature [C]
Effi
cien
cy [-
]CarnotLiquidKalinaORC
42
Húsavík geothermal power plant
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
22
43
General design parameters
44
Summary
• Commissioned in the summer 2000• Running at ~1500 kW net 2000 - 2001• Running at ~1700 kW net since
November 2001• Final acceptance certificate issued• Total investment cost 3,7 MEUR
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
23
45
Processdiagram
G121 °C
80 °C
90 kg/s1950 kW
5 °C
24 °C
118 °C31 bara
5,5 bara
12 °C
67 °C
16,3 kg/s
173 kg/s
0,81 NH3
130 kW
11,2 kg/s
46
The power plant
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
24
47
Design
• Standard industrial components• Turbine-generator from KKK, Germany• Electrical components CE marked• Heat exchangers from USA• Most of the tanks made in Iceland• Installed by a local contractor
48
Equipment
• Evaporator, shell and tube, 1600 m2
• Separator• Turbine• Recuperators• Condenser, plate, 2 x 750 m2
• Hotwell• Circulation pump
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
25
49
Auxillary equipment
• Ammonia storage tank• Demineralized water tank• Blow down tank
50
Thermal equilibrium
Power output
Power input
14.000 kWCooling water
1.700 kWElectricity, net
15.700 kWBrine
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
26
51
Start-up problems
• Separator• Evaporator• Axial sealing of the turbine, ammonia leakage• Condensers• Miscellaneous
» Main pump» Safety valves» Magnetite
52
Separator
• Difficult to measure the performance of the separator
• Mist eliminator module wrongly installed• New separator installed November 2001
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
27
53
Evaporator
• Ammonia leaking into the hot water• All tubes rolled again in November
2001, no leakages since
54
Axial sealing
• One axial sealing on the low pressure side
• N2 used in the sealing• The sealing has been replaced once• Leakage caused by carry-over from
separator
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
28
55
Condensers
• Plate heat exchangers• Important to mix liquid and vapor• Spraying nozzles modified• Increase power output by improving
performance of the condensers
56
Miscellaneous
• Safety valves, leakage• Circulation pump, seals and
guides/bushings for shaft• Improvment of spraying system in
recuperator 2• Magnetite (Fe3O4)
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
29
57
Turbine corrosion
• Ferritic material in the turbine corrodes• Austenitec material is unharmed• Corrosion is from the turbine control
valve until abt 1 metre after the turbine• Influence of rotor magnetic field?
» Elimination of ferritic material in the steam side of the turbine
58
Operating experience
• Maximizing the output by optimizing the strength of the NH3 – H2O solution
• Prevent air entering the system• Improvement of flushing and filters• The system is stable in operation
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
30
59
Lessons learned
• The technology is proven• Standard equipment has been adjusted• The plant is running according to specs• Individual equipment still may be
improved to increase output• Engineering details will improve future
plants
60
What do we have?
• New thermodynamic cycle with better efficiency in particular when the heat source cooled down while heat is extracted
• Theoretical and technical descriptions of the processes involved.
• Mathematical models that have been tested up against Husavik Plant
• Known media and known machinery
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
31
61
The knowledge
• Real experience and know how from the Husavik plant
• Mathematical models and ,,steam tables”, worked out in cooperation with University of Iceland.
• Over 30 years of experience in electrical generation from low and mid heat sources (Geothermal)
62
Kalina references
• Canoga Park, USA, demonstration plant 3 – 6 MW, 1991-1997
• Fukuoka, Japan, incineration plant, 4,5 MW, 1999
• Sumitomo, Japan, waste heat recovery from a steel plant, 3,1 MW
• Husavik, Iceland, geothermal plant, 2,0 MW, 2000
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
32
Krafla 1970, 60 MWel
Svartsengi 1980, 37 MWel, 60 MWdh
Nesjavellir 1990, 90 MWel, 250 MWdh
64
Were does it fit?
• Temperature above 150°C and good size stands on its own.
• Cogeneration of electricity and water for district heating.
• Good cold end helps• Environmental issues and subsidizing
changes these values
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
33
65
Budgetary unit price
• Budgetary price for 500 kW unit 720.000 USD
• Equals 1440 USD/kW• Turbine/generator most costly unit (30-
35%)• Presuming cooling water available
66
Opportunities
Where there may be hot water or other fluid/gas available at temperatures between 120 and 300°C
• Geothermal• Waste heat
» Industrial processes» Gas and diesel engine exhaust
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
34
67
Kalina plant benefits
• Energy cost efficient• Environmental issues• Green energy• Reduced emissions• Standard off the shelf equipment
68
General marketconditions
• General market price 4 eurocents/kWh• General pay-back time 4 years required• General investment 1000 USD/kW
• So how can 1440 USD/kW be competitive?
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
35
69
German law
“Green” electrical power generation bonus
• 2002-2004 1,74 eurocents/kWh• 2005-2006 1,69 eurocents/kWh• 2007-2008 1,64 eurocents/kWh• 2009-2010 1,59 eurocents/kWhProvided plant in operation before end of 2005
70
Competitive investment ?
• Given 1,65 eurocents/kWh bonus• 4 year pay-back period• Additional investment acceptable for
“green” energy USD 528.000/MW• Total acceptable investment cost thus
USD 1.528.000/MW or 764.000 USD for our 500 kW unit.
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
36
71
“Green” energy?
• Is waste heat recovery=green energy?• Waste heat recovery reduces emissions• CO2 quotas pricing as high as 20-25
USD/tonCO2
• Average emission 800 gCO2/kWh in fossil fuel plants
• Equals values of 1,8-2,2 eurocents/kWh
72
Summary
• Given green energy bonuses or CO2 evaluation the investment in a Kalina waste heat recovery electrical generating plant is feasible
• The investment is competitive to other investments for industrial improvements
• Pay-back time of 4 years in a green energy technology is short
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
37
73
Introduction
• Finite source heat capacity lowers this upper bound due to the reduction of the source temperature as heat is removed from the source
• This results in high cost for such low temperature power plants, as they have to handle large heat flows
• The Kalina cycle is a novel approach to increase this efficiency
74
The Models
• It is assumed that a fluid is available at temperatures ranging from 100 to 150°C
• A heat customer is assumed for the primary outlet water at the temperature of 80°C
• Primary flow of 50 kg/s is assumed• Cooling water source is assumed at 15°C,
and cooling water outlet is fixed at 30°C• It is assumed that a cooling water pump has
to overcome a pressure loss of 1 bar on the cooling water side in the condenser
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
38
75
Solution
• The software Engineering Equation Solver (EES) is used to run the models
• The cost model keeps the logarithmic mean temperature difference for each heat exchanger at the same value as found in the cycle data for the tenders for the Husavik power plant
• Estimated cost figures for individual components were then added together in order to obtain the final cost value
76
Process assumptions
• The OCR model is based on a system without regeneration
• Isopenthane is assumed as a working fluid
• A Kalina cycle for generation of saturated vapour for the turbine is used
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
39
77
ORC flow sheet
Production well
Injection well
Turbine
Condenser
Pump
Boiler
78
Kalina flow sheet
Production well
Injection well
Separator
Turbine
Condenser
Boiler
Throttling valve
Cooling tower
Regenerator
Pump
Brine hx
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
40
79
Kalina assumptions
• This cycle will be limited by the dew point of the mixture, that is when the boiling of the mixture is complete, and no liquid remains at the boiler outlet
• The bubble temperature of the mixture has to be lower or equal to the primary fluid outlet temperature to ensure safe operation
• The feasible region will be in the area between the 70 and 80°C bubble contours
• The feasible area regarding the dew temperature is limited to a value some 2-4°C lower than the maximum temperature of the primary fluid
Bubble temperature
1020
3040
0.6
0.8
120
40
60
80
100
Pressure [bar]
Bubble temperature
Ammonia ratio [-]
Bub
ble
tem
pera
ture
[°C
]
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
41
Dew temperature
1020
3040
0.6
0.8
150
100
150
200
Pressure [bar]
Dew temperature
Ammonia ratio [-]
Dew
tem
pera
ture
[°C
]
Bubble contours
10 15 20 25 30 35 400.65
0.7
0.75
0.8
0.85
0.9
0.95
1
Pressure [bar]
Amm
onia
ratio
[-]
Bubble temperature contours
3040
4050
5060
6070
70
70
80
80
80
90
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
42
Dew contours
10 15 20 25 30 35 400.65
0.7
0.75
0.8
0.85
0.9
0.95
1
Pressure [bar]
Am
mon
ia ra
tio [-
]
Dew temperature contours
80 90
100100
110
110110
120
120120 120
130
130
130130
140
140
14014
0
150
150
150
160
160
160
170
170 180
190
84
More Kalina
• Both high pressure level and ammonia content are design variables in the Kalina cycle
• This gives flexibility in the design of the cycle, and requires as well a certain design strategy
• The plant can be designed for maximum power• or with strong demands on the investment cost• The cost is 100 for the lowest cost, and the power
100 for the highest power• An x denotes an infeasible solution, the cycle will not
be able to run at these ammonia content – pressure combinations
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
43
Cost contours
15 20 25 30 35 400.65
0.7
0.75
0.8
0.85
0.9
0.95
1
Pressure [bar]
Am
mon
ia ra
tio [-
]
Cost contours, 100°C source
101
101
101
102102
102
102103
103
103
103103
104104
104
104
105
105
105
105
106
106
106
106
107
107
107
107
107
108
108
108
108
109
109
109
109
110
110
110
110
110
Power contours
20 25 30 35 400.65
0.7
0.75
0.8
0.85
0.9
0.95
1
Pressure [bar]
Am
mon
ia ra
tio [-
]
Power contours, 100°C source
9090
90 9090
90
90 9090
90
90
909090
9191
91
9191
91 91
91 9191
91
9191
92
92
9292 92
929292
92
9292
92
93
93
9393
9393
9393
9393
93
9494
94
9494
9494
94
9494 94
9595
95
9595
95 9595
9595
96
96
96
9696
96
9696
9797
97
9797
97
9797
9898
98
9898
99
9999
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
44
Cost surface
1020
3040
0.6
0.8
10
200
400
600
Pressure [bar]
Cost function
Ammonia ratio [-]
Cos
t
Power surface
1020
3040
0.6
0.8
10
50
100
Pressure [bar]
Power function
Ammonia ratio [-]
Pow
er
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
45
89
Discussion
• The best power and best cost points are different
• The lowest cost is at 32 bar, 92% ammonia, but highest power is at 34 bar and 88% ammonia
• This leads to the definition of two different Kalinacycles, the best power and the best cost cycles, with different pressure and ammonia content
15 20 25 30 35 400.65
0.7
0.75
0.8
0.85
0.9
0.95
1
Pressure [bar]
Am
mon
ia ra
tio [-
]
Cost contours, 120°C source
101
101
101
102
102
102
102 102102
103
103
103 103 103
104104
104
104104
105
105
105
105105
106
106106
106
107
1071 0 7
107
108
108108
108
108 109
109
109109
110
110
110
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
46
22 24 26 28 30 32 34 36 38 400.65
0.7
0.75
0.8
0.85
0.9
0.95
1
Pressure [bar]
Am
mon
ia ra
tio [-
]
Power contours, 120°C source
7070
72
72
74
74
76
76
7678
78
78
8080
80
82
82
82
84
84
8484
86
8686
88
8888
90
9090
91
9191
92
9292
93
9393
9494
95
95
96
9697 98
99
18 20 22 24 26 28 30 32 34 360.65
0.7
0.75
0.8
0.85
0.9
0.95
1
Pressure [bar]
Am
mon
ia ra
tio [-
]
Cost contours, 150°C source
101
101101
102
102
102
103103
103 103
104
104
104 104
105
105
105
106
106
106
107
107
107
108
108
108
108
109
109
109
110
110
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
47
22 24 26 28 30 32 34 36 38 400.65
0.7
0.75
0.8
0.85
0.9
0.95
1
Pressure [bar]
Am
mon
ia ra
tio [-
]
Power contours, 150°C source
70
70
72
7274
74
7 676
7878
8080
82
82
84
84
86 8890
9192
9394
9596
97
94
Comparison
• Two ORC companies made a tender in the Husavik bid
• Manufacturer A offered a high power, high cost power plant, where manufacturer B took a more conservative approach
• Following is a comparison with both the low cost and high power Kalina power plants
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
48
Cost comparison
100 110 120 130 140 1501000
1200
1400
1600
1800
2000
2200
2400
Source inlet temperature [°C]
Net
cos
t [$/
kW]
Kalina vs ORC cost comparison
ORC BORC A
Kalina HP
Kalina LC
Power comparison
100 110 120 130 140 1500
500
1000
1500
2000
Source inlet temperature [°C]
Net
pow
er [k
W]
Kalina vs ORC power comparison
Kalina HP
Kalina LC
ORC A
ORC B
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
49
97
Results
• The maximum power generated for a given source is greater for the Kalinacycle
• Kalina cycle is well positioned against an ORC cycle for applications with high utilization time, a base load application
98
Results II
• A heat consumer is beneficial for the ORC cycle as it results in less temperature change of the primary fluid during the boiling process
• The Kalina cycle has the boiling or vaporization of the fluid happening over a temperature range up to 100°C, which is beneficial when the primary fluid return temperature has to be minimized
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com
50
99
The debate
• The theoretical efficiency and cost/production ratio are better for Kalina as shown above
• Other arguments like difficulties with machinery and lesser operational security or uptime are only temporary discussion items, as always for a new technology
• These arguments were exactly the same between conventional flash cycle and ORC when the latter popped up 30 years ago with better efficiency but little track record
100
Conclusion
• The Kalina cycle is thermodynamically superior or equal to the ORC cycle
• There is no black magic behind the Kalina cycle
• The startup problems that have either been solved, or are solvable
Created with FinePrint pdfFactory (beta version) -- details at www.fineprint.com