Short Notes on Carbon Cycle, Nitrogen Cycle and Sulphur Cycle
Moisseytsev #45 Recent SCO2 Cycle Analysis at...
Transcript of Moisseytsev #45 Recent SCO2 Cycle Analysis at...
Recent Developments in S-CO2
Cycle
Dynamic Modeling and Analysis at ANL
Anton Moisseytsev and Jim Sienicki
Argonne National Laboratory, USA
The 4th International Symposium - Supercritical CO2
Power Cycles
Pittsburgh, PA
September 9-10, 2014
Contents
� The presentation provides an update on recent analyses of the S-CO2
Brayton
cycle at ANL
– Since the last S-CO2
Symposium in 2011
– The majority of the results has been already presented at various conferences
� The analysis presented is this paper has been carried out using the Plant Dynamics
Code (PDC) developed at ANL
– Most of the S-CO2
cycle control analysis carried out at ANL and presented here has been
done in application of the cycle as an energy converter for Sodium-Cooled Fast Reactors
(SFRs)
� Recent Progress
– Coupling of the PDC with the SAS4A/SASSYS-1 code
– Dynamic simulation and control of the S-CO2
cycle
– S-CO2
cycle control with active reactor control
– ANL Plant Dynamics Code validation
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
2
ANL Plant Dynamics Code (PDC)
� Specifically developed for analysis of S-CO2
cycle
– One-dimensional system level transient analysis code
– Targets the specific features of the cycle
• Operation close to the critical point
• Recompression cycle (if needed)
– Real CO2
properties
• Properties variation in HX’s and turbomachinery
• No ideal gas assumptions
• Compressibility effects
� Incorporates S-CO2
cycle control mechanisms and logic
� Incorporates steady-state design code to determine cycle initial conditions
� Integrated “semi-automatic” turbomachinery design
– Design and performance subroutines for both turbine and compressor
� Fast solution scheme runs efficiently on a PC
– Taylor series
– Almost real time (for slow transients)
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
3
Coupling of the PDC with the SAS4A/SASSYS-1 Code
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
4
SAS4A/SASSYS-1 Code
� Developed by Argonne National Laboratory
� Leading capability for analysis of liquid-metal-cooled reactors at system level
– Primary use: safety analysis
� Incorporates detailed reactor as well as primary and intermediate loop thermal
hydraulics
– Forced and natural circulation, pump models, heat exchangers
� Reactivity coefficients specific to fast reactors
– Core axial and radial expansions, control rod expansion, coolant density, etc.
– Radial-, axial-, and temperature-dependent distributions in transients
� Includes balance-of-plant model
– Currently limited to steam cycles
� Coupling PDC with SAS4A/SASSYS-1 would combine analysis capabilities for the
reactor and S-CO2
BOP
– Covering the entire plant
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
5
Coupling PDC with SAS4A/SASSYS-1 code
� Coupling PDC with SAS4A/SASSYS-1 would combine analysis capabilities for the
reactor and S-CO2
BOP
– Covering the entire plant
� However, the access to SAS4A/SASSYS-1 source code is restricted (Export
Controlled)
– Previously, the two codes were run separately with iterative update of input files
� New coupling scheme uses SAS4A/SASSYS-1 PC executable file
– Does not require access to the SAS4A/SASSYS-1 source code
– Utilizes Fortran capabilities to work with EXE files
– Takes advantage of SAS4A/SASSYS-1 restart capability
� The data transfer between PDC and SAS4A/SASSYS-1
– Occurs at each SAS4A/SASSYS-1 time step
– Is done by reading SAS4A/SASSYS-1 output files and writing input files
– Involves intermediate sodium conditions at RHX inlet and outlet
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
6
PDC-SAS4A/SASSYS-1 Coupling
� New coupling scheme uses
SAS4A/SASSYS-1 PC executable file
– Does not require access to the
SAS4A/SASSYS-1 source code
– Utilizes Fortran capabilities to work
with EXE files
– Takes advantage of SAS4A/SASSYS-1
restart capability
– Could be applied to other heat source
codes
� The data transfer between PDC and
SAS4A/SASSYS-1
– Occurs at each SAS4A/SASSYS-1 time
step
– Is done by reading SAS4A/SASSYS-1
output files and writing input files
– Involves intermediate sodium
conditions at RHX inlet and outlet
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
7
SAS4A/SASSYS-1 – Reactor Side
PDC – S-CO2 Side
Run steady-state calculations
Use steady-state sodium RHX-outlet
temperature Update restart
input file
Run one time step transient calculations
Sodium RHX-inlet temperature and
flow rate
Run transient calculations
Current time = SAS4A/
SASSYS-1 time
No
Sodium RHX-outlet temperature
Extrapolate sodium
RHX-outlet temperature
Run steady-state calculations
Maximum time
reached
End
Start
No
Update input tables for sodium temperature and flow rate
Yes
Yes
Dynamic Simulation and Control of the S-CO2Cycle
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
8
S-CO2Cycle Control Analyses
� Load Following
– Goal: to show that the cycle can be controlled to change power level
– Investigate and optimize control mechanisms
� Decay heat removal
– Goal: demonstrate that the cycle can be used in low power regime
– Investigate and optimize control mechanisms
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
9
S-CO2Cycle Control
� Included into transient PDC equations
� Relies on combination of control mechanism
– Turbine bypass [17]
• Fastest power control (on pressure and flow)
• Least efficient
– Inventory control [18]
• The most efficient (preserves TM velocity triangles
for ideal gas)
• Slow (introduce disturbance to compressor conditions)
• Limited by tank volume
• Decreases compressor surge margin
– Turbine throttling [16]
• More efficient than TBP, but less efficient than INV
• Introduces an extra pressure rise with valve closing (not good for full pressure)
– Cooler bypass [19]
– Cooling water flow rate [14]
– Compressor surge control
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
10
S-CO2Cycle Control: Control Strategy
� Use inventory control when possible
– Subject to storage tank volume limitation
� Turbine throttling
– Outside INV range
� Turbine bypass for small
load changes and
fine-tuning
� Minimum temperature
control
– Always
� Shaft speed control
– For decay heat removal
– After disconnection from
the grid
� Compressor surge control
– When needed
– Low power
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
11
Demonstration of S-CO2Cycle Control with PDC
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
12
� Simulation in two stages
– First stage: Load following from 100% down to 0%
– Second Stage: Transition to decay heat removal
First Stage
� The electrical grid demand was set to reduce linearly at 5%/min rate from 100% to
0%
– 1,200 s transient time
– The only external input
� S-CO2
cycle automatic controls adjust the conditions on the BOP side to maintain
balance between the net generator output and the grid demand
� Autonomous reactor operation is assumed
– Sodium-cooled fast reactor (ABR-1000)
– No reactor power control by control rods
• Only rely on reactivity feedbacks
– No control of primary and intermediate pumps
Results of Transient Simulations: Load Following
� Grid demand decreases linearly to 0 (external input)
� Net generator output follows the grid demand very closely
– Achieved by control actions on S-CO2
side
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
13
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
0 200 400 600 800 1000 1200
W,
FRA
CT
ION
OF N
OM
INA
L G
EN
ER
ATO
R P
OW
ER
TIME, s
TURBINE AND COMPRESSORS WORK AND GENERATOR OUTPUT
W_Turb
W_Comp1
W_Comp2
W_gen
W_grid
0
10
20
30
40
50
60
70
80
90
100
0 200 400 600 800 1000 1200
VA
LV
E O
PEN
FLO
W A
REA
, %
TIME, s
VALVES CONTROL ACTION
f_TBPv
f_INViv
f_INVov
f_TINv
f_C2Ov
Results of Transient Simulations: Load Following
� Reduced heat removal by CO2
affects the temperatures on the sodium side of
Na-CO2
HX
– Sodium inlet temperature is defined by heat balance on the reactor side
� On the reactor side, changing intermediate and then primary loop temperatures
introduce reactivity to the core
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
14
300
350
400
450
500
550
0 200 400 600 800 1000 1200
TEM
PER
AT
UR
E, oC
TIME, s
BRAYTON CYCLE TEMPERATURES: RHX
Hot_i
Hot_o
Cold_i
Cold_o
-0.10
-0.08
-0.06
-0.04
-0.02
0.00
0.02
0.04
0.06
0.08
0.10
0 200 400 600 800 1000 1200
REA
CT
IVIT
Y, $
TIME, s
CORE REACTIVITY
NET EXTERNAL DOPPLER AX. EXP.
RAD. EXP. CRDL EXP. COOLANT
Results of Transient Simulations: Load Following
� The net effect of reactivity feedbacks is reduction of reactor power
– Which closely follows the heat removal by CO2
side
– Even without active reactor power control
� The peak reactor temperatures stay below the nominal values
– Result of strong negative reactivity feedbacks of ABR-1000 SFR core
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
15
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0 200 400 600 800 1000 1200
NO
RM
ALIZ
ED
PO
WER O
R F
LO
W
TIME, s
CORE POWER AND FLOW
POWER DECAY POWER PEAK CHANNEL FLOW
300
350
400
450
500
550
600
650
700
750
0 200 400 600 800 1000 1200
TEM
PERAT
URE, oC
TIME, s
PEAK CORE TEMPERATURES
PEAK FUEL PEAK CLAD
PEAK COOLANT INLET
Transition to the Decay Heat Removal Mode
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
16
� Goal: find S-CO2
cycle control which would allow plant operation at low reactor
power
– 6% or less
� Simulation was continued after the generator power reached 0% level
– Same transient, Stage 2 starts at 1,200 s
� Generator is assumed to be disconnected from the grid
– Zero grid demand
– Ideally, the control system would maintain zero net generator output
• If there is enough reserve power in turbine to drive the compressors
– Turbomachinery shaft speed is assumed to reduce linearly to 20%
� No other changes in external input
– Still autonomous reactor operation
� Minimum temperature control is no longer required to maintain minimum
temperature at design point
– As long as the temperature stays above the critical value
Results of Transient Simulations: Transition to Decay Heat
� Slowing down shaft reduces both turbine and compressors work
– Still, turbine provides sufficient power to drive the compressors
– Zero net generator output is maintained
� The automatic cycle control maintains zero generator power
– To match zero load demand
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
17
-1%
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
10%
1200 1500 1800 2100 2400 2700
W,
FR
AC
TIO
N O
F N
OM
INA
L G
EN
ER
ATO
R P
OW
ER
TIME, s
TURBINE AND COMPRESSORS WORK AND GENERATOR OUTPUT
W_Turb
W_Comp1
W_Comp2
W_gen
W_grid
0
10
20
30
40
50
60
70
80
90
100
1200 1500 1800 2100 2400 2700
VA
LV
E O
PEN
FLO
W A
REA
, %
TIME, s
VALVES CONTROL ACTION
f_TBPv
f_INViv
f_INVov
f_TINv
f_C2Ov
� Compressor surge/stall approach is detected at about 2,100 s
– Triggering surge control flow
– Still, the flow is not too high to increase the compressor work significantly
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
18
Results of Transient Simulations: Transition to Decay Heat
0%
10%
20%
30%
40%
50%
60%
70%
80%
1200 1500 1800 2100 2400 2700
STA
LL M
ARG
IN
TIME, s
COMPRESSORS STALL MARGIN
Comp1
Comp2
-0.01
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
1200 1500 1800 2100 2400 2700
m,
RELAT
IVE T
O N
OM
INA
L T
UR
BIN
E
TIME, s
BRAYTON CYCLE FLOW RATES: SURGE CONTROL
Comp1
Comp2
� Temperatures above the critical point are maintained
– Even with disabled cooler bypass control
� The heat removal rate by the cycle reduces to about 3% nominal
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
19
Results of Transient Simulations: Transition to Decay Heat
30.9
31.4
31.9
32.4
32.9
33.4
33.9
34.4
34.9
1200 1500 1800 2100 2400 2700
TEM
PERATU
RE, oC
TIME, s
COMPRESSOR #1 INLET TEMPERATURE
T_c_i
T_crit
0%
5%
10%
15%
20%
25%
30%
35%
1200 1500 1800 2100 2400 2700
PO
WER,
FRA
CTIO
N O
F F
ULL R
EA
CTO
R P
OW
ER
TIME, s
HEAT BALANCE IN RHX
Q_RHX_Rx
Q_RHX_CO2
S-CO2Cycle Control with Active Reactor Control
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
20
Active Reactor Control in PDC-SAS4A/SASSYS-1 Codes
� In load following transients, the changes start on BOP side
– E.g., turbine bypass valve opening in response to reducing grid load
– Reactor “sees” changes through CO2
heat removal variation (reduction) in the
intermediate sodium-to-CO2
reactor heat exchanger (RHX)
� To maintain intermediate sodium temperatures, sodium flow rate needs to be
adjusted (reduced)
– IHTS pump control
� This will affect IHX
temperatures
– Primary pump control
� With changed (low) primary
coolant flow, reactor power
needs to be changed
– To maintain core-outlet
temperature
– By means of the control rod
movements
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
21
6
5
3
4
2
Reactor Control Logic
Controllable Parameter Measured Value Controlled By Control
Intermediate loop cold-
side temperatures
RHX-outlet Na
temperature
Intermediate sodium
flow rate
Intermediate sodium
pump torque
Primary loop cold-side
temperatures
IHX-outlet primary Na
temperature
Primary sodium flow
rate
Primary sodium
pump torque
Hot-side temperaturesCore-average outlet
temperatureCore power (reactivity)
External core
reactivity
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
22
� Active reactor control logic was implemented in the PDC
– Control action is communicated to SAS4A/SASSYS-1
� In ideal world, the temperatures will be maintained at steady-state levels
everywhere
– Power/flow=1
� In reality, system delays (and other limitations) will prevent that
– Are there S-CO2
cycle limitations?
Results of Active Reactor Control (40% linear load reduction)
� Active control showed better results
– Less temperature variation
– Better power-to-flow ratio
– Higher cycle efficiency
� However, these results were
obtained for only 40% load
reduction
– What happens below 60%
load?
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
23
320
325
330
335
340
345
350
0 200 400 600 800 1000 1200
TEM
PER
AT
UR
E, oC
TIME, s
BRAYTON CYCLE TEMPERATURES: RHX
Hot_o Cold_i
CO2Temperature at RHX Inlet
� In highly-recuperated S-CO2
cycle, RHX-
inlet temperature is defined by HTR
performance
� When flow rate increases, heat
transfer in HTR is reduced, RHX-inlet
temperature ↓– During turbine bypass
• 0-120 s
� With decreasing flow rate, RHX-inlet
temperature ↑– With inventory control
� When CO2
temperature at RHX cold
end reaches steady-state value for
sodium temperature, ability to
maintain that temperature is lost
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
24
320
325
330
335
340
345
350
0 200 400 600 800 1000 1200
TEM
PER
AT
UR
E, oC
TIME, s
BRAYTON CYCLE TEMPERATURES: RHX
Hot_o Cold_i
CO2Temperature at RHX Inlet
� Several approaches to control the temperature (prevent from increasing) were
investigated
– Reduced inventory control action
– Recuperator bypass
– Relaxed sodium temperature requirements
– Reduced core-outlet temperature
� Control on S-CO2
cycle side resulted in too low cycle efficiency
– Plus, there are additional limits
� Results with relaxed temperature limits are acceptable
– If high structure temperatures (say, 400 oC) can be tolerated
� Reduced core-outlet temperature is recommended as a preferred option
– All temperature limits are satisfied
• There are short-term effects
– Cycle efficiency drop from lower turbine-inlet temperature is not significant
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
25
ANL Plant Dynamics Code Validation
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
26
ANL Plant Dynamics Code V&V History
� S-CO2
cycle has never been commercially operated
� Earlier V&V was limited to benchmark calculations against similar codes elsewhere
� Later, limited code validation on individual components
– Heat exchangers (PCHE test loop at ANL)
– Turbomachinery (Ideal gas loop at SNL)
– Small-scale S-CO2
compressor test loop at Barber-Nichols Inc.
� Recently, Barber-Nichols, Inc. (BNI) and Sandia National Labs (SNL) constructed
and operated small-scale closed split-flow S-CO2
cycle loop
– SNL Loop will provide a configuration for PDC testing and validation of modeling
• CO2
working fluid
• Equipment designed to operate close to the critical point
• Recompression cycle configuration
– Recent PDC validation effort is focused on simulating the SNL Loop and data
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
27
July 14, 2011 SNL Loop Setup
� Single TAC
� Two recuperators
– HTR is “overdesigned” for this setup -> very little, if any, heat transfer in LTR
– Only HTR is simulated in the PDC
� 4 heaters
– 390 kW input limit at
BNI facility
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
28
SNL Loop Data and Modeling
� Some specifics of the experimental setup (for model validation purposes) were
identified
– Significant heat loss in pipes and in turbine
• The heat loss in turbine volute is not measured
– The loop could not be run for a long time to achieve “perfect” heat balance
• Required for steady-state initialization in the PDC
– Lack of information on PCHE internal configuration
– No water flow rate measurements
– Temperature measurement uncertainty (1.1 oC) is significant
� Some features were dealt with modeling assumptions
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
29
Transient Simulation: Assumptions
� The LTR was excluded from the PDC model
� The electrical heaters are simulated with shell-and-tube heat exchangers
– Sodium flow on the hot side for the steady-state model
– Direct heat input into the HX “tubes” in transients
� All of the heat losses in the system were simulated to occur in the HTR-heater pipe
– Including turbomachinery heat loss
– The amount of the heat loss was determined to achieve a heat balance throughout the
system at steady-state conditions
– In a transient, the heat loss is scaled with the changing CO2
temperatures in the pipe
� Other than the heat loss, no heat transfer is simulated in the pipes in the PDC
– Also includes the absence of heat exchange between the CO2
and the pipe walls
� The TAC drain flow is not included into the PDC steady-state and transient models
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
30
Steady-State Calculations
� Good agreement was achieved for
steady-state
– Required significant heat loss
simulation (19 kW) at heater inlet
• Simulates all heat losses in the
cycle
• Also compensates for not
achieving perfect heat balance in
HTR by 3356 s
– Skipping LTR did not affect
temperatures much
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
31
Time 3355.6 s1.779 kg/s T, °C
31.3 P, MPa232.7 231.3 7.4639.712 9.611 RPM
31.5 28.07.465 #N/A
210.7 46.3142.2 7.724 9.8679.751
207.2 45.4 45.6 46.1 45.9 33.97.464 9.784 9.772 9.893 7.496 #N/A
#N/A kg/s145.1 45.7 45.5 45.79.540 7.523 7.512 7.482
35,000
Heater
Alt
Cooler
CT
HTRLTR
1.859 kg/s T, °C0.0 31.5 P, MPa
232.2 232.1 7.451 Q, kW9.786 9.759 RPM192.20 35,000
31.7 28.07.475 0.228
16.50210.3 46.0 153.0
145.4 7.803 9.8509.794
209.5 45.9 45.8 33.07.532 9.838 7.487 0.101
7.301 kg/s-19.0 153.8 46.2
9.834 7.527
0.00
PDC Steady-State Results
33.30
15.00
401.2
Heater
Alt
Cooler
CT
HTRLTR
PDC
Data
Dynamic Simulation of SNL S-CO2Loop with the PDC
� Transient is defined by the experimental data through the external input:
� Initial conditions: “steady-state” at 3356 s
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
32
34,000
36,000
38,000
40,000
42,000
44,000
46,000
48,000
50,000
52,000
54,000
0 500 1000 1500 2000 2500
SHA
FT R
OTA
TIO
NA
L SP
EED
, R
PM
TIME, s
SHAFT SPEED
RPM A
PDC (Input)
200
220
240
260
280
300
320
340
360
0 500 1000 1500 2000 2500
TEM
PER
ATU
RE,
oC
TIME, s
HEATER OUTLET TEMPERATURE
Target Inlet Temp
PDC Input
27
27.5
28
28.5
29
29.5
0 500 1000 1500 2000 2500
TEM
PER
ATU
RE,
oC
TIME, s
WATER INLET TEMPERATURE
T900
PDC Input
0.998
1
1.002
1.004
1.006
1.008
1.01
0 500 1000 1500 2000 2500
TEM
PER
ATU
RE,
NO
RM
ALI
ZED
TO
ST
EAD
Y ST
ATE
VALU
E
TIME, s
COOLER OUTLET TEMPERATURE
Experimental Value
PDC Input
Shaft Speed Water Inlet Temperature
Target Heater Outlet TemperatureTarget Cooler Outlet Temperature
Transient Results Presentation
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
33
T600
PDC
� Compare all available experimental measurements with the corresponding PDC
predictions
– Temperatures, pressures, density, flow rates
� The following slides show such comparisons for each individual parameter
� Few results are presented here
– Differences are briefly discussed
– See the paper for more detailed comparison
Measured data (e.g., Temperature at point 600)
Corresponding code results
Transient Results Comparison
� Heater outlet temperature is predicted very accurately
– Results of the automatic control
– Short-term behavior is achieved by control’s PID coefficients optimization
– In some cases, there is not enough reserve capacity in the heater
• Discussed below
� Turbine inlet temperature is very close to the heater outlet
– There is, however, a somewhat noticeable time delay in the experimental data
• Not showing in the PDC – result of not simulating pipe heat capacity
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
34
200
220
240
260
280
300
320
340
360
0 500 1000 1500 2000 2500
TEM
PER
AT
UR
E, oC
TIME, s
HEATER OUTLET TEMPERATURE
T600
PDC
200
220
240
260
280
300
320
340
360
0 500 1000 1500 2000 2500
TEM
PERATU
RE, oC
TIME, s
TURBINE INLET TEMPERATURE
T100A
PDC
T600
PDC
1
2
3
4
1
2
3
4
40
45
50
55
60
65
0 500 1000 1500 2000 2500
TEM
PER
AT
UR
E, oC
TIME, s
HTR HOT SIDE OUTLET TEMPERATURE
T202
PDC
Transient Results Comparison
� Turbine outlet temperature is predicted accurately earlier in the transient
– But diverges with higher turbine speed
• Overprediction of turbine performance at higher speeds
� HTR hot side outlet temperature is slightly overpredicted by the PDC
– Overprediction of HTR performance
– Maximum difference is only 1.5 oC
• Measurement ΔT = 1.1 oC
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
35
1
2
3
200
220
240
260
280
300
320
0 500 1000 1500 2000 2500
TEM
PER
AT
UR
E, oC
TIME, s
TURBINE OUTLET TEMPERATURE
T200A
PDC
12
3
Transient Results Comparison
� Cooler outlet temperature is very close
– Maximum difference is only 0.3 oC
– Measured temperature was an input for PDC
• Verifies control performance, not necessarily cooler model
� Compressor outlet temperature is predicted very accurately
– Maximum difference is only 1 oC
• Within the measurement uncertainty
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
36
31
31.5
32
32.5
33
33.5
34
34.5
0 500 1000 1500 2000 2500
TEM
PER
AT
UR
E, oC
TIME, s
COOLER OUTLET TEMPERATURE
T399
PDC
40
45
50
55
60
65
0 500 1000 1500 2000 2500
TEM
PERATU
RE, oC
TIME, s
COMPRESSOR OUTLET TEMPERATURE
T500A
PDC
Transient Results Comparison
� Compressor outlet pressure is predicted accurately earlier and late in the
transient
– Pressure is overpredicted in the middle section
• High rotational speeds
� Turbine inlet pressure is consistently overpredicted by the code
– In part, due to compressor performance overprediction
– Also, underprediction of the pressure drops
• LTR
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
37
1
2
9
9.5
10
10.5
11
11.5
12
12.5
13
13.5
0 500 1000 1500 2000 2500
PR
ESSU
RE,
MPa
TIME, s
COMPRESSOR OUTLET PRESSURE
P500A
PDC1
2
1
9
9.5
10
10.5
11
11.5
12
12.5
13
13.5
0 500 1000 1500 2000 2500
PR
ESSU
RE,
MPa
TIME, s
TURBINE INLET PRESSURE
P100A
PDC
Transient Results Comparison
� Low side pressures show much better agreement
– Especially, for the compressor inlet
� Compressor inlet density prediction is close
– The only density measurement in the loop
– Some difference at later stages are possibly due to difference in the cooler-outlet
temperature
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
38
1
2
7.2
7.4
7.6
7.8
8
8.2
8.4
8.6
8.8
9
0 500 1000 1500 2000 2500
PR
ESSU
RE,
MPa
TIME, s
BRAYTON CYCLE PRESSURES: LOWP200A
Turb-out
(PDC)
P400A
Comp-in
(PDC)
1
340
350
360
370
380
390
400
410
420
0 500 1000 1500 2000 2500
DEN
SIT
Y, k
g/m
3
TIME, s
COMPRESSOR INLET DENSITY
DensityA
PDC
2
Transient Results Comparison
� CO2
flow rate is overpredicted by the code for most of the transient
– By about 5%
– Consistent with steady-state results
• Flow rate was calculated from the heat balance
• Perfect heat balance was not achieved at 3356 s
– Also, code predicts much faster change in the flow rate
• Possibly, due to heat capacity of turbomachinery, which is ignored in the PDC by virtue of using
the turbomachinery maps
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
39
1.7
1.8
1.9
2
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
2.9
0 500 1000 1500 2000 2500
FLO
W R
AT
E,
kg/s
TIME, s
BRAYTON CYCLE FLOW RATES
MassFlowA
Turb
(PDC)
Comp
(PDC)
Lessons Learned from SNL Loop Simulation (So Far)
� The PDC transient results are close to the experimental data
– The existing modeling capabilities may be sufficient
� Lack of information on PCHE internal configuration
– E.g., channel diameter, zigzag angle
� The loop could not be run for a long time to achieve “perfect” heat balance
– Required for steady-state initialization in the PDC
– “Steady-state” at supercritical conditions is highly recommended
� Heat loss in pipes, recuperators, and turbine is difficult to account for
– The heat loss in turbine volute is not measured
� Water flow rate measurements are needed for cooler model validation
� Complete understanding of the control action is required for modeling
� Temperature measurement uncertainty (1.1 oC) is significant
– Density is only measured at the compressor inlet
� Turbomachinery performance is overpredicted at higher speeds
– Need loss correlations better suited for small-scale TM
� Note that all of these effects are specific to the SNL small-scale loop and become
negligible for commercial-scale S-CO2
cycles
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
40
Summary
� Plant Dynamics Code is developed at ANL for analysis of S-CO2
cycles
– Addresses specific features of the cycle
� A new approach was developed to couple the PDC with SAS4A/SASSYS-1
– Coupled reactor-BOP calculations
� A control strategy for the S-CO2
Brayton cycle has been developed
– Combination of various control mechanisms
– Enables load following from 100% to 0% and transition to decay heat removal mode
� Active reactor control is implemented in the coupled PDC-SAS4A/SASSYS-1 codes
– Aspects of S-CO2
cycle operation on the reactor control are discovered and investigated
� PDC validation is ongoing
– Effort is concentrated on modeling the SNL S-CO2
Loop
– Lessons learned regarding simulation of the experimental facilities
"Recent Developments in S-CO2 Cycle Dynamic Modeling and Analysis at ANL" by A. Moisseytsev, 4th S-CO2 Symposium, Pittsburgh, PA, September 9-10, 2014
41