Rankine Cycle + HRSG #554
Replies: 1 comment 7 replies
-
|
There's some problem with two branches that split and then merge, and the equalization of pressures. One of these branches is the economizer bypass, and the other is the sweet-water condenser (for superheated steam attemperation). Apart from that, the code assumes that the drum component internally, implicitly equalizes all pressures, as a merge does (or a split). That's why the pr1=1 equation is not indicated in the LPEV1 (or HPEV1). EDIT: By deleting pr1=1 in the HPEC1 one of these problems should disappear. But when I tried to do so in HE_COND, obviously an equation is missing, and I can't figure which. Python Codedef export(__file__, results):
import os
path = os.path.dirname(os.path.abspath(__file__))
text_file = os.path.join(path, 'results.txt')
formatted_text = results.to_string(index=False, justify='left')
with open(text_file, 'w') as file:
file.write(formatted_text)
from tespy.networks import Network
from tespy.components import (
CycleCloser,
Turbine,
Pump,
Valve,
Source,
Sink,
Splitter,
Merge,
SimpleHeatExchanger,
HeatExchanger,
Drum,
)
from tespy.connections import Bus
from tespy.tools import ExergyAnalysis
my_plant = Network()
my_plant.set_attr(
T_unit='C',
p_unit='bar',
h_unit='kJ / kg',
m_unit='t / h',
s_unit='kJ / kgK',
iterinfo=False
)
powergen = Bus("electrical power output")
my_plant.add_busses(powergen)
pamb = 1.013 # bar(a)
Tamb = 17 # C
cc = CycleCloser('cycle closer HPDRUM to ST')
st_ext = Turbine('extraction st', eta_s=0.9)
st_cond = Turbine('cond st', eta_s=0.9)
att_pump = Pump('att_pump', eta_s=0.7)
cond_pump = Pump('cond_pump', eta_s=0.7)
sg_pump = Pump('steam generator pump', eta_s=0.7)
lp_pump=Pump('low pressure pump', eta_s=0.7)
cond_st_cv = Valve('cond st cv', pr=0.85)
water_inlet_1 = Source('Process condensate return')
water_inlet_2 = Source('Water makeup')
turbine_exhaust = Source('Hot gases source')
wst = Sink('LPDRUM saturated steam waste')
wst_2 = Sink('HPDRUM saturated liquid waste')
process_steam = Sink('process IP steam')
gases_outlet = Sink('hot gases outlet')
sp_1 = Splitter('st inner splitter')
sp_2 = Splitter('condensate header')
sp_3 = Splitter('HPEC1 splitter')
sp_4 = Splitter('HPDRUM outlet')
bd_1 = Splitter('LPDRUM splitter to waste')
bd_2= Splitter('HPDRUM splitter to waste')
me_1 = Merge('Attemper')
me_3 = Merge('Water from reposition & condensate mixture', num_in=3)
me_4 = Merge('merge at the inlet of sweet water condenser in SG')
me_5 = Merge('attemp HPS2')
HE = HeatExchanger('heat exchanger', pr1=0.9, pr2=0.9)
LPEV1 = HeatExchanger('LPEV1', pr2=1)
LPDRUM = Drum('LP Drum')
#HPEC1 = HeatExchanger('HPEC1', pr1=1, pr2=1)
HPEC1 = HeatExchanger('HPEC1', pr2=1)
HPEC2 = HeatExchanger('HPEC2',pr1=1, pr2=1)
HPEC3 = HeatExchanger('HPEC3',pr1=1, pr2=1)
HPEV1 = HeatExchanger('HPEV1', pr2=1)
HPS1 = HeatExchanger('HPS1', pr1=1, pr2=1)
HPS2 = HeatExchanger('HPS2', pr1=1, pr2=1)
HPDRUM = Drum('HP Drum')
condenser = SimpleHeatExchanger('condenser', pr=1)
#HE_cond = HeatExchanger('sweet water condenser', pr1=1, pr2=1)
HE_cond = HeatExchanger('sweet water condenser', pr1=1)
powergen.add_comps(
{"comp": st_ext, "char": 0.97, "base": "component"},
{"comp": st_cond, "char": 0.97, "base": "component"},
{"comp": att_pump, "char": 0.97, "base": "bus"},
{"comp": cond_pump, "char": 0.97, "base": "bus"},
{"comp": sg_pump, "char": 0.97, "base": "bus"},
)
from tespy.connections import (Connection, Ref)
N = 50
c = [None] * N
names = [None] * N
# caudal_vapor_sobrecalentado = 2*102.3 # t/h
c[0] = Connection(cc, 'out1', st_ext, 'in1', label='1')
c[0].set_attr(
#m=caudal_vapor_sobrecalentado,
fluid={'water': 1}
)
names[0] = 'hp steam'
c[1] = Connection(st_ext, 'out1', sp_1, 'in1', label='2')
names[1] = 'extraction st outlet'
caudal_extraccion_turbina = 160 # t/h
c[2] = Connection(sp_1, 'out2', me_1, 'in2', label='3')
c[2].set_attr(m=caudal_extraccion_turbina)
names[2] = 'Steam from TV to attemp'
c[3] = Connection(sp_1, 'out1', cond_st_cv, 'in1', label='4')
names[3] = 'cond st cv inlet'
c[4] = Connection(cond_st_cv, 'out1', st_cond, 'in1', label='5')
names[4] = 'cond st inlet'
c[5] = Connection(st_cond, 'out1', condenser, 'in1', label='6')
names[5] = 'condenser inlet'
c[6] = Connection(condenser, 'out1', sp_2, 'in1', label='7')
c[6].set_attr(x=0, T=Tamb + 18)
names[6] = 'condenser outlet'
c[7] = Connection(sp_2, 'out1', cond_pump, 'in1', label='8')
names[7] = 'cond pump inlet'
c[8] = Connection(sp_2, 'out2', att_pump, 'in1', label='9')
names[8] = 'att pump inlet'
c[9] = Connection(att_pump,'out1', me_1,'in1',label='10')
names[9] = 'att pump outlet'
presion_vapor_proceso = 13 # bar(a)
grado_sobrecalentamiento = 10 # C
c[10] = Connection(me_1, 'out1',process_steam,'in1',label='11')
c[10].set_attr(
Td_bp=grado_sobrecalentamiento,
p=presion_vapor_proceso
)
names[10] = 'process steam'
c[11] = Connection(cond_pump,'out1', me_3, 'in1', label='12')
c[11].set_attr(p=pamb)
names[11] = 'cond pump outlet'
retorno_condensado = 0.75 # dim
c[12] = Connection(water_inlet_1, 'out1', me_3, 'in2', label='13')
c[12].set_attr( m=Ref(c[10], retorno_condensado, 0), fluid={'water': 1})
names[12] = 'Condensate return from process'
c[13] = Connection(water_inlet_2,'out1',me_3,'in3',label='14')
c[13].set_attr(T=Tamb)
names[13] = 'Water makeup'
c[14] = Connection(me_3,'out1', lp_pump,'in1', label='15')
names[14] = 'Water mixture feed to low pressure pump'
c[15] = Connection(lp_pump,'out1', HE,'in2', label='16')
c[15].set_attr(p=2.3, T=55)
names[15] = 'Water mixture feed to heat exchanger'
c[16] = Connection(HE,'out2', LPDRUM,'in1',label='17')
c[16].set_attr(T=100)
names[16] = 'heated water feed to LPDRUM'
c[17] = Connection(HE,'out1', sg_pump,'in1', label='18')
names[17] = 'Cooled water feed to HP pump'
c[18] = Connection(sg_pump, 'out1', sp_3,'in1', label='19')
c[18].set_attr(p=88.52)
names[18] = 'pressurized water to HPEC1'
c[19] = Connection(sp_3, 'out1', me_4, 'in2', label='20')
c[19].set_attr(m=0)
names[19] = 'pressurized water to sg condenser'
c[20] = Connection(sp_3, 'out2', HPEC1, 'in1', label='21')
names[20] = 'pressurized water to HPEC1'
c[21] = Connection(HPEC1, 'out1', HPEC2, 'in1', label='22')
c[21].set_attr(T=116.1)
names[21] = 'pressurized water to HPEC2'
c[22] = Connection(HPEC2, 'out1', me_4, 'in1', label='23')
c[22].set_attr(T=230.9)
names[22] = 'outlet of HPEC2 to merge with cold flow'
c[23] = Connection(me_4, 'out1', HE_cond, 'in1', label='24')
c[23].set_attr(p=88.52)
names[23] = 'outlet of merge to sweet water condenser'
c[24] = Connection(HE_cond, 'out1', HPEC3, 'in1', label='25')
names[24] = 'outlet of sweet water condenser to HPEC3'
c[25] = Connection(HPEC3, 'out1', HPDRUM, 'in1', label='26')
c[25].set_attr(T=289.9)
names[25] = 'outlet of HPEC3 to HP drum'
c[26] = Connection(HPDRUM, 'out1', bd_2, 'in1', label='27')
names[26] = 'saturated liquid outlet of HPDRUM to Splitter'
c[27] = Connection(bd_2, 'out1', HPEV1, 'in1', label='28')
names[27] = 'saturated liquid outlet of HPDRUM to HPEV1'
c[28] = Connection(bd_2, 'out2', wst_2, 'in1', label='29')
names[28] = 'saturated liquid outlet of HPDRUM to waste'
c[28].set_attr(m=2)
c[29] = Connection(HPEV1, 'out1', HPDRUM, 'in2', label='30')
c[29].set_attr(x=0.05)
names[29] = 'saturated steam outlet of HPEV1 to HPDRUM'
c[30] = Connection(HPDRUM, 'out2', sp_4 , 'in1', label='31')
names[30] = 'saturated steam outlet of HPDRUM'
c[31] = Connection(sp_4, 'out1', HPS1, 'in1', label='32')
names[31] = 'saturated steam inlet to HPS1'
c[32] = Connection(HPS1, 'out1', me_5, 'in1', label='33')
c[32].set_attr(T=528.6)
names[32] = 'Steam outlet of HPS1 to attemp'
c[33] = Connection(sp_4, 'out2', HE_cond, 'in2', label='34')
names[33] = 'Steam derivation of HPDRUM to sweet water condenser'
c[34] = Connection(HE_cond, 'out2', me_5, 'in2', label='35')
c[34].set_attr(x=0)
names[34] = 'Sweet water condenser outlet to attemp'
c[35] = Connection(me_5, 'out1', HPS2, 'in1', label='36')
c[35].set_attr(T=428.8)
names[35] = 'Attemp outlet to HPS2'
c[36] = Connection(HPS2, 'out1', cc, 'in1', label='37')
temperatura_vapor_sobrecalentado = 480 # C
c[36].set_attr(T=temperatura_vapor_sobrecalentado)
names[36] = 'Superheated steam to steam turbine'
c[37] = Connection(LPDRUM,'out1', bd_1,'in1', label='38')
names[37] = 'Saturated liquid outlet of LPDRUM to LPEV1'
c[38] = Connection(bd_1, 'out1', HE, 'in1', label='39')
names[38] = 'LPDRUM saturated liquid outlet to HE'
c[39] = Connection(bd_1, 'out2', LPEV1, 'in1', label='40')
names[39] = 'Saturated liquid inlet to LPEV1'
c[40] = Connection(LPEV1, 'out1', LPDRUM, 'in2', label='41')
c[40].set_attr(x=0.05)
names[40] = 'Saturated steam inlet to LPDRUM'
c[41] = Connection(LPDRUM, 'out2', wst, 'in1', label='42')
names[41] = 'Saturated steam outlet to vent'
c[41].set_attr(m=2*0.05)
c[42] = Connection(turbine_exhaust, 'out1', HPS2, 'in2', label='43')
x_gh_CO2 = 0.061 # dim
x_gh_H2O = 0.07768 # dim
x_gh_N2 = 0.7364 # dim
x_gh_O2 = 0.1126 # dim
x_gh_Ar = 1 - x_gh_CO2 - x_gh_H2O - x_gh_N2 - x_gh_O2
T_gh = 712.9 # C
G_gh = 2*481.98 # t/h
p_gh = pamb # bar(a)
c[42].set_attr(
m=G_gh,
T=T_gh,
p=p_gh,
fluid={'O2': x_gh_O2, 'N2': x_gh_N2, 'CO2': x_gh_CO2, 'water': x_gh_H2O, 'Ar': x_gh_Ar}
)
names[42] = 'Hot gases input to HPS2'
c[43] = Connection(HPS2, 'out2', HPS1, 'in2', label='44')
names[43] = 'Hot gases input to HPS1'
c[44] = Connection(HPS1, 'out2', HPEV1, 'in2', label='45')
names[44] = 'Hot gases input to HPEV1'
c[45] = Connection(HPEV1, 'out2', HPEC3, 'in2', label='46')
names[45] = 'Hot gases input to HPEC3'
c[46] = Connection(HPEC3, 'out2', HPEC2, 'in2', label='47')
names[46] = 'Hot gases input to HPEC2'
c[47] = Connection(HPEC2, 'out2', LPEV1 , 'in2', label='48')
names[47] = 'Hot gases input to LPEV1'
c[48] = Connection(LPEV1, 'out2', HPEC1 , 'in2', label='49')
names[48] = 'Hot gases input to HPEC1'
c[49] = Connection(HPEC1, 'out2', gases_outlet , 'in1', label='50')
names[49] = 'Hot gases outlet'
for j in range(0,N):
my_plant.add_conns(c[j])
my_plant.solve(mode='design')
my_plant.print_results()
bus_results = my_plant.results['electrical power output']
df_results_for_conns = my_plant.results['Connection']
df_results_for_conns['denomination'] = names
results = df_results_for_conns[['denomination',
'p','p_unit',
'T','T_unit',
'h','h_unit',
's','s_unit',
'x',
'm','m_unit']]
start_index_for_gases = 42
results_gases = results[start_index_for_gases:]
results_gases = results_gases[['denomination',
'p','p_unit',
'T','T_unit',
'h','h_unit',
's','s_unit',
'm','m_unit']]
results = results[0:start_index_for_gases]
results = results.drop(results.index[0])
results = results.round(3)
export(__file__, results) |
Beta Was this translation helpful? Give feedback.
-
|
Wow, impressive! And in such a short time! A bit beyond my current expertise, I must say. The use of initial guesses is quite interesting. This boiler has two loops: the economizer bypass and the HP desuperheater. The economizer bypass actually represents a performance downgrade for the boiler. The original idea was to allow it to produce less than 70 tph of steam (the amount produced when no supplemental firing is used). The boiler could generate down to 60 tph, using this loop, and, with the help of the diverter damper—and considering this setup uses two HRSGs—the CHP plant could produce almost any steam flow between 40 tph and 210 tph, which is a very wide range. The gas turbine output was 55 MW each (ISO performance). The real economizer has eight stages. Here there are just three or four, to simplify. In this model, a control valve is missing in this bypass to balance the pressure drop between the economizer and the bypass. In fact, this CHP plant was commissioned just last month and the startup was last week. I have plenty of process data, but of course, I can’t share it without anonymizing it. I was responsible for the thermal cycle engineering, and then I took part in the basic engineering (and in the detail engineering, to some extent). Interestingly, this boiler operated with what could be called a “negative approach” (the temperature coming out of the economizer was higher than that of the HP drum). This issue was managed by increasing the economizer pressure. Although this solution is inefficient from an energy standpoint, I believe it was chosen for operational safety, likely to reduce the risk of economizer steaming. The sweetwater condenser offers an excellent solution for HP steam desuperheating, in my experience. A more complete model would ideally include the temperature control valve. Also, pegging steam (deriving hp steam to lp drum) wasn't modelled, aiming to simplify (this is an undergraduate final project). The blowdown disposal, and the blowdown flashing weren't modelled as well. The data in this model was set somehow without order, just to match the process data. We should choose better how to define the thermodynamic states. It should be done with set points and control loops in mind. Here are a few things I learned while working on this with my student (who was really the one doing the hard work):
Besides, it tasted unrealistic to me that the condensate pump, makeup pump and process condensate pump all had the same discharge pressure (caused by the merge component). I know that this is a mixing chamber, as we find in thermodynamic problems, but it might be a good idea a component that does energy balance but with no requirement for pressure. But some bad choices could lead to unreal solutions and entropy destruction (with small inlet pressure and high outlet pressure).
We'll check the results with the manufacturer data. We'll continue refining the setup and testing the library’s capabilities—possibly exploring exergy analysis, off-load performance, and exporting to AutoCAD for automatic generation of PFDs. Please take this feedback as a piece of gratitude, and apologies if my writing reveals rude or if the comments are out of place. I'm satisfied with the library. With practice, I believe it's possible to achieve a high coding speed. I'll try to contribute to the community, and I'll be happy to review a new version of the code if it's needed. Thanks, again!! |
Beta Was this translation helpful? Give feedback.
-
|
Thank you for all the insights, very interesting! Would you mind, if we somehow arrange a web conference at some point in the future? I created a draft (#558) and would like to make a review, when I have the first complete draft ready. The tutorial will be placed on this page https://tespy--558.org.readthedocs.build/en/558/tutorials/hrsg_process_steam.html#tespy-tutorial-hrsg-label.
Thank you :) Best Francesco |
Beta Was this translation helpful? Give feedback.
-
This again seems to be an issue with starting values. With the converged solution of step 1 you can fix the mass flow 11 instead of mass flow 3. Btw.: Note, that I changed the numbering on the connections in the pull request.
Maybe this is not clear enough from the documentation and from the examples then? Are there particular components you are referring to?
This would be a MergeWithPressureloss?
You mean a new class
We could introduce a new component: DiabaticHeatExchanger. This component could be modeled using something like a kA factor describing the heat transfer coefficient to the ambient while using the mean temperature of the shell side fluid and the ambient temperature or a heat loss directly proportional to the total amount of heat transferred.
I think that would not be too hard, the API to connect tespy with pygmo is available. Once you have a stable model, optimization should be quite straight-forward.
Those could be calculated based on a design case. I guess those information are nowhere specified, but deriving them with the help of the model is fine in my opinion.
You could also fix a certain mass flow here, e.g. in relation to the primary mass flow.
Then it would just be naming the component in a different way? The fact that we have two-phase fluid (just very near to liquid) is still correct, or? |
Beta Was this translation helpful? Give feedback.
-
|
Hi, First, I believe it would be better to add a (control) valve in the branch that goes from HP EC1 to the merge after HP EC2. The pressure ratio (pr) for that valve would be a calculated parameter. This way, the model has the flexibility to select a different pressure ratio for the economizer stages. Without this valve, the only valid value for the pressure ratio would be pr1 = 1, which limits the model's functionality. The idea is to be able to model the boiler capacity to underperform by bypassing the economizer last stages. A similar issue occurs with the sweetwater condenser from HP COND to the merge after HP SH1. Otherwise, the head loss in HPSH1 must be set to one, which is highly unrealistic. I acknowledge that these are flaws in the original model I provided. Results: This is with G_gh = 2 × 481.98 t/h (which corresponds to c[42]). However, the main steam production of 361.2 t/h remains unchanged despite variations in the flue gas flow. This behavior is incorrect. It gets worse: when G_gh = 481.98 t/h, the model returns NaN values for the flue gas connections. The expected result is around 75 t/h when G_gh = 482 t/h and T_gh = 560°C. The HRSG should produce up to 110 t/h when T_gh is approximately 750°C (I am recalling this from memory and should double-check). In all cases, the flue gas outlet (i.e., the stack) should have a temperature between 95°C and 115-120°C. However, the model shows a stack temperature of 225°C, which is incorrect. Regarding parameter specification, I believe there’s room for a cleaner setup than what I initially provided. I'll write more about this later, along with other pending topics. Finally, feel free to contact me at [email protected] to arrange a virtual meeting. |
Beta Was this translation helpful? Give feedback.
-
I will do so in about a week or two :). Thank you |
Beta Was this translation helpful? Give feedback.
-
Issue with NaN Results in TESPy Model
Hi, community. Hope you're all great.
Together with a student (I'm his tutor), we are developing a model of a HRSG and a small Rankine Cycle. The process data is from real pieces of equipment.
The (ackward) truth is that we are stuck. We cannot find why the system shows
NaNin the results.TPP - PFD (2).pdf
The process fluid diagram is attached. It contains the connections' numbers (almost every connection is represented, with the small exception of a valve inside the steam turbine and the gas path).
I’m fully aware that debugging is really a pain in the foot, and that this code is still a work-in-progress in terms of style. I am posting to ask for help, and I’ll continue improving readiness and clarity.
Thanks in advance! 🙏
PS: Apologies for the bad English and the Spanish-English messy mixture
Python Code
Beta Was this translation helpful? Give feedback.
All reactions