Compressed Air Energy Storage (CAES) Modelisation

[baptistedq]

Hello everyone,
I'm Baptiste, a french student in preparatory class for engeener schools.
For a project, i need to study Compressed Air Energy Storage (CAES) systems, and in particular Adiabatic CAES. I wanted to create a model on python and I found the Tespy module which seems to be really adapted to what I needed. I followed the tutorials and tried to model an adiabatic CAES system but I'm facing some difficulties in my work.
To explain quicly, a CAES is a system in which electric power is used to compress a gas in a chamber/cavern, where it stays until new power is needeed. Then, the gaz is decompressed through a turbine which create electric power. It is a system to store power. The adiabatic CAES is a CAES in which the heat created by the compression is kept in a heat exchanger and given back to the gas when it is decompressed.
Here is a schema of what I want to do :

First, has anyone ever tried to create the model of a CAES ?

I've tried two different codes, using SimpleHeatExchanger and HeatExchanger.

Here is the first code :

from tespy.networks import Network
from tespy.components import (Source, Compressor, Sink,HeatExchangerSimple, Turbine)
from tespy.connections import Connection, Ref, Bus
from tespy.tools import document_models

fluid_list=['air']
nw1=Network(fluids=fluid_list, p_unit='bar', T_unit='K')

cp=Compressor('compresseur')
air=Source('air source')
tu=Turbine('turbine')
ac=Sink('compressed air')
he=HeatExchangerSimple('heat exchanger')
he2=HeatExchangerSimple('2nd heat exchanger')

c1=Connection(air,"out1", cp, "in1", label="1")
c2=Connection(cp,"out1", he,"in1", label="2")
c3=Connection(he,"out1", he2,"in1", label="3")
c4=Connection(he2,"out1", tu, "in1", label="4")
c5=Connection(tu,"out1",ac,"in1", label="5")

nw1.add_conns(c1,c2,c3,c4,c5)

cp.set_attr(pr=10, eta_s=0.8)
c1.set_attr(p=1, T=198, v=5, fluid={'air':1}) 
he.set_attr(pr=0.85, L=1,D=0.2, Tamb=198) 
c3.set_attr(T=220)      # cooled down gas temperature
he2.set_attr(pr=0.85, L=1, D=0.2, Tamb=198)
c4.set_attr(T=400)      #heated gas temperature
tu.set_attr(pr=0.15,eta_s=0.8)


nw1.solve('design')
nw1.print_results()

I've got results whith this one but I don't really know if the model matches the model that I want to do.

My second code, with the casual heat exchanger is the following :

from tespy.networks import Network
from tespy.components import (Source, Compressor, Sink,HeatExchanger, Turbine)
from tespy.connections import Connection, Ref, Bus
from tespy.tools import document_models

fluid_list=['air','water']
nw1=Network(fluids=fluid_list, p_unit='bar', T_unit='K')

cp=Compressor('compressor')
air=Source('air source')
eauinit=Source('water source ')
eauout=Sink('water sink')
tu=Turbine('turbine')
ad=Sink('depressured air')
he=HeatExchanger('heat exchanger')
he2=HeatExchanger('2nd heat exchanger')

c1=Connection(air,"out1", cp, "in1", label="c1")
c2=Connection(cp,"out1", he,"in1", label="c2")
c3=Connection(he,"out1", he2,"in2", label="c3")
c4=Connection(he2,"out2", tu, "in1", label="c4")
c5=Connection(tu,"out1",ad,"in1", label="c5")
c6=Connection(eauinit,"out1", he,"in2", label="c6")
c7=Connection(he,"out2", he2,"in1", label="c7")
c8=Connection(he2,"out1", eauout,"in1", label="c8")

nw1.add_conns(c1,c2,c3,c4,c5,c6,c7,c8)

cp.set_attr(pr=10, eta_s=0.8)
c1.set_attr(fluid={'air': 1, 'water': 0}, p=1, T=198, v=20) 
c2.set_attr(fluid={'air': 1, 'water': 0})
#he.set_attr(pr1=0.85)                                 #pressure ratio in the heat exchanger maybe needed but tey where too many parameters
c3.set_attr(fluid={'air': 1, 'water': 0},T=220)      # cooled down gas temperature
#he2.set_attr(pr2=0.85)                                #same as he 
c4.set_attr(fluid={'air': 1, 'water': 0},T=400)      #heated gas temperature

tu.set_attr(pr=0.15)
c6.set_attr(fluid={'air': 0, 'water': 1}, T=168 ,v=2)

nw1.solve('design')
nw1.print_results()

This code doesn't run. Apparently, I shouldn't give both pressure and temperature as initial parameters for the gas and I don't understand how to replace it whith enthropy.

What I want is to be able to know pressure, temperature, air flow speed at the entrance, and how much power I can get in the output (turbine). Parameters in the middle of the systems can be adapted for it to work, it just requires that the temperature of the gas in the tank is not too high.

I tried my best to understand tespy language and to use it but it is kind of hard for me given that I'm just discovering it and that i've got to understand in english !
Of course I don't expect anyone to do it from scratch for me because that would be a lot of work, but if anyone have ever worked on a CAES model, or if I don't use the tespy modules right, let me know.
Thank you very much for reading me.
Baptiste

PS : sorry if there are english mistakes :)

[fwitte]

@baptistedq,

thanks for reaching out and welcome to the TESPy and oemof community :)!

Actually, CAES was one of the projects I worked wiht TESPy originally a couple of years ago, it is very possible to model these systems with the software. I can give you a quick overview on the approach I would take for this topic:

• You should create two separate TESPy models: one for charging and one for discharging the storage, you would not operate them simultaneously anyways.
  • Build the models in a class or a function and run the models individually with the input parameters required (e.g. target mass flow or target power of the individual section of the plant).
• Create two storage classes, one for the heat storage one for the cavern storage.
  • The carvern storage should be able to calculate the temperature and the pressure inside the cavern based on the amount of air injected into the cavern with its temperature or the amout of air extracted from the cavern.
  • The heat storage should be able to calculate the amout of energy stored based on the energy inputted from the injection cooling process or based on the energy retracted for the expansion heating process.

Try to build these things up step by step, everything at once might be quite a lot to do. If the time fits for you, I'd invite you to discuss this topic at the next online user meeting on May 15th.

Best

Francesco

[baptistedq]

@fwitte

Hello, thank you for your answer ! I'm sorry it took me so long to answer, I'm in the middle of a period of exam.

I'm going to follow your advices and see if I succed to get something !
I'll be pleased to attend to the meeting, i'll see if I can be there.

Best,

Baptiste

[baptistedq]

Hi, I have one other question if you can also help me :
Is there somwhere I can see the calculation used to define the differents modules ? (I've seen the source code of the modules but it seems very hard to get the calculations from this)
Because, for example, with the compressor, I don't get the same results at all between the Tespy model and my calculations.

My calculs are the following for a compression from air at T1=283 K and p1=1 bar to air at p2= 10 bar :
(I've put eta_s=1 so it is a reversible compression)

where :
P= power (W)
m= mass flow (kg/s) I took m=1
R=perfects gaz constant (J/K/mol) R=8.314
M=air molar mass (kg/mol) M=29*10E-3
y=Cp/Cv=1.4

with this calculation I get P=9.1*10e8 W whereas the model gives me P=2.65*10e5 W

from tespy.components import Sink, Source, HeatExchangerSimple, Compressor
from tespy.connections import Connection
from tespy.networks import Network



fluids = ['air']
nw = Network(fluids=fluids)
nw.set_attr(p_unit='bar', T_unit='K', h_unit='kJ / kg', m_unit='kg / s', iterinfo=False)

so1 = Source('source 1')
si1 = Sink('sink 1')
heat_sink = HeatExchangerSimple('heat sink')
cp = Compressor('compressor')

cp.set_attr(pr=10,eta_s=1)
heat_sink.set_attr(Tamb=283, pr=0.95, design=['pr'])



c1 = Connection(so1, 'out1', cp, 'in1', label='c1')
c2 = Connection(cp, 'out1', heat_sink, 'in1', label='c2')
c3 = Connection(heat_sink, 'out1', si1, 'in1', label='c3')
nw.add_conns(c1, c2, c3)
c1.set_attr(fluid={'air': 1}, m=1, T=283, p=1)
c3.set_attr(T=300, design=['T'])

nw.solve('design')
nw.print_results()

I understand there might be some gaps between the calculations I make (which are modelised as if air was a perfect gas) but here there is a huge gap !

Thank you for reading me,
Cordialy

Baptiste

[fwitte]

I am not sure, I do think your formula is incorrect. It should be

P = m*c_p*T_1*[(p_2/p_1)^((y-1)/y)-1]

for ideal gases, for example see https://en.m.wikipedia.org/wiki/Isentropic_process

Best