Tech Science Press

DOI: 10.32604/ee.2023.027593

ARTICLE

Modeling of Large-Scale Hydrogen Storage System Considering Capacity

Attenuation and Analysis of Its Efficiency Characteristics

Junhui Li1 , Haotian Zhang1 , Cuiping Li1, *, Xingxu Zhu1 , Ruitong Liu2 , Fangwei Duan2 and
Yongming Peng3
1
Key Laboratory of Modern Power System Simulation and Control & Renewable Energy Technology,
Ministry of Education (Northeast Electric Power University), Jilin, 132012, China
2
Ministry of Science and Technology, Electric Power Research Institute of State Grid Liaoning Electric Power Co., Ltd.,
Shenyang, 110006, China
3
Northeast Electric Power Dispatching Control Center, Northeast Branch of State Grid Corporation of China,
Shenyang, 110180, China
* Corresponding Author: Cuiping Li. Email: licuipingabc@163.com
Received: 05 November 2022 Accepted: 16 January 2023 Published: 25 January 2024

ABSTRACT
In the existing power system with a large-scale hydrogen storage system, there are problems such as low efficiency of
electric-hydrogen-electricity conversion and single modeling of the hydrogen storage system. In order to improve
the hydrogen utilization rate of hydrogen storage system in the process of participating in the power grid operation,
and speed up the process of electric-hydrogen-electricity conversion. This article provides a detailed introduction
to the mathematical and electrical models of various components of the hydrogen storage unit, and also establishes
a charging and discharging efficiency model that considers the temperature and internal gas partial pressure of the
hydrogen storage unit. These models are of great significance for studying and optimizing gas storage technology.
Through these models, the performance of gas storage units can be better understood and improved. These studies
are very helpful for improving energy storage efficiency and sustainable development. The factors affecting the
charge-discharge efficiency of hydrogen storage units are analyzed. By integrating the models of each unit and
considering the capacity degradation of the hydrogen storage system, we can construct an efficiency model for a
large hydrogen storage system and power conversion system. In addition, the simulation models of the hydrogen
production system and hydrogen consumption system were established in MATLAB/Simulink. The accuracy and
effectiveness of the simulation model were proved by comparing the output voltage variation curve of the simulation
with the polarization curve of the typical hydrogen production system and hydrogen consumption system. The
results show that the charge-discharge efficiency of the hydrogen storage unit increases with the increase of
operating temperature, and H2 and O2 partial voltage have little influence on the charge-discharge efficiency. In
the process of power conversion system converter rectification operation, its efficiency decreases with the increase
of temperature, while in the process of inverter operation, power conversion system efficiency increases with the
increase of temperature. Combined with the efficiency of each hydrogen storage unit and power conversion system
converter, the upper limit of the capacity loss of different hydrogen storage units was set. The optimal charge-
discharge efficiency of the hydrogen storage system was obtained by using the Cplex solver at 36.46% and 66.34%.

KEYWORDS
Hydrogen storage system; simulation modeling; electrolyzer; fuel cell; capacity loss

This work is licensed under a Creative Commons Attribution 4.0 International

License, which permits unrestricted use, distribution, and reproduction in any
medium, provided the original work is properly cited.
292 EE, 2024, vol.121, no.2

1 Introduction
The Guiding Opinions on Accelerating the Development of New Energy Storage issued by the
National Development and Reform Commission and the National Energy Administration in July 2021
[1], stating that hydrogen energy is regarded as “new energy storage”, which indicates that hydrogen
energy storage has great development prospects. Compared with other electrochemical energy storage,
hydrogen energy storage is a green energy storage technology, which can be used in scenarios such as
absorbing renewable energy, grid peak shaving and valley filling, and microgrid. However, there are
few studies on the charge-discharge efficiency, hydrogen production and hydrogen consumption flow
rates of hydrogen storage systems. In order to better understand the performance of hydrogen storage
systems (HSS) and provide a basis for their participation in power system peak shaving and capacity
allocation [2], it is necessary to conduct in-depth research and analysis of the internal characteristics
of the entire hydrogen storage system.
The electrochemical reaction of the HSS during charging and discharging is a nonlinear process
affected by many factors. In terms of electrical engineering applications, in order to facilitate research,
common electrical components are used, and the external characteristics of the voltage and current
of the hydrogen storage system during the charging and discharging process are connected, and an
equivalent circuit model is constructed to clearly describe the charging and discharging characteristics
of the HSS. In terms of battery characteristics analysis and modeling, Li et al. [3] constructed
a third-order dynamic model of lead-acid batteries, which provided a basis for the rational use
of lead-acid batteries and state monitoring; He et al. [4] introduced three modeling methods for
the operating efficiency and life decay of electrochemical energy storage systems; Feng et al. [5]
comprehensively considered factors such as battery capacity, temperature, and rate, and analyzed
the power battery terminal voltage and ohmic polarization, electrical The relationship between
chemical polarization, concentration polarization, unbalanced potential, hysteresis characteristics and
other factors, a dynamic comprehensive equivalent circuit model is proposed to analyze the battery
characteristics; Li et al. [6] based on the working mechanism and external characteristics of vanadium
redox batteries, we have established a battery simulation model. Through this model, we studied
the voltage and current loss characteristics of vanadium redox batteries under rated power charging
and discharging modes, and measured the changes in charging and discharging efficiency at low
temperatures under different power levels; Liu [7] analyzed a charging and discharging efficiency model
for lithium iron phosphate batteries was established and the factors affecting efficiency were analyzed.
For the research on the characteristics of each part of the hydrogen storage system, Liu et al. [8]
introduced the characteristics of hydrogen energy storage and the current mathematical model of
electricity-hydrogen energy storage; Qin et al. [9] established a wind-hydrogen coupled system model,
and has Modeling of hydrogen storage and hydrogen consumption; Ren et al. [10] considered the physi-
cal characteristics of the HSS, and established a high-pressure hydrogen storage tank operation model
based on Fan’s equation; Gu et al. [11] detailed description of the various characteristics of proton
exchange membrane fuel cells and construction of its simulation model; Farrukh et al. [12] established
mathematical models of electrolytic cells and fuel cells, and studied the effects of various parameters on
system energy efficiency and external energy efficiency from an energy perspective. At the same time,
we also considered the participation of hydrogen storage systems; Cai et al. [13] mainly focused on the
hydrogen consumption and durability characteristics of fuel cells, and proposes an optimal control
method for multi series fuel cell systems that considers system hydrogen consumption and durability,
in order to reduce system hydrogen consumption and extend service life; Cai et al. [14] combined the
electrochemical characteristics, gas flow, and temperature of proton exchange membrane fuel cells,
taking into account the efficiency and lifespan of fuel cells, and establishes a customized dynamic
EE, 2024, vol.121, no.2 293

power generation system model based on PSCAD/EMTDC, and a grid-connected control strategy is
proposed; Cai et al. [15] analyzed the power exchange characteristics of heterogeneous energy sources
in the wind/solar hydrogen production system, establishes a homogenization analysis model of the
wind/solar hydrogen production system, analyzes the operating point law and the operating point
index range, analyzes the system operation domain and makes a reasonable analysis.
For the existing research on hydrogen storage system, Amir et al. [16] analyzed an innovative
combination of solar cells has been established based on parabolic trough collectors and proton
exchange membrane electrolyzers and fuel cell as a new green hydrogen production configuration from
the perspective of thermodynamics and actual economy; Seyed et al. [17] proposed A new hybrid
energy production/storage system has been proposed, including compressed air energy storage, a
Brayton cycle driven by a heliostat, and a hydrogen production device. This system can reduce the
pollutant emissions of compressed air energy storage technology, and has the advantages of high
efficiency and environmental protection; Rama et al. [18] mainly introduced the importance of existing
porous materials, polymers, metals and composite metal hydrides in solid hydrogen storage, and
emphasized the importance of catalysts; a new hydrogen production system has been constructed using
medical waste and zero carbon emissions biogas [19]. The system consists of plasma gasification, solid
oxide fuel cells, steam turbine cycles, and autothermal reforming. This system can efficiently utilize
waste resources and achieve a zero carbon emission hydrogen production process; research has been
conducted on comprehensive efficiency [20], hydrogen production rate, and economic efficiency, and
the use of cement waste heat through the Rankine cycle to generate the required electrical energy for
the operation of PEM electrolytic cells; the power allocation degradation problem in multi stack fuel
cell systems was considered to meet the load power requirements [21].
Research on System Capacity Decay Characteristics, Liu et al. [22] used FLUENT software
to establish a numerical model for the leakage and diffusion of high-pressure hydrogen storage
tanks, and studied its influencing factors. This method can be used to simulate the leakage and
diffusion process of high-pressure hydrogen storage tanks, which helps to evaluate and optimize
relevant safety measures. Li et al. [23] tested the durability of fuel cells through power generation-
electrolysis cycle experiments, and analyzed the reasons for fuel cell performance degradation through
electrochemical impedance spectroscopy and microphysical characterization methods. Zhu et al. [24]
has been established to describe the aging degree of metal hydrogen storage tanks, based on the pressure
composition temperature curve of the hydrogenation process of metal hydrogen storage tanks. This
model can be used to evaluate the aging of hydrogen storage tanks and provide a reference for the
maintenance and replacement of hydrogen storage tanks. Xu et al. [25] took 18650 NCM811 lithium-
ion battery as the research object to study the influencing factors of the capacity decay of lithium-ion
battery. Zhao et al. [26] conducted node tests on lithium titanate batteries for a specific number of
cycles, and used the Arrhenius formula to establish a capacity decay characteristic curve.
Current research may not fully consider the factors of the entire hydrogen storage system when
modeling hydrogen storage systems, and there are few literature describing the relationship between
temperature and pressure on the charging and discharging efficiency of hydrogen storage systems.
In addition, there is a lack of detailed description of the factors that affect the capacity decay
characteristics of hydrogen storage systems. Therefore, future research can continue to explore the
modeling of the entire hydrogen storage system, and delve into the effects of temperature and pressure
on the charging and discharging efficiency of the hydrogen storage system, while analyzing and
describing the factors that affect the capacity attenuation characteristics of the hydrogen storage
system.
294 EE, 2024, vol.121, no.2

In this paper, the capacity attenuation characteristic equation applicable to the hydrogen storage
system is constructed first considering the system capacity attenuation. On this basis, the simulation
models of each part and the whole of the large-scale hydrogen storage system are established, and the
mathematical models and electrical models of each unit of the hydrogen storage system are introduced.
The physical meaning of each part is analyzed. The hydrogen storage system is systematically simulated
and analyzed in MATLAB/Simulink using the electrical model. The validity of the simulation
model is proved by comparing the polarization curves of typical hydrogen production and hydrogen
consumption systems. Finally, the optimal charging and discharging efficiency of hydrogen storage
system considering the capacity attenuation factor is obtained by synthesizing the unit model.

2 Network Structure of Large-Scale Hydrogen Storage System

The simplified structure of the power system with hydrogen storage system is shown in Fig. 1.
The DC bus connects the large-scale hydrogen storage system with the public power grid and the
photovoltaic power generation system. The large-scale hydrogen storage system constructed in this
paper refers to a large-scale system consisting of a collection of multiple internal hydrogen storage
units and an external power conversion system (PCS) converter. Each hydrogen storage unit includes
a hydrogen production unit (PEM electrolyzer), a hydrogen storage tank and a hydrogen consumption
unit (Proton exchange membrane fuel cell) [27]. Each electrolyzer unit and fuel cell unit are connected
in series and parallel. Hydrogen production needs to meet the needs of local industries. By connecting
the photovoltaic system, cheap renewable energy can be used to power the electrolyzer to obtain
hydrogen and oxygen, which can be stored in the hydrogen storage tank. Acceptable electrical energy
is delivered to the grid and can also reduce the amount of electricity purchased from the grid [28].
Since the actual electricity-to-hydrogen-to-electricity conversion efficiency is not high, it is necessary
to study the characteristics of each part of the hydrogen storage system more specifically to improve
its energy conversion efficiency within the artificially controllable range [29]. Based on this, this paper
combines the mathematical model and electrical model of each part of the hydrogen storage system
for simulation.

Figure 1: Structure of power system with hydrogen storage system

EE, 2024, vol.121, no.2 295

3 Simulation Model of Large-Scale Hydrogen Storage System

3.1 Simulation Model of Hydrogen Storage System
The hydrogen storage system consists of multiple electrolytic cell units, hydrogen storage tanks,
and multiple fuel cell units. After the system is connected to the power grid, each electrolytic cell
unit generates hydrogen gas and compresses it into a hydrogen storage tank. When electrical energy
is needed, each fuel cell unit converts the hydrogen stored in the hydrogen storage tank into electrical
energy.
Due to the different internal characteristics of each electrolyzer unit and fuel cell unit, and when
the internal temperature and gas pressure of the hydrogen storage system are different, the charge and
discharge efficiency will also change accordingly. Considering the series-parallel combination of the
monomer model, the construction is composed of the large-scale hydrogen storage system is shown in
Fig. 2.

Figure 2: Simplified structure of large-scale hydrogen storage system

Assume that the temperature set  and pressure  set of each single cell in the series-parallel
combination are Ti,j = T1,1 , · · · , Tm,n and pi,j = p1,1 , · · · , pcm,n . The corresponding cell input voltage
c c c c c

   
and the efficiency of each cell are Ui,jel = U1,1 el
, · · · , Um,n
el
and ηi,jel = η1,1
el
, · · · , ηm,n
el
.
 
The temperature and pressure set for each single fuel cell are Ti,jd = T1,1 d
, · · · , Tm,n
d
and pdi,j =
 
pd1,1 , · · · , pdm,n , the corresponding fuel cell thermodynamic electromotive force and fuel cell efficiency
   
are Ei,jner = E1,1 ner
, · · · , Em,n
ner
and ηi,jfc = η1,1
fc
, · · · , ηm,n
fc
.
m and n are the number of parallel and series connected large-scale hydrogen storage systems,
respectively.
296 EE, 2024, vol.121, no.2

From this, the formulas for the charge and discharge efficiency of large-scale hydrogen storage
systems are obtained as follows:
m
 n 
m

n

ηc = Ui,jel ηi,jel Ui,jel (1)

i=1 j=1 i=1 j=1


m n 
mn

ηd = ner
E ηi,j
i,j
fc
Ei,jner (2)
i=1 j=1 i=1 j=1

where ηc and ηd are the charging and discharging efficiencies, respectively.

The energy conversion efficiency of PCS (Power Conversion System) in hydrogen storage systems
during external charging and discharging processes will also have an impact on the efficiency of
the entire large-scale hydrogen storage system. In addition to considering the impact of internal
characteristics of large hydrogen storage systems on efficiency, it is also necessary to consider the
impact of PCS energy conversion efficiency on the efficiency of the entire hydrogen storage system.
The efficiency of PCS has a nonlinear relationship with the output power of the hydrogen storage
system [30], The efficiency of PCS is as follows:
 
PA
ηpcs = x +y (3)
PN
where PA /PN is the per-unit value of the output power, and x, y are the piecewise linearization
coefficients.
In summary, the total charge-discharge efficiency of the large-scale hydrogen storage system is:
ηc,z = ηc ηPCS
(4)
ηd,z = ηd ηPCS

In the actual production process, the capacity of the hydrogen storage system will gradually
decrease due to a decrease in the number of cycles. The changes in working temperature and internal
pressure within the system will have an impact on the absorption and release of hydrogen in the
hydrogen storage system, and thus affect the capacity of the hydrogen storage system. In order to
better fit the engineering practice, The capacity decay effect formula [22–24] that affects the capacity
decay of the hydrogen storage system is:
r1= 3.714 × cycle + 0.1651
r2= −7.585 × cycle + 360
(5)
r3= 0.232 × T − 8.927
r4= −0.164 × T + 14.26
 
r1 r2
Qless = 1 − + (6)
282.74 (1 + exp (r3 × p + r4 )) 282.74
where Qloss is the capacity loss rate, r influence coefficient of reaction temperature and number of cycles
on capacity loss, T is the working temperature, p is the working pressure. cycle is the number of battery
cycles.
According to the above formula, the total life loss of the hydrogen storage system is:

N

Qz_loss = Qless,r /N (7)

r=1
EE, 2024, vol.121, no.2 297

where Qless,r is the capacity loss of hydrogen storage units, and N is the number of hydrogen storage
units.

3.2 Hydrogen Storage Unit Model

3.2.1 Mathematical Model
1) Electrolyzer Model
The electrolytic cell voltage model is derived from the relationship between the electrolysis voltage
and electrolytic current [31], and the single electrolytic cell voltage model can be expressed by the
following formula:
Uel = E + Uel,act + Uel,ohm (8)
where E is the open circuit voltage of the electrolytic cell, Uel,act is the activation polarization
overvoltage, and Uel,ohm is the ohmic polarization overvoltage.
In an electrolytic cell, the open circuit voltage is related to temperature and the activity of each
component involved in the reaction. The open circuit voltage of the electrolytic cell can be derived
through the Nernst equation. This equation is used to calculate the equilibrium voltage of a specific
redox pair on the electrode relative to the standard potential:

RTel PH2 PO0.5
E = E0 + ln 2
(9)
zF aH2 O
where E0 is the standard electromotive force, Tel is the cell temperature, aH2 O is the water activity
between the anode and the electrolyte, PH2 is the partial pressure of hydrogen at the anode, and PO2 is
the partial pressure of oxygen at the cathode.
The standard electromotive force expression is:
E0 = ΔG/zF (10)
where ΔG is the Gibbs free energy change of the electrochemical reaction process; z is the number of
electrons transferred per reaction; F is the Faraday constant.
 
t1 + t2 /Tel + t3 /Tel 2
Uel,act = s1 + s2 Tel + s3 Tel ln
2
Iel + 1 (11)
Ael
where Tel is the temperature of the electrolytic cell; Ael is the area of the electrolysis module; Iel is the
DC current; sn is the electrode overvoltage coefficient (n = 1, 2, 3), and tp is the electrode overvoltage
coefficient (p = 1, 2, 3).
The ohmic polarization overvoltage is:
r1 + r2 Tel
Uel,ohm = Iel (12)
Ael
where r1 and r2 are the ohmic resistance parameters of the electrolyte.
The cell efficiency formula is equal to the ratio of the input voltage inside the cell to the voltage
at the cell port.
ηel = E/Uel (13)
298 EE, 2024, vol.121, no.2

2) Fuel cell model

The output voltage of a single fuel cell [32] port can be expressed as:
Ufc = Enernst − Ufc,act − Ufc,ohm − Ufc,conc (14)
where Enernst is the thermodynamic electromotive force; Ufc,act is the activation overvoltage; Ufc,ohm is the
ohmic overvoltage; Ufc,conc is the concentration overvoltage.
Thermodynamic electromotive force means that the work done to transfer electrons during the H2
combustion reaction is equal to the Gibbs free energy released during the reaction without considering
the loss, expressed by the Nernst equation and the Gibbs free energy change.
ΔG ΔS RTfc 1
Enernst = + (Tfc − Tref ) + ln PH2 + ln PO2 (15)
2F 2F 2F 2
where ΔG is the Gibbs free energy change of the chemical reaction; F is the Faraday constant; ΔS is
the change in entropy; Tfc is the fuel cell temperature; Tref is the reference temperature; R is the gas
constant; PO2 is the partial pressure of hydrogen at the anode catalyst/gas interface.
The activation polarization overvoltage is mainly manifested as a slow rate phenomenon when the
electrode surface is just about to activate the electrochemical reaction, which is generally expressed by
a fixed parameter.
Uact = [ξ1 + ξ2 Tfc + ξ3 Tfc ln (CO2 ) + ξ4 × Tfc × ln (Ifc )] (16)
CO2 = PO2 /5.08 × 106 exp (−498/Tfc ) (17)
where ζ1234 is the empirical parameter; Ifc is the current of the fuel cell module; CO2 is the oxygen
concentration of the cathode gas and liquid surface; PO2 is the oxygen interface partial pressure of
the fuel cell.
The ohmic overvoltage of the PEMFC module is mainly caused by the voltage drop generated by
the membrane impedance corresponding to the proton membrane, which is equivalent to the voltage
drop generated by the internal resistance of the fuel cell.
Uohm = Ifc Robm = Ifc (rM l/A + Rc ) (18)
    2  2.5 
Ifc Tfc Ifc
181.6 1 + 0.03 + 0.062
A 303 A
rM =     (19)
Ifc Tfc − 303
λ − 0.634 − 3 exp 4.18
A Tfc
where l is the thickness of the membrane; rM is the resistivity of the membrane; Rc is the resistance that
prevents protons from passing through the membrane; A is the effective area of the membrane.
The concentration overvoltage of the PEMFC module mainly occurs in the working state of
high current. At this time, the electrochemical reaction is too fast, which causes the slow diffusion of
reactants or products and the restriction of mass transfer of reactants. The mathematical expression
is as follows:
Uconc = −B ln (1 − J/Jmax ) (20)
where J and Jmax are the current density and the maximum current density, respectively; B is the
equation constant, which is determined by the battery operating conditions.
EE, 2024, vol.121, no.2 299

The current density is:

J = Ifc /A (21)

The fuel cell efficiency formula is equal to the ratio of the output voltage of the fuel cell port to
the thermodynamic electromotive force of the internal reaction.
ηfc = Ufc /Enernst (22)

3) Hydrogen storage tank model

The hydrogen production flow rate of a single electrolyzer is [33]:
Iel Iel E Pc,in
Vel,H2 = = = (23)
zF zFE zFE
The principle of hydrogen consumption molar flow rate is the same as above:
Ifc Ifc Enernst Pd,out
Vfc− H2 = = = (24)
zF zFEnernst zFEnernst
Hydrogen storage:
 t2
MH 2 = Vel− H2 − Vfc− H2 dt (25)
t1

3.2.2 Electrical Model

1) Equivalent circuit model of electrolyzer
The electrolytic cell model can be equivalent to the charging process, so the electrolytic cell can
be considered as an equivalent power source. In an electrolytic cell, internal current flows from the
positive pole of the power supply to the negative pole. However, the direction of overvoltage caused by
activation polarization loss and ohmic overvoltage is opposite to the direction of open circuit voltage.
That is to say, the effects of activated polarization and ohmic overvoltage will require a higher voltage
than the open circuit voltage for the actual electrolysis process to proceed. Ohmic loss is mainly caused
by the current passing through electrode materials, connecting components, liquid junction potential
when in contact with different solutions, and the resistance of ions in the electrolyte. The relationship
between Ohm voltage loss and current follows Ohm’s law, which means that voltage loss is equal to
current multiplied by resistance. Polarization loss is mainly caused by the deviation of cell equilibrium
potential caused by changes in concentration in the electrolytic cell. When the electrolyte concentration
changes, polarization loss can lead to a difference between the working voltage and open circuit voltage
of the electrolytic cell. In summary, ohmic loss and polarization loss are the two main factors causing
voltage loss in the electrolysis process.
According to the above analysis, the equivalent circuit model of the electrolytic cell [34] is shown
in Fig. 3.
According to the equivalent circuit and mathematical model, a simulation model is built through
MATLAB/Simulink [35], as shown in Fig. 4. Refer to Appendix A for each model parameter. In order
to reflect the characteristics of the electrolytic cell model, the voltage and current constraints during
charging and discharging are not considered for the time being.
300 EE, 2024, vol.121, no.2

Iel
Ract Rohm
Polarization Ohmic

Iel
resistance resistance
Cact E
Uel Electrode Cell open circuit
Single cell capacitance voltage
voltage

Figure 3: Electrolyzer equivalent circuit model

Figure 4: Electrolyzer simulation model

EE, 2024, vol.121, no.2 301

2) Equivalent circuit model of fuel cell

Since a series of physical and chemical changes will occur on the positive and negative electrodes
of the fuel cell during operation, each process will generate a certain resistance. In order to make
the reaction on the electrodes continue to occur, it needs to consume its own energy to overcome the
resistance. During this process, the electrode potential shift phenomenon will occur, and the losses
generated during the reaction include activation polarization voltage, ohmic polarization voltage and
concentration overvoltage. The corresponding fuel cell equivalent circuit model [36] is shown in Fig. 5.

Ract Ifc
Rohm
Polarization
resistance
Ohmic
resistance
Cact
Ucell Polarized
capacitance
Single cell
voltage Enernst
thermodynamic
Rconc Concentration electromotive
resistance force

Figure 5: Equivalent circuit diagram of single fuel cell

The fuel cell simulation model structure [37] is shown in Fig. 6. The fuel cell reaction is equivalent
to the discharge process in the entire hydrogen storage system, where the thermodynamic electromotive
force is equivalent to the internal voltage of the fuel cell, and Ufc is equivalent to the output voltage of
the fuel cell.

4 Example Analysis
In order to verify the validity of the model established in this paper, some parameters are set as
shown in Appendix A [38]. Combined with the simulation model of the PEM electrolyzer and Proton
exchange membrane fuel cell built in Simulink, the variation of the voltage of the electrolyzer and the
fuel cell with the current is obtained. Figs. 7 and 8 respectively show the variation of the input voltage of
the electrolyzer and the output voltage of the fuel cell with current and the typical polarization curves
of the above-mentioned electrolyzer and fuel cell. It can be seen from the figure that the variation trend
of the voltage and current of the electrolyzer and the fuel cell model constructed in this paper and the
typical polarization curves of the two, which shows the correctness and effectiveness of the simulation
model established in this paper.
302 EE, 2024, vol.121, no.2

Figure 6: Fuel cell simulation structure diagram

EE, 2024, vol.121, no.2 303

Figure 7: Comparison between the input voltage and current curve of the electrolyzer and the typical
polarization curve

Figure 8: Comparison of output voltage vs. current curve of fuel cell and typical polarization curve

4.1 Analysis of the Influence of Hydrogen Storage Unit Efficiency

In order to analyze the influence of temperature and partial gas pressure of each part of the
hydrogen storage unit on the charging and discharging efficiency of the hydrogen storage unit, the
hydrogen storage unit is charged and discharged under the constant current of 10, 30, 50 and 70 A,
that is, the current density is 0.04, 0.12, 0.2 and 0.28 A/cm2 . Figs. 9 and 10 show the variation curves of
hydrogen storage unit charge-discharge efficiency with operating temperature and gas partial pressure
under each constant current density charge-discharge state.
304 EE, 2024, vol.121, no.2

Figure 9: Charge efficiency curves at constant current density

Figure 10: Discharge efficiency curves under constant current density

It can be seen from Figs. 9 and 10 that under the condition of constant current density and
constant gas partial pressure, the charge discharge efficiency of the hydrogen storage system increases
with the increase of operating temperature, because the increase of temperature enhances the activity
of the catalyst. When the system operating temperature and gas partial pressure are constant, the
charging efficiency decreases with the increase of current density, because the increase of current
leads to the increase of activation polarization overvoltage and ohmic polarization overvoltage in the
electrolytic cell reaction, which leads to the decrease of efficiency; The discharge efficiency decreases
with the increase of current density because the increase of current leads to the increase of activation
over-voltage, ohmic over-voltage and concentration over-voltage during the fuel cell reaction process,
which reduces the efficiency. At constant operating temperature and current density, with the increase
of H2 and O2 partial pressure, the charging and discharging efficiency of the hydrogen storage system
will increase, but the range is not large. Because changing the gas partial pressure will only affect the
open circuit voltage of the electrolytic cell and the thermodynamic electromotive force and activation
overvoltage of the fuel cell, H2 and O2 partial pressure have little effect on the charging and discharging
EE, 2024, vol.121, no.2 305

efficiency. With the increase of current, the voltage loss of electrolytic cell and fuel cell will increase,
thus reducing the charge discharge efficiency. Theoretically, within the upper and lower limits of the
current temperature and gas partial pressure settings, the optimal charging and discharging efficiency
of the hydrogen storage unit can be 98.5% and 86.16%, respectively.
The PCS converter efficiency is strongly related to the output power of the large-scale hydrogen
storage system. When considering the influence of the internal characteristics of the hydrogen storage
system on the system charge and discharge efficiency, changes in the internal temperature and pressure
of the system will also affect the power and lead to changes in the PCS efficiency. Figs. 11 and 12
describe the influence curves of temperature and pressure on PCS efficiency in the rectification process
and the inversion process, respectively.

Figure 11: Influence curve of PCS efficiency in rectification process

Figure 12: Inverting process PCS efficiency influence curve system

In the rectification stage of the PCS converter, as the temperature increases, the efficiency of
the PCS converter will decrease, while the pressure has little effect on it. This is because during the
rectification operation, power flows to the PCS converter, causing high current and low voltage,
306 EE, 2024, vol.121, no.2

resulting in poor heat dissipation conditions inside the converter. However, an increase in temperature
will further reduce the heat dissipation effect, thus negatively affecting the rectification efficiency.
In the inverter operation of the PCS converter, the PCS efficiency increases with increasing
temperature and slightly increases with increasing pressure. This is because during the operation of
the inverter, power is output from the PCS, and at this time, the current is low and the voltage is high,
making the requirements for heat dissipation relatively low. At the same time, a high-temperature
environment is beneficial for improving the performance of certain components, thereby improving
inverter efficiency. The reason why the impact of pressure on efficiency is relatively small may be due
to the small impact of pressure changes on the working state of circuit components.

4.2 Analysis of the Influence of Hydrogen Storage Efficiency

In this paper, it is assumed that the capacity of the hydrogen storage tank is infinite, and the
upper limit of the capacity decay rate is set to limit the quality of hydrogen that the hydrogen storage
tank can hold. Since the partial pressure of the gas has little effect on the hydrogen storage unit, it is
assumed that the partial pressure of the gas and the pressure in the tank are constant at 3 atm, the
number of cycles of the hydrogen storage system is 2000 times, and the working environment is set
to a constant temperature of 25°C. At this time, the capacity loss of the electrolyzer and the fuel cell
increases with the increase of the internal working temperature, and the change curve of the capacity
loss with temperature is shown in Fig. 13.

Figure 13: Capacity loss vs. temperature curve

Theoretically, if the temperature is kept at a high level, higher charge-discharge efficiency can
be obtained, but in the actual production process, the capacity of the electrolyzer and fuel cell will
decrease with the increase of temperature and the number of cycles. In order to be more practical,
different upper limits of capacity decay are set for the hydrogen storage unit, as shown in Table 1,
combined with the charge-discharge efficiency of the hydrogen storage unit and the influence curve
of the PCS efficiency, and the Cplex solver is used to obtain the optimal value of the efficiency of the
hydrogen storage system and the corresponding operating temperature.
EE, 2024, vol.121, no.2 307

Table 1: Hydrogen storage unit capacity decay upper limit

Electrolyzer unit Capacity loss upper Fuel cell unit number Capacity loss upper
number limit/% limit/%
1–10 70 1–10 30
11–20 60 11–20 40
21–30 50 21–30 50
31–40 40 31–40 60
41–50 30 41–50 70

It can be seen from Section 4.1 that the higher the working temperature of each unit electrolyzer
and fuel cell, the higher the charge and discharge efficiency of the corresponding unit. Through
calculation, it can be seen from Figs. 14 and 15 that after considering the capacity loss of the hydrogen
storage unit, in the large-scale hydrogen storage system, each unit electrolyzer and fuel cell can no
longer maintain the upper temperature limit. Through the solution, it is concluded that the optimal
charging efficiency of the internal system of the hydrogen storage system is 63.46%, the PCS efficiency
of the charging process is 97.57%, and the total charging efficiency of the system is 61.92%. The
capacity loss of each unit is maintained within a controllable range, and the total system loss is 50%;
The temperature range of a single fuel cell unit is between 22.9°C and 29.1°C, and it decreases with the
decrease of the upper limit of capacity loss. Correspondingly, the optimal internal discharge efficiency
of the hydrogen storage system is 68.34%, and the PCS efficiency during the discharge process is
97.09%. The total discharge efficiency of the system is 66.34%, and the total system capacity loss
is 49.9%.

Figure 14: Calculation result of charging efficiency of hydrogen storage system

308 EE, 2024, vol.121, no.2

Figure 15: Calculation result of discharge efficiency of hydrogen storage system

5 Conclusion
In this paper, firstly, the attenuation characteristics of the battery are considered, and the
simulation model of each part of the large-scale hydrogen storage system is systematically constructed
in detail. Secondly, the steady-state simulation model of each component of the hydrogen storage unit
is constructed, as well as the charge-discharge efficiency model and PCS efficiency model considering
the temperature of the hydrogen storage system and the internal gas partial pressure. Through
the simulation and modeling of the hydrogen storage system in MATLAB/Simulink, the efficiency
characteristics of the hydrogen storage system and hydrogen storage unit are analyzed. With the goal
of optimizing the efficiency of the large-scale hydrogen storage system, the results obtained by using
the Cplex solver are as follows:
(1) The overall efficiency of a large-scale hydrogen storage system is affected by the internal
characteristics of each hydrogen storage unit that makes up the system, including charging and
discharging current, operating temperature, gas partial pressure and hydrogen storage tank capacity
attenuation, as well as the efficiency of the PCS converter during AC/DC conversion with the outside
world.
(2) Properly increasing the internal working temperature of the hydrogen storage unit can improve
the charging and discharging efficiency of the system and the PCS efficiency during inverter operation,
but will reduce the PCS efficiency during rectifier operation; Increasing H2 and O2 partial pressure can
increase the efficiency of the hydrogen storage unit and PCS, but the increase is small, indicating that
the pressure has little effect on the efficiency of the hydrogen storage system. The efficiency of the
hydrogen storage unit will increase with the increase of charge-discharge current.
(3) The overall efficiency of the large-scale hydrogen storage system is linear with the system loss
and working temperature. The capacity decay rate of the hydrogen storage tank increases with the
temperature and changes fastest between 20°C and 30°C. Although the high working temperature will
affect the capacity attenuation of the system, it will increase the charging and discharging efficiency.
Under the above conditions, the optimal total charging efficiency of the system is 63.46%, and the
total discharge efficiency is 66.34%.
EE, 2024, vol.121, no.2 309

To sum up the actual situation, in order to obtain appropriate charging and discharging efficiency
and ensure that the system capacity loss is within a certain range, it is necessary to reasonably design
the parameters.

Acknowledgement: The authors acknowledge the reviewers for providing valuable comments and
helpful suggestions to improve the manuscript.

Funding Statement: This work was supported by the Jilin Province Higher Education Teaching Reform
Research Project Funding (Contract No. 2020285O73B005E).
Author Contributions: The authors confirm their contribution to the paper as follows: study concep-
tion: Junhui Li; design, data collection: Haotian Zhang; analysis and interpretation of results: Cuiping
Li and Xingxu Zhu; draft manuscript preparation: Ruitong Liu, Fangwei Duan and Yongming Peng.
All authors reviewed the results and approved the final version of the manuscript.
Availability of Data and Materials: All data generated or analysed during this study are included in
this published article.
Conflicts of Interest: The authors declare that they have no conflicts of interest to report regarding the
present study.

References
1. Development and Reform Commission (2021). Guiding opinions of the national development and reform
commission and the energy administration on accelerating the development of new energy storage. Bulletin
of the State Council of the People’s Republic of China, pp. 43–46.
2. Li, J. H., Zhang, J. H., Mu, G., Ge, Y. F., Yan, G. G. et al. (2019). Hierarchical optimal scheduling for deep
peak shaving of energy storage-assisted thermal power units. Power Grid Technology, 43(11), 3961–3970.
3. Li, J. H., Jiao, J., Yan, G. G., Mu, G., Gao, C. L. et al. (2013). Simulation study on the third-order dynamic
model of lead-acid battery. Journal of Northeast Electric Power University, 33(1), 103–108.
4. He, H. J., Zhang, N., Du, E. S., Ge, Y., Kang, C. Q. (2020). A review of the operation efficiency and life decay
modeling methods of large-scale electrochemical energy storage on the grid side. Automation of Electric
Power Systems, 44(12), 193–207.
5. Feng, J. Q., Si, Y. W., Wu, L., Huang, K. F., Zhang, X. (2020). Characteristic analysis of power battery
based on dynamic comprehensive equivalent circuit model. Energy Storage Science and Technology, 9(3),
986–992.
6. Li, J. H., Hu, D. C., Mu, G., Wang, S., Zhang, Z. S. et al. (2020). Optimal control strategy for large-scale
VRB energy storage auxiliary power system in peak shaving. International Journal of Electrical Power &
Energy Systems, 120, 106007.
7. Liu, Y. (2021). Power distribution strategy of electrochemical energy storage power station participating
in secondary frequency regulation of power system (Master Thesis). Northeast Electric Power University,
China.
8. Liu, D. B., Yuan, Y., Li, S. C., Cao, H. J., Jin, Z. T. et al. (2022). A review of capacity allocation in systems
with renewable energy using hydrogen energy storage. Electrical Measurement and Instrumentation, 59(12),
1–13.
9. Qin, M. Z., Zhang, G. Y., Qi, D. L. (2016). Modeling and simulation of wind power-hydrogen energy
coupling system. Electronic Technology, 45(8), 18–23.
310 EE, 2024, vol.121, no.2

10. Ren, Z. Y., Luo, X., Qin, H. L., Jiang, Y. P., Yang, Z. X. et al. (2022). Mid- and long-term optimal operation
of a hydrogen-containing regional integrated energy system considering the physical properties of hydrogen
storage. Power Grid Technology, 46(9), 3324–3333.
11. Gu, T., Yuan, Y. (2021). Simulation modeling and analysis of proton exchange membrane fuel cells. Power
Technology, 45(4), 459–462.
12. Farrukh, K., Ibrahim, D., Marc, A. R. (2016). Analysis and assessment of an integrated hydrogen energy
system. International Journal of Hydrogen Energy, 41(19), 7960–7967.
13. Cai, L. D., Li, Q., Liu, Q., Wang, T. H., Yang, W. et al. (2022). An optimal control method for multi-stack fuel
cells considering system hydrogen consumption and durability. Chinese Journal of Electrical Engineering,
42(10), 3670–3680.
14. Cai, G. W., Kong, L. G. (2016). PEMFC grid-connected power generation system modeling and optimal
control. China Electric Power, 49(11), 75–81.
15. Cai, G. W., Bian, Y. D., Kong, L. G., Song, J., Xu, H. (2020). Homogeneous modeling of wind/light hydrogen
production system. China Electric Power, 53(10), 59–65.
16. Amir, R. R., Seyed, M. A., Mohammad, H. N., Ehsanolah, A., Mahdi, S. (2022). A green hydrogen energy
storage concept based on parabolic trough collector and proton exchange membrane lectrolyzer/fuel cell:
Thermodynamic and exergoeconomic analyses with multi-objective optimization. International Journal of
Hydrogen Energy, 47(62), 26468–26489.
17. Seyed, M. A., Amir, R. R., Ahmad, A. (2021). Comprehensive assessment and multi-objective optimization
of a green concept based on a combination of hydrogen and compressed air energy storage (CAES) systems.
Renewable and Sustainable Energy Reviews, 142, 1364–0321.
18. Rama, C. M., Paresh, K. (2022). Silicon nanostructures for solid-state hydrogen storage: A review.
International Journal of Hydrogen Energy, 48(4), 1401–1439.
19. Zhao, X. Y., Chen, H., Zheng, Q. W., Liu, J., Pan, P. Y. et al. (2022). Thermo-economic analysis of a
novel hydrogen production system using medical waste and biogas with zero carbon emission. Energy, 265,
126333.
20. Mohamed, N., Hamdy, H. (2022). Assessment of hydrogen production from waste heat using hybrid systems
of Rankine cycle with proton exchange membrane/solid oxide electrolyzer. International Journal of Hydrogen
Energy, 48(20), 0360–3199.
21. Liang, Y. F., Liang, Q. C., Zhao, J. F., He, J. N. (2022). Downgrade power allocation for multi-fuel cell
system (MFCS) based on minimum hydrogen consumption. Energy Reports, 8, 15574–15583.
22. Liu, Y. L., Zheng, J. Y., Xu, P., Zhao, Y. Z., Chen, H. G. (2008). Numerical simulation of the effect of ambient
temperature on leakage and diffusion of high-pressure hydrogen storage tanks. Journal of Engineering
Thermophysics, 5, 770–772.
23. Li, P. G., Qiu, D. K., Shen, S. Y., Peng, L. F. (2021). Research on performance degradation of integrated
regenerative fuel cell. Power Technology, 45(6), 749–752+796.
24. Zhu, D., Chabane, D., Ait-Amirat, Y., N’Diaye, A., Djerdir, A. (2017). Estimation of the state of charge
of a hydride hydrogen tank for vehicle applications. 2017 IEEE Vehicle Power and Propulsion Conference
(VPPC), pp. 1–6. New York, USA.
25. Xu, Z., Li, X. C., Jia, L. Z., Chen, B., Dai, Z. Q. et al. (2022). The influence of overcharge cycle on the
capacity attenuation and safety of lithium-ion batteries. Energy Storage Science and Technology, 11(12),
3978–3986.
26. Zhao, J. N., Jin, Y., Wang, S. J., Guo, X. J. (2017). Temperature characteristics and capacity decay model of
commercial Li4 Ti5 O12 batteries. Chinese Journal of Electrical Engineering, 37(17), 5069–5075+5226.
27. Yuan, T. J., Cao, J. L. (2022). Capacity optimization of wind-hydrogen low-carbon energy system consider-
ing wind power-load uncertainty. High Voltage Technology, 48(6), 2037–2044.
EE, 2024, vol.121, no.2 311

28. Cai, Q. Q., Xiao, Y., Zhu, Y. Q. (2021). A day-ahead economic coordination dispatching model of electric
heating microgrid considering electricity-to-hydrogen and fuel cells. Electric Power Automation Equipment,
41(10), 107–112+161.
29. Li, Z., Zhang, R., Sun, H. X., Zhang, W. D., Mei, C. X. (2022). A review of key technologies for renewable
energy multi-energy complementary production-storage-hydrogen transport. Journal of Electrotechnical
Technology, 36(3), 446–462.
30. Yan, G. G., Liu, Y., Duan, S. M., Li, H. B., Mu, G. et al. (2020). Power distribution strategy of battery
energy storage unit group participating in secondary frequency regulation of power system. Automation of
Electric Power Systems, 44(14), 26–34.
31. Zhang, C. Z. (2009). Modeling and simulation research of solar water electrolysis hydrogen production system
(Master Thesis). Southwest Jiaotong University, China.
32. L, Q. (2011). Research on the modeling and control method of proton exchange membrane fuel cell system
(Ph.D. Thesis). Southwest Jiaotong University, China.
33. Cai, G. W., Chen, C., Kong, L. G., Peng, L., Li, Z. X. (2018). Control strategy of wind hydrogen coupling
combined network system. Journal of Solar Energy, 39(10), 2970–2980.
34. Wu, J., Yin, Z., Li, H., Zhao, B., Ji, Q. et al. (2021). High-frequency equivalent circuit model of lithium-ion
battery based on fractional order theory. Journal of Electrotechnical Technology, 36(18), 3902–3910.
35. Mi, L. X. (2021). Study on the electrolysis characteristics of wind power electrolysis water hydrogen production
system (Master Thesis). Xinjiang Agricultural University, China.
36. Yang, X. B., Zheng, Y. J., Gao, W. K., Ren, D. S., Han, X. B. et al. (2021). Power state estimation of high
specific energy energy storage lithium battery system based on improved equivalent circuit model. Power
Grid Technology, 45(1), 57–66.
37. Liu, X. R., Jiang, Y., Zhang, X. X., Chen, W. R. (2021). A review of three-dimensional numerical simulation
of proton exchange membrane fuel cells. Chinese Journal of Electrical Engineering, 41(21), 7352–7370.
38. Kong, L. G., Cai, G. W., Chen, C., Xing, L. (2017). Modeling and grid connection control of active
permanent magnet direct drive wind turbine based on hydrogen energy storage. Journal of Electrotechnical
Engineering, 32(18), 276–285.

Appendix A

Appendix A: Electrolyzer and fuel cell model parameters table

Parameter name Parameter expression Numerical value Unit
The number of electrons z 2 –
transferred
Faraday constant F 96.487 kJ/(V·mol)
Gibbs ΔG 236.48 kJ/mol
Water activity aH2 O 1 –
Electrolyzer module area Acell 0.25 m2
Electrolyzer temperature Tel 25 °C
Number of serial modules of Nel 80 –
electrolyzers
Ohm parameter r1 0.000 073 Ωm2
Ohm parameter r2 −0.000 000111 Ωm2 °C−1
(Continued)
312 EE, 2024, vol.121, no.2

Table A1 (continued)
Parameter name Parameter expression Numerical value Unit
Electrode overvoltage s1 0.16 V
parameters
Electrode overvoltage s2 0.001 38 V/°C
parameters
Electrode overvoltage s3 −0.000 016 V/°C2
parameters
Electrode overvoltage t1 0.016 m2 A−1
parameters
Electrode overvoltage t2 −1.3 m2 A−1 °C
parameters
Electrode overvoltage t3 421 m2 A−1 °C−2
parameters
Faraday efficiency coefficient a1 99.5 %
Faraday efficiency coefficient a2 −9.578 8 m2 /A
Faraday efficiency coefficient a3 −0.055 5 m2 °C/A
Faraday efficiency coefficient a4 1 502.71 m4 /A
Faraday efficiency coefficient a5 −70.8 m4 °C/A
H2 high calorific value HHVH2 286 kJ/mol
Entropy change ΔS 0.164 kJ/(K·mol)
Reference temperature Tref 298.15 K
Gas constant R 8.314 41 ∗ 10−3 kJ/(K·mol)
Empirical parameters ξ1 0.951 4 V
Empirical parameters ξ2 −0.003 12 V/K
Empirical parameters ξ3 −7.4 ∗ 10−5 V·cm3 /(K·mol)
Empirical parameters ξ4 1.87 ∗ 10−4 V/A/K
Water content λ 10 –
Film resistivity rM – Ω·cm
Battery operating factor B 0.016 –
Maximum battery current Jmax 1.5 A/cm2
density
PEMFC thickness l 51 ∗ 10−4 cm
Membrane effective area A 250 cm2
Impeding protons through the Rc 0.000 3 Ω
membrane impedance
PCS efficiency piecewise a (inversion process) 0.01797 (P/PN = 0.4∼1) –
linearization coefficient 0.07134 (P/PN = 0.2∼0.4)
0.299 (P/PN = 0.1∼0.2)
PCS efficiency piecewise a (rectification process) 0.01592 (P/PN = 0.4∼1) –
linearization coefficient 0.06107 (P/PN = 0.2∼0.4)
0.2797 (P/PN = 0.1∼0.2)
(Continued)
EE, 2024, vol.121, no.2 313

Table A1 (continued)
Parameter name Parameter expression Numerical value Unit
PCS efficiency piecewise b (inversion process) 0.9533 (P/PN = 0.4∼1) –
linearization coefficient 0.0.9316 (P/PN = 0.2∼0.4)
0.0.8855 (P/PN = 0.1∼0.2)
PCS efficiency piecewise b (rectification process) 0.96 (P/PN = 0.4∼1) –
linearization coefficient 0.9418 (P/PN = 0.2∼0.4)
0.8976 (P/PN = 0.1∼0.2)