Energy Reports 9 (2023) 4789–4798

Contents lists available at ScienceDirect

Energy Reports
journal homepage: www.elsevier.com/locate/egyr

Research on the adaptability of proton exchange membrane

electrolysis in green hydrogen–electric coupling system under
multi-operating conditions
∗
Yannan Dong a , Shaohua Ma a , , Zijiao Han a,b , Jintong Bai a , Qiwei Wang a
a
School of Electrical Engineering, Shenyang University of Technology, Shenyang 110870, China
b
State Grid Liaoning Electric Power Co., Ltd., Shenyang 110000, China

article info a b s t r a c t

Article history: The green hydrogen–electric coupling system can consume locally generated renewable energy,
Received 30 December 2022 thereby improving energy utilization and enabling zero-carbon power supply within a certain range.
Received in revised form 19 March 2023 This study focuses on a green hydrogen–electric coupling system that integrates photovoltaic, en-
Accepted 28 March 2023
ergy storage, and proton exchange membrane electrolysis (PEME). Firstly, the impact of operating
Available online 10 April 2023
temperature, power quality, and grid auxiliary services on the characteristics of the electrolysis cell
Keywords: is analyzed, and a voltage model and energy model for the cell are established. Secondly, a multi-
Green hydrogen–electric coupling system operating conditions adaptability experiment for PEME grid-connected operation is designed. A test
PEME platform consisting of a grid simulator, simulated photovoltaic power generation system, lithium
Multi-operating conditions adaptability battery energy storage system, PEME, and measurement and acquisition device is then built. Finally,
Test platform experiments are conducted to simulate multi-operating conditions such as temperature changes,
voltage fluctuations, frequency offsets, harmonic pollution, and current adjustment speed. The energy
efficiency and consumption is calculated based on the recorded data, and the results are helpful to
guide the operation of the system.
© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction equipment in producing hydrogen through water electrolysis, as

it uses external DC electric energy to split water into hydrogen
The energy industry has a significant impact on the ecological and oxygen. It is connected to the energy grid through AC/DC or
environment. In light of the national ‘‘30·60’’ carbon peak and DC/DC converters (Joonas et al., 2019).
carbon neutrality targets, it has become imperative to reduce The green hydrogen–electrolysis coupling system makes elec-
reliance on traditional fossil fuels and increase the capacity of trolysis as a flexible regulation load in future grid systems (Yang
renewable energy sources (Houhe et al., 2021; Ziming et al., et al., 2022b). However, due to the dynamic characteristics of
2020; Yang et al., 2023a). Hydrogen is a zero-carbon, clean en- renewable energy systems, the work state, efficiency and en-
ergy source with high energy density, high conversion efficiency
ergy consumption of the electrolyzer are affected by changes
and green credentials. It has the potential to play a key role
in voltage, frequency and working temperature. Therefore, it is
in future industries and transportation, forming an integral part
necessary to conduct multi-condition adaptability research on
of the global development of a low-carbon, green energy sys-
the electrolyzer, analyze its performance under different condi-
tem (Zijiao et al., 2022; Dodds et al., 2015; Rabiee et al., 2021).
Electrolyzing water for hydrogen production using renewable tions, and provide theoretical support for its efficient operation.
energy has garnered attention from both researchers and industry Multiple references have proposed approaches to improve the
professionals. Due to its ability to absorb variable energy outputs, electrolysis process for hydrogen production. For instance, Zhong-
making the green hydrogen–electric coupling system a promis- ming et al. (2022) proposes the thinking of developing hydrogen–
ing means of hydrogen production (Jiarong et al., 2019; Hiromi electrolysis fusion, which uses hydrogen energy as a link between
et al., 2021; Tayfun and Ali Cem, 2018). This system includes production and use upstream and downstream. And participates
the electrical and long-term spatial coupling between the elec- in grid auxiliary services to achieve its demand–supply balance
trolysis cells and the energy grid. The electrolysis cell is a vital and flexible application supply function. Nan (2017) provides
theoretical support for improving the process of water electrol-
∗ Corresponding author. ysis for hydrogen production by considering changes in working
E-mail address: [email protected] (S. Ma). temperature and electrolyte circulation. Xiaojun et al. (2021)

https://doi.org/10.1016/j.egyr.2023.03.119
2352-4847/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
Y. Dong, S. Ma, Z. Han et al. Energy Reports 9 (2023) 4789–4798

summarizes the power adjustment and start–stop characteris-

tics of the electrolysis cell, then proposes a coordinated control
strategy for an array of alkaline electrolysis cells. Frano (2005)
examines the performance of proton exchange membrane elec-
trolysis (PEME) in a renewable energy system by considering fac-
tors such as PEME size, intermittent operation, output pressure,
oxygen production, and water consumption. Godula-Jopek (2015)
conducts a systematic study of various electrolysis cell tech-
nologies for water electrolysis for hydrogen production. Which
considering their technical advantages, reaction modes, model-
ing methods, and future prospects. Abdin et al. (2015) presents
an advanced model for PEME, which enables a better under-
standing of the contributions of internal components, including Fig. 1. Schematic diagram of green hydrogen–electric coupling system.
electrodes and membranes to the electrolysis voltage. This model
can offer guidance for researching control strategies. Ruomei
et al. (2023) addresses the issue of temperature overcompen-
the nonlinearity of components in the system affect the elec-
sation phenomenon in the stack temperature caused by load
trical power quality during operation. And changes in tempera-
fluctuations during green hydrogen production. Then studies the
ture, frequency and voltage during the operation process affect
thermal modeling and controller design of alkaline electrolysis
the system energy efficiency. In addition, the response speed
systems under dynamic conditions. By using a flow-forward PID
of the electrolysis cell also affect the operation of the green
controller and a model predictive controller, the temperature
hydrogen–electric coupling system as a flexible resource in the
overcompensation can be reduced and the system efficiency im-
power system.
proved. Jiarong (2023) researches the operation of large-scale
In summary, this paper first analyzes the influence of multiple
power-to-hydrogen systems and proposes to link the grid-side
operating conditions on the structure of the green hydrogen–
performance with traditional control, and summarizes typical
electric coupling system. Then, it establishes the voltage model,
operating modes under different load conditions for practical
energy model, and energy efficiency model of the electrolysis cell.
applications. Haluk (2006) describes a dynamic model of PEME
Finally, the grid-connected operating conditions, such as changes
tanks suitable for verifying control strategies for efficient and
in working temperature, voltage fluctuations, frequency offsets,
reliable operation of the electrolysis cell. This dynamic model can
and current regulation speed, are tested through experiments. By
be integrated with a renewable energy system model to design,
comparing the experimental data with the theoretical character-
analyze, and optimize sustainable energy systems.
istics and analyzing the grid-connected characteristics, we were
The investment cost of the electrolysis cell power converter is
able to obtain insight into the impact of different operating condi-
approximately 15% of the PEME water electrolysis system (Joonas
tions on the efficiency and energy consumption of the electrolysis
et al., 2020). Due to the need to maintain low voltage and high
current characteristics of electrolysis water, thyristors and diodes cell, filling the gap in the current research on the grid-connected
are typically used as converters (Alfredo et al., 2013). However, characteristics of electrolysis cells for hydrogen production in re-
this conversion process causes additional thermal loss in the elec- newable energy systems. This study provides a valuable reference
trolysis cell and produces harmonics in the DC current and volt- for further research on its adaptability and participation in grid
age of the electrolysis cell (Friedrich-W. et al., 2019). Xiaoqiang auxiliary services in renewable energy systems.
et al. (2021) examines the characteristics and technical indicators The remaining research content of our work is arranged as
of the electrolysis cell for hydrogen production and summarizes follows: Chapter 2 presents the structure of the green hydrogen–
the topological structure of the electrolysis cell power electronic electrical coupling system and analyzes its multi-operating con-
converter. It also discusses its potential application, existing prob- dition characteristics; Chapter 3 establishes the electrolysis cell
lems, and development direction in energy electrolysis hydrogen model, which includes both the voltage model and energy model;
production system. Burin et al. (2021) develops a PEME simulator Chapter 4 describes the experimental design used to investigate
that simulates the characteristics of a real electrolysis cell. And is the adaptability of PEME to various operating conditions during
used to test new PEME power electronic converters to avoid dam- grid operation, such as changes in electrolysis cell temperature,
aging the real electrolysis cell for experimental purposes. Zsolt grid-connected power quality, and participation in grid auxiliary
and Árpád Bence (2017) investigates the impact of continuous services; and Chapter 5 draws conclusions.
changes in voltage and current on the electrode reaction, battery
performance, and electrolysis efficiency of a water electrolysis 2. The structure and multi-operating conditions characteris-
cell under DC conditions. To accomplish this, they subject the tics of green hydrogen–electric coupling system
electrolysis cell to sinusoidal waves with increasing amplitude,
frequency, and phase shift during a series of experiments. Buiten- 2.1. The structure of green hydrogen–electric coupling system
dach et al. (2021) studies the influence of different ripple current
characteristics on PEME efficiency by experimental determination The green hydrogen–electric coupling system is a crucial el-
of the relationship between electrolysis power consumption and ement in the advancement and implementation of zero-carbon
ripple current. power supply. As the energy supply sector transitions to a greener
Based on the above mentioned analysis, scholars’ research on approach, the system is anticipated to emerge as a vital energy
electrolysis cells is mostly focused on itself, catalysts, electrode supply mechanism, capable of enhancing the storage capacity
materials or power electronic converter topology. However, few of renewable energy and increasing the flexibility of the power
researchers have studied the grid-connected characteristics of system. The schematic diagram of the system structure is shown
PEME connected to renewable energy systems. The uncertainty of in Fig. 1.
renewable energy output results in different grid characteristics In Fig. 1 illustrates a typical green hydrogen–electric coupling
that impact on the grid-connected adaptability of the electrol- system, comprising photovoltaic generation, lithium battery en-
ysis cell. The electrolysis characteristics of the electrolyzer and ergy storage, PEME, hydrogen storage tanks, and hydrogen and
4790
Y. Dong, S. Ma, Z. Han et al. Energy Reports 9 (2023) 4789–4798

electric loads. To account for the high proportion of renewable

energy systems, the operation state of the electrolysis cell, po-
tential operating conditions are analyzed from three aspects:
changes in the operation temperature of the electrolysis cell,
power quality of grid-connected electricity and participation in
grid auxiliary services.
Fig. 2. The output voltage model of inverter.
2.2. Electrolysis cell operating temperature change condition

The operation temperature change condition of the electrol-

ysis cell is a critical factor for hydrogen production. Electrolysis voltage by ±7%; and the single-phase common connection point
of water is a widely used method to produce hydrogen, which voltage deviation should not exceed +7%, −10%.
involves obtaining hydrogen and oxygen gases through the re- In addition, the high proportion of renewable energy systems
action of water electrolysis. The electrolysis process of water may cause fluctuations in the electric frequency of the power grid,
can be characterized using the reversible voltage Vref and the which can affect the electrolysis system. The normal frequency
thermoneutral voltage Vthe , which are expressed by Eq. (1) and deviation should not exceed 0.5 Hz, and the system frequency
Eq. (2), respectively. deviation by impact load changes to ±0.2 Hz (Jun et al., 2016).
Therefore, it is necessary to calculate the electrolyzer efficiency
∆Gd (T , P ) in the green hydrogen–electrolysis coupling system under the
Vref (T , P ) = (1)
nF disturbance of voltage and frequency on the grid side. Analyzing
∆H (T , P ) the impact of voltage fluctuation and frequency deviation on the
Vthe (T , P ) = (2)
nF operation of the electrolyzer equipment and provide guidance for
Where, n represents the number of charge transfers in electrolysis the long-term safe and efficient operation of the electrolyzer in
of water molecules (n = 2); F represents the Faraday constant the system.
(F = 96485 C mol−1 ); E represents the free energy voltage re-
lated to the electrolysis reaction. ∆Gd represents the free energy 2.3.2. Electrolysis cell grid-connected harmonic disturbance
change related to the reaction; T represents the temperature; P The green hydrogen–electric coupling system contains vari-
represents the pressure. ous power electronic devices, including photovoltaic inverters,
The electrolysis reaction of water is dependent on the working wind turbine converters, and PEME rectifiers, which can increase
voltage Vwork (T, P) of the electrolysis cell, which determines its harmonic voltage and current in the system due to their physi-
different states. If Vwork (T, P) < Vref (T, P), there is no electrolysis cal characteristics and rectification methods. Additionally, factors
reaction. While Vref (T, P) < Vwork (T, P) < Vthe (T, P) initiates the such as changes in irradiance and wind speed can also affect the
process, but not enough to maintain isothermal. The electrolysis output of renewable energy in the system.
cell is an open system, it often absorbs the heat needed for the The waveform of the grid-side voltage waveform typically de-
process from the surrounding environment. When Vwork (T, P) > pends on the control strategy of the renewable energy generation
Vref (T, P), the current begins to flow, and the heat is generated by equipment and its grid-side inverter. The model of the grid-side
different irreversible sources internally. Moreover, when Vwork (T, inverter output voltage is established as shown in Fig. 2.
P) > Vref (T, P), the current density increases and generates more The mathematical model can be expressed by Eq. (3):
heat. The electrolysis heat reaction also affects the working volt- v0 = vc − Z0 (s) i (3)
age. In the process of electrolyzing water to produce hydrogen,
the changes in enthalpy, entropy, and free energy associated with Where, ν0 represents the output voltage of the inverter; νc repre-
the water splitting reaction are functions of temperature and sents the reference voltage of the renewable energy power grid;
pressure. Therefore, thermoneutral voltage is a function of the Z0 represents the output impedance of the inverter.
electrolysis temperature and pressure and energy efficiency, and In the green hydrogen–electric coupling system, the connec-
it is necessary to measure and calculate its electrolysis energy tion of the PEME hydrogen load causes harmonic distortion in
efficiency under temperature change conditions. It is essential the current waveform, which results in a corresponding harmonic
to control the appropriate working temperature to improve the drop in the voltage after i flowing through Z0 . When the harmonic
energy efficiency of the electrolysis cell equipment in the green content in νc is low or almost non-existent, there is a significant
hydrogen–electrolysis coupling system. harmonic distortion in νc . At this time, the voltage harmonic
content in νc is high, which adversely affects the quality of the AC
2.3. Electrolysis cell operating electric power quality conditions bus electrical energy and increasing the DC side current voltage
ripple. Subsequently, this increase the energy consumption of
2.3.1. Electrical grid voltage fluctuation and frequency deviation hydrogen production and costs. To enhance the overall energy
Voltage fluctuations in green hydrogen–electric coupling sys- efficiency of the green hydrogen–electrolysis coupling system and
tems are typically caused by rapid changes in the effective value reduce operational costs, it is necessary to improve the DC power
of the grid voltage resulting from start–stop and output changes quality supplied to the electrolyzer and reduce switching and
in renewable energy power plants, as well as the switching of conduction losses in the power converter of electrolyzer.
compensation capacitors. Due to the dynamic fluctuation of wind
speed (or solar irradiance) in wind farms (or photovoltaic power 2.4. Electrolysis cells participating in grid auxiliary services
plants), there is a large random fluctuation in the output power
of renewable energy grid-connected power plants (Yang et al., The power system requires different standby resources for
2023c), which negatively impacts the quality of electricity in energy storage to handle the uncertainty of renewable energy
the green hydrogen–electric coupling system. The direct cause of and load (Yang et al., 2023b), which can vary in effectiveness
voltage fluctuations in the grid is typically output power fluctua- and response times. To meet these needs, the green hydrogen–
tion. The voltage deviation indicator for the three-phase common electric coupling system investigates the response characteristics
connection point is that the voltage does not exceed the nominal of electrolysis cells and participate in grid auxiliary services.
4791
Y. Dong, S. Ma, Z. Han et al. Energy Reports 9 (2023) 4789–4798

Table 1
Classification of reserve resources with their response time and frequency band. 1
Parameters Response time T Overall reaction : H2 O → H2 + O2 (6)
2
30 s real-time response standby T ≤ 30 s
AGC standby 30 s < T ≤ 3 min
The electrolysis voltage of PEME is represented by the sum
10 min rotation for backup 3 min < T ≤ 10 min of the reversible voltage of the electrolyzer and other additional
30 min spin for standby 10 min < T ≤ 30 min overvoltages, which can be expressed by Eq.(7) (Zini and Tar-
60 min rotation for standby 30 min < T ≤ 60 min tarini, 2015).
Cold standby T > 60 min
Downward adjustment for standby 10 min < T ≤ 30 min Vcell = Vrev + Vohm + Vact + Vcon (7)
Where, Vcell represents the electrolysis voltage; Vrev represents
the reversible voltage, which is the minimum voltage of the water
splitting reaction during the electrolysis process; Vohm represents
the ohmic overvoltage, which is mainly composed of the resis-
tance to electron flow caused by the electrodes, bipolar plates,
collectors, and the resistance to ion flow caused by the electrolyte
and membrane; Vact represents the activation overvoltage, which
is mainly caused by the activation properties of the electrode
surface chemical reactions, and its value is determined by the
electrochemical performance of the electrode material and has a
strong non-linear relationship with the current; Vcon represents
the concentration voltage caused by the mass transfer process,
which is ignored in the study because the current density in the
PEME is not beyond the mass transfer limit.
1/2
( )
MTcell pH2 · pO
0 2
Vrev = Vref + ln (8)
zF pH2 O
0
Where, Vref represents the reversible potential voltage in stan-
Fig. 3. Schematic diagram of hydrogen production from electrolytic water of
PEME.
dard conditions; M represents the gas constant (8.31446212
Jmol−1 K−1 ); Tcell represents the temperature of the electrolytic
1/2
cell; pH2 , pO , and pH2 O represent the partial pressure of hydro-
2
gen, the partial pressure of oxygen, and the partial pressure of
Table 1 displays the division of standby resources and their cor-
water, respectively.
responding response times and frequency bands based on the
The ohmic overvoltage is represented as Eq. (9).
requirements of the system (Xiaolin et al., 2019).
In addition, due to the different time scale requirements for Vohm = Rcell jcell (9)
peaking and frequency regulation in grid auxiliary services. The
backup with a response time of less than 10 min can be used as Where, Rcell represents the resistance of the electrolytic cell, jcell
a frequency regulation auxiliary service backup, and the backup represents the current density (Acm−2 ), and the ohmic overvolt-
with a response time of more than 10 min can be used as a peak age is directly proportional to the resistance of the electrolytic
load auxiliary service backup. The hydrogen–electric coupling cell.
system participating in peak load, frequency regulation, and other The activation overvoltage is represented as Eq. (10).
grid auxiliary services improves the flexibility. The PEME uses jcell
PEM as the electrolyte and pure water as the reactant. And the Vact = α arcsin h( ) (10)
j0
high purity of the hydrogen produced due to the low hydrogen
permeability of the PEM, as well as its ability to adapt to the Where, α represents the electrode charge transfer coefficient, j0
rapidly changing renewable energy power input. So it is impor- represents the exchange current density at the electrode sur-
tant to study the feasibility of the electrolyzer participating in face, which is typically defined as a function of temperature and
grid auxiliary services from the perspectives of start-up time and obtained through experimentation.
response time. Therefore, the simplified model for PEME electrolysis voltage
can be expressed by Eq. (11).
3. The PEME voltage and energy model modeling in green 1/2
( )
MTcell pH2 · pO jcell
0 2
hydrogen–electric coupling system Vcell = Vref + ln + Rcell jcell +α arcsin h( ) (11)
zF pH2 O j0
PEME has been widely studied due to its strong adaptability, From the perspective of electrolysis reaction energy, assum-
high electrolysis efficiency, and ease of coupling with renewable ing that H2 and O2 are ideal gases and the volume of H2 O is
energy generation (Roberto Moreno et al., 2021). A PEME consists non-compressible in Eq. (6). According to the first principle of
of two electrodes and a polymer membrane, and its principle for thermodynamics, the amount of electricity (nFE) required to de-
electrolyzing water to produce hydrogen is shown in Fig. 3. compose 1 mole of water at equilibrium is equal to the Gibbs free
In PEME electrolysis of water to produce hydrogen, the cath- energy change of the water decomposition reaction (∆Gd ), which
ode produces hydrogen gas while the anode produces oxygen gas. is represented by Eq. (12).
The electrode reactions can be expressed by Eq. (4)–Eq. (6) (Zsolt
and Árpád Bence, 2016): ∆Gd − nFE = 0; ∆Gd > 0 (12)
1 The electrolysis reaction at constant temperature and pressure
Anode : H2 O → 2H+ + O2 + 2e− (4)
2 can be expressed by Eq. (13).
Cathode : 2H+ + 2e− → H2 (5) ∆Gd (T , P ) = ∆Hd (T , P ) − T ∆Sd (T , P ) > 0 (13)
4792
Y. Dong, S. Ma, Z. Han et al. Energy Reports 9 (2023) 4789–4798

Fig. 4. The multi-operating conditions adaptability test circuit diagram of the green hydrogen–electric coupling system PEME.

Where, ∆Hd (T, P), ∆Sd (T, P) and ∆Gd (T, P) represent the changes Table 2
in enthalpy, entropy, and Gibbs free energy during electrolysis at PEME parameters.

T and P, respectively. Parameters Value

The reference electrolysis voltage is typically calculated under Hydrogen production m3 /h (1 atm, 20 ◦ C) 2(50% ∼ 100% adjustable)
standard conditions of temperature (T0 = 298 K) and pressure Oxygen production m3 /h (1 atm, 20 ◦ C) 1
Work pressure MPa 1.6
(P0 = 1 bar). The standard conditions assume that water is in the
Hydrogen purity% ≥99.999
liquid state, H2 and O2 are in the gaseous state. On the basis of Oxygen purity% ≥98.5
the standard Gibbs free energy ∆Gd0 , enthalpy change ∆Hd0 and Number of cells in the electrolyzer/piece 22
entropy change ∆Sd0 as shown in Eq. (14) to Eq. (16), further cal- Device size mm 1500 × 1200 × 2000
culation of the reversible voltage Vref0 and thermoneutral voltage Hydrogen water content/g/m3 H2 ≤4
Vthe0 of the electrolysis reaction under standard conditions can be
performed.
∆Gd0 (H2 O) = +237.22 kJ mol−1 The AC efficiency is represented by the ratio of DC energy
(14)
⇒ Vref0 = ∆Gd02F(H2 O) = 1.2293 V ≈ 1.230 V consumption to AC energy consumption, as indicated by Eq. (21).

∆Hd0 (H2 O) = +285.840 kJ mol−1 100Edc

(15)
⇒ Vthe0 = ∆Hd02F(H2 O) = 1.4813 V ≈ 1.481 V ηac = (21)
Eac
∆Sd0 (H2 O) = +163.15 J mol K −1 −1
(16) Where, ηac represents the AC efficiency of the electrolytic cell.
Under standard conditions, 2 × 96500 C of electricity can
be used to electrolyze 1 mole of water to produce 1 mole of 4. Experiment and result
hydrogen and 1/2 mole of oxygen. The volume of 1 mole of
hydrogen gas at standard conditions is 22.43 × 10−3 m3 . Thus, 4.1. PEME grid-connected operation experiment design
the theoretical amount of electricity required to produce 1 m3
of hydrogen gas under standard conditions can be expressed by In actual operation, the green hydrogen–electric coupling sys-
Eq. (17). tem PEME has complex structure. The grid-connection adaptabil-
ity of PEME to renewable energy grid cannot be completed by
2 × 96500 × 1000
= 2390 A h/m3 (17) calculation, it needs to be verified by experiment. To this end, This
3600 × 22.43 paper set up a test platform comprising a grid simulator, a sim-
Combining the above equations, voltage efficiency can be fur- ulated photovoltaic power generation system, a lithium battery
ther calculated by Eq. (18). energy storage system, PEME, and a measurement acquisition
1.481 device. The multi-operating conditions adaptability test circuit
ηdc = (18) diagram of the green hydrogen–electric coupling system PEME is
Vcell
shown in Fig. 4.
Where, ηdc represents the direct current voltage efficiency of the The simulated AC grid power supply consists of simulated
electrolytic cell. photovoltaic power generation system, the lithium battery energy
The direct current energy consumption Edc can be calculated storage system, and the PEME is connected through a rectifier.
by Eq. (19). The DEWE5000 data acquisition system is used to measure the
2390Vcell AC/DC voltage and current, while the FLUKE563 matrix LCD in-
Edc = (19) frared thermometer is used for temperature measurements. The
1000n
PEME electrolytic cell is developed by the 718 research institute
Where, n represents the number of cells in the electrolytic cell.
of CSIC, and has a hydrogen gas output 2 m3 /h, purity ≥99.999%,
The alternating current energy consumption Eac can be calcu-
pressure 1.6 MPa, dew point −50 ◦ C, and cooling method is
lated by Eq. (20).
air cooling. The detailed parameters of the device are shown in
P
( )
2390 1000
Table 2.
Eac = Idc
(20) Subsequently, the experiments are conducted to verify the
n potential operating conditions of the electrolytic cell of the green
Where, P represents the active power of the electrolytic cell for hydrogen–electric coupling system as analyzed in the previous
hydrogen production. chapters.
4793
Y. Dong, S. Ma, Z. Han et al. Energy Reports 9 (2023) 4789–4798

Fig. 5. Test results of PEME temperature change operation. Fig. 6. Test results of PEME voltage fluctuation operation.

4.2. The operating temperature affects the experimental results

The hydrogen–electric coupling system test platform is

equipped with the PEME, and experiments were conducted to
investigate the effect of temperature on the electrolytic cell’s per-
formance. The test conditions were set as follows: cold starting
the cell at room temperature, and running it at a constant DC
current of 100 A. The surface temperature of the cell is measured
every minute using the thermometer, while the PEME data ac-
quisition signal was recorded to calculate the DC efficiency, AC
efficiency, AC energy consumption, and DC energy consumption.
The results as shown in Fig. 5.
As shown in Fig. 5, as the working temperature of PEME
increases, its DC electrolysis efficiency significantly increases, the
AC energy consumption and DC energy consumption both signifi-
cantly decrease, and the AC efficiency changes little. For instance,
when the temperature of the electrolytic cell increases from 30 ◦ C
to 50 ◦ C, the DC electrolysis efficiency increases by 7.05%, while
the AC and DC energy consumption decreases by 4.29% and 4.97%,
respectively. The reason is that as the temperature increases, the Fig. 7. Test results of PEME frequency offset operation.
activation of the electrolytic catalyst in the electrolytic cell causes
the electrolytic potential of the electrode to decrease, resulting in
a decrease in the DC electrolysis voltage of the electrolytic cell, the AC side voltage effective value deviates from the baseline
and subsequently reducing its DC energy consumption. There- value to 200–240 V, the AC efficiency decreases by 3.45%, but
fore, maintaining a higher working temperature can effectively there is almost no impact on the DC electrolysis efficiency, AC
improve the efficiency and reduce the energy consumption and energy consumption, and DC electrolysis energy consumption.
hydrogen production cost of the PEME in the hydrogen–electric Since the grid voltage must be kept within the specified standard
coupling system.
range under normal working conditions, the influence of simu-
lated grid-side voltage fluctuation on the operation of the PEME
4.3. Voltage fluctuations and frequency excursions affect the exper-
hydrogen–electric coupling system can be almost ignored.
imental results
Further testing was conducted to assess the impact of fre-
quency deviation on the operation of the PEME in the hydrogen–
The working temperature has an impact on the results of the
electric coupling system. The frequency deviation operating con-
experiment. PEME was connected to the hydrogen–electric cou-
pling system test platform and the voltage fluctuation conditions ditions were set as follows: the simulated grid frequency was set
were set as follows: the RMS value of the AC side simulated to 50 Hz, 49 Hz, 48 Hz, 50 Hz, 51 Hz, and 52 Hz, and the DC
grid voltage was sequentially set to 220 V, 210 V, 200 V, 230 V, current was maintained at a constant current of 100 A. The PEME
and 240 V, and the DC current was kept at a constant current data acquisition signals were recorded and the DC efficiency, AC
of 100 A. The PEME data acquisition signals were recorded and efficiency, AC energy consumption, and DC energy consumption
the DC efficiency, AC efficiency, AC energy consumption, and DC were calculated as shown in Fig. 7.
energy consumption were calculated as shown in Fig. 6. As shown in Fig. 7, the change in the simulation grid frequency
The results of the experiments show that as the simulated grid of the hydrogen–electric coupling system has almost no impact
voltage fluctuates from the baseline value, the AC efficiency of the on AC efficiency, DC efficiency, AC energy consumption, and DC
PEME green hydrogen–electric coupling system decreases. When energy consumption of PEME.
4794
Y. Dong, S. Ma, Z. Han et al. Energy Reports 9 (2023) 4789–4798

Table 3
PEME efficiency under temperature, frequency offset and voltage fluctuation conditions.
Conditions AC efficiency DC efficiency
Temperature Frequency Voltage T (◦ C) f (Hz) V (V) T (◦ C) f (Hz) V (V)
(◦ C) offset (Hz) fluctuation (V)
30.6 50 220 86.00 87.2 91 85.00 94 86.9
33.6 49 210 85.30 87.2 91 88.00 94 86.4
36.9 48 200 85.20 87.2 90 89.00 94 86.02
39.4 50 230 85.20 87.2 91 90.00 94 84.4
45.2 51 240 85.13 87 91 90.00 94 83.7
50.3 52 / 85.02 86.9 / 91.00 /

Table 4
PEME energy consumption under temperature, frequency offset and voltage fluctuation conditions.
Conditions AC energy consumption DC energy consumption
Temperature Frequency Voltage T (◦ C) f (Hz) V (V) T (◦ C) f (Hz) V (V)
(◦ C) offset (Hz) fluctuation (V)
30.6 50 220 5.35 4.79 5.01 4.63 4.25 4.38
33.6 49 210 5.23 4.79 5.01 4.50 4.25 4.40
36.9 48 200 5.23 4.79 5.12 4.47 4.25 4.45
39.4 50 230 5.12 4.79 5.12 4.45 4.25 4.37
45.2 51 240 5.12 4.79 5.12 4.42 4.25 4.36
50.3 52 / 5.12 4.79 / 4.39 4.25 /

The change of AC efficiency, DC efficiency, AC energy con-

sumption, and DC energy consumption of PEME under temper-
ature change, frequency deviation, and voltage fluctuation condi-
tions are summarized as shown in Tables 3 and 4 based on the
experimental results.
In conclusion, the experiments demonstrate that the operating
temperature has a impact on the efficiency of PEME, with the DC
efficiency increasing, the AC energy consumption and DC energy
consumption both decreasing as temperature increases. However,
the impact of allowable frequency offset and voltage fluctuation
on the grid side on the efficiency of PEME can be ignored. There-
fore, for the green hydrogen–electric coupling system application,
it is important to reasonably control the operating temperature of
PEME to enhance working efficiency, reduce energy consumption
in hydrogen production process.

4.4. The power quality of harmonic conditions affects the experimen-

tal results

In the test platform of the green hydrogen–electric coupling Fig. 8. Test results of PEME DC current waveforms in Case 1, Case 2.
system, the operation state was switched to island mode. The
lithium battery energy storage system was started and set to
constant voltage and constant frequency control. The simulated the DC voltage waveform is significantly improved under Case 2
photovoltaic power generation system was connected, and the as compared to Case 1.
PEME was connected to the hydrogen–electric coupling system As shown in Fig. 9, the DC voltage ripple of the PEME in the
test platform. Further set the following operating conditions:
hydrogen–electric coupling system is reduced and the voltage
Case 1: Set the PEME hydrogen production load input current
waveform is significantly improved under Case 2 compared to
to 100 A, 125 A, 150 A, 175 A, and 200 A, and the control strategy
Case 1.
did not include the harmonic control method.
The DC and AC energy consumption of PEME are calculated
Case 2: Set the PEME hydrogen production load input current
using Eqs. (19) and (20) under two conditions, and the result is
to 100 A, 125 A, 150 A, 175 A, and 200 A, and the control strategy
shown in Fig. 10.
included the harmonic control method mentioned in Yannan et al.
(2022). As shown in Fig. 10, with the DC current of PEME increases
In Fig. 8 illustrates a comparison of the DC side current wave- from 100 A to 200 A, the AC and DC energy consumption of
form when the PEME hydrogen production load input current is the hydrogen–electrical coupling system both decrease after the
set to 175 A in Case 1 and Case 2. harmonic control. The quality of electrical energy has a significant
In Fig. 8 shows a noticeable reduction in the DC current ripple impact on the energy consumption of PEME when the DC current
and significant improvement in the DC current waveform of the is large. At the maximum DC current is 200 A, the AC energy
PEME hydrogen–electric coupling system in Case 2 compared to consumption after control is 4.55% lower than before, and the DC
Case 1. energy consumption is 7.07% lower than before. Therefore, the
Similarly, Fig. 9 indicates that the DC voltage ripple of the above experimental results confirm that the cost of electrolyzing
PEME in the hydrogen–electric coupling system is reduced and water to hydrogen can be lowered by improving the quality of
4795
Y. Dong, S. Ma, Z. Han et al. Energy Reports 9 (2023) 4789–4798

Fig. 9. Test results of PEME DC voltage waveforms in Case 1, Case 2. Fig. 11. PEME segmentation adjustment response time.

Fig. 12. The response speed test results under maximum power startup
conditions.
Fig. 10. Test results of PEME energy consumption change in Case 1, Case 2.

issuing a new instruction to the current value returning to a stable

electrical energy on the input side of PEME to reduce the energy state was recorded, and the DC efficiency, AC efficiency, AC energy
consumption of the electrolytic pool. consumption, and DC energy consumption after returning to a
stable state were recorded. The response speed test results are
4.5. The current regulation speed of the auxiliary service of the shown in Fig. 11.
power grid affects the experimental results As shown in Fig. 11, the average segmented adjustment re-
sponse speed of PEME is 143 s, which is suitable for serving as
The maximum power start-up response speed and segmented a controllable load with minute-level response and participating
adjustment response speed of the PEME were tested after con- in peak shaving service and AGC backup auxiliary services for
necting it to the hydrogen–electric coupling system test platform. the power grid. As the current increases, the response speed
For the maximum power start-up response speed test, the slows down. In addition, the response speed test results under
current instruction was set to 200 A from an initial current of maximum power startup conditions are shown in Fig. 12.
20 A. And the time taken for the current to return to a stable On the other hand, the maximum power start-up response
state is measured. To test the segmented adjustment response speed of PEME, when set to its maximum working current, is
speed test, the response speed was first tested for an upward approximately 386 s, as shown in Fig. 12. Due to the long start-
adjustment. The PEME was run in a constant current mode of up time of the equipment, it is recommended to avoid frequently
50 A-100 A-150 A-200 A, with a new current value given every starting and stopping the equipment when participating in grid
5 min. Then, the response speed was tested for a downward auxiliary services. The change in energy consumption and effi-
adjustment, with the PEME run in a constant current mode of ciency of the PEME in the green hydrogen–electrolysis coupling
200 A-150 A-100 A-50 A, with a new current value given every system with segmented regulation is shown in Fig. 13.
5 min. To avoid the influence of temperature, the PEME tempera- As depicted in Fig. 13, when the direct current side current of
ture was raised to 50 ◦ C before testing. The time difference from the green hydrogen–electrical coupling system PEME increases,
4796
Y. Dong, S. Ma, Z. Han et al. Energy Reports 9 (2023) 4789–4798

of different types of frequency modulation services in the grid.

This will further promote the application of green hydrogen pro-
duction in green hydrogen–electric coupling systems, enhance
the flexibility of the power system, and reduce system carbon
emissions (Yang et al., 2022a).

Funding

This work was supported by the Liaoning Electric Power Co.,

Ltd, China. Science and Technology Project of State Grid, China
under Grant No. 2021YF-81, Liaoning BaiQianWan Talents
Program.

Declaration of competing interest

The authors declare that they have no known competing finan-

cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
Fig. 13. Segment regulation of energy consumption and efficiency.
Data availability

Data will be made available on request.

the alternating current efficiency of PEME will increase, while
the direct current efficiency will decrease. Additionally, both the References
alternating current consumption and the direct current consump-
tion will increase. In engineering applications, the size of the Abdin, Z., Webb, C.J., Gray, E. MacA, 2015. Modelling and simulation of a proton
direct current side current can be controlled to achieve the oper- exchange membrane (PEM) electrolyser cell. Int. J. Hydrogen Energy 40 (39),
ating state of PEME that meets different hydrogen gas demands 13243–13257. http://dx.doi.org/10.1016/j.ijhydene.2015.07.129.
Alfredo, Ursúa, Pablo, Sanchis, Luis, Marroyo, 2013. Electric conditioning and effi-
and improve energy utilization efficiency.
ciency of hydrogen production systems and their integration with renewable
energies. Renew. Hydrog. Technol. 333–360. http://dx.doi.org/10.1016/B978-
5. Conclusion 0-444-56352-1.00014-3.
Buitendach, Henning P.C., Rupert, Gouws, Martinson, Christiaan A., Carel, Min-
naar, Dmitri, Bessarabov, 2021. Effect of a ripple current on the efficiency
This paper aims to fill the research gap by exploring the adapt-
of a PEM electrolyser. Results Eng. 10, 100216. http://dx.doi.org/10.1016/j.
ability of green hydrogen–electric coupling systems to varying rineng.2021.100216.
grid conditions, with a specific focus on analyzing the perfor- Burin, Yodwong, Damien, Guilbert, Melika, Hinaje, Matheepot, Phattanasak,
mance of PEME under different operating conditions. Through the Wattana, Kaewmanee, Gianpaolo, Vitale, 2021. Proton exchange membrane
electrolyzer emulator for power electronics testing applications. Processes 9
experiment and result, the following conclusions are drawn:
(03), 498. http://dx.doi.org/10.3390/pr9030498.
(1) Increasing the operating temperature of the PEME can Dodds, Paul E., Iain, Staffell, Hawkes, Adam D., Francis, Li, Philipp, Grünewald,
enhance the electrolysis efficiency, decrease energy consumption, Will, McDowall, Paul, Ekins, 2015. Hydrogen and fuel cell technologies for
and lower the cost of hydrogen production. From 30 ◦ C to 50 ◦ C heating: a review. Int. J. Hydrogen Energy 40 (05), 2065–2083. http://dx.doi.
the AC and DC energy consumption decreases by 4.29% and 4.97%. org/10.1016/j.ijhydene.2014.11.059.
Frano, Barbir, 2005. PEM electrolysis for production of hydrogen from renewable
(2) Voltage and frequency changes have a minimal impact energy sources. Sol. Energy 78 (05), 661–669. http://dx.doi.org/10.1016/j.
on the energy consumption and efficiency of PEME. Deviations solener.2004.09.003.
within the system’s specified range do not significantly affect the Friedrich-W., Speckmann, Steffen, Bintz, Kai Peter, Birke, 2019. Influence of
operation. However, harmonic distortion on the AC side of the rectifiers on the energy demand and gas quality of alkaline electrolysis
systems in dynamic operation. Appl. Energy 250, 855–863. http://dx.doi.org/
grid can affect the cell’s operating voltage, DC energy consump-
10.1016/j.apenergy.2019.05.014.
tion, and the cost of electrolysis hydrogen production. When the Godula-Jopek, Agata, 2015. Hydrogen Production: By Electrolysis. Wiley-VCH
PEME DC side current was 200 A, the AC energy consumption was Verlag GmbH & Co. KGaA, Weinheim.
reduced by 4.55% and the DC energy consumption was reduced by Haluk, Görgün, 2006. Dynamic modelling of a proton exchange membrane (PEM)
7.07%. Improving the electrical energy quality on the grid side of electrolyzer. Int. J. Hydrogen Energy 31 (01), 29–38. http://dx.doi.org/10.
1016/j.ijhydene.2005.04.001.
the green hydrogen–electrolysis coupling system can effectively Hiromi, Yamamoto, Hanako, Fujioka, Kunihiko, Okano, 2021. Cost analysis of
reduce the cost of green hydrogen production. stable electric and hydrogen energy supplies derived from 100% variable
(3) The response speed of PEME during maximum power start- renewable resources systems. Renew. Energy 178, 1165–1173. http://dx.doi.
up is approximately 386 s, and the average response speed during org/10.1016/j.renene.2021.06.061.
Houhe, Chen, Wenling, Mao, Rufeng, Zhang, Wenfang, Yu, 2021. Low-carbon
segmented adjustment is around 143 s. This feature allows it to
optimal scheduling of a power system source-load considering coordination
be used as a load-balancing service and AGC standby, fulfilling based on carbon emission flow theory. Power Syst. Prot. Control 49 (10),
the requirements of grid auxiliary services. Additionally, PEME 1–11. http://dx.doi.org/10.19783/j.cnki.pspc.200932.
can serve as peak adjustment and frequency adjustment, as well Jiarong, Li, 2023. A multi-stack power-to-hydrogen load control framework for
the power factor-constrained integration in volatile peak shaving conditions.
as a backup electric load in renewable energy power systems,
http://dx.doi.org/10.48550/arXiv.2301.09578.
enhancing system flexibility. Jiarong, Li, Jin, Lin, Yonghua, Song, Xuetao, Xing, Chen, Fu, 2019. Operation
Further research may focus on improving the performance of optimization of power to hydrogen and heat (p2hh) in adn coordinated with
the electrolyzer converter and reduce electrolysis energy con- the district heating network. IEEE Trans. Sustain. Energy 10 (04), 1672–1683.
sumption, thus further reducing the electricity cost of the system. http://dx.doi.org/10.1109/TSTE.2018.2868827.
Joonas, Koponen, Vesa, Ruuskanen, Antti, Kosonen, Markku, Niemelä, Jero, Ahola,
Additionally, studying the performance of the converter and im- 2019. Effect of converter topology on the specific energy consumption of
proving the catalytic material of the electrolyzer can enhance alkaline water electrolyzers. IEEE Trans. Power Electron. 34 (07), 6171–6182.
the response speed of the electrolyzer to meet the requirements http://dx.doi.org/10.1109/TPEL.2018.2876636.

4797
Y. Dong, S. Ma, Z. Han et al. Energy Reports 9 (2023) 4789–4798

Joonas, Koponen, Vesa, Ruuskanen, Michael, Hehemann, Edward, Rauls, Yang, Li, Fanjin, Bu, Yuanzheng, Li, Long, Chao, 2023a. Optimal scheduling
Antti, Kosonen, Jero, Ahola, Detlef, Stolten, 2020. Effect of power quality on of island integrated energy systems considering multi-uncertainties and
the design of proton exchange membrane water electrolysis systems. Appl. hydrothermal simultaneous transmission: A deep reinforcement learning
Energy 279, 115791. http://dx.doi.org/10.1016/j.apenergy.2020.115791. approach. Appl. Energy 333, 120540. http://dx.doi.org/10.1016/j.apenergy.
Jun, Zeng, X.U., Dongdong, Huafang, Guo, Chendi, Li, Junfeng, Liu, 2016. Renew- 2022.120540.
able energy-oriented micro-grid power quality characteristic analysis and Yang, Li, Meng, Han, Mohammad, Shahidehpour, Jiazheng, Li, Chao, Long, 2023b.
comprehensive evaluation. Power Syst. Prot. Control 44 (19), 10–16. Data-driven distributionally robust scheduling of community integrated
Nan, Ning, 2017. Research on hydrogen generation system by water electrolysis energy systems with uncertain renewable generations considering integrated
under wide power fluctuation. Ship Sci. Technol. 39 (11), 133–136. demand response. Appl. Energy 335, 120749. http://dx.doi.org/10.1016/j.
Rabiee, Abbas, Keane, Andrew, Soroudi, Alireza, 2021. Green hydrogen: A new apenergy.2023.120749.
flexibility source for security constrained scheduling of power systems with Yang, Li, Ruinong, Wang, Yuanzheng, Li, Meng, Zhang, Long, Chao, 2023c. Wind
renewable energies. Int. J. Hydrogen Energy 46 (37), 19270–19284. http: power forecasting considering data privacy protection: A federated deep
//dx.doi.org/10.1016/j.ijhydene.2021.03.080. reinforcement learning approach. Appl. Energy 329, 120291. http://dx.doi.
Roberto Moreno, Soriano, Nuria, Rojas, Emilio, Nieto, Rosa de org/10.1016/j.apenergy.2022.120291.
Guadalupe, González-Huerta, Juan Manuel, Sandoval-Pineda, 2021. Yang, Li, Wang, Bin, Zhen, Yang, Jiazheng, Li, Guoqing, Li, 2022b. Optimal
Influence of the gasket materials on the clamping pressure distribution scheduling of integrated demand response-enabled community-integrated
in a PEM water electrolyzer: Bolt torques and operation mode in energy systems in uncertain environments. IEEE Trans. Ind. Appl. 58 (2),
pre-conditioning. Int. J. Hydrogen Energy 46 (51), 25944–25953. 2640–2651. http://dx.doi.org/10.1109/TIA.2021.3106573.
http://dx.doi.org/10.1016/j.ijhydene.2021.03.076. Yannan, Dong, Shaohua, Ma, Zijiao, Han, Henan, Dong, Xiangjun, Li, 2022. A
Ruomei, Qi, Jiarong, Li, Jin, Lin, Yonghua, Song, Jiepeng, Wang, Qiangqiang, Cui, comprehensive virtual synchronous generator control strategy for harmonic
Yiwei, Qiu, Ming, Tang, Jian, Wang, 2023. Thermal modeling and controller and imbalance voltage suppression of multi-inverter parallel microgrid.
design of an alkaline electrolysis system under dynamic operating conditions. Electronics 11 (03), 492. http://dx.doi.org/10.3390/electronics11030492.
Appl. Energy 332, 120551. http://dx.doi.org/10.1016/j.apenergy.2022.120551. Zhongming, Du, Jinyang, Zheng, Jianfeng, Dai, Jianfeng, Shi, Zhengli, Hua, Bo, Li,
Tayfun, zgür, Ali Cem, Yakary lmaz, 2018. A review: exergy analysis of PEM Tongfeng, Zhang, Mengjing, Hou, 2022. Construction of green-hydrogen
and PEM fuel cell based CHP systems. Int. J. Hydrogen Energy 43 (38), supply system in China: reflections and suggestions. Strateg. Study CAE 24
17993–18000. http://dx.doi.org/10.1016/j.ijhydene.2018.01.106. (06), 64–71.
Xiaojun, Shen, Congying, Nie, Hong, Lv, 2021. Coordination control strategy of Zijiao, Han, Shun, Yuan, Yannan, Dong, Shaohua, Ma, Yudong, Bian, Xinyu, Mao,
wind power-hydrogen alkaline electrolyzer bank considering electrothermal 2022. Research on the flexibility margin of an electric-hydrogen coupling
characteristics. Trans. China Electrotech. Soc. 36 (03), 463–472. http://dx.doi. energy block based on model predictive control. Front. Energy Res. 10,
org/10.19595/j.cnki.1000-6753.tces.200283. 879244. http://dx.doi.org/10.3389/fenrg.2022.879244.
Xiaolin, Ge, Yan, Jin, Shu, Xia, Yang, Fu, 2019. A coordinated optimization Ziming, Ma, Haiwang, Zhong, Qing, Xia, Chongqing, Kang, Liming, Jin, 2020.
scheduling of wind-hydro-thermal power system based on requirement Constraint relaxation-based day-ahead market mechanism design to promote
of peak load and frequency regulation. Power Syst. Technol. 43 (11), the renewable energy accommodation. Energy 198, 117204. http://dx.doi.org/
3917–3926. http://dx.doi.org/10.13335/j.1000-3673.pst.2018.2042. 10.1016/j.energy.2020.117204.
Xiaoqiang, Guo, Yupeng, Wei, Yanming, Wan, You, Zhou, Lingguo, Kong, Zini, (Y)Gabriele, Tartarini, Paolo, 2015. Solar Hydrogen Energy Systems: Science
Zhe, Zhang, Zhigang, Lu, Changchun, Hua, Mariusz, Malinowski, 2021. Review and Technology for the Hydrogen Economy. China Machine Press, Beijing.
on power electronic converters for producing hydrogen from renewable Zsolt, Dobó, Árpád Bence, Palotás, 2016. Impact of the voltage fluctuation of the
energy sources. Autom. Electr. Power Syst. 45 (20), 185–199. http://dx.doi. power supply on the efficiency of alkaline water electrolysis. Int. J. Hydrogen
org/10.7500/AEPS20201101004. Energy 41 (28), 11849–11856. http://dx.doi.org/10.1016/j.ijhydene.2016.05.
Yang, Li, Fanjin, Bu, Jiankai, Gao, Guoqing, Li, 2022a. Optimal dispatch of low- 141.
carbon integrated energy system considering nuclear heating and carbon Zsolt, Dobó, Árpád Bence, Palotás, 2017. Impact of the current fluctuation on
trading. J. Clean. Prod. 378, 134540. http://dx.doi.org/10.1016/j.jclepro.2022. the efficiency of alkaline water electrolysis. Int. J. Hydrogen Energy 42 (09),
134540. 5649–5656. http://dx.doi.org/10.1016/j.ijhydene.2016.11.142.

4798