1-s2.0-S2352152X23015670-main

Journal of Energy Storage 72 (2023) 108170
Contents lists available at ScienceDirect
Journal of Energy Storage

journal homepage: www.elsevier.com/locate/est
Review article
Control strategy review for hydrogen-renewable energy power system

Md. Arafat Hossain a , Md. Rashidul Islam a ,∗, Md. Alamgir Hossain b , M.J. Hossain c
a
Department of Electrical and Electronic Engineering, Rajshahi University of Engineering & Technology, Rajshahi, 6204, Bangladesh
b
Queensland Micro-and Nano-technology Centre, Griffith University, 4111, Queensland, Australia
c
School of Electrical and Data Engineering, University of Technology Sydney, Ultimo NSW 2007, New South Wales, Australia
ARTICLE INFO ABSTRACT
Keywords: Hydrogen is emerging as a crucial component for the advancement and integration of renewable energy
Hydrogen sources (RESs) within modern power systems. It plays a vital role as an energy storage system (ESS), ensuring
Fuel cell stability and reliability in the power grid. Due to its high energy density, large storage capacity, and fast
Electrolyzer
operational capabilities, hydrogen is rapidly gaining popularity as a preferred choice among other ESSs. In
Renewable energy
a hydrogen energy system, hydrogen stored in the hydrogen storage system is converted into direct current
Energy storage system
Control system
(DC) power by a hydrogen fuel cell during energy shortages in the power system. Conversely, an electrolyzer
converts excess power from the grid into hydrogen, which is stored in the hydrogen storage system. These
operations necessitate the implementation of various control systems to facilitate the connection of fuel cells
and electrolyzers to the power grid. These control systems include power electronic converters and DC link
regulation control systems. This review article presents a comprehensive examination of the current challenges
related to the control aspects of integrating hydrogen energy storage technologies into modern power grids. It
provides an in-depth analysis of the advancements and deployments of control systems associated with these
technologies, along with the barriers and prospects they entail. By focusing on the integration of renewable
energy sources, this study aims to assist researchers, engineers, and policymakers in gaining a comprehensive
understanding of the current state and future possibilities of hydrogen energy storage systems.
1. Introduction worldwide capacity for wind energy increased by 60.4 GW in 2019,

which represents a 19% upsurge compared to the preceding year [9].
Energy consumption in the form of electricity becomes the linch- Moreover, as global cross-border conflict compromises the security of
pin of modern civilization. The amount of consumption of electricity the energy supply, individual states are trying to RES goal with utmost
has dramatically increased because it reflects socioeconomic stability priority [10–12]. However, these RESs influx which is surfacing to
in the face of continuous growth of population since the industrial replace fossil fuel to produce electricity put forth various challenges
revolution [1,2]. Traditionally, hydrocarbon (fossil fuel) has been the to the grid because of its intermittency [13–15].
principal energy source for meeting electrical energy demands. How- Energy Storage Systems (ESSs) serve the key role in overcoming
ever, due to the incessant amount of consumption of hydrocarbon, the problems in the grid posed by the penetration of RESs [16–18].
the world is witnessing the gruesome depletion of these sources [3– The services provided by them encompass a wide range of offerings,
5]. Furthermore, this phenomenon results in a substantial elevation of
including short-time scale balancing and reserve capacity, ancillary
atmospheric carbon dioxide (CO2 ) as well as other greenhouse gases,
services intended to enhance grid stability as well as long-term energy
thereby contributing to the phenomenon of global warming [5–7].
storage and grid restoration in the event of a blackout [19–21]. There
Therefore, Renewable Energy Sources (RESs) manifest as an alternative
are various kinds of ESSs based on how they store energy in them-
solution to resolve the energy crisis and climate change. Solar, wind,
selves: thermal, electromagnetic, electromechanical, electrochemical,
hydro, and geothermal are some of the popular RESs across the globe
etc. Among them, electrochemical types of ESS such as hydrogen are
because of their ubiquity. In terms of capacity additions, solar photo-
voltaic (PV) has been the fastest-growing technology in recent years; emerging in the power system domain [22–24]. Hydrogen is gaining
nonetheless, even the record 150 GW added in 2021 is just roughly one- momentum as ESS not only because it is storable in scale but also
third of the average annual additions in the net zero scenario milestones highly transportable [25,26]. The higher heating value and lower heat-
from 2022 through 2030 [8]. Based on the most recent report, the ing value are subsequently 141.8 MJ/kg and 120 MJ/kg at 298K,
∗ Corresponding author.
E-mail address: [email protected] (M.R. Islam).
https://doi.org/10.1016/j.est.2023.108170
Received 5 April 2023; Received in revised form 22 June 2023; Accepted 24 June 2023
Available online 11 July 2023
2352-152X/© 2023 Elsevier Ltd. All rights reserved.
M.A. Hossain et al. Journal of Energy Storage 72 (2023) 108170
and easy to design. However, due to the nonlinear nature of large power
Abbreviations networks, fuel cells, electrolyzers, and storage systems, they cannot
ANFIS Adaptive neuro-fuzzy inference system provide efficient operation [60]. Robust model-based control systems
CNN Convolutional neural network such as sliding mode controllers, and backstepping-based nonlinear
controllers based DC/DC converters are more fail-safe, and control
DNN Dense neural network
inputs are independent of the operating point [47,61,62]. However,
ESS Energy storage system
some shortcomings are associated with them: chattering effect, para-
FC Fuel cell
metric uncertainty, and heavy computational burden. Apart from the
GFM Grid-forming inverter model-based control system, there are numerous types of model-free
IBR Inverter based resources controller-based converters are proposed. Among these, the model
NARMA-L2 Nonlinear auto regressive moving average predictive control system is a mathematical optimization-based control
with exogenous inputs approach that predicts the future behaviour of a system using a mathe-
P2G Power-to-gas matical model and optimizes the control actions based on the predicted
PEM Proton exchange membrane behaviour. Numerous accounts show that it achieves low operation cost
PLL Phase-lock loop with better efficient operation with a significant increment in the health
QRL Quick reaching law of fuel cells and electrolyzers [63,64]. Nonetheless, it stipulates a high
RES Renewable energy source computational burden, as well as truncation of the system model, may
lead to overall instability. In contrast, fuzzy logic control system-based
SG Synchronous generators
DC/DC converters can perform with more robust stability over a large
IC Incremental conductance
margin [65]. However, there is no theoretical stability guarantee in the
LSTM Long short term memory case of the fuzzy logic controller. Feedforward neural network-based
SMC Sliding mode controller DC/DC converters show better performance and can cope with uncer-
TSMC Terminal sliding mode controller tainty due to tuning PI controllers with the appropriate training data
set [66]. Nevertheless, it performs poorly in fast-changing operating
conditions [67]. Some state-of-the-art control systems such as genetic
algorithm-based adaptive neuro-fuzzy inference system that does not
which is higher than most of the fuels [27–29]. Besides, hydrogen require the first principle model of a DC/DC converter shows the
is a nontoxic gas which makes the operation safer [30]. In addition, flexible operation and high accuracy [68]. Hydrogen storage and flow
hydrogen generated from RESs is considered to be clean hydrogen control are also crucial for hydrogen-based power systems. Frequency
that can lessen the emission of greenhouse gases [31–33]. All these domain controller-based hydrogen storage and flow controls reportedly
indispensable advantages of hydrogen are consolidating the hydrogen perform well [59,69], though it can exhibit sluggish response. Kalman
economy day by day [34–36]. Hence, hydrogen is adopted more and filter-based energy management system employs hydrogen storage con-
more in the power systems as energy reserve [37–39]. trol with high efficiency. However, model accuracy can put the overall
Hydrogen fuel cells and hydrogen electrolyzers are the main ap- response at stake. Dynamic programming-based hydrogen storage con-
paratus that are mainly used in the power system for balancing pur- trol systems can obtain global optimal points but it requires apriori
poses [40–42]. A fuel cell is an electrochemical mechanism that gener- experience [70]. The most critical issue of a hydrogen-based power
ates electrical power through the reaction of hydrogen and oxygen. It is system can be referred to power-sharing strategy. The power electronic
made up of two primary parts: the anode, which oxidizes hydrogen, and apparatus which is responsible for it is inverted. The power-sharing
the cathode, which reduces oxygen [23,43]. The electrons produced by strategy can be broadly classified as grid-following power sharing
hydrogen oxidation go through an external load, producing an electric and grid-forming power sharing. Grid-following power sharing is the
current. Afterwards, the electrons return to the cathode and mix with most widely used method reported in the literature. It can decouple
oxygen to make water. It is highly efficient, converting over 60% of the the active and reactive power control however this type of control
energy stored in hydrogen into electricity [44,45]. It produces no emis- system reduces the stability of the classical power grid [58]. On the
sions other than water vapour, making it a clean energy source [46]. other hand, a grid-forming strategy can help to achieve higher re-
It is used in the power systems to increase resilience and reliability of newable penetration [71]. Therefore, a comprehensive review paper
power systems such as meeting power deficiency in the utility grid [47], is needed to critically analyse all the control systems proposed in
storing fluctuating energy of wind turbine and solar [48], serving as the literature for fuel cells and electrolyzers in power systems. This
a distributed energy source during a long power outage [49,50]. A paper aims to bring together all the scattered proposed controllers and
hydrogen electrolyzer is defined as a device that uses electrical energy provide a comprehensive overview of the control systems used in these
to separate water into hydrogen and oxygen [51,52]. The process of technologies.
electrolysis is employed to split water molecules into their constituent There are a number of review accounts with different focuses re-
hydrogen and oxygen atoms by passing an electrical current through garding hydrogen energy systems that can be found in the literature.
it [53,54]. The hydrogen is collected at the cathode, while the oxygen Considering factors such as production and storage techniques, as well
is collected at the anode. It is used in the utility grid as a power-to- as the associated safety and risk considerations of hydrogen technology,
gas (P2G) programme. Like fuel cells, they also serve a crucial function it is being suggested that hydrogen could be a feasible substitute for
in grid operation such as fast frequency response, ancillary services, conventional car fuel and also serve as a means for large-scale energy
virtual inertia, storing excess power in the form of hydrogen, etc [55– storage in power systems [72–74]. A comprehensive technical overview
57]. Thereby, to realize these services, various control systems have of hydrogen fuelling stations is provided in [75]. P2G pilot projects
been proposed available in the literature. are examined in [76] where renewable sources are used to produce
At the outset, a power electronic device such as DC/DC converter hydrogen by the electrolyzer. Risks and safety concerns regarding
is required to integrate hydrogen technology into a utility-scale power hydrogen production and storage methods are studied in [77]. State-of-
system. Various accounts can be found on the DC/DC converter control the-art and future direction of hydrogen storage into clathrate hydrates
system pertaining to the stable operation of the power system and is delineated in [78]. Three steps in P2G plants and their future perspec-
increment of the life-cycle of devices regarding hydrogen technology. tives are uniquely presented in [79]. The impact of different catalyst
Analytical controllers such as PI controllers based DC/DC converters materials on the performance of electrolyzers is discussed in [80].
are widely proposed [42,57–59]. These control systems are versatile Projected P2G deployment scenarios at various geographical locations
2
are surveyed in [22,81,82]. Different metal hydride storages for hy- for the design and optimization of practical hydrogen-based
drogen as an energy vector is documented in [83,84]. A brief review power systems. Additionally, it offers significant insights into the
of the techno-economic aspect of hydrogen is provided in [85] with implementation of efficient control techniques.
the implication that increasing production, setting standardization and (d) Ultimately, future scopes, insights, and guidelines are presented
favourable policies will exhort the adoption of hydrogen technology. so that prospective researchers can seamlessly conduct their
Risk and reliability analysis of the state-of-the-art hydrogen storage and respective research unfaltering which will accelerate the decar-
reliability are reviewed in [86,87]. Hydrogen generation processes and bonization of this world.
related technologies are looked into in [88]. Hydrogen as an energy
The bibliometric survey is carried out on the principle of inclusion
vector and a road map for Europe to implement hydrogen technology is
and exclusion where credible papers are collected from google scholar,
reviewed extensively in [89,90]. Technological advancement regarding
science direct, IEEE xplore and semantic scholar. Over 10,000 credible
green hydrogen is outlined in [91–93], and a geopolitical roadmap
papers were extracted based on the keywords ‘‘hydrogen’’, ‘‘energy
for different countries around the world for hydrogen development is
storage’’, and ‘‘renewable energy’’. Afterwards, those papers are filtered
discussed. Physical modelling of the Proton Exchange Membrane (PEM)
based on year ranges from 1980 to 2023. Then the papers written in
electrolyzer is provided in [23,94] as guidelines for beginners. Various
other than English are taken away. Again papers are filtered out based
hydrogen electrolyzer models are studied in [95]. In [96], hydrogen
on the keywords ‘‘electrolyzer’’, and ‘‘fuel cell’’. The final screening
storage in microgrid systems is comprehensively studied. The microgrid
is done by identifying articles that are in line with the objectives of
based on hydrogen fuel cell, electrolyzer and storage, and subsequently
this paper by extracting papers on the basis of searching by keywords
design and planning of it, is reported detailed in [97,98]. The Energy
‘control system’ and ‘hybrid grid’ in the abstract and conclusions. The
Management System (EMS) for alkaline electrolyzer and hybrid storage
sequential diagram of this procedure is provided in Fig. 1.
system is discussed in [99]. The effects of converter topology on fuel
The structure of this paper is organized in the following way.
cells or electrolyzers are studied in [52,100,101]. Health-conscious
Section 2 describes the superiority of hydrogen ESS among other ESSs
fuel cell energy management techniques are discussed by the authors
in terms of efficacy, cost, and technology readiness level for the power
of [102]. In [103], the control system PEM fuel cell and hydrogen
systems compared to other ESSs. Mathematical modelling of the PEM
storage regarding avoiding fuel starvation is critically reviewed.
fuel cell and electrolyzer cell is presented in Section 3. Section 4
There are a good amount of research has been done on the con-
provides the primary focus of this article, wherein a comprehensive
trol system of hydrogen-based power systems and plenty of scientific
exposition of the control systems is presented. Future trends and some
papers are documented in various peer review journals. Thereby, it
generic advice have been provided in Section 5 based on all the studies
becomes essential for a review paper that organized all those con-
done in the previous section. Conclusions are drawn in Section 6.
trol systems under a single study model so that new researchers can
delve into this topic easily by tracking past development and state-
of-the-art. The other review papers extensively review on properties 2. Hydrogen as energy storage element in power system
of hydrogen, generation and distribution pathways, hydrogen-based
economy, hydrogen storage system, modelling of fuel cells and elec- The ongoing replacement of Synchronous Generators (SGs) with
trolyzers, and power electronic converter competency with fuel cells inverters, driven by the increasing penetration of RESs, gives rise to in-
and electrolyzers. However, better to the authors’ knowledge, there is stability concerns characterized by volatile power generation. Besides,
no comprehensive review paper investigating the control systems of phasing out SGs reduces system inertia proportionately culminating
fuel cells and electrolyzers within a unified study model. This review in a lower inertia system described in Fig. 2. Though a reduction in
paper aims to address this lack in the literature by providing a com- mechanical inertia seems to be an efficient system, it can vehemently
prehensive analysis and synthesis of the current state of knowledge cause frequency instability of the system [104] that leads to an increase
on the topic, thus contributing to the advancement of the field and in the rate of change of frequency and a low-frequency nadir. As a
bridging the existing knowledge gap. The major contribution of this result, frequency relays may trip, resulting in under-frequency load
review paper is stated in the following points. shedding and, in the worst case scenario, cascading outages [105–
107]. Reportedly, international organizations across the world have
(a) This work presents a comparative assessment of different ESSs, documented a continuing reduction in their system’s inertial response
highlighting the superior performance of hydrogen-based ESSs and urge increased inertial response [108]. For grid-connected systems,
compared to other available alternatives. The review thoroughly IEEE advises using a frequency operating standard of ±0.036 Hz [104].
investigates the characteristics, advantages, and limitations of When the system frequency falls below 59.3 Hz, it is advised to activate
various ESSs and asserts that hydrogen-based systems have the the first level of the under-frequency load shedding [104]. ESSs are
potential to overcome the current limitations of other ESSs. the most appropriate solutions to the problems brought about by the
Moreover, they offer a more sustainable and efficient solution high integration of RESs. Moreover, the power system is vulnerable
for storing renewable energy. to natural disasters such as wildfire, earthquakes, tornados, tsunamis,
(b) The review paper provides a comprehensive mathematical mod- etc., or any human intended haphazard like war; ESSs can provide a
elling approach for Proton Exchange Membrane (PEM) fuel cells sustainable solution by forming pocket microgrids preventing a long
and electrolyzers, which are crucial components of hydrogen- power outage [109]. An energy storage system is defined as a system
based power systems. The developed models in this review can that stores energy set aside for future use. There are several different
be utilized to simulate and optimize the performance of PEM types of energy storage systems, each with its own set of advantages
fuel cells and electrolyzers, thereby enhancing their efficiency, and disadvantages. Fig. 3 shows different types of ESS used in the
reliability, and cost-effectiveness within hydrogen-based power power system.
systems. Hydrogen functions as an electrochemical ESS, serving as an energy
(c) This paper introduces a power system model based on hydro- carrier rather than an energy source. Its utilization spans diverse ap-
gen, with a specific focus on investigating control challenges plications such as electricity generation, transportation, and industrial
related to the integration of hydrogen ESSs. The review identi- processes, leveraging its attributes of cleanliness, abundance, and re-
fies the complexities associated with managing power flow and newability. While it is widely recognized as a promising solution for
ensuring system stability and proposes control strategies and carbon reduction, the alignment of hydrogen’s environmental benefits
algorithms to address these challenges. The developed study with this objective relies on factors such as the feedstock employed and
model presented in this work serves as a valuable reference the intended end-use application. In a study conducted by [88], the
3
Fig. 1. A schematic of the sequential articles selection process from different sources.
Fig. 2. Transition towards an inverter-dominated power system.
authors propose a concept for categorizing produced hydrogen based energy from a fuel, such as hydrogen, into electricity. Fuel cells operate
on its quality. Fig. 4 illustrates the colour grading system employed by passing fuel over a catalyst, such as platinum, which causes the
to delineate different production methods. Green and blue hydrogen fuel to break down into protons and electrons. In this process, protons
represents the most desirable categories in terms of achieving carbon traverse a PEM towards the cathode, while electrons are compelled
neutrality. Green hydrogen gives impetus to the researchers to make to travel through an external circuit, thereby producing an electric
an application in power systems. In the power system, hydrogen can current. Afterwards, the electrons and protons join together with air-
be used as a way to store excess electricity from RESs, such as wind supplied oxygen at the cathode to generate water, resulting in the
and solar, which can then be converted back into electricity when it completion of the electrochemical reaction [23,24,111].
is needed [110]. This process, known as electrolysis, involves using From Fig. 6, it can be deduced that hydrogen provides long-term
an electric current to split water into hydrogen and oxygen [22,80]. and other scale solutions compared to other ESS devices in the power
PEM, solid oxide, and alkaline electrolyzers are already on the market system landscape [112]. Therefore, hydrogen ESS is capable of supply-
or are in advanced stages of development and are anticipated to do ing electrical loads during a long power outage [113]. Additionally, it
so in the near future. A fuel cell is a device that converts chemical serves as frequency stabilization in the face of a high rate of change
4
Fig. 3. Catalogue of storage technology.
Fig. 4. H2 production colour code.
Of frequency power system [55,56,114]. It also reduces the intermit- with an electrolyzer, a local power grid, and hydrogen pipelines can
tency of renewable energy sources [46,47,58]. It is worth mentioning accelerate the transformation process towards a carbon-neutral resi-
that the main disposal from this system is water which does not dential community [115]. The emergence of hydrogen-powered cars
necessarily need to be discarded. On top of that, modularity and presents a transformative opportunity in the transportation sector, as
scalability give hydrogen-based power systems a new dimension. Stud-
they offer numerous advantages over conventional internal combustion
ies have been conducted shows that the integration of distributed
engine vehicles. Hydrogen fuel cell vehicles produce zero tailpipe emis-
renewable systems with H2 energy infrastructure offers a promising
sions, mitigating air pollution and reducing greenhouse gas emissions,
opportunity to improve regional energy flexibility, enhance power
reliability, ensure operational robustness, and promote system sustain- thus contributing to combating climate change [116]. Furthermore,
ability. The community-level regional energy system that integrates hydrogen-powered vehicles have longer driving ranges and shorter
electrical, thermal, and hydrogen interactions, involving low-rise single refuelling times compared to electric vehicles, addressing the issue
houses, rooftop PV systems, hydrogen vehicles, a hydrogen station of limited driving range and extended charging durations. A modern
5
Fig. 5. Hydrogen-based community: a modern vehicle energy network.
Fig. 6. An analysis of the relationship between the discharge time and capacity of various energy storage technologies.
vehicle energy network is presented in Fig. 5 that insinuates all re- there are regulatory and policy challenges, such as the need to establish
newable sources can significantly reduce carbon emission by bringing standards and regulations for the use of hydrogen in the power system.
in hydrogen-based vehicles. In [117], a transportation-based energy Notwithstanding these challenges, hydrogen is anticipated to assume a
migration framework is formed that extends a neighbourhood scale to progressively significant role in the future power system, aligning with
a city scale. There are already many ongoing pilot projects going on the escalating demand for clean and renewable energy sources [20].
across the world tabulated in Table 1 that give a clear indication that
this technology is highly reliable in terms of technological readiness 3. Dynamical modelling of PEM fuel cell and PEM electrolyzer
level. There are also some challenges to the widespread adoption of
hydrogen in the power system. One of the main challenges is the cost of Dynamical modelling of PEM fuel cells and electrolyzers is im-
production, which is currently higher than that of fossil fuels. There are portant because it allows for the simulation and prediction of their
also technical challenges, such as the need for advanced infrastructure behaviour over time. This information can be used to optimize their
to support the production, storage, and transport of hydrogen. Finally, design, control, and operation, leading to improved performance and
6
Table 1
H2 pilot projects.
Project name Country Electrolyzer Hydrogen storage Fuel cell Ref.
Type Power Pressure Volume capacity Type Power

[KW] bar [N m^3] [KW]
SWB Germany Alkaline 100 30 5000 PAFC 80 [118,119]

INTA Spain Alkaline 5 10 24/−9 PAFC/PEM 10/7.5 [120]
SCHATZ USA Alkaline 6 8 60 PEM 1.5 [121]
Solar hydrogen pilot plant Finland Alkaline 0.8 25 200 PAFC 0.5 [122]
SCHATZ solar hydrogen USA Alkaline 6 5.7 30 [123,124]

project
SWB Project, Neunburg Germany Alkaline 100 5000 30 [119]

vorm Wald
Freiburg Solar House Germany PEM 2 15 30 [125]

SHDP Germany Alkaline 26 137 3000 PEM 6 [126]
PHOEBUS Germany Alkaline 26 120 3000 PEM 5.6 [120,127]
PHOEBUS, Julich Germany Alkaline 26 2100 120 [127]
SAPHYS Europe Alkaline 5 200 120 PEM 3 [126]
Solar house Germany PEM 2 28 400 PEM 3.5 [128]
SAPHYS Italy Alkaline 5 300 20 [129]
Laboratory Plant Stralsund Germany Alkaline 20 200 25 [130]
FIRST Spain PEM 1 30 _ PEM 0.42 [126]

Grimstad Renewable Norway Alkaline 50 8 15 [131]
Energy Park
HRI Canada Alkaline 5 120 40 PEM 5 [132]

Laboratory Plant HRI Argentina Alkaline 5 35 10 [133]
Quebec
HARI UK Alkaline 36 137 2856 PEM 7 [123,134]

FIRST Showcase II Spain PEM 1 70 30 [135]
Laboratory Plant IFE Norway PEM 1.5 14 16 [136]

Kjeller
HARI UK Alkaline 36 200 2856 PEM 7 [134]

PVFCSYS Sophia Antipolis France Alkaline 3.6 0.4 10 [137]
PVFCSYS Agrate Italy Alkaline 3.4 4 10 [138]

Utsira Island Norway Alkaline 50 2400 200 [136]
CEC USA PEM 3.35 15 5.4 PEM 2.4 [139]
HELMETH Europe SOEC 15 30 5.4 [140,141]
RENOVAGAS Spain Alkaline 15 25 2 [142]
DemoSNG Sweden PEM 60 _ _ [143]
Wind2hydrogen Australia PEM 100 163 14 [144]
Levenmouth Community UK Alkaline 250 _ 10 100 [145]
Energy Project LCEP
HyDeploy UK PEM 500 20 _ [146]

STORE&GO Germany PEM 1000 [142]
efficiency. In addition, dynamical modelling helps to identify potential PEM fuel cells and electrolyzers since it is necessary for understanding
issues that may arise during operation and provides insight into the their physical characteristics. In contemporary power systems, there is
underlying physical and chemical processes that take place within the a growing need for enhanced attributes such as rapid responsiveness,
fuel cell or electrolyzer. Furthermore, this information can be used modularity, increased efficiency, and heightened safety. Among these
to inform the development of control strategies, which are crucial
criteria, PEM fuel cells outperform alkaline fuel cells, despite the latter’s
for ensuring the safe and stable operation of these devices. With this
prevailing commercial popularity [65,147,148]. Consequently, it can
knowledge, engineers and scientists can evaluate the impact of different
parameters, such as pressure, temperature, and current density, on be logically inferred that PEM fuel cells represent a promising direction
the performance of the fuel cell or electrolyzer, and make informed for future advancements. Thus, the scope of this study is deliberately
decisions about how to optimize their design and operation. This restricted to solely encompassing the modelling aspects of PEM fuel
section is committed to a concise review of the dynamical modelling of cells and electrolyzers. Furthermore, this paper focuses primarily on the
7
control system of fuel cells and electrolyzers, intentionally excluding a

comprehensive examination of various fuel cell models.
The mathematical model of the PEM fuel cell gets established with
adequate accounts [24,43,51,149–153]. Like other fuel cells, the PEM
fuel cell also produced electricity by the following reactions ((1), (2))
happening at the anode and cathode.
H2 ⟶ 2H+ + 2e− (1)
1
2H+ + O + 2e− ⟶ H2 O (2)
2 2
The theoretical value of an open circuit voltage can be expressed as the
following equation.
𝛥𝑔 𝑓 𝑔 H2 + 12 𝑔 O2 − 𝑔 H2 O
𝐸=− = (3)
2𝐹 2𝐹
the symbol 𝑔𝑓 denotes the Gibbs free energy, while 𝐹 represents the
Faraday constant. Despite the potential for extracting complete energy
from PEM fuel cells, there are inevitable losses during operation. The
primary losses in PEM fuel cells are activation loss, concentration
loss, and ohmic loss. The Tafel equation can be used to mathemati-
cally express activation loss at the electrode, as was shown in a prior
study [154].
( ) Fig. 7. Modelling the PEM fuel cell: a simplified equivalent circuit approach.
𝑖
𝛥𝐸𝑎 = 𝐴 ln (4)
𝑖𝑜
where 𝑖 is current density; 𝐴 and 𝑖𝑜 are constant. The equation is
applicable only when the current density is less than 𝑖𝑜 . Inadequate
transport of reactant to the electrode surface results in concentration
loss, which can be represented by the following equation according
to [43].
( )
𝑖
𝛥𝐸𝜌 = −𝐵 ln 1 − (5)
𝑖1
where 𝐵 is constant which is dependent on the PEM fuel cell and 𝑖1 is a
limitation of current density. The ohmic loss due to the resistance of the
electrode and electrolyte can be modelled as the following equation.
𝛥𝐸𝛺 = 𝐼𝑅𝑓 𝑐 (6)
where 𝑅𝑓 𝑐 is the resistance of PEM fuel cells’ electrode and electrolyte.

Two charged layers of opposite polarity formed at the electrode of the
PEM fuel cell behave like an electrical capacitor known as the double-
layer charging effect [43]. The terminal voltage of the PEM fuel cell
(FC) can be expressed as follows.
𝑉𝑓 𝑐 = 𝐸 − 𝛥𝐸𝑎 − 𝛥𝐸𝜌 − 𝛥𝐸𝛺 (7)
Evidence shows that the significant losses in PEM fuel cells are ohmic
loss and activation loss. For simplicity, the concentration loss can be
neglected. The simplified equivalent circuit model of a PEM fuel cell is
drawn in Fig. 7.
PEM electrolyzer performs the opposite reactions of the PEM fuel
cell. PEM electrolyzer model is also established in the literature and
some research is still ongoing [23,51,155,156]. It splits the pumped
water into oxygen (O2 ), protons (H+ ), and electrons (e− ) at the anode
given in reaction (8). The proton travelled through the proton exchange
membrane and combined with the electron at the cathode producing
hydrogen shown in reaction (9). Fig. 8. Modelling the PEM electrolyzer: a simplified equivalent circuit.
1
H2 O ⟶ 2H+ + O + 2e− (8)
2 2
electrodes during chemical reactions, voltage is lost, which causes this
2H+ + 2e− ⟶ H2 (9) phenomenon. The activation energy is necessary for both the anode
Fig. 8 depicts an equivalent circuit model of a PEM electrolyzer cell and cathode as a result of the activation overpotential can be expressed
which is empirically synthesized in [157]. The circuit contains two by applying the Butler–Volmer equation, as demonstrated by numerous
parallel RC branches for the anode and cathode which represent exces- researchers [158,159].
sive potential. The excess potential required to trigger proton transfer ( )
and electrochemical kinetics in a PEM electrolyzer is defined as ac- 𝑅𝑇 𝑖𝑎
𝑉𝑎𝑎𝑐𝑡 = log (10)
tivation over-potential. When electrons are transported to or from 𝛼𝑎 𝑧𝐹 𝑖0,𝑎
8
Fig. 9. Schematic of grid-connected ELZ/FC system.
( )
𝑅𝑇 𝑖𝑐 the hydrogen as chemical energy during the electrolysis process. The
𝑉𝑐𝑎𝑐𝑡 = log (11)
𝛼𝑐 𝑧𝐹 𝑖0,𝑐 fuel cell stack and electrolyzer cell stack are connected to the DC
In the given equations, 𝑉𝑎𝑎𝑐𝑡 and 𝑉𝑐𝑎𝑐𝑡 represent the activation over- bus through boost and buck converters respectively. DC bus and AC
potential of the anode and cathode respectively. The universal gas bus are interfaced with a bidirectional DC/AC inverter; as a result, it
constant is represented by 𝑅, while 𝑧 represents the stoichiometric co- can send power both on the DC side and AC side. As it is discussed
efficient that denotes the number of electrons transferred in the overall earlier, this arrangement of the hydrogen power system acts as an
semi-reactions as defined by Faraday’s law. The charge transfer coeffi- energy buffer that dials down the intermittency of renewable sources
cients are denoted by 𝛼𝑎 and 𝛼𝑐 . In the context of a PEM electrolyzer, as well as increases the resilience of the power system. Whenever there
the ohmic over-potential can be defined as the electrical resistance is excessive power available in the grid is consumed by the electrolyzer
that arises from the movement of electrons as well as the electronic cell stack and stored in the form of hydrogen in the storage tank; on
resistance inherent in the system. The magnitude of this ohmic over- the other hand, if there is no adequate power supply in the grid, the
potential is contingent upon the specific type of PEM employed and the fuel cell stack supply that amount of power back to the grid. It leads
nature of the electrode material in use. In addition, the resistance at- to the solution regarding frequency stabilization during the occurrence
tributable to the membrane, either in the form of membrane resistance of any power mismatch. The fuel cell stack and electrolyzer stack both
or ionic resistance, can be regarded as a component of the ohmic over- can be connected to a H2 tank through H2 pipelines. This is why the
potential and is directly related to the level of hindrance encountered system is called ELZ/FC since they function conjugately. This conjugate
by proton transport through the PEM. 𝑅𝑚𝑒𝑚 in the schematic model system can be expressed in Fig. 9.
of the PEM electrolyzer accounts for membrane resistance. The over The control objectives associated with the system depicted in Fig. 9
potential resulting from the membrane’s ohmic resistance is presented
are enlisted below:
below:
(i) Control of buck and boost converters
𝑉𝛺𝑚𝑒𝑚 = 𝑅𝑚𝑒𝑚 𝑖0 (12)
(ii) DC link voltage regulation
where, 𝑅𝑚𝑒𝑚 is the ionic resistance. The cell in the PEM electrolyzer (iii) Control technique of pressure tank
schematic accounts for reversible potential expressed as (iv) Active and reactive power sharing control strategies
𝛥𝐺
𝑉𝑟𝑒𝑣 = (13) The rest of this section is divided into subsections which will cover the
𝑛𝐹
review of the control system in regard to the control objectives list.
The number of electrons transferred per mole of hydrogen (𝑛) deter-
mines the Gibbs free energy (𝛥𝐺) in a chemical reaction involving
hydrogen. 4.1. Control techniques of ELZ/FC’s buck and boost converter
4. Control system of H𝟐 integrated power system Buck and boost converters are power electronic devices that are
used to adjust the voltage and current of the power supplied to the
The fuel cell utilizes hydrogen as fuel to generate electrical power, electrolyzers or fuel cells. They are critical components in these systems
while the electrolyzer consumes electrical power and stores it within because they ensure that the electrolyzers or fuel cells receive the
9
Fig. 10. H2 FC/ELZ integrated DC/DC converter.
correct amount of power and operate at the optimal voltage and current
levels. Proper selection of topology of these converters is extremely im-
portant because the health of the stack of both fuel cell and electrolyzer
depends on it [52,160,161]. The slow diffusion and charge transfer
activities inside the cell are hampered by the low-frequency current
ripple caused by the nonlinearity of the electrical load [162–164]. On
the other hand, the high-frequency current ripples generated by the
DC/DC converter element have less of an impact on the performance
of the stack because there is double-layer capacitance at the anode and
cathode electrode interfaces [52,165,166]. Control techniques for the
buck and boost converters are important because they ensure that the
converters operate efficiently and accurately. They regulate the output
voltage and current of the converters and prevent any fluctuations or
deviations that could affect the performance of the electrolyzer or fuel
cell. In addition, control techniques help to optimize the performance
Fig. 11. A schematic representation of the control system for a DC/DC converter in
of the converters and to reduce their energy losses, which can result in its generic form.
improved efficiency and longer instrument life.
Whether it is an electrolyzer cell stack or a fuel cell stack, all
stacks consist of their respective cells which are connected in a series
on the difference between the planned setpoint and the actual output,
and parallel manner. Throughout this paper, the term electrolyzer cell
proportional control is used. The proportional gain (𝐾𝑝 ) determines the
stack will be referred to as ELZ and fuel cell stack will be referred to
degree of correction applied to the control output for a given error.
as FC. The DC link voltage is too high for the ELZ and too low for
Integral control takes into account the accumulated error over time,
the FC stack for faithful operation. A buck converter reduces voltage
which allows for a gradual correction of long-term deviations from the
down to the operation level of the ELZ. On the other hand, a boost
setpoint. The integral gain (𝐾𝑖 ) determines the rate of correction of the
converter boosts the voltage of the FC so that it can send power to the
accumulated error. Together, the proportional and integral components
DC bus. The DC voltage can range from 48–480V [167]. This voltage
of the PI controller work to minimize the error between the desired and
can vary concerning the fuel cell and electrolyzer’s rating, the overall
actual values of the system being controlled.
system where it will be integrated, and other environmental param-
Refs. [42,57–59] account for the PI controller design of ELZ and FC.
eters. From [168,169], the conclusion can be drawn that in terms of
complexity, reliability, and cost the conventional non-isolated DC/DC In [42,59], both papers adopt the same control loop portrayed in Fig. 13
converters weigh better. On top of that, due to empirical adequacy and regarding the current control of the ELZ and FC. It is worth noting that,
rich control literature, this review paper is set its scope only to these in [42], the authors used Alkaline ELZ and PEM FC; in [59], the authors
traditional converters. Both a non-isolated DC/DC buck converter and used Alkaline ELZ and Solid Oxide (SO) FC. 𝑃𝑑(𝑟𝑒𝑓 ) , 𝑖𝑒𝑙𝑧(𝑟𝑒𝑓 ) and 𝑖𝑓 𝑐(𝑟𝑒𝑓 )
the boost converter illustrated in Fig. 10 comprise an IGBT switch, are the reference power for both ELZ and FC, ELZ’s reference current
a diode, an inductor and a resistor but there are differences in the and FC’s reference current respectively. The current in ELZ and FC are
arrangement of the components. The ELZ acts as an electrical load of opposite in the direction flowing in the system. That is why 𝑃𝑑(𝑟𝑒𝑓 ) is
the buck converter, and the FC acts as an electrical source of boost held as positive in the case of FC and negative in the case of ELZ. The
converter. Fig. 11 produces the perception of a generic DC/DC con- associated parameters are held as positive constants.
verter control system. Numerous control methods resonate in the power In [42], the authors provide a design procedure for tuning param-
system literature shown in Fig. 12. Within the category of favourable eters of the respective PI controllers of Alkaline ELZ and PEM FC.
controllers, certain ones have been suggested specifically for DC/DC Because of being constant voltage, 𝑉𝑒𝑙𝑧 and 𝑉𝑓 𝑐 are neglected. On
converters, serving as an interface between H2 FC/ELZ systems and the top of that, the dynamic parts of both the electrolyzer and fuel cell
electrical network. The next subsections are dedicated to reviewing the are linearized as resistors neglected considering the worst damping
study of those controllers regarding FC/ELZ integrated power systems. condition. Adopting an averaged switching model, the authors provide
the following small signal control diagram represented in Fig. 14. The
4.1.1. PI controlled method transfer function (TF) of the ELZ’s buck converter and TF of FC’s
The most popular and widely used controller is the proportional– boost converter, therefore, will be the same. According to the authors’
integral (PI) controller because it is effective and simple to develop argument, the symmetric design offers improved control of the DC link
without requiring extensive mathematical or physical modelling. Based voltage as both the ELZ and FC current loops function as an inner loop
10
Fig. 12. Different controllers proposed in DC/DC converter control system literature.
for the DC voltage loop. The open loop TF is obtained by applying KVL the real axis or the imaginary axis. The authors also have prescribed a
in Fig. 10(a) or Fig. 10(b) as the following. larger bandwidth for the faster response of the system. Fig. 15 shows a
𝑖𝑒𝑙𝑧 (𝑠) 1 faster response is achieved through larger closed-loop bandwidth and
= (14) small bandwidth yields a slow response.
𝑉1 (𝑠) 𝑅𝑑 + 𝑠𝐿𝑑
Esmaeilian et al. design the controllers for alkaline ELZ and PEM
Therefore, the closed-loop transfer function, 𝐺𝑐 will be: FC slightly differently than the aforementioned method [58]. In this
𝑖𝑒𝑙𝑧 (𝑠) 𝑘𝑝𝑒𝑙𝑧 𝑠 + 𝑘𝑖𝑒𝑙𝑧 paper, the authors have used two PI cascade loops where the outer
𝐺𝑐 = = (15)
𝑖𝑒𝑙𝑧(𝑟𝑒𝑓 ) (𝑠) 2
𝐿𝑑 𝑠 + (𝑅𝑑 + 𝑘𝑝𝑒𝑙𝑧 )𝑠 + 𝑘𝑖𝑒𝑙𝑧 loop generates a reference for the ELZ and FC’s current that is to be
regulated in the inner loop. The transfer functions for the ELZ and FC
Second-order TF characterizes this closed-loop TF. The PI controller are obtained by state space average modelling. Figs. 16(a) and 16(b)
gains can be established by selecting the characteristic roots that have a are the control loops proposed in [58].
higher damping oscillation frequency and critical damping factor. This While the previous methods do not take into account the inter-
is achieved through careful calibration. The following are the roots of nal dynamics of the ELZ or FC, Dozein et al. develop an open loop
the qualities. transfer function for both alkaline and PEM ELZ through which an
√
𝑅𝑑 + 𝑘𝑝𝑒𝑙𝑧 ± (𝑅𝑑 + 𝑘𝑝𝑒𝑙𝑧 )2 − 4𝐿𝑑 𝑘𝑖𝑒𝑙𝑧 electrical circuit model is synthesized [57]. A second-order transfer
𝛼1,2 = − (16) function is developed according to control system theory and comes
2𝐿𝑑 up with the electrical circuit model depicted in Fig. 17. Where the
The choice of 𝑘𝑝𝑒𝑙𝑧 and 𝑘𝑖𝑒𝑙𝑧 will produce polls at different locations inductance and capacitance, 𝐿𝑠 and 𝐶𝑂𝑃 , arise from the time delay of
in the S-plane. The determinant determines whether the poles stay on the step response of the alkaline electrolyzer. In the case of the PEM
11
Fig. 13. PI controlled (a) ELZ and (b) FC.
Fig. 14. Small signal control loop for PI controller design.
Fig. 15. Speed of response versus system bandwidths.
electrolyzer, however, the authors adopt the same circuit model of the the circuit modelling incorporates two energy storage elements, the
alkaline electrolyzer stack but neglect 𝐿𝑠 because comparatively, PEM two-state variables dictate the course of action regarding the controller
ELZ offers a faster dynamical response. In fact, PEM is claimed to be design. Neglecting the resistance of buck converter 𝑅𝑑 , the following
a better candidate for virtual inertia and primary frequency response state equations can be obtained [57].
provisions. The capacitance, 𝐶𝑂𝑃 represents electrolysis cell physical
d𝑖𝑒𝑙𝑧
features obtained from [56]. The parallel resistance 𝑅𝑚 is indicative of (𝐿𝑠 + 𝐿𝑑 ) = d𝑉𝑑𝑐 − 𝑉𝑒𝑙𝑧 , (17)
d𝑡
the transport loss resulting from the push–pull interactions of charges.
The internal resistance 𝑅𝑖 is a nonlinear resistance combination of d𝑉𝐸𝐷𝐿 𝑉
stack material and gas concentration that is obtained from [48]. As 𝐶𝑂𝑃 = 𝑖𝑒𝑙𝑧 − 𝐸𝐷𝐿 (18)
d𝑡 𝑅𝑚
12
Fig. 16. Cascaded PI controlled ELZ and FC.
makes the power electronic system an appropriate system where SMC

can be implemented. A good amount of research documents are spring-
ing up in which different types of SMCs are implemented for different
types of converters. In [171–173], the foundation of implanting the
DC/DC buck converter with an SMC system is laid. Likewise, in [174–
176], the SMC design for DC/DC boost converter is proposed. All of
these features and seminal works dedicating the design of SMC for
DC/DC converters give researchers the impetus to develop methods that
included hydrogen FC/ELZ systems [177–179].
Global terminal sliding mode controller: Terminal sliding mode con-
troller (TSMC) is an evolution of the conventional sliding mode con-
troller which was proposed back in 1992 [180,181]. Conventional
sliding mode controller is notorious for the chattering phenomenon
Fig. 17. Alkaline ELZ obtained from the transfer function.
that occurs in output response, and the response contains a significant
amount of steady-state error. Besides, its sliding manifold is linear and
the speed of convergence of states is slower as it approaches near equi-
Like Esmaeilian et al. [58], in this paper, the authors also use two librium than the initial position. TSMC alleviates these shortcomings
PI controllers for alkaline ELZ: one for voltage control and the other of classical SMC. Fig. 22 represents the sliding manifolds of SMC and
one for current control because of the two-state variables depicted in TSMC. TSMC-controlled DC/DC converters are robust in terms of nature
Fig. 18(a). On the flip side, as 𝐿𝑠 is neglected and electrical double
and the converging rate is also faster than the conventional SMC with
layer (EDL) is negligible concerning PEM ELZ, the outer voltage loop
a lower chattering phenomenon. On the flip side, all TSMC controller
is neglected. The gains of the PI controller can be obtained by a small
suffers from singularity problem. Global terminal sliding mode con-
signal model synthesized from Eq. ((17), (18)) [19]. The subsequent
troller [182], nonsingular terminal sliding mode controller [183–185],
figure, Fig. 19, represents the small signal model for the PI controller
integral terminal sliding mode controller [148,186,187] can overcome
to be designed. Consequently, a closed-loop transfer function 𝐼 𝐼(𝑠) (𝑠)
𝑒𝑙𝑧(𝑟𝑒𝑓 ) the limitation of TSMC.
can be expressed by employing a PI controller with parameters 𝑘𝑝𝑖 and Armghan et al. in [47] propose Quick Reaching Law (QRL) based
𝑘𝑖𝑖 as follows: global terminal sliding mode controller for wind/hydrogen/battery DC
2𝜁𝑖 𝜔𝑖 𝑠 + 𝜔2𝑖 microgrid. In this work, the authors use PEM FC/ELZ system to support
𝐼(𝑠) 𝐶𝑖 (𝑠)𝐺𝑖 (𝑠)
= = (19) the DC grid. Therefore, DC/DC non-isolated converters illustrated in
𝐼𝑒𝑙𝑧(𝑟𝑒𝑓 ) (𝑠) 1 + 𝐶𝑖 (𝑠)𝐺𝑖 (𝑠) 𝑠 + 2𝜁𝑖 𝜔𝑖 𝑠 + 𝜔2𝑖
2
Fig. 10 is used to interface the PEM equipment with the electrical grid,
Like, the design of the PI controller for the current control loop, the PI and a QRL-based TSMC is designed for the switching of those con-
controller for the voltage control loop can be designed by analysing a verters. For the dynamical equation of the DC/DC boost converter and
small signal model for the alkaline ELZ [19]. DC/DC buck converter, the authors have used the following equations.
4.1.2. Sliding mode control techniques d𝑖𝑓 𝑐 𝑅𝑑 𝑢𝑓 𝑐 − 1 𝑉𝑓 𝑐

A sliding mode controller (SMC) is a prominent nonlinear controller = 𝑖 − 𝑉𝑑𝑐 + (20)
d𝑡 𝐿𝑑 𝑓 𝑐 𝐿𝑑 𝐿𝑑
that brings the state of the system from any arbitrary position to the
d𝑖𝑒𝑙𝑧 𝑅 𝑢 𝑉
equilibrium position. Rather than using a continuous control law, SMC = − 𝑑 𝑖𝑒𝑙𝑧 + 𝑒𝑙𝑧 𝑉𝑑𝑐 − 𝑒𝑙𝑧 (21)
traditionally uses discontinuous law for sliding around the designed d𝑡 𝐿𝑑 𝐿𝑑 𝐿𝑑
sliding manifold depicted in Fig. 20 [170]. This is why SMC is called where 𝑉𝑓 𝑐 is the voltage developed across the PEM FC encompasses
the variable structure feedback control method. From the standard pro- equilibrium potential, activation voltage, ohmic voltage and concen-
cedure of designing SMC given in Fig. 21, it can be said that designing tration voltage [188]. 𝑉𝑒𝑙𝑧 is the voltage drop across the PEM ELZ.
of SMC is comparatively easier than other nonlinear controllers; besides It comprises open circuit voltage, activation polarization, and ohmic
SMC provides robustness against external disturbance. Typically, SMC polarization [152]. Followed by the dynamical modelling, Armghan
performs well in those systems rife with discontinuous switching. That et al. design the QRL-based TSMC which is presented in Fig. 23. The 𝑖∗𝑓 𝑐 ,
13
Fig. 18. PI controlled DC/DC converter for alkaline and PEM electrolyzer.
Fig. 19. Small signal modelling for the current control loop.
14
Fig. 20. Trajectory of a conventional SMC system.
Fig. 21. Standard procedure of designing SMC for a nonlinear system.
and 𝑖∗𝑒𝑙𝑧 are reference currents obtained from the high-level supervisory reduces the chattering effects. Derbeli et al. proposes a robust higher-
EMS. As it is evident from the sliding surface selected in step 2, a sliding order twisting algorithm-based sliding mode controller for performance
surface is a nonlinear sliding manifold, where 𝜂 > 0, 𝜆 > 0, 𝜃 < 0, improvement of FC [61]. The sliding manifold is chosen by the authors
and 𝜃 = 𝑞𝑝 , where both 𝑝 and 𝑞 are positive odd numbers. The quick- as follows:
reaching ability is achieved from the switching control input 𝑢𝑠 which
𝑠=𝑒+𝜆 𝑒 ⋅ dt (24)
has been defined as follows: ∫
where, 𝑒 in Eq. (24) is tracking error defined as 𝑒 = 𝑖𝑓 𝑐 − 𝑖∗𝑓 𝑐 . The
𝑢𝑠 = −𝜑𝑠𝑖𝑔𝑛(𝑆) (22)
switching control law is chosen according to [191] as the following.
( ) 𝑢 = 𝑢𝑒𝑞 + 𝑢𝑠𝑤 (25)
𝜑 = 𝐶1 𝛼 |𝑆| − 1 + 𝐶2 |𝑆|𝛽 + (𝑄 + 𝜂) (23)
𝐶 The authors consider the following certain second-order system.
where 𝐶1 &𝐶2 > 0, 𝛽 = 1 + 𝐶2 , 𝜂 > 0, 0 < 𝛼 < 1, 𝜂 > 0 and |𝜖| ≤ 𝑄. The {
1
avoidance of the singularity problem can be ensured due to the fact 𝑦̇ 1 = 𝑦2
(26)
that 𝛽 is consistently greater than 1. 𝑦̇ 2 = 𝜙(𝑥, 𝑡) + 𝛾(𝑥, 𝑡)𝜈
Twisted algorithm based higher order sliding mode controller: Higher In Eq. (26), 𝑥 is state parameter of FC denoted as 𝑥 = [𝑖𝑓 𝑐 𝑉𝑑𝑐 ], 𝑦 is the
order SMC introduced by Emeryanov et al. [189] which has a sliding output of FC voltage that is 𝑉𝑑𝑐 , 𝑦1 and 𝑦2 are the local coordinate which
set of the order of 𝑟 has the ability of finite time convergent like the represent 𝑠 and 𝑠. ̇ The 𝛾(𝑥, 𝑡) and 𝜙(𝑥, 𝑡) are smooth functions derived
TSMC [184]. It exhibits robustness and uncertainty handling capacity from Eq. (28), 𝜈 represents the first derivative of control input 𝜈 = 𝑢. ̇
with a lower chattering effect than SMC [190]. Twisting algorithm The 𝑢̇ is discontinuous control, and it drives the sliding manifold in the
15
Fig. 22. The phase portraits of conventional SMC and TSMC.
Fig. 23. QRC-SMC-based H2 FC/ELZ DC/DC converter design steps.
second-order sliding surface 𝑆 2 . The second derivative of the sliding Eq. (31) represents the overall equation.
surface of Eq. (24) can be expressed as the following.
𝑣 = 𝑣𝑒𝑞 + 𝑣𝑠𝑤 (31)
1 1 The following equation is chosen as the Lyapunov candidate function
𝑦̇ 2 = 𝑠̈ = (𝑣̇ − 𝑥̇ 2 ) + 𝜆𝑒̇ + (𝑥̇ 𝑢 − 𝑥2 𝑢)
̇ (27)
𝐿𝑑 𝐿𝑑 2 given below:
where 𝑣 is equal to the voltage of FC, 𝑉𝑓 𝑐 . Now from this equation, the 1 1
𝑉2 = 𝑘 𝑠2 + 𝑠̇ 2 (32)
following expression is derived. 2 1 2
It is well known that stability will be guaranteed if and only if the first
1 [ ] 1
𝑦̇ 2 = (𝑢 − 1)𝑥̇ 2 + 𝜆(𝑢 − 1)𝑥2 + 𝜆𝑣 + 𝑣̇ + 𝑥 ⋅𝑣 (28) derivative of the Lyapunov candidate is semi-negative definite.
𝐿𝑑 𝐿𝑑 2
⏟⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏟⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏞⏟ ⏟⏟⏟ 𝑉̇ 2 = 𝑘1 𝑠𝑠̇ + 𝑠̇ 𝑠̈
𝜙 𝛾
= 𝑘1 𝑠𝑠̇ + 𝑠(𝜙
̇ + 𝛾𝑣)
( )
The switching control law is adopted from the twisting algorithm 1 [ ] 1
= 𝑠̇ 𝑘1 𝑠 + (𝑢 − 1)𝑥̇ 2 + 𝑣̇ + 𝜆(𝑢 − 1)𝑥2 + 𝜆𝑣 + 𝑥2 𝑣 (33)
because it reduces the chattering phenomenon. Derbeli et al. adopt the 𝐿 𝐿
switching law expressed in Eq. (29). = −𝑠𝑘̇ 1 𝛽sign(𝑠)
̇
𝑘1 𝐿𝑑 = −|𝑠|𝑘
̇ 1𝛽 < 0
𝑣𝑠𝑤 = − (𝑠 + 𝛽sign(𝑠))
̇ (29)
𝑥2 According to the equation presented in (33), it can be deduced that
the stability of the entire system is ensured, given that the state of
where 𝛽 and 𝑘1 are the positive constant. Thereby, the following can
the system reaches equilibrium within a finite period. Moreover, the
be used to represent the equivalent control law [61]. utilization of the higher-order SMC twisting algorithm technique results
𝜙 1 [ ] in a significant reduction in the current oscillation, which is estimated
𝑣𝑒𝑞 = − =− (𝑢 − 1)𝑥̇ 2 + 𝑣̇ + 𝜆(𝑢 − 1)𝑥2 + 𝜆(𝑢 − 1)𝑥2 + 𝜆𝑣 (30) to be as low as 6%.
𝛾 𝑥2
16
4.1.3. Backstepping controller The first derivative of the overall Lyapunov candidate is the following
Backstepping is a recursive Lyapunov-based strategy that was put (44).
forth at the start of the 1990s [192,193]. Krstic, Kanellakopoulos, and
𝑉̇ = 𝑉̇ 1 + 𝑒2 𝑒̇ 2
Kokotovic covered the method in depth in [194]. One of the principal ( ) (45)
benefits of employing the backstepping control methodology is its = −𝑏1 𝑒21 + 𝑒2 𝑒̇ 2 − (1 − 𝑢)𝑒1
capacity to prevent the elimination of valuable nonlinear dynamics, Global asymptotic stability is achieved only if the first derivative of 𝑒2
thus enabling the attainment of both regulation and tracking proper- is chosen as:
ties [195–197]. Backstepping is the process of creating a controller
iteratively by treating some of the state variables as ‘‘virtual controls’’ 𝑒̇ 2 = −𝑏2 𝑒2 + (1 − 𝑢)𝑒1 (46)
and creating intermediary control laws for them [198–201]. As re-
where 𝑏2 is a positive constant. Finally, the control law is achieved
ported in [202,203], the recursive procedure employed in this study
by combining Eq. (39) & Eq. (46). The overall system designed by the
facilitates the development of a Lyapunov function for the overall sys-
authors is presented in Fig. 24 [62].
tem, encompassing the parameter estimations. This approach renders
the verification of the aforementioned characteristics unavoidable. 1−𝑢 { }
𝑢̇ 1 = 𝑏2 𝑒2 − (1 − 𝑢)𝑒1 + 𝛹 (47)
Derbeli et al. propose a backstepping controller for achieving 𝛾1
smooth maximum power point tracking [62]. This method achieves a
4.1.4. Model predictive controller
robust solution with Lyapunov stability that can improve the lifetime
The Model Predictive Controller (MPC) is a class of algorithms
of FC. The tracking error is defined as follows.
that computes a controlled variable profile using an explicit process
𝑒1 = 𝑖𝑓 𝑐 − 𝑖𝑀𝑃 𝑃 (34) model to optimize a predicted performance objective over a future
time interval while adhering to predetermined limitations. It can be
To track MPPT reference perfectly, 𝑒1 needs to vanish. The derivative
characterized as a model-based, predictive-based, optimized base, and
of 𝑒1 is
feedback-based control strategy [204]. In the process sector, it is one
𝑒̇ 1 = 𝑖̇ 𝑓 𝑐 − 𝑖̇ 𝑀𝑃 𝑃 of the most widely used sophisticated control strategies [205]. It first
𝑉𝑑𝑐 𝑉𝑓 𝑐 (35) started in the chemical process industry but now finds applications in a
= −(1 − 𝑢) + − 𝑖̇ 𝑀𝑃 𝑃 wide range of fields including electrical and mechanical systems [206].
𝐿𝑑 𝐿𝑑
As time passes, MPC is employed in the power system more than before
where 𝑖̇ 𝑓 𝑐 , is replaced with the state of the average of the conventional
𝑉 because of the advancement of digital microcontrollers [207,208].
non-isolated converter, 𝐿𝑑𝑐 is considered as a virtual variable. The Extensive research is going on in the domain of power electronics [209–
𝑑
Lyapunov function chosen by the authors is the following 212] pushes forth the community towards substantial use of renewable
1 2 energy [213]. There are various accounts can be found regarding micro-
𝑉1 = 𝑒 (36)
2 1 grids using MPC. The extensive research on DC/AC inverters [214–216]
First derivative of 𝑉1 is the following and DC/DC converters [217–219] designed by MPC will usher in a
( ) promising path of 100% renewable integration.
𝑉 𝑉𝑓 𝑐
𝑉̇ 1 = 𝑒1 𝑒̇ 1 = 𝑒1 −(1 − 𝑢) 𝑑𝑐 + − 𝑖̇ 𝑀𝑃 𝑃 (37) Fig. 25 sheds light on traditional aspects of MPC. From Fig. 25(a),
𝐿𝑑 𝐿𝑑
it can be deduced that MPC is a discrete-time control that requires a
𝑉
Now, Eq. (37) shows that 𝑒1 can be made zero if and only if 𝐿𝑑𝑐 = 𝛾1 , dynamic process model. The observer measures the output; calculates
𝑑
where 𝛾1 is termed as stabilization factor defined as the following. the plant states accordingly and feeds to the controller. Afterwards,
( ) it uses explicit prediction of the future based on the past input and
1 𝑉𝑓 𝑐
𝛾1 = 𝑏1 𝑒 1 + − 𝑖̇ 𝑀𝑃 𝑃 (38) measures the error from the reference shown in Fig. 25(b). Based on
1−𝑢 𝐿𝑑
the error signal it performs an online optimization and generates input
𝑉𝑑𝑐
where 𝑏 is held as a positive constant. Since, 𝐿𝑑
is a virtual variable, for the present steps. This process will be continuing until the plant
a tracking error 𝑒2 is set as the following reaches the desired reference optimally.
𝑉𝑑𝑐 In [63], Trifkovic et al. proposes MPC for the PEM ELZ/FC inte-
𝑒2 = − 𝛾1 (39) grated into a renewable energy hybrid system at Lambton College,
𝐿𝑑
Canada. The authors argue that as MPC inherently deals with system
Combining Eq. ((38) & (39)), the following expression can be obtained.
constraints, PEM ELZ/FC does not suffer severe degradation by an
𝑒̇ 1 = −𝑏1 𝑒1 − (1 − 𝑢)𝑒2 (40) abrupt change in load. The authors tailor the non-linear dynamical
model to the linear model in the following form.
Consequently, the following can be used to rewrite the Lyapunov {
function. 𝑥(𝑘 + 1) = 𝐴𝑥(𝑘) + 𝐵𝑢(𝑘) + 𝐷𝑑(𝑘)
(48)
𝑦(𝑘) = 𝐶𝑥(𝑘)
𝑉̇ 1 = −𝑏1 𝑒21 − (1 − 𝑢)𝑒1 𝑒2 (41)
where 𝑘 is the sampling time, and 𝐴, 𝐵, 𝐶 and 𝐷 are matrices with ap-
The aim of the control system is to reduce the magnitudes of the propriate dimension. 𝑥, 𝑦, 𝑢 and 𝑑 represent the model state, measured
variables 𝑒1 and 𝑒2 . Using Eq. ((38) & (43)), the following expression output, control input and disturbance accordingly. Eq. (49) & Eq. (50)
is derived. represent the state space model variables of the electrolyzer and fuel
𝑢̇ cell respectively.
𝑒̇ 2 = − 𝛾 +𝛹 (42)
1−𝑢 1
where, ⎧ 𝑥𝑒𝑙𝑧 = [𝛿𝑁O𝑎 , 𝛿𝑁H𝑎 O , 𝛿𝑁H𝑐 , 𝛿𝑁H𝑐 O ]𝑇
⎪ 2 2 2 2
( ) ( ) ⎪ 𝑦𝑒𝑙𝑧 = [𝛿𝑃𝑒𝑙𝑧 , 𝛿𝑉𝑒𝑙𝑧 , 𝛿𝑁H2 , 𝛿𝑝H2 ]𝑇
1 𝑉̇ 𝑓 𝑐 1 1−𝑢 𝑉 ⎨ (49)
𝛹= 𝑏21 𝑒1 + (1 − 𝑢)𝑏1 𝑒2 − + 𝐼̇ 𝑀𝑃 𝑃 + 𝑖𝑓 𝑐 − 𝑑𝑐 . ⎪ 𝑢𝑒𝑙𝑧 = [𝛿𝐼𝑒𝑙𝑧 ]
1−𝑢 𝐿𝑑 𝐿𝑑 𝐶 𝑅𝐶 ⎪
⎩ 𝑢𝑒𝑙𝑧 = [𝛿𝑇𝑒𝑙𝑧 ]
(43)
⎧ 𝑥𝑓 𝑐 = [𝛿𝑝𝑓 𝑐 , 𝛿𝑁H𝑎 , 𝛿𝑁H𝑎 O , 𝛿𝑁N𝑐 ]𝑇
To obtain stabilization control 𝑢 the overall Lyapunov function candi- ⎪ 2 2 2
⎪ 𝑦𝑓 𝑐 = [𝛿𝑃𝑓 𝑐 , 𝛿𝑉𝑓 𝑐 ]
date proposed by the author is following ⎨ (50)
⎪ 𝑢𝑓 𝑐 = [𝛿𝐼𝑓 𝑐 ]
1 1 1 ⎪ 𝑢𝑓 𝑐 = [𝛿𝑇𝑓𝑐 𝑐 ]
𝑉 = 𝑉1 + 𝑒22 = 𝑒21 + 𝑒22 (44) ⎩
2 2 2
17
Fig. 24. Backstepping controlled PEM FC.
Fig. 25. A traditional view of the model predictive controller.
In the context of this study, the notation 𝛿 refers to the deviation PEM electrolyzer, 7 Nm3 metal hydride hydrogen storage tank and a
from the operating point, while the superscripts 𝑎 and 𝑐 correspond to 367 Ah lead–acid battery as main components. Fig. 26(a) represents
the anode and cathode, respectively. 𝑃𝑒𝑙𝑧 and 𝑃𝑓 𝑐 represent consumed the implicit MPC diagram proposed in [64]. The main objective of
and generated power respectfully as a control variable. To reduce the the proposed design is to generate reference signals for the modified
following finite-horizon control and performance index, the authors variables to mitigate the discrepancy between the generated and de-
choose the following cost function [63]. manded power, which is a common occurrence in renewable energy
[ ]2 systems. The system is controlled through four variables, namely, fuel
⎧ ∑𝐻𝑃
⎪ 𝑘=1
𝑊 𝑦 𝑦(𝑘) − 𝑦(𝑘)𝑟𝑒𝑓 cell power production (𝑃𝑓 𝑐 ), electrolyzer power consumption (𝑃𝑒𝑙𝑧 ),
⎪ [ ]2
⎪ ∑𝐻𝐶 power exchange with the grid (𝑃𝑔𝑟𝑖𝑑 ), and battery power (𝑃𝑏 ), which
min 𝐽 (𝑥(𝑡), 𝑢(𝑡), 𝑡) = ⎨ + 𝑘=1 𝑊𝑢 𝑢(𝑘) − 𝑢(𝑘)𝑟𝑒𝑓 regulate the battery power (𝑃𝑏𝑎𝑡𝑡 ). Since the battery in the testbed is
𝑢
⎪ [ ]2 not controlled by any converter, this variable has been made to equal
⎪ ∑𝐻 (51)
⎪ + 𝐶
𝑊𝑢 𝛥𝑢𝑘|(𝑘 + 1) 0 by setting a large weight in the cost function. The controller must
⎩ 𝑘=1
adjust for the imbalance between power generated and demand using
⎧ 𝑦(𝑘) < 𝑦(𝑘) < 𝑦(𝑘) the remaining controlled variables (𝑃𝑒𝑙𝑧 , 𝑃𝑓 𝑐 , and 𝑃𝑔𝑟𝑖𝑑 ), distributing
⎪ 𝑙𝑙 𝑢𝑙
s.t. ⎨ 𝑢(𝑘)𝑙𝑙 < 𝑢(𝑘) < 𝑢(𝑘)𝑢𝑙 excess or deficit energy across the microgrid units as efficiently as
⎪ 𝛥𝑢(𝑘)𝑙𝑙 < 𝛥𝑢(𝑘) < 𝑢(𝑘)𝑢𝑙 possible. The authors prescribed that if the battery has a dedicated
⎩
power converter, it is regarded as one of the modified variables. If it
In this study, the variables 𝑊𝑦 and 𝑊𝑢 denote the input and output
is possible, the controller stores some energy near the required output
weight scalers, respectively, while 𝐻𝑃 and 𝐻𝐶 represent the prediction
references (𝑆𝑂𝐶𝑟𝑒𝑓 , 𝑀𝐻𝐿𝑟𝑒𝑓 ) during operation. Authors put forward a
and control horizons. The constraints are imposed on the upper and
hierarchy of control systems illustrated in Fig. 26(b) in which the first
lower bounds of the power of ELZ and FC, denoted by (𝑦𝑢𝑙 , 𝑦𝑙𝑙 ), the
stage determines which MPC will be in operation based on the weather
upper and lower limits of the cell current, indicated as (𝑢𝑢𝑙 , 𝑢𝑙𝑙 ), and
the rate of change in the ELZ and FC cell current, represented by (𝛥𝑢). under which the renewable sources PVs and wind farms are operating.
Moreover, the upper and lower limits (𝑦𝑢𝑙 , 𝑦𝑙𝑙 ) demarcate the operating The 𝑋𝑏𝑘 is a Boolean variable that map weather pattern. The implicit
range of the ELZ and FC power. MPC, which is in charge of the microgrid’s dependable, secure, and
Valverde et al. introduce an implicit MPC that is capable of rapidly cost-effective functioning, is located in the second layer. In the context
and efficiently calculating optimal set-points for power management. of hydrogen-integrated microgrids, the integration of fuel cells poses
Notably, the proposed MPC is formulated to consider the effects of a considerable challenge in relation to load management. Specifically,
degradation and equipment constraints. Their findings are published the dispatch command must be established at a level that not only
in [64]. In contrast to this review paper testbed, here the authors used accommodates rapid fluctuations in load but also satisfies specific life-
a battery ESS. The test bench comprises a 1.5 kW PEM fuel cell, 1 kW cycle objectives that are closely tied to load profile considerations. The
18
Fig. 26. Implicit model predictive controller.
linearized state-space modelling is chosen as the following: DC/DC converter for 7 kW fuel cell sourcing 50 Ω [65]. Unlike the
fixed step IC MPPT, in this article, a variable step IC MPPT scaled by
̇ + 1) = 𝐴𝑑 𝑥(𝑘) + 𝐵𝑑 𝑢(𝑘) + 𝐵𝑑′ 𝑤(𝑘)
𝑥(𝑘
(52) the fuzzy logic controller is designed for the fuel cell depicted in Fig. 28.
𝑦(𝑘) = 𝐶𝑑 𝑥(𝑘) The variable step size of defined as the following
where 𝑘 represents the sampling time; 𝑥(𝑘) ∈ R𝑛 is states of system,
𝑑𝑘 = 𝑑𝑘−1 + 𝑆𝐹 𝛥𝑑 (54)
𝑢(𝑘) ∈ R𝑚 is control input and 𝑦(𝑘) ∈ R𝑝 is controlled output. Where
these can be defined as the following where 𝑑𝑘 and 𝑑𝑘−1 is duty cycle at instants 𝑘 and 𝑘 − 1; 𝛥𝐷 is the fixed
| 𝑃 −𝑃 |
⎛low frequency battery current⎞ ⎛ 𝑃𝑓 𝑐 ⎞ ( ) step size. 𝑆𝐹 is the scaling factor defined as 𝑆𝐹 = || 𝑖𝑘 −𝑖 𝑘−1 ||, where
⎜ battery charging current ⎟ ⎜𝑃 ⎟ | 𝑘 𝑘−1 |
⎜ ⎟ 𝑆𝑂𝐶 𝑃𝑘 and 𝑃𝑘−1 are the output power at instants 𝑘 and 𝑘 − 1; 𝑖𝑘 and 𝑖𝑘−1
𝑥(𝑘) = ; 𝑢(𝑘) = ⎜ 𝑒𝑙𝑧 ⎟ ; 𝑦(𝑘) =
⎜ battery discharging current ⎟ ⎜𝑃𝑔𝑟𝑖𝑑 ⎟ 𝑀𝐻𝐿 are current at instants 𝑘 and 𝑘 − 1. Authors consider input as 𝑆𝐹 as
⎜ ⎟ ⎜ ⎟
⎝ hydrogen flow rate ⎠ ⎝ 𝑃𝑏𝑎𝑡𝑡 ⎠ an error crisp, 𝐸𝑘 , goes into the fuzzification process. The input and
A function 𝐽 that depends on current and upcoming control signals will output variables are both divided into five fuzzy sets. For the inference,
show how the system is approaching the desired behaviour [64]: Mamdani inference is adopted. The lookup table consists of 25 fuzzy
{ ∑𝑁 rules for mapping input with output. The Centroid method is adopted
̂ + 𝑘|𝑡) − 𝑤(𝑡 + 𝑘|𝑡)]2
𝑘=1 𝛿(𝑘) [𝑦(𝑡 for the defuzzification process for converting the fuzzy output into crisp
min 𝐽 (𝑥(𝑡), 𝑢(𝑡), 𝑡) = ∑ 𝑁𝑢
𝑢 + 𝑘=1 𝜆(𝑘) [(𝛿𝑢(𝑡 + 𝑘 − 1|𝑡))]2 output which will drive the converter. This method is able to keep its
⎧ current ripple reduction between 70.93% and 100%.
𝑃𝑒𝑧min ≤ 𝑃𝑒𝑧 ≤ 𝑃𝑒𝑧max
⎪
⎪ 𝑃𝑓 𝑐min ≤ 𝑃𝑓 𝑐 ≤ 𝑃𝑓 𝑐max
4.1.6. Feedforward neural network control
⎪ 𝑃𝑔𝑟𝑖𝑑min ≤ 𝑃𝑔𝑟𝑖𝑑 ≤ 𝑃𝑔𝑖𝑟𝑑max
⎪ An artificial neural network called a feedforward neural network
⎪ 𝑃𝑏𝑎𝑡𝑡min ≤ 𝑃𝑏𝑎𝑡𝑡 ≤ 𝑃𝑏𝑎𝑡𝑡max (53) processes data using a number of layers [227,228] where each con-
⎪ 𝛥𝑃𝑒𝑧min ≤ 𝛥𝑃𝑒𝑧 ≤ 𝛥𝑃𝑒𝑧max
s.t. ⎨ nection between layers carries a weight that indicates how much of
⎪ 𝛥𝑃𝑓 𝑐min ≤ 𝛥𝑃𝑓 𝑐 ≤ 𝛥𝑃𝑓 𝑐max
an impact it will have on the output. The input layer receives raw
⎪ 𝛥𝑃𝑔𝑟𝑖𝑑min ≤ 𝛥𝑃𝑔𝑟𝑖𝑑 ≤ 𝛥𝑃𝑔𝑟𝑖𝑑max
⎪ data from the hidden levels, which then changes it via mathematical
⎪ 𝛥𝑃𝑏𝑎𝑡𝑡min ≤ 𝛥𝑃𝑏𝑎𝑡𝑡 ≤ 𝛥𝑃𝑏𝑎𝑡𝑡max operations into something more useful for the output layer. A control
⎪ 𝑆𝑂𝐶min ≤ 𝑆𝑂𝐶 ≤ 𝑆𝑂𝐶max system based on this algorithm is denoted as a feedforward neural
⎪ 𝑀𝐻𝐿min ≤ 𝑀𝐻𝐿 ≤ 𝑀𝐻𝐿max
⎩ network [229,230]. Since it is a model-free control system that can
In this context, 𝑤 denotes the sequence of the reference trajectory, handle very complex systems, it is also popular in the converter control
while 𝜆 and 𝛿 are the weighting factors. The primary objective of the system domain [231–233].
controller is to calculate the control input 𝑢 at each sampling time in a Harrag et al. design a feedforward neural network incremental-
manner that minimizes the value of 𝐽 . based variable step size MPPT controller for regulating PEM FC with
DC/DC converter [66]. Fig. 29 shows the diagram of the control system
4.1.5. Fuzzy logic control proposed by the authors. The input variables that are selected are the
Fuzzy logic control is a heuristic approach based on fuzzy sets which PEM fuel cell’s current 𝐼𝑓 𝑐 and voltage 𝑉𝑓 𝑐 . The output layer gives duty
are classes of objects in which the transition from membership function cycle 𝑑 that drives the DC/DC converter of the system. Mean squared
to non-membership function is smooth [220–222]. It does not require error is chosen for training the data set. This proposed method is able
knowledge of the precise dynamical model of the plant. Hence, it can to suppress 100% current ripple.
handle complex, nonlinear and undefined systems efficiently [223–
226]. Fig. 27 shows a schematic of a fuzzy logic controller. Beginning 4.1.7. Genetic algorithm-based adaptive neuro-fuzzy inference system
with defining the input and output variables of the fuzzy controller, An Adaptive Neuro-fuzzy Inference System (ANFIS) is defined as a
the fuzzy rules are formulated in linguistic terms [2]. With that, the type of intelligent model-free control system that combines the learning
crisp data are converted to fuzzy data by the fuzzification process. The and adaptation capabilities of neural networks with the rule-based
Inference engine uses a lookup table that maps the input fuzzy data reasoning of fuzzy logic based on Takagi–Sugeno fuzzy inference sys-
with the output fuzzy data. Afterwards, defuzzification translates the tem [234–236]. This concept developed back in the 1990s [237].
fuzzy output to crisp output which will drive the plant. It renders advantages of both fuzzy rule-based control systems and
Harrang et al. proposes variable step fuzzy based Incremental neural networks in a single framework. The Genetic Algorithm (GA)
Conductance (IC) Maximum Power Point Tracking (MPPT) controller is a type of population-based metaheuristic algorithm that belongs to
19
Fig. 27. General control block diagram of fuzzy logic controller.
Fig. 28. Variable-sized incremental conductance steps maximum power point tracking using fuzzy scaling.
attained. This yields the optimal training data set for ANFIS which is
trained by the backpropagation (BP) algorithm. The proposed ANFIS
is depicted in Fig. 31 containing five layers. The first layer termed
the fuzzification layer converts crip pressure and temperature data into
fuzzy data by the following equation
1 1
𝑂𝐴 = 𝜇𝐴𝑖 (𝑇 ) and 𝑂𝐵 = 𝜇𝐵𝑖 (𝑃 ) (55)
𝑖 𝑖
in which 𝑃 and 𝑇 denote the input parameters for node 𝑖 (where 𝑖 takes
the values of 1, 2, 3, …), 𝑚 refers to the membership function, 𝑂𝐴 1 and
𝑖
1
𝑂𝐵 denote the membership function degrees, and 𝐴𝑖 and 𝐵𝑖 signify the
𝑖
linguistic terms of the inputs. Layer 2 denotes the multiplication layer.
1 1
𝑊𝑖 = 𝑂𝐴 × 𝑂𝐵 (56)
𝑖 𝑖
where 𝑊𝑖 is the multiplied signals and output. Layer 3 is the normaliza-

tion layer where upstream signals are normalized by 𝑊𝑖 . 𝑊𝑖 is defined
as the following
𝑊𝑖
Fig. 29. Architecture of feedforward neural network system. 𝑊𝑖 = (57)
𝑤1 + 𝑤2 + ⋯ + 𝑤𝑖
Layer 4 contains the defuzzification layer with an adaptive node. The
fuzzy data is transformed into crisp data by computing the following
the broader category of Evolutionary Algorithms (EAs). The GA draws
equation
inspiration from natural selection processes and operates by iteratively
generating a population of potential solutions that evolve and improve 𝑂𝑖4 = 𝑊 𝑖 (𝑝𝑖 𝑇 + 𝑞𝑖 𝑃 + 𝑟𝑖 ) (58)
over successive generations [238–240]. GA imitates Darwin’s notion of
evolution of survival of the fittest in nature [241]. J.H. Holland made Layer 5 is the output layer where signals are summed up and produce
the GA proposal in 1992 [242]. the 𝑉𝑟𝑒𝑓 . The overall output (𝑂5 ) is;
∑
Savrun et al. blend ANFIS control system with GA to improve 𝑂5 = 𝑊𝑖 𝑓𝑖 (59)
extraction of power from fuel cell [68]. The overall control system is 𝑘=1
structured in three steps: (1) generation of the optimal dataset by GA The authors use input parameter variation taken ranges from 313 K
for ANFIS; (2) obtaining proper membership function generated by GA —328 K for temperature and 0.8 bar —1.4 bar for pressure.
and producing reference voltage for the subsequent step; (3) with the Considering the unique characteristics of alkaline ELZ fuel cells,
reference voltage, generation of duty cycle by PI controller to realize further research and development may be required to tailor the control
MPPT. The flowchart depicted in Fig. 30 illustrates the successive methods to suit their specific requirements and response characteristics.
stages involved in the implementation of GA. The GA is initialized By exploring the similarities and differences between different fuel cell
by populating its initial population with random reference voltage types, valuable knowledge, and lessons learned from one type of fuel
values that span from 0 V to 50 V. GA accounts for the system model cell can potentially be applied or adapted to enhance the performance
and constraints, and subsequently, a corresponding fitness function is and control strategies of other fuel cell technologies. Therefore, while
20
Fig. 30. GA-ANFIS controller: sequential steps of working procedure.
Fig. 31. Architecture of the controller based on GA-ANFIS.
21
Table 2
Comparison among various DC/DC converters in the context of H2 ELZ/FC.
Reference Method Advantages Disadvantages
i. Flexible and easy i. Does not take account

[42] PI
to design the detail modelling
ii. Operates in a
small region
i. Use of LPF makes i. Cannot respond to
[58] PI
system resilient against external disturbance
high frequency changes load variance
ii. High settling time,
steady-state error,
overshoots
i. Consider the i. Fails to provide
[57] PI dynamics of ELZ robust service
against physical
constraints
ii. Able to provide ii. Steady-state error remains
fast frequency response fault clearance
iii. Ensures frequency
control ancillary service
Quick reaching i. Finite-time convergent i. Heavy computation burden
[47] law based global ii. Reduces chattering ii. Chattering is not
terminal sliding completely eradicated
mode controller iii. Robust and nonlinear
control system
Twisted algorithm i. Capability of chattering i. Complex design
[61] based higher order reduction
sliding mode controller ii. Robust to uncertainties ii. Unable to use for
and disturbances first-order systems.
iii. Finite time convergence iii. Complex stability
demonstration
i. Robust method i. Fails to provide adaptability
[62] Backstepping controller
against parametric uncertainty
ii. First convergent
i. Online optimization i. Linearizing dynamical model
[63] Model predictive controller makes the process easier can lead to instability
ii. Taking account of ii. Careful choosing of control
constraint sets increases horizon and prediction
health of ELZ/FC horizon is required
iii. Low-level heuristic approach
i. Appropriate for i. High computational burden
Implicit model
[64] real-time testing
predictive controller
ii. Reduces operation cost ii. Does not account
for optimal scheduling
for long-term
generation planning
iii. Smooth power reference
i. Robust and i. Stability is not guaranteed
[65] Fuzzy logic controller
non-linear control
ii. Stability over ii. IC yields tracking
large variations error for fast changing
operating condition
Feedforward neural i. Perform better i. It requires excessive
[66]
network controller than traditional data for training
PI controller and learning
ii. Tracking error increases
for fast changing operating
condition
Genetic algorithm-based i. Flexible operation i. Two-stage operation
[68] adaptive neuro-fuzzy and high accuracy
inference system ii. Reduction in
steady state error
iii. GA provides
optimum data set
the control methods and laws discussed in the paper primarily focus 4.2. Control techniques of DC link voltage
on PEM fuel cells, they can serve as a reference point and provide a
foundation for further investigation and adaptation for other fuel cell The proper regulation of the DC link voltage is a crucial aspect
types, including alkaline ELZ fuel cells. of inverter operation, as the voltage level of the DC link, serving as
Table 2 is the summary the subsection where all proposed methods an intermediary component between the DC source and the inverter,
for DC/DC converter for H2 integrated ELZ/FC are tabulated along with directly governs the output frequency and voltage of the inverter.
its advantages and shortcomings. Ensuring the consistency of the DC link voltage is vital for the steady
22
Fig. 32. DC link voltage control by PI controller.
and efficient functioning of the inverter. Any deviation of the DC link as illustrated in Fig. 32. To ensure stability, it is recommended that
voltage from its designated level can result in output voltage and the outer loop bandwidth be at least one-fifth that of the current loop.
frequency alterations, causing power quality concerns, such as voltage This is because the inner power stage can be considered a unity link.
sags/swells and harmonic distortion. By tuning the PI controller, as derived from Fig. 32(b), and selecting
DC link voltage control is also important for ensuring the safe and appropriate roots of the closed-loop transfer function, it is possible to
reliable operation of the inverter. If the DC link voltage exceeds its achieve both regulations of DC voltage and system stability.
maximum allowable value, it can cause damage to the components and
reduce the lifespan of the inverter. On the other hand, if the DC link
4.3. Control techniques for storage tank
voltage drops below its minimum allowable value, the inverter may
shut down to protect itself, which can result in an interruption of the
power supply. That is why the DC link voltage control plays a key role The pressure of hydrogen stored in a tank is a critical factor in
in grid-connected ELZ/FC systems. This subsection is dedicated to the determining the energy density and stability of the stored hydrogen,
review of different DC link control systems. and it must be carefully controlled to avoid potential hazards and to
Quan et al. developed a PI controller to regulate the DC link ensure that the hydrogen can be used effectively. If the pressure inside
voltage [42]. Neglecting the instantaneous power of the inductor, the a hydrogen tank exceeds its maximum allowable value, it can lead to a
authors choose 𝐶𝑑 can be expressed as follows. dangerous over-pressure situation that can cause the tank to fail or ex-
plode. On the other hand, if the pressure inside a hydrogen tank drops
2
1 d𝑢𝑑𝑐 too low, it can reduce the energy density of the stored hydrogen and
𝐶 = 𝑃𝑑 − 𝑃𝑖 = 𝑃𝑐 (60)
2 𝑑 d𝑡 make it difficult to dispense the hydrogen efficiently. Therefore, control
Within the current framework, the variable 𝑃𝑖 signifies the momentary techniques for hydrogen pressure tanks are necessary to maintain the
power output of the inverter, while 𝑃𝑑 represents the instantaneous pressure within safe and optimal ranges and to ensure that the hydro-
power generated by the Fuel Cell/Electrolyzer (FC/ELZ). Additionally, gen can be stored and used safely and efficiently. Hydrogen storage
the variable 𝑃𝐶 denotes the energy stored in the DC link capacitor, method can be categorically classified into two classes depending on
23
Fig. 33. Different types of hydrogen storage methods.
metal hydride unit to provide hydrogen to the fuel cell. To achieve

the desired outcomes of regulating the flow of H2 and maintaining
optimal temperature, the authors have employed a thermal coupling
system. This system utilizes two PI controllers, as depicted in Fig. 35,
to effectively fulfil the aforementioned objectives.
The first control loop stabilizes the temperature of H2 by taking it
from the fuel cell and feeding it back to the storage unit. Conversely,
the second loop maintains the flow rate of hydrogen to the fuel cell. H2
Fig. 34. Storage control system of hydrogen. discharge rate 𝑚𝑜𝑢𝑡 can be represented by the following function.
√ ( ) 𝛾+1
𝛾 𝛾 − 1 2 2−2𝛾
𝑚𝑜𝑢𝑡 = 𝑃 𝐴𝑡ℎ 𝑀 1+ 𝑀𝑒 (63)
its volumes, pressure and storage duration [91,243]. A classification 𝑅𝑢 𝑇 𝑒 2
of hydrogen storage methods is presented in Fig. 33. Typically control where 𝐴𝑡ℎ is area of exit valve, 𝛾 is the specific heat ratio, 𝑀𝑒 represents
techniques regarding hydrogen storage involve sensors and controllers Mach number at the valve exit, and 𝑅𝑢 refers to hydrogen gas constant.
that monitor the pressure inside the tank and adjust it as necessary to 𝑇𝑤 is the temperature of circulating water at the storage tank that is
maintain the desired pressure level. Some well-known control methods kept at 343.15 K by setting the reference 𝑇𝑤𝑠𝑒𝑡 at this temperature.
are briefly described in the following. Zhang et al. put forward a hybrid energy storage system that em-
Mendis et al. provide a control system to store hydrogen that ploys the Kalman filter methodology for the regulation of the energy
is produced and consumed by electrolyzer and fuel cell depicted in storage system [244]. The proposed system incorporates hydrogen
1
Fig. 34 [59]. A first-order transfer function, 𝑠𝜏+1 , (typically 𝜏 = 3) is energy storage as a means to mitigate long-term power fluctuations,
adopted to represent the dynamics of the compressor. while a fast storage device such as an electric double-layer capacitor
is utilized for short-term compensation. The authors adopt a protocol
𝑝𝑡 = 𝑝𝑒𝑙𝑧 − 𝑝𝑓 𝑐 (61)
outlined in [244] wherein the energy content of the stored hydrogen is
utilized, instead of its mass, to govern the flow of H2 .
𝑁H2 𝑅𝑇𝑡
𝑝𝑡 − 𝑝𝑖 = 𝑧 (62) Fig. 36 illustrates the structure of the control method proposed
𝑀H2 𝑉 in [244]. Based on the energy stored into H2 and the fuel cell’s
The pressure is calculated using Eq. (61) and (62). The equation pre- power, the Kalman filter speculates the required energy in the electrical
sented in the context pertains to the calculation of hydrogen storage network and regulates the flow of H2 accordingly.
tank parameters, where 𝑝𝑡 , 𝑝𝑒𝑙𝑧 , and 𝑝𝑓 𝑐 represent the pressures of the Huangfu et al. propose a subsection bi-objective optimization dy-
tank, electrolyzer, and fuel cell, respectively. The symbol 𝑁H2 refers to namic programming-based energy management system that solves a
the rate of delivery of hydrogen moles per second to the tank, whereas global optimal operating point that accounts for control of hydrogen
𝑝𝑖 signifies the initial pressure of the tank. The stated equation takes storage unit [70]. Fig. 37 shows the structure of the proposed control
into account various physical parameters, including the universal gas system. The objective of this system is to optimize all control parame-
constant denoted by 𝑅, the volume of the tank indicated by 𝑉 , the ters, minimizing solar power curtailment while efficiently managing the
operational temperature of the tank signified by 𝑇𝑡 , the molar mass of flow of H2 in accordance with the capacity of the H2 container. The
hydrogen represented by 𝑀H2 , and the compressibility factor denoted expression for the cumulative hydrogen storage within the hydrogen
by 𝑧. storage system is defined by:
Liu et al. have developed a control system aimed at conserving 𝑡𝑜 ( )
energy in the storage of H2 [69]. The storage system incorporates a 𝑀𝑡𝑎𝑛𝑘H = 𝑀𝑂 + 𝑓𝑒𝑙𝑧 − 𝑓𝑓 𝑐 𝑑𝑡 (64)
2 ∫𝑡1
24
Fig. 35. Block diagram of the closed-loop system of hydrogen storage based on thermal coupling.
Fig. 36. Structure of hybrid energy storage system method.
Fig. 37. Block diagram of the optimal energy management strategy with subsection bi-objective optimization dynamic programming for hybrid energy system.
25
Fig. 38. Flow chart of subsection bi-objective optimization dynamic programming method to find the optimal solution.
Table 3
Comparison among different methods of the control system of the storage system of H2 .
Reference Method Achievements Drawbacks
Transfer function i. Easy to design i. Inaccurate formulation
[59]
based control system of transfer function of
compressor may
lead instability.
ii. Less computational
burden
PI based i. Reduces heat loss i. Generalization of
[69]
thermal coupling of fuel cell fitting function
may increase
computational burden
ii. Efficiency of ii. Two PI controller
hydrogen-to-power tuning is required.
increased
[244] Kalman filter i. Can address both i. Output can vary
long term and based on noise
short time power imbalance modelling
Subsection bi-objective i. Rule based judgement i. Technical experience
[70]
optimization dynamic reduces complexity is needed priorly to
programming set rule based judgement
ii. Dynamic program obtains
global optimal point
The variable 𝑀𝑡𝑎𝑛𝑘H represents the quantity of hydrogen reserves, in the diagram in Fig. 38. A comparison among various control systems
2
while 𝑀𝑂 signifies the initial amount of reserves. The hydrogen con- of hydrogen storage is presented in Table 3.
sumption rate by FC and hydrogen production by ELZ are expressed as
the following equations. 4.4. Active and reactive power sharing
𝑁
𝑓𝑒𝑙𝑧 = 𝜂𝑒𝑙𝑧 𝑒𝑙𝑧 𝐼𝑒𝑙𝑧 (65)
2𝐹 In a power system, it is important to share both active and reactive
power effectively because imbalances in either active or reactive power
𝑁 𝑓 𝑐 𝑀H 2 can lead to stability problems and power quality issues, such as voltage
𝑓𝑓 𝑐 = 𝐼𝑓 𝑐 (66)
𝑛𝐹 sags/swells and harmonic distortion [245]. For example, if there is an
𝜂𝑒𝑙𝑧 represents the efficiency of electrolyzer. The number of fuel cells excess of reactive power in the system, it can cause the voltage to
and electrolyzer are denoted as 𝑁𝑒𝑙𝑧 and 𝑁𝑓 𝑐 . 𝑀H2 is the molar mass become unstable and increase the risk of power outages. On the other
of hydrogen, 𝑛 is the number of electrons transferred. In accordance hand, if there is a deficiency of reactive power, it can cause the voltage
with the principles of the ideal gas law, the pressure of the tank can be to become too low, which can lead to decreased power quality and
derived using the subsequent equation [70]. reduced efficiency.
Almost all kind of renewable resources that produces electricity
𝑀𝑡 𝑅𝑇
𝑃𝑡 = (67) use power inverters for transferring power to the AC grid. These grid-
𝑉𝑡 connected Inverter Based Resources (IBR) play a crucial role in sharing
The pressure of the tank is denoted as 𝑃𝑡 , while the volume of the tank active and reactive power which eventually culminates in power system
is represented by 𝑉𝑡 . In order to facilitate a more convenient depiction stability issues [246,247]. Typically, grid-connected inverters incorpo-
of the remaining hydrogen reserve within the tank, the term metal tank rate a Phase-Lock Loop (PLL) mechanism that monitors the voltage
hydrogen level is introduced as Eq. (68). magnitude and angle of the grid and injects active and reactive power
into the grid [71,185,248]. This type of IBRs known as Grid Following
𝑃𝑡
𝑀𝐻𝐿 = × 100 (68) Inverters (GFLs) shown in Fig. 39(a). However, over the course of time,
𝑃max
as a gradual increase of IBRs in power systems is observed, an array
where 𝑀𝐻𝐿 denotes metal tank hydrogen level. The variable 𝑃max of questions are being raised regarding seamless power transfer with
signifies the pressure resulting from the maximum hydrogen storage overall stability [71,106,249]. Continuous decommissioning of SGs-
capacity of the tank. The proposed optimization method is illustrated based power plants results in a substantial decrease in inertia which
26
Fig. 39. Grid connected inverter power sharing control system.
Table 4
Comparison between Grid-following control and Grid-forming control.
Criteria Grid-following control Grid-forming control
Grid status Assumes that grid is operating in Assumes it has responsibility to
normal condition form and maintain healthy grid
Basic control Supplies a predetermined quantity of Establishes grid voltage and frequency
objectives power to an operational power grid
Parameters to be Direct-quadrature (dq) vector control Control of voltage magnitude

controlled of current injected into the grid and frequency/phase
Phase-locked loop It is required It may use PLL control

requirements
Black-start Cannot black-start Can black-start a power system

capability
System inertia Reduces system overall inertia Provides virtual inertia

leading to instability
Power sharing Independent regulation of Perfectly decoupled control does
active and reactive power not exist
leads to frequency instability [195,250]. Moreover, there is no well- In addition, grid-following inverters must be able to coordinate with
defined method of how the IBRs will operate with each other regarding other inverters and energy storage devices in the power system in
power sharing. The conventional SGs have black-start capability after order to provide grid support and to ensure that the system operates
being hit by a blackout which GFLs do not have [249,251]. As a matter smoothly and reliably. This requires advanced control techniques that
of fact that these hurdles can be bolstered by a new level of inverter can manage power flow and coordinate the operation of multiple
power-sharing control paradigm called Grid Forming Inverter (GFM) devices in the system. Esmaeilian et al. designs a GFL with a closed loop
shown in Fig. 39(b). It maintains a voltage level of its own without PLL that connects H2 resources with an external grid [58] presented in
taking any reference from feedback that arbitrates the issues posed Fig. 40. If interfacing reactance can be truncated, the DC link voltage
by GFLs [252–254]. The key difference between GFLs and GFMs is can be written as the following.
presented in Table 4. 3
Since the H2 ELZ/FC system primarily acts as an energy buffer that (𝑢 𝑖 + 𝑢𝑠𝑞 𝑖𝑞 )
𝑃𝑑𝑐 = 𝑢𝑠𝑎 𝑖𝑎 + 𝑢𝑠𝑏 𝑖𝑏 + 𝑢𝑠𝑐 𝑖𝑐 = (69)
2 𝑠𝑑 𝑑
stores surpass energy in the DC grid in form of H2 and supplies power To align the grid voltage 𝑢𝑎𝑏𝑐 in line with the 𝑑 frame, the voltage
back upon deficiency in the grid. Therefore, a bidirectional inverter component in 𝑞 axis, 𝑢𝑠𝑞 has to be made zero. Thereby, the following
is needed so that power can transfer from both sides. The H2 ELZ/FC current reference signal can be generated
integrated power system is introduced as a means of facilitating power-
sharing via bidirectional inverters. Specifically, two dominant forms of 2
𝑖𝑑(𝑟𝑒𝑓 ) = 𝑃 (70)
power-sharing control, namely grid following control and grid forming 𝑢𝑠𝑑 𝑟𝑒𝑓
control, are proposed within this context. The following subsections The reactive power control is a crucial aspect of power generation
will contain these two IBRs power-sharing methods that are proposed systems, as it ensures that the output voltage from the IBR remains
for the H2 ELZ/FC integrated power system. constant at a reference value, denoted as 𝑈𝑐∗ . In this context, the
instantaneous complex power theory is employed to determine the
4.4.1. Control techniques of grid following inverter reference reactive power, which is expressed as follows.
A grid-following inverter must be able to accurately track the volt- 3
𝑄𝑟𝑒𝑓 = (−𝑢𝑠𝑑 𝑖𝑞 + 𝑢𝑠𝑞 𝑖𝑑 ) (71)
age and frequency of the power system in order to avoid causing 2
disturbances to the grid and to ensure that it operates within acceptable Now, considering, 𝑢𝑠𝑞 to be zero, the reference current to be tracked in
limits [71]. If the voltage and frequency of the inverter deviate from the 𝑞 axis can be written as the following.
those of the grid, it can cause power quality problems and potentially
2
lead to blackouts. Control techniques for grid-following inverters help 𝑖𝑞𝑟𝑒𝑓 = − 𝑄 (72)
3𝑢𝑠𝑑 𝑟𝑒𝑓
to regulate their output voltage and frequency and to maintain the
stability and reliability of the power system. Usually, feedback control The modulation signals are defined as the following equation [71].
algorithms are utilized as control techniques, which detect the grid’s 2
𝑚𝑑 = 𝑢 (73)
voltage and frequency and adapt the inverter’s output correspondingly. 𝑉𝑑𝑐 𝑡𝑑
27
Fig. 40. Control diagram of GFL inverter for H2 integration to the grid.
2 Table 5
𝑚𝑑 = 𝑢 (74)
𝑉𝑑𝑐 𝑡𝑞 Sources and tested power system types.
Sources DC Hybrid Standalone Grid connected
where, 𝑢𝑡𝑑 and 𝑢𝑡𝑞 represent terminal voltage of IBR. The authors use
resources from [255] and come up with a third-order harmonic injected [42] ✓ ✓
[58] ✓ ✓
PWM strategy.
[57] ✓ ✓
[47] ✓ ✓
[61] ✓ ✓
4.4.2. Control techniques of grid forming inverter [62] ✓ ✓
Grid-forming inverters are a class of power electronic devices that [63] ✓ ✓
are capable of producing and regulating both the voltage and frequency [64] ✓ ✓
of an electrical power system, and they are becoming increasingly uti- [65] ✓ ✓
[66] ✓ ✓
lized for integrating RESs into power grids, as documented in previous
[68] ✓ ✓
studies [71,106,256]. A GFM must be able to accurately control its [264] ✓ ✓
output voltage and frequency in order to maintain the stability and [244] ✓ ✓
reliability of the power system. For example, if the output voltage and [70] ✓ ✓
frequency of the inverter are not controlled properly, it can cause power
system instability and potentially lead to blackouts [107,257,258].
Control techniques for GFMs help to regulate their output voltage and coefficient of the controller based on this small signal model, the con-
frequency and to maintain the stability and reliability of the power trol objective can be achieved. The following equation, Eq. (75) implies
system. These control techniques typically involve feedback control al- that the active power sending depends mainly on the power angle and
gorithms that sense the voltage and frequency of the power system and reactive power flow depends on the voltage difference between two
adjust the output of the inverter accordingly [185,259]. Furthermore, buses [42].
in order to furnish grid support and ensure the smooth and dependable
𝑈𝐶 − 𝑈𝑔 𝛥𝛿 𝑈𝐶 (𝑈𝐶 − 𝑈𝑔 )
operation of the system, it is imperative that GFMs possess the ability 𝑃𝑖 + 𝑗𝑄𝑖 ≃ +𝑗 (75)
to synchronize with other inverters and energy storage devices within 𝑋 𝑋
the power system. This requires advanced control techniques that can where 𝑈𝐶 and 𝑈𝑔 are the voltage of the inverter and the grid subse-
manage power flow and coordinate the operation of multiple devices quently; 𝑋 is the reactance of the line. On the contrary, the 𝑄 − 𝑉
in the system. GFM gains momentum in the context of microgrid droop is realized through an integrator controller. The whole control
and distributed energy resources for not only the mere capability of system for the proposed control is shown in Fig. 43.
riding through unintentional islanding but also for the fact that GFM The control systems discussed in this section have been formulated,
is the way forward for the 100% renewable integration. In [247,257], evaluated, and validated by various authors across diverse power sys-
various types of control strategies regarding GFM IBRs techniques are tem configurations. These systems encompass both pure DC systems
being documented. Fig. 41 shows the classification of GFM control and hybrid systems, as well as standalone and grid-connected systems.
techniques. It turns out that, there are a few accounts that can be found Consequently, when a method is assessed outside its designated power
in the literature where GFM-based IBR is proposed in H2 integrated system, disparities in the obtained results may arise. To provide a
power system. comprehensive overview, Table 5 has been constructed, enumerating
Active power–frequency droop (𝑃 − 𝑓 ) and reactive power–voltage the different methods and the corresponding power system types in
droop (𝑄 − 𝑉 ) are one of the paths to realize GFM in a low resistance which they have been examined.
feeder [260–262]. In [42], authors propose 𝑃 − 𝑓 and 𝑄 − 𝑉 droop
control for active and reactive power sharing for H2 ELZ/FC integrated 5. Future trend
power system that emanated from [105,263]. Fig. 42(a) represents how
IBR will deliver power upon increasing or decreasing power demand This section propounds the future scope of the research conducted
based on the droop coefficient. Fig. 42(b) represents the small signal into hydrogen-powered electrical systems. Given the inherent proper-
model that is being designed in [42]. By appropriately selecting the ties of hydrogen, it may be posited that the widespread implementation
28
Fig. 41. Classification of grid forming inverter.
Fig. 42. The 𝑃 − 𝑓 droop control.
Fig. 43. Control diagram of GFM inverter for H2 integration to the grid.
29
Fig. 44. Map of future scopes of control system hydrogen-based power system.
of hydrogen-based ESSs holds the potential to guarantee the establish- TSMC, nested hierarchical TSMC, and adaptive TSMC can be
ment of forthcoming renewable sources. Therefore, continuous active employed which can significantly reduce chattering problems
research should be done in this area. Insofar as this paper covers the and deal with singularity issues. Besides, single-integral and
control system, some future scopes are given that categorically fall double-integral SMC-based hydrogen systems can be proposed
under this paper in the following points. due to their simplicity. Unlike the backstepping controller, the
adaptive backstepping controller can handle parametric uncer-
1. The control system of the DC/DC converter can be improved tainty remarkably. So, backstepping can be used for driving
further by designing a more efficient and robust controller. those DC/DC converters also. As far as the authors’ knowledge
Heuristic tuning methods such as Ziegler–Nichols, Cohen–Coon, robust MPC has not yet been proposed for the DC/DC converter
Åström–Hägglund and Tyreus Luyben-based PI controllers effi- for hydrogen-integrated power systems. Constraint-tightening
cacy can be examined. Advanced TSMCs such as higher-order
30
MPC, multi-stage MPC, and tube-enhanced multi-stage MPC are successful operation of a hydrogen-based power system, as highlighted
some of the robust MPC schemes that can be implemented in this review through the presentation of PEM FC and ELZ dynamical
for improving results. There is plentiful research scope to im- models. The review of DC/DC converter control systems, power-sharing
prove results by designing an advanced neural network. Con- strategies, DC link voltage regulation, and tank storage control provides
volution Neural Network (CNN), Reinforcement Learning (RL), valuable insights for individual researchers and engineers seeking to
Deep reinforcement learning (DRL), Long Short Term Mem- improve the technological advancement of control system issues for H2
ory (LSTM), Support Vector Machine (SVM), least-squares SVM, ESS power systems. The works discussed in this review paper provide
Dense Neural Network (DNN), CNN-LSTM, Nonlinear Auto Re- an organized and structured summary of the state of the art in this field,
gressive Moving Average (NARMA-L2) neural controller, au- as well as predictions for the future and general guidance for additional
toregressive with extra input controller, iterative learning con- study and advancement. Continued research and development in this
troller, active disturbance rejection controller, etc. are yet to area will be critical in enabling the widespread adoption of renewable
be proposed in this study. Learning-based algorithms such as energy sources and achieving a more sustainable future. This organized
extremum-seeking control can be exploited. Various optimiza- and structured review will benefit individual researchers and engineers
tion techniques such as grey wolf, particle swarm, cellular evo- in understanding the technological advancement with regard to control
lutionary algorithm, spiral optimization algorithm, grammatical system issues for H2 ESS power systems.
evolution, sparse Bayesian learning, Bayesian nonparametric ap-
proach, Markov chain Monte Carlo, etc. can be applied to tuning Declaration of competing interest
the control parameters of the aforementioned control system.
2. DC link generates the reference signal to be tracked by DC/DC The authors declare that they have no known competing finan-
or DC/AC converter. So, an immense focus should be cast on cial interests or personal relationships that could have appeared to
future research dedicating control of the DC link control system. influence the work reported in this paper.
Since GFMs are gaining favour among researchers for future
power systems, GFMs control systems such as improved 𝑃 ∕𝑓 Data availability
and 𝑄∕𝑉 droop control, virtual synchronous machine, matching
control, etc. can be studied by regulating DC link parameters. Data will be made available on request.
Parallaley, a stratum of this research pathway should focus
on exploring nonlinear, robust, adaptive, intelligent controllers References
that will effectively regulate DC link parameters. Additionally,
the DC link acts as a bridge between the DC and AC buses. [1] K. Alanne, S. Cao, An overview of the concept and technology of ubiqui-
Therefore, various protection devices such as static synchronous tous energy, Appl. Energy 238 (2019) 284–302, http://dx.doi.org/10.1016/j.
apenergy.2019.01.100, URL https://www.sciencedirect.com/science/article/pii/
compensators, flywheel ESSs, supercapacitors, superconducting S0306261919301011.
magnetic ESSs, superconducting fault current limiters, etc. can [2] M.R. Islam, J. Hasan, M.R.R. Shipon, M.A.H. Sadi, A. Abuhussein, T.K. Roy,
be applied to protect the fuel cell and electrolyzer stacks from Neuro fuzzy logic controlled parallel resonance type fault current limiter to
being damaged. improve the fault ride through capability of DFIG based wind farm, IEEE Access
8 (2020) 115314–115334, http://dx.doi.org/10.1109/ACCESS.2020.3000462.
3. Though the storage control system falls under the mechanical
[3] N. Abas, A. Kalair, N. Khan, Review of fossil fuels and future energy tech-
domain, it pivotally determines the speed of the whole oper- nologies, Futures 69 (2015) 31–49, http://dx.doi.org/10.1016/j.futures.2015.
ation. Therefore extensive future on this should be conducted 03.003.
so that the conventional system can be improved significantly. [4] Y. Tian, Grid-Connected Energy Storage Systems: Benefits, Planning and
The fluid dynamics of hydrogen in the pipeline and storage tank Operation, Michigan State University, 2018, http://dx.doi.org/10.25335/
M5SQ8QN4M.
should be studied in detail. Other than that of the open-loop [5] A.L. Dicks, D.A. Rand, Fuel Cell Systems Explained, John Wiley & Sons, 2018.
control system, reduced order modelling, optimal control and [6] H. Heidari, S.T. Katircioğlu, L. Saeidpour, Economic growth, CO2 emissions, and
closed-loop control systems can be explored to avoid starvation. energy consumption in the five ASEAN countries, Int. J. Electr. Power Energy
4. Active and reactive power-sharing is the utmost important thing Syst. 64 (2015) 785–791, http://dx.doi.org/10.1016/j.ijepes.2014.07.081.
[7] M.O. Faruque, M.A.J. Rabby, M.A. Hossain, M.R. Islam, M.M.U. Rashid, S.
for the overall system. The way fuel cell stacks supply loads
Muyeen, A comparative analysis to forecast carbon dioxide emissions, Energy
and consumes electricity by electrolyzer defines the overall sys- Rep. 8 (2022) 8046–8060, http://dx.doi.org/10.1016/j.egyr.2022.06.025.
tem stability. So, the reliability and resiliency of the proposed [8] Renewables, Tech. Rep., International Energy Agency, 2022, URL https://www.
method should be further investigated. Since mass-scale hydro- iea.org/fuels-and-technologies/renewables.
gen power systems are not deployed at the national level, it is [9] M.R. Islam, M.A. Hossain, J. Hasan, T.K. Roy, M.A.H. Sadi, Double integral
sliding mode controller based bridge-type flux-coupling non-superconducting
high time to set standards for increasing the pace of achieving fault current limiter to protect DFIG-based multi-machine power system under
renewable goals. The documented methods of all controllers transient-state, Int. J. Electr. Power Energy Syst. 142 (PA) (2022) 108271,
should be deployed in a weak and medium weak grid so that http://dx.doi.org/10.1016/j.ijepes.2022.108271.
overall stability can be studied. Subsequently, the revealed lim- [10] B. Steffen, A. Patt, A historical turning point? Early evidence on how the
Russia-Ukraine war changes public support for clean energy policies, Energy
itation has to be overcome with new solutions.
Res. Soc. Sci. 91 (2022) 102758, http://dx.doi.org/10.1016/j.erss.2022.102758,
A map is drawn to illustrate the future scope in an organized manner URL https://www.sciencedirect.com/science/article/pii/S2214629622002614.
[11] M. Umar, Y. Riaz, I. Yousaf, Impact of Russian-Ukraine war on clean
in Fig. 44. This will help the researchers to carry out research in an energy, conventional energy, and metal markets: Evidence from event
efficient way and helps to extend this discourse effectively. study approach, Resour. Policy 79 (2022) 102966, http://dx.doi.org/10.1016/
j.resourpol.2022.102966, URL https://www.sciencedirect.com/science/article/
pii/S0301420722004093.
6. Conclusion
[12] J. Hasan, M.R. Islam, M.R. Islam, A.Z. Kouzani, M.A.P. Mahmud, A capacitive
bridge-type superconducting fault current limiter to improve the transient
The development of a hydrogen energy storage system represents a performance of DFIG/PV/SG-Based hybrid power system, IEEE Trans. Appl.
crucial step towards achieving 100% renewable energy systems while Supercond. 31 (8) (2021) 1–5, http://dx.doi.org/10.1109/TASC.2021.3094422.
ensuring power system reliability and stability. The superiority of H2 [13] E. Kötter, L. Schneider, F. Sehnke, K. Ohnmeiss, R. Schröer, The future
electric power system: Impact of Power-to-Gas by interacting with other
ESS over other energy storage systems has been justified, underscoring renewable energy components, J. Energy Storage 5 (2016) 113–119, http:
the importance of further research and development in this area. The //dx.doi.org/10.1016/j.est.2015.11.012, URL https://www.sciencedirect.com/
dynamical modelling of fuel cells and electrolyzers is also critical to the science/article/pii/S2352152X15300347.
31
[14] G. Pleßmann, M. Erdmann, M. Hlusiak, C. Breyer, Global energy stor- [34] V. Oldenbroek, L.A. Verhoef, A.J.M. van Wijk, Fuel cell electric vehicle as a
age demand for a 100% renewable electricity supply, Energy Procedia 46 power plant: Fully renewable integrated transport and energy system design
(2014) 22–31, http://dx.doi.org/10.1016/j.egypro.2014.01.154, URL https:// and analysis for smart city areas, Int. J. Hydrogen Energy 42 (12) (2017)
www.sciencedirect.com/science/article/pii/S1876610214001702. 8166–8196, http://dx.doi.org/10.1016/j.ijhydene.2017.01.155, URL https://
[15] P. Colbertaldo, S.B. Agustin, S. Campanari, J. Brouwer, Impact of hydrogen www.sciencedirect.com/science/article/pii/S036031991730321X.
energy storage on California electric power system: Towards 100% renewable [35] B. Lyseng, T. Niet, J. English, V. Keller, K. Palmer-Wilson, B. Robertson, A.
electricity, Int. J. Hydrogen Energy 44 (19) (2019) 9558–9576, http://dx. Rowe, P. Wild, System-level Power-to-Gas energy storage for high penetrations
doi.org/10.1016/j.ijhydene.2018.11.062, URL https://www.sciencedirect.com/ of variable renewables, Int. J. Hydrogen Energy 43 (4) (2018) 1966–1979, http:
science/article/pii/S0360319918336449. //dx.doi.org/10.1016/j.ijhydene.2017.11.162, URL https://www.sciencedirect.
[16] X. Liu, M. Shahidehpour, Z. Li, X. Liu, Y. Cao, Z. Bie, Microgrids for enhancing com/science/article/pii/S0360319917345809.
the power grid resilience in extreme conditions, IEEE Trans. Smart Grid 8 (2) [36] J.O. Bockris, The hydrogen economy: Its history, Int. J. Hydrogen Energy 38
(2017) 589–597, http://dx.doi.org/10.1109/TSG.2016.2579999. (6) (2013) 2579–2588, http://dx.doi.org/10.1016/j.ijhydene.2012.12.026, URL
[17] K.S. Ratnam, K. Palanisamy, G. Yang, Future low-inertia power systems: https://www.sciencedirect.com/science/article/pii/S036031991202664X.
Requirements, issues, and solutions - A review, Renew. Sustain. Energy Rev. [37] S.Z. Hassan, S. Mumtaz, T. Kamal, L. Khan, Performance of grid-integrated
124 (2020) 109773, http://dx.doi.org/10.1016/j.rser.2020.109773, URL https: photovoltaic/fuel cell/ electrolyzer/battery hybrid power system, in: 2015
//www.sciencedirect.com/science/article/pii/S1364032120300691. Power Generation System and Renewable Energy Technologies, PGSRET, 2015,
[18] M.R. Islam, M.M. Hasan, F.R. Badal, S.K. Das, S.K. Ghosh, A blended improved pp. 1–8, http://dx.doi.org/10.1109/PGSRET.2015.7312249.
H5 topology with ILQG controller to augment the performance of microgrid [38] Y. Li, K. Liu, X. Liao, S. Zhu, Q. Huai, A virtual impedance control strategy for
system for grid-connected operations, IEEE Access 8 (2020) 69639–69660, improving the stability and dynamic performance of VSC-HVDC operation in
http://dx.doi.org/10.1109/ACCESS.2020.2986213. bidirectional power flow mode, Appl. Sci. (Switzerland) 9 (15) (2019) 16–19,
[19] F. Díaz-gonzález, Energy Storage in Power Systems, John Wiley & Sons, Ltd, http://dx.doi.org/10.3390/app9153184.
Chichester, UK, 2016, pp. 143–162, URL https://onlinelibrary.wiley.com/doi/ [39] M. Nasser, H. Hassan, Assessment of standalone streetlighting energy storage
10.1002/9781118971291.ch5. systems based on hydrogen of hybrid PV/electrolyzer/fuel cell/desalination and
[20] G. Rostirolla, L. Grange, T. Minh-Thuyen, P. Stolf, J.M. Pierson, G. Da Costa, PV/batteries, J. Energy Storage 63 (2023) 106985, http://dx.doi.org/10.1016/
G. Baudic, M. Haddad, A. Kassab, J.M. Nicod, L. Philippe, V. Rehn-Sonigo, j.est.2023.106985.
R. Roche, B. Celik, S. Caux, J. Lecuivre, A survey of challenges and solutions [40] C.J. Hatziadoniu, A.A. Lobo, F. Pourboghrat, M. Daneshdoost, A simplified
for the integration of renewable energy in datacenters, Renew. Sustain. Energy dynamic model of grid-connected fuel-cell generators, IEEE Trans. Power Deliv.
Rev. 155 (September 2021) (2022) 111787, http://dx.doi.org/10.1016/j.rser. 17 (2) (2002) 467–473, http://dx.doi.org/10.1109/61.997919.
2021.111787.
[41] S.M. Mohiuddin, M.A. Mahmud, A.M. Haruni, H.R. Pota, Design and imple-
[21] M. Faisal, M.A. Hannan, S. Member, F. Blaabjerg, Review of energy storage sys- mentation of partial feedback linearizing controller for grid-connected fuel
tem technologies in microgrid applications : Issues and challenges, IEEE Access cell systems, Int. J. Electr. Power Energy Syst. 93 (2017) 414–425, http:
6 (2018) 35143–35164, http://dx.doi.org/10.1109/ACCESS.2018.2841407. //dx.doi.org/10.1016/j.ijepes.2017.06.015.
[22] X. Xing, J. Lin, Y. Song, Y. Zhou, S. Mu, Q. Hu, Modeling and operation of
[42] X. Quan, Q. Hu, X. Dou, Z. Wu, L. Zhu, W. Li, Control of grid-forming
the Power-to-Gas system for renewables integration : A review, CSEE J. Power
application for fuel cell/electrolyser system, IET Renew. Power Gener. 14 (17)
Energy Syst. 4 (2) (2018) 168–178, http://dx.doi.org/10.17775/CSEEJPES.
(2020) 3368–3374, http://dx.doi.org/10.1049/iet-rpg.2020.0508.
2018.00260.
[43] C. Wang, S. Member, M.H. Nehrir, S. Member, S.R. Shaw, Dynamic Models
[23] D.S. Falcão, A.M. Pinto, A review on PEM electrolyzer modelling: Guidelines
and Model Validation for PEM Fuel Cells Using Electrical Circuits, IEEE Trans.
for beginners, J. Clean. Prod. 261 (2020) http://dx.doi.org/10.1016/j.jclepro.
Energy Convers. 20 (2) (2005) 442–451, http://dx.doi.org/10.1109/TEC.2004.
2020.121184.
842357.
[24] X. Chen, J. Xu, C. Yang, Y. Fang, W. Li, Y. Zhang, Z. Wan, X. Wang,
[44] N. Sazali, W.N. Wan Salleh, A.S. Jamaludin, M.N. Mhd Razali, New perspectives
Thermodynamic and economic study of PEMFC stack considering degradation
on fuel cell technology: A brief review, Membranes 10 (5) (2020) http://
characteristic, Energy Convers. Manage. 235 (January) (2021) 114016, http:
dx.doi.org/10.3390/membranes10050099, URL https://www.mdpi.com/2077-
//dx.doi.org/10.1016/j.enconman.2021.114016.
0375/10/5/99.
[25] D. Apostolou, Optimisation of a hydrogen production–storage–re-powering
[45] S. Mekhilef, R. Saidur, A. Safari, Comparative study of different fuel cell
system participating in electricity and transportation markets. A case study
technologies, Renew. Sustain. Energy Rev. 16 (1) (2012) 981–989, http:
for Denmark, Appl. Energy 265 (2020) 114800, http://dx.doi.org/10.1016/j.
//dx.doi.org/10.1016/j.rser.2011.09.020, URL https://www.sciencedirect.com/
apenergy.2020.114800.
science/article/pii/S1364032111004709.
[26] H. Blanco, A. Faaij, A review at the role of storage in energy systems with
a focus on Power to Gas and long-term storage, Renew. Sustain. Energy [46] D. Yu, Y. Mao, B. Gu, S. Nojavan, K. Jermsittiparsert, M. Nasseri, A new LQG
Rev. 81 (2018) 1049–1086, http://dx.doi.org/10.1016/j.rser.2017.07.062, URL optimal control strategy applied on a hybrid wind turbine/solid oxide fuel cell/
https://www.sciencedirect.com/science/article/pii/S1364032117311310. in the presence of the interval uncertainties, Sustain. Energy Grids Netw. 21
(2020) 100296, http://dx.doi.org/10.1016/j.segan.2019.100296.
[27] I. Vincent, D. Bessarabov, Low cost hydrogen production by anion exchange
membrane electrolysis: A review, Renew. Sustain. Energy Rev. 81 (2018) [47] H. Armghan, M. Yang, N. Ali, A. Armghan, A. Alanazi, Quick reaching law
1690–1704, http://dx.doi.org/10.1016/j.rser.2017.05.258, URL https://www. based global terminal sliding mode control for wind/hydrogen/battery DC
sciencedirect.com/science/article/pii/S1364032117309127. microgrid, Appl. Energy 316 (March) (2022) 119050, http://dx.doi.org/10.
[28] J. Liu, X. Duan, Z. Yuan, Q. Liu, Q. Tang, Experimental study on the 1016/j.apenergy.2022.119050.
performance, combustion and emission characteristics of a high compression [48] T. Zhou, B. Francois, Modeling and control design of hydrogen production
ratio heavy-duty spark-ignition engine fuelled with liquefied methane gas and process for an active hydrogen/wind hybrid power system, Int. J. Hydrogen
hydrogen blend, Appl. Therm. Eng. 124 (2017) 585–594, http://dx.doi.org/ Energy 34 (1) (2009) 21–30, http://dx.doi.org/10.1016/j.ijhydene.2008.10.030.
10.1016/j.applthermaleng.2017.06.067, URL https://www.sciencedirect.com/ [49] R. Rodriguez, G. Osma, D. Bouquain, J. Solano, G. Ordoñez, R. Roche, D.
science/article/pii/S1359431117320963. Paire, D. Hissel, Sizing of a fuel cell–battery backup system for a university
[29] D. Parra, X. Zhang, C. Bauer, M.K. Patel, An integrated techno-economic and building based on the probability of the power outages length, Energy Rep.
life cycle environmental assessment of Power-to-Gas systems, Appl. Energy 8 (2022) 708–722, http://dx.doi.org/10.1016/j.egyr.2022.07.108, URL https:
193 (2017) 440–454, http://dx.doi.org/10.1016/j.apenergy.2017.02.063, URL //www.sciencedirect.com/science/article/pii/S2352484722013774.
https://www.sciencedirect.com/science/article/pii/S0306261917302064. [50] G. Sukumara, A. Parthasarathy, V. Shankar, Fuel cell based uninterrupted
[30] Safe Use of Hydrogen | Department of Energy, 2023, URL https://www.energy. power sources, in: Proceedings of Second International Conference on Power
gov/eere/fuelcells/safe-use-hydrogen. Electronics and Drive Systems, Vol. 2, 1997, pp. 728–733, http://dx.doi.org/
[31] S.B. Walker, D. van Lanen, M. Fowler, U. Mukherjee, Economic analysis with 10.1109/PEDS.1997.627449.
respect to Power-to-Gas energy storage with consideration of various market [51] F.J. Folgado, I. González, A.J. Calderón, Simulation platform for the assessment
mechanisms, Int. J. Hydrogen Energy 41 (19) (2016) 7754–7765, http://dx. of PEM electrolyzer models oriented to implement digital Replicas, Energy
doi.org/10.1016/j.ijhydene.2015.12.214, URL https://www.sciencedirect.com/ Convers. Manage. 267 (June) (2022) http://dx.doi.org/10.1016/j.enconman.
science/article/pii/S0360319916308771. 2022.115917.
[32] P. Olivier, C. Bourasseau, P.B. Bouamama, Low-temperature electrolysis system [52] J. Koponen, V. Ruuskanen, A. Kosonen, M. Niemelä, J. Ahola, Effect of
modelling: A review, Renew. Sustain. Energy Rev. 78 (2017) 280–300, http: converter topology on the specific energy consumption of alkaline water
//dx.doi.org/10.1016/j.rser.2017.03.099, URL https://www.sciencedirect.com/ electrolyzers, IEEE Trans. Power Electron. 34 (7) (2019) 6171–6182, http:
science/article/pii/S136403211730432X. //dx.doi.org/10.1109/TPEL.2018.2876636.
[33] B. Simonis, M. Newborough, Sizing and operating Power-to-Gas systems to [53] A. Ursúa, I. San Martín, E.L. Barrios, P. Sanchis, Stand-alone operation of
absorb excess renewable electricity, Int. J. Hydrogen Energy 42 (34) (2017) an alkaline water electrolyser fed by wind and photovoltaic systems, Int. J.
21635–21647, http://dx.doi.org/10.1016/j.ijhydene.2017.07.121, URL https:// Hydrogen Energy 38 (35) (2013) 14952–14967, http://dx.doi.org/10.1016/j.
www.sciencedirect.com/science/article/pii/S0360319917329506. ijhydene.2013.09.085.
32
[54] J. Milewski, G. Guandalini, S. Campanari, Modeling an alkaline electrolysis [75] M. Genovese, P. Fragiacomo, Hydrogen refueling station: Overview of the
cell through reduced-order and loss-estimate approaches, J. Power Sources 269 technological status and research enhancement, J. Energy Storage 61 (2023)
(2014) 203–211, http://dx.doi.org/10.1016/j.jpowsour.2014.06.138. 106758, http://dx.doi.org/10.1016/j.est.2023.106758.
[55] C. Huang, Y. Zong, S. You, C. Træholt, Analytical modeling and control of [76] G. Gahleitner, Hydrogen from renewable electricity: An international review of
grid-scale alkaline electrolyzer plant for frequency support in wind-dominated Power-to-Gas pilot plants for stationary applications, Int. J. Hydrogen Energy
electricity-hydrogen systems, IEEE Trans. Sustain. Energy 14 (1) (2023) 38 (5) (2013) 2039–2061, http://dx.doi.org/10.1016/j.ijhydene.2012.12.010.
217–232, http://dx.doi.org/10.1109/TSTE.2022.3208361. [77] S. Dutta, A review on production, storage of hydrogen and its utilization
[56] M.G. Dozein, A. Jalali, P. Mancarella, Fast frequency response from utility-scale as an energy resource, J. Ind. Eng. Chem. 20 (4) (2014) 1148–1156, http:
hydrogen electrolyzers, IEEE Trans. Sustain. Energy 12 (3) (2021) 1707–1717, //dx.doi.org/10.1016/j.jiec.2013.07.037, URL https://www.sciencedirect.com/
http://dx.doi.org/10.1109/TSTE.2021.3063245. science/article/pii/S1226086X13003481.
[57] M.G. Dozein, A.M.D. Corato, S. Member, Virtual inertia response and frequency [78] H.P. Veluswamy, R. Kumar, P. Linga, Hydrogen storage in clathrate hydrates:
control ancillary services from hydrogen electrolyzers, IEEE Trans. Power Syst. Current state of the art and future directions, Appl. Energy 122 (2014) 112–132,
(2022) 1–12, http://dx.doi.org/10.1109/TPWRS.2022.3181853. http://dx.doi.org/10.1016/j.apenergy.2014.01.063.
[58] H.R. Esmaeilian, R. Fadaeinedjad, Resolving power quality issues raised [79] M. Bailera, P. Lisbona, L.M. Romeo, S. Espatolero, Power to Gas projects review:
by aerodynamic aspects of wind turbine in isolated microgrids using fuel Lab, pilot and demo plants for storing renewable energy and CO2, Renew.
cell/electrolyzer system, IEEE Trans. Sustain. Energy 7 (3) (2016) 1274–1283, Sustain. Energy Rev. 69 (November 2016) (2017) 292–312, http://dx.doi.org/
http://dx.doi.org/10.1109/TSTE.2016.2544102. 10.1016/j.rser.2016.11.130.
[59] N. Mendis, K.M. Muttaqi, S. Perera, S. Kamalasadan, An effective power [80] F.M. Sapountzi, J.M. Gracia, C.J.J. Weststrate, H.O. Fredriksson, J.W. Nie-
management strategy for a wind-diesel-hydrogen-Based Remote Area power mantsverdriet, Electrocatalysts for the generation of hydrogen, oxygen and
supply system to meet fluctuating demands under generation uncertainty, IEEE synthesis gas, Prog. Energy Combust. Sci. 58 (2017) 1–35, http://dx.doi.org/
Trans. Ind. Appl. 51 (2) (2015) 1228–1238, http://dx.doi.org/10.1109/TIA. 10.1016/j.pecs.2016.09.001.
2014.2356013. [81] V. Eveloy, T. Gebreegziabher, A review of projected Power-to-Gas deploy-
[60] M.R. Islam, J. Hasan, M.R. Islam, A.Z. Kouzani, M.A.P. Mahmud, Transient ment scenarios, Energies 11 (7) (2018) 1824, http://dx.doi.org/10.3390/
performance augmentation of DFIG based wind farms by nonlinear control en11071824, URL http://www.mdpi.com/1996-1073/11/7/1824.
of flux-coupling-type superconducting fault current limiter, IEEE Trans. Appl. [82] A.R. Dahiru, A. Vuokila, M. Huuhtanen, Recent development in Power-to-X: Part
Supercond. 31 (8) (2021) 1–5, http://dx.doi.org/10.1109/TASC.2021.3091061. I-A review on techno-economic analysis, J. Energy Storage 56 (2022) 105861,
[61] M. Derbeli, O. Barambones, M. Farhat, J.A. Ramos-Hernanz, L. Sbita, Robust http://dx.doi.org/10.1016/j.est.2022.105861.
high order sliding mode control for performance improvement of PEM fuel [83] J.O. Abe, A.P. Popoola, E. Ajenifuja, O.M. Popoola, Hydrogen energy, economy
cell power systems, Int. J. Hydrogen Energy 45 (53) (2020) 29222–29234, and storage: Review and recommendation, Int. J. Hydrogen Energy 44 (29)
http://dx.doi.org/10.1016/j.ijhydene.2020.07.172. (2019) 15072–15086, http://dx.doi.org/10.1016/j.ijhydene.2019.04.068.
[84] D. Tang, G.-L. Tan, G.-W. Li, J.-G. Liang, S.M. Ahmad, A. Bahadur, M. Hu-
[62] M. Derbeli, O. Barambones, L. Sbita, A robust maximum power point tracking
mayun, H. Ullah, A. Khan, M. Bououdina, State-of-the-art hydrogen generation
control method for a PEM fuel cell power system, Appl. Sci. 8 (12) (2018)
techniques and storage methods: A critical review, J. Energy Storage 64 (2023)
2449, http://dx.doi.org/10.3390/app8122449.
107196, http://dx.doi.org/10.1016/j.est.2023.107196.
[63] M. Trifkovic, M. Sheikhzadeh, K. Nigim, P. Daoutidis, Modeling and control of
[85] D. Parra, L. Valverde, F.J. Pino, M.K. Patel, A review on the role, cost and value
a renewable hybrid energy system with hydrogen storage, IEEE Trans. Control
of hydrogen energy systems for deep decarbonisation, Renew. Sustain. Energy
Syst. Technol. 22 (1) (2014) 169–179, http://dx.doi.org/10.1109/TCST.2013.
Rev. 101 (2019) 279–294, http://dx.doi.org/10.1016/j.rser.2018.11.010.
2248156.
[86] R. Moradi, K.M. Groth, Hydrogen storage and delivery: Review of the state of
[64] L. Valverde, C. Bordons, F. Rosa, Integration of fuel cell technologies in
the art technologies and risk and reliability analysis, Int. J. Hydrogen Energy 44
renewable-energy-based microgrids optimizing operational costs and durability,
(23) (2019) 12254–12269, http://dx.doi.org/10.1016/j.ijhydene.2019.03.041.
IEEE Trans. Ind. Electron. 63 (1) (2016) 167–177, http://dx.doi.org/10.1109/
[87] S.R. Thiyagarajan, H. Emadi, A. Hussain, P. Patange, M. Watson, A compre-
TIE.2015.2465355, URL https://ieeexplore.ieee.org/document/7181662/.
hensive review of the mechanisms and efficiency of underground hydrogen
[65] A. Harrag, S. Messalti, How fuzzy logic can improve PEM fuel cell MPPT
storage, J. Energy Storage 51 (2022) 104490, http://dx.doi.org/10.1016/j.est.
performances? Int. J. Hydrogen Energy 43 (1) (2018) 537–550, http://dx.doi.
2022.104490.
org/10.1016/j.ijhydene.2017.04.093.
[88] F. Dawood, M. Anda, G.M. Shafiullah, Hydrogen production for energy: An
[66] A. Harrag, H. Bahri, Novel neural network IC-based variable step size fuel
overview, Int. J. Hydrogen Energy 45 (7) (2020) 3847–3869, http://dx.doi.
cell MPPT controller: Performance, efficiency and lifetime improvement, Int.
org/10.1016/j.ijhydene.2019.12.059.
J. Hydrogen Energy 42 (5) (2017) 3549–3563, http://dx.doi.org/10.1016/j.
[89] D. Guilbert, G. Vitale, Hydrogen as a clean and sustainable energy vector for
ijhydene.2016.12.079.
global transition from fossil-based to zero-carbon, Clean Technol. 3 (4) (2021)
[67] M.R. Islam, J. Hasan, M.M. Hasan, M.N. Huda, M.A. Hossain Sadi, A. AbuHus-
881–909, http://dx.doi.org/10.3390/cleantechnol3040051.
sein, Performance improvement of DFIG-based wind farms using NARMA-L2 [90] M.M. Bertsiou, E. Baltas, Power to hydrogen and power to water using wind
controlled bridge-type flux coupling non-superconducting fault current limiter, energy, Wind 2 (2) (2022) 305–324, http://dx.doi.org/10.3390/wind2020017.
IET Gener. Transm. Distrib. 14 (26) (2020) 6580–6593, http://dx.doi.org/10. [91] B.E. Lebrouhi, J.J. Djoupo, B. Lamrani, K. Benabdelaziz, T. Kousksou, Global
1049/iet-gtd.2019.1917. hydrogen development - A technological and geopolitical overview, Int. J.
[68] M.M. Savrun, M. İnci, Adaptive neuro-fuzzy inference system combined with Hydrogen Energy 47 (11) (2022) 7016–7048, http://dx.doi.org/10.1016/j.
genetic algorithm to improve power extraction capability in fuel cell applica- ijhydene.2021.12.076.
tions, J. Clean. Prod. 299 (2021) 126944, http://dx.doi.org/10.1016/j.jclepro. [92] A. Razmjoo, S. Mirjalili, M. Aliehyaei, P.A. Østergaard, A. Ahmadi, M. Majidi
2021.126944. Nezhad, Development of smart energy systems for communities: Technologies,
[69] J. Liu, F. Yang, Z. Wu, Z. Zhang, Energy saving control and operation strategy policies and applications, Energy 248 (2022) 123540, http://dx.doi.org/10.
analysis of thermal coupling system of fuel cell and metal hydride tank, Int. J. 1016/j.energy.2022.123540.
Hydrogen Energy (xxxx) (2023) http://dx.doi.org/10.1016/j.ijhydene.2023.03. [93] L. Pingkuo, H. Xue, Comparative analysis on similarities and differences of
072. hydrogen energy development in the World’s top 4 largest economies: A
[70] J.P. Torreglosa, P. García, L.M. Fernández, F. Jurado, Energy novel framework, Int. J. Hydrogen Energy 47 (16) (2022) 9485–9503, http:
dispatching based on predictive controller of an off-grid wind //dx.doi.org/10.1016/j.ijhydene.2022.01.038.
turbine/photovoltaic/hydrogen/battery hybrid system, Renew. Energy 74 [94] B. Yodwong, D. Guilbert, M. Phattanasak, W. Kaewmanee, M. Hinaje, G. Vitale,
(2015) 326–336, http://dx.doi.org/10.1016/j.renene.2014.08.010. Proton exchange membrane electrolyzer modeling for power electronics control:
[71] D. Pattabiraman, R.H. Lasseter, T.M. Jahns, Comparison of grid following and A short review, C 6 (2) (2020) 29, http://dx.doi.org/10.3390/c6020029.
grid forming control for a high inverter penetration power system, in: IEEE [95] R. Sarrias-Mena, L.M. Fernández-Ramírez, C.A. García-Vázquez, F. Jurado,
Power and Energy Society General Meeting, Vol. 2018-August, IEEE, 2018, pp. Electrolyzer models for hydrogen production from wind energy systems, Int.
1–5, http://dx.doi.org/10.1109/PESGM.2018.8586162. J. Hydrogen Energy 40 (7) (2015) 2927–2938, http://dx.doi.org/10.1016/j.
[72] K. Mazloomi, C. Gomes, Hydrogen as an energy carrier: Prospects and chal- ijhydene.2014.12.125.
lenges, Renew. Sustain. Energy Rev. 16 (5) (2012) 3024–3033, http://dx.doi. [96] M. Shi, W. Wang, Y. Han, Y. Huang, Research on comprehensive benefit of
org/10.1016/j.rser.2012.02.028. hydrogen storage in microgrid system, Renew. Energy 194 (2022) 621–635,
[73] S. Dutta, A review on production, storage of hydrogen and its utilization http://dx.doi.org/10.1016/j.renene.2022.05.125.
as an energy resource, J. Ind. Eng. Chem. 20 (4) (2014) 1148–1156, http: [97] L. Valverde, F. Rosa, C. Bordons, Design, planning and management of a
//dx.doi.org/10.1016/j.jiec.2013.07.037. hydrogen-based microgrid, IEEE Trans. Ind. Inform. 9 (3) (2013) 1398–1404,
[74] C. Smith, J. Mouli-Castillo, D. van der Horst, S. Haszeldine, M. Lane, Towards a http://dx.doi.org/10.1109/TII.2013.2246576.
100% hydrogen domestic gas network: Regulatory and commercial barriers to [98] M.B. Hossain, M.R. Islam, K.M. Muttaqi, D. Sutanto, A.P. Agalgaonkar, Ad-
the first demonstrator project in the United Kingdom, Int. J. Hydrogen Energy vancement of fuel cells and electrolyzers technologies and their applications
47 (55) (2022) 23071–23083, http://dx.doi.org/10.1016/j.ijhydene.2022.05. to renewable-rich power grids, J. Energy Storage 62 (2023) 106842, http:
123. //dx.doi.org/10.1016/j.est.2023.106842.
33
[99] W. Pei, X. Zhang, W. Deng, C. Tang, L. Yao, Review of operational control [120] C. Meurer, H. Barthels, W.A. Brocke, B. Emonts, H.G. Groehn, PHOEBUS - An
strategy for DC microgrids with electric-hydrogen hybrid storage systems, CSEE autonomous supply system with renewable energy: Six years of operational
J. Power Energy Syst. 8 (2) (2022) 329–346, http://dx.doi.org/10.17775/ experience and advanced concepts, Sol. Energy 67 (1–3) (1999) 131–138,
CSEEJPES.2021.06960. http://dx.doi.org/10.1016/s0038-092x(00)00043-8.
[100] S.A. Gorji, Reconfigurable quadratic converters for electrolyzers utilized in DC [121] C. Chamberlin, P. Lehman, R. Reid, T. Herron, Preliminary results of the schatz
microgrids, IEEE Access 10 (2022) 109677–109687, http://dx.doi.org/10.1109/ fuel cell research project, Adv. Hydrogen Energy 10 (1994) 1659.
ACCESS.2022.3214581. [122] H. Miland, Operational Experience and Control Strategies for a Stand-Alone
[101] F.F. Wang, R. Chen, Z. Dong, Power electronics: A critical enabler of future Power System Based on Renewable Energy and Hydrogen (Ph.D. thesis),
hydrogen–electric systems for aviation, IEEE Electrif. Mag. 10 (2) (2022) 57–68, Norwegian University of Science and Technology, 2005.
http://dx.doi.org/10.1109/MELE.2022.3165954. [123] A. Yilanci, I. Dincer, H.K. Ozturk, A review on solar-hydrogen/fuel cell hybrid
[102] M. Yue, S. Jemei, R. Gouriveau, N. Zerhouni, Review on health-conscious energy systems for stationary applications, Prog. Energy Combust. Sci. 35
energy management strategies for fuel cell hybrid electric vehicles: Degradation (3) (2009) 231–244, http://dx.doi.org/10.1016/j.pecs.2008.07.004, URL https:
models and strategies, Int. J. Hydrogen Energy 44 (13) (2019) 6844–6861, //www.sciencedirect.com/science/article/pii/S0360128508000439.
http://dx.doi.org/10.1016/j.ijhydene.2019.01.190. [124] P.A. Lehman, C.E. Chamberlin, G. Pauletto, M.A. Rocheleau, Operating experi-
ence with a photovoltaic-hydrogen energy system, Int. J. Hydrogen Energy 22
[103] W. Daud, R. Rosli, E. Majlan, S. Hamid, R. Mohamed, T. Husaini, PEM fuel
(5) (1997) 465–470, http://dx.doi.org/10.1016/s0360-3199(96)00127-9.
cell system control: A review, Renew. Energy 113 (2017) 620–638, http:
[125] K. Voss, A.P.D. Goetzberger, G. Bopp, A. Häberle, A. Heinzel, H. Lehmberg,
//dx.doi.org/10.1016/j.renene.2017.06.027.
The self-sufficient solar house in Freiburg—Results of 3 years of operation, Sol.
[104] U. Tamrakar, D. Shrestha, M. Maharjan, B.P. Bhattarai, T.M. Hansen, R.
Energy 58 (1996) 17–23, http://dx.doi.org/10.1016/0038-092X(96)00046-1.
Tonkoski, Virtual inertia: Current trends and future directions, Appl. Sci.
[126] T. Schucan, Case Studies of integrated hydrogen systems. International Energy
(Switzerland) 7 (7) (2017) 1–29, http://dx.doi.org/10.3390/app7070654.
Agency Hydrogen Implementing Agreement, Final report for Subtask A of
[105] L. Harnefors, M. Hinkkanen, U. Riaz, F.M. Rahman, L. Zhang, Robust analytic
task 11-Integrated Systems, Tech. Rep., International Energy Agency Hydrogen
design of power-synchronization control, IEEE Trans. Ind. Electron. 66 (8)
Implementing Agreement, Golden, CO (US), 1999, http://dx.doi.org/10.2172/
(2019) 5810–5819, http://dx.doi.org/10.1109/TIE.2018.2874584.
775587.
[106] J. Matevosyan, B. Badrzadeh, T. Prevost, E. Quitmann, D. Ramasubramanian, [127] H. Barthels, W.A. Brocke, K. Bonhoff, H.G. Groehn, G. Heuts, M. Lennartz,
H. Urdal, S. Achilles, J. Macdowell, S.H. Huang, V. Vital, J. Osullivan, R. Quint, H. Mai, J. Mergel, L. Schmid, P. Ritzenhoff, Phoebus-Jülich: An autonomous
Grid-forming inverters: Are they the key for high renewable penetration? IEEE energy supply system comprising photovoltaics, electrolytic hydrogen, fuel cell,
Power Energy Mag. 17 (6) (2019) 89–98, http://dx.doi.org/10.1109/MPE.2019. Int. J. Hydrogen Energy 23 (4) (1998) 295–301, http://dx.doi.org/10.1016/
2933072. S0360-3199(97)00055-4, URL https://www.sciencedirect.com/science/article/
[107] R.H. Lasseter, Z. Chen, D. Pattabiraman, Grid-forming inverters: A critical asset pii/S0360319997000554.
for the power grid, IEEE J. Emerg. Sel. Top. Power Electron. 8 (2) (2020) [128] K. Voss, A. Goetzberger, G. Bopp, A. Häberle, A. Heinzel, H. Lehmberg, The self-
925–935, http://dx.doi.org/10.1109/JESTPE.2019.2959271. sufficient solar house in Freiburg - Results of 3 years of operation, Solar Energy
[108] J. Matevosyan, S. Sharma, S.-H. Huang, D. Woodfin, K. Ragsdale, S. Moorty, 58 (1–3) (1996) 17–23, http://dx.doi.org/10.1016/0038-092X(96)00046-1.
P. Wattles, W. Li, Proposed future ancillary services in electric reliability [129] H. Barthels, W. Brocke, K. Bonhoff, H. Groehn, G. Heuts, M. Lennartz, H. Mai,
council of Texas, in: 2015 IEEE Eindhoven PowerTech, 2015, pp. 1–6, http: J. Mergel, L. Schmid, P. Ritzenhoff, Phoebus-Jülich: An autonomous energy
//dx.doi.org/10.1109/PTC.2015.7232743. supply system comprising photovoltaics, electrolytic hydrogen, fuel cell, Int.
[109] R.H. Byrne, Energy STorage For Resilience Applications, Tech. Rep., Sandia J. Hydrogen Energy 23 (4) (1998) 295–301, http://dx.doi.org/10.1016/S0360-
National Lab.(SNL-NM), Albuquerque, NM (United States), 2019. 3199(97)00055-4.
[110] P. Marocco, D. Ferrero, A. Lanzini, M. Santarelli, The role of hydrogen in the [130] F. Menzl, Windmill-Electrolyser System for Hydrogen Production at Stralsund.
optimal design of off-grid hybrid renewable energy systems, J. Energy Storage IEA Report, DE-18435, Stralsund, Germany, 2003.
46 (2022) 103893, http://dx.doi.org/10.1016/j.est.2021.103893. [131] M. Ahlhaus, M. Ollick, H. Nielsen, Disintegration of energy crops under thermal
[111] S. Pascuzzi, A.S. Anifantis, I. Blanco, G.S. Mugnozza, Electrolyzer performance treatment; Zersetzungsverhalten von Energiepflanzen bei thermischer Nutzung,
analysis of an integrated hydrogen power system for greenhouse heating a 2004.
case study, Sustainability (Switzerland) 8 (7) (2016) 1–15, http://dx.doi.org/ [132] K. Agbossou, M.L. Kolhe, J. Hamelin, É. Bernier, T.K. Bose, Electrolytic hydro-
10.3390/su8070629. gen based renewable energy system with oxygen recovery and re-utilization,
[112] J. Moore, B. Shabani, A critical study of stationary energy storage policies Renew. Energy 29 (8) (2004) 1305–1318, http://dx.doi.org/10.1016/j.renene.
in Australia in an international context: The role of hydrogen and battery 2003.12.006.
technologies, Energies 9 (9) (2016) http://dx.doi.org/10.3390/en9090674. [133] S. Kélouwani, K. Agbossou, R. Chahine, Model for energy conversion in
renewable energy system with hydrogen storage, J. Power Sources 140 (2)
[113] S. Fukaume, Y. Nagasaki, M. Tsuda, Stable power supply of an independent
(2005) 392–399, http://dx.doi.org/10.1016/j.jpowsour.2004.08.019.
power source for a remote island using a hybrid energy storage system
[134] R. Gammon, A. Roy, J. Barton, M. Little, Hydrogen and renewables integration
composed of electric and hydrogen energy storage systems, Int. J. Hydrogen
( Hari ), 2006.
Energy 47 (29) (2022) 13887–13899, http://dx.doi.org/10.1016/j.ijhydene.
[135] A. Chaparro, J. Soler, M. Escudero, E. De Ceballos, U. Wittstadt, L. Daza, Data
2022.02.142.
results and operational experience with a solar hydrogen system, J. Power
[114] A. Ibáñez-Rioja, P. Puranen, L. Järvinen, A. Kosonen, V. Ruuskanen, J.
Sources 144 (1) (2005) 165–169, http://dx.doi.org/10.1016/j.jpowsour.2004.
Ahola, J. Koponen, Simulation methodology for an off-grid solar–battery–water
12.044.
electrolyzer plant: Simultaneous optimization of component capacities and
[136] Ø. Ulleberg, T. Nakken, A. Ete, The wind/hydrogen demonstration system at
system control, Appl. Energy 307 (2022) 118157, http://dx.doi.org/10.1016/j.
Utsira in Norway: Evaluation of system performance using operational data and
apenergy.2021.118157.
updated hydrogen energy system modeling tools, Int. J. Hydrogen Energy 35
[115] Y. He, Y. Zhou, J. Yuan, Z. Liu, Z. Wang, G. Zhang, Transformation towards a
carbon-neutral residential community with hydrogen economy and advanced
[137] Z. Chehade, C. Mansilla, P. Lucchese, S. Hilliard, J. Proost, Review and
energy management strategies, Energy Convers. Manage. 249 (May) (2021)
analysis of demonstration projects on power-to-X pathways in the world, Int.
114834, http://dx.doi.org/10.1016/j.enconman.2021.114834.
J. Hydrogen Energy 44 (51) (2019) 27637–27655, http://dx.doi.org/10.1016/
[116] Y. He, Y. Zhou, Z. Wang, J. Liu, Z. Liu, G. Zhang, Quantification on fuel j.ijhydene.2019.08.260.
cell degradation and techno-economic analysis of a hydrogen-based grid- [138] M. Thema, F. Bauer, M. Sterner, Power-to-Gas: Electrolysis and methanation
interactive residential energy sharing network with fuel-cell-powered vehicles, status review, Renew. Sustain. Energy Rev. 112 (May) (2019) 775–787, http:
Appl. Energy 303 (2021) 117444, http://dx.doi.org/10.1016/j.apenergy.2021. //dx.doi.org/10.1016/j.rser.2019.06.030.
117444. [139] E. Cetin, A. Yilanci, Y. Oner, M. Colak, I. Kasikci, H.K. Ozturk, Electrical
[117] Y. He, Y. Zhou, J. Liu, Z. Liu, G. Zhang, An inter-city energy migration analysis of a hybrid photovoltaic-hydrogen/fuel cell energy system in Denizli,
framework for regional energy balance through daily commuting fuel-cell Turkey, Energy Build. 41 (9) (2009) 975–981, http://dx.doi.org/10.1016/j.
vehicles, Appl. Energy 324 (November 2021) (2022) 119714, http://dx.doi. enbuild.2009.04.004.
org/10.1016/j.apenergy.2022.119714. [140] M. Landgraf, C.P. Officer, Power to Gas: Storing the wind and sun in natural
[118] T. Schucan, Case studies of integrated hydrogen systems. International energy gas, Press Release-Karlsruhe Inst. Technol. (044) (2014) 1–2.
agency hydrogen implementing agreement, final report for subtask a of task [141] O. Posdziech, K. Schwarze, J. Brabandt, Efficient hydrogen production for
11 - integrated systems, 1999, Other Information: PBD: 31 Dec 1999, URL industry and electricity storage via high-temperature electrolysis, Int. J. Hydro-
https://digital.library.unt.edu/ark:/67531/metadc715026/. gen Energy 44 (35) (2019) 19089–19101, http://dx.doi.org/10.1016/j.ijhydene.
[119] A. Szyszka, Ten years of solar hydrogen demonstration project at Neunburg 2018.05.169.
vorm Wald, Germany, Int. J. Hydrogen Energy 23 (10) (1998) 849–860, http:// [142] C. Wulf, J. Linßen, P. Zapp, Review of Power-to-Gas projects in Europe, En-
dx.doi.org/10.1016/S0360-3199(97)00172-9, URL https://www.sciencedirect. ergy Procedia 155 (2018) 367–378, http://dx.doi.org/10.1016/j.egypro.2018.
com/science/article/pii/S0360319997001729. 11.041.
34
[143] M. Landgraf, C.P. Officer, Flexible methane production from electricity and [166] G. Fontes, C. Turpin, S. Astier, T.A. Meynard, Interactions between fuel cells
bio-mass, Press Release-Karlsruhe Inst. Technol. (2014). and power converters: Influence of current harmonics on a fuel cell stack,
[144] M. Sartory, E. Wallnöfer-Ogris, P. Salman, T. Fellinger, M. Justl, A. Trattner, IEEE Trans. Power Electron. 22 (2) (2007) 670–678, http://dx.doi.org/10.1109/
M. Klell, Theoretical and experimental analysis of an asymmetric high pressure TPEL.2006.890008.
PEM water electrolyser up to 155 bar, Int. J. Hydrogen Energy 42 (52) [167] A. Kolli, A. Gaillard, A. De Bernardinis, O. Bethoux, D. Hissel, Z. Khatir,
(2017) 30493–30508, http://dx.doi.org/10.1016/j.ijhydene.2017.10.112, URL A review on DC/DC converter architectures for power fuel cell applications,
https://www.sciencedirect.com/science/article/pii/S0360319917341538. Energy Convers. Manage. 105 (2015) 716–730, http://dx.doi.org/10.1016/j.
[145] M. Persson, D. Mignard, D. Hogg, Data averaging for modelling energy enconman.2015.07.060.
storage: What information is hidden? in: 2018 53rd International Universities [168] J.E. Valdez-Resendiz, V.M. Sanchez, J.C. Rosas-Caro, J.C. Mayo-Maldonado,
Power Engineering Conference, UPEC, 2018, pp. 1–5, http://dx.doi.org/10. J.M. Sierra, R. Barbosa, Continuous input-current buck-boost DC-DC converter
1109/UPEC.2018.8541917. for PEM fuel cell applications, Int. J. Hydrogen Energy 42 (51) (2017)
[146] T. Isaac, HyDeploy: The UK’s first hydrogen blending deployment project, Clean 30389–30399, http://dx.doi.org/10.1016/j.ijhydene.2017.10.077.
Energy 3 (2) (2019) 114–125, http://dx.doi.org/10.1093/ce/zkz006. [169] D. Guilbert, S.M. Collura, A. Scipioni, DC/DC converter topologies for elec-
[147] W.K. Na, B. Gou, Feedback-linearization-based nonlinear control for PEM fuel trolyzers: State-of-the-art and remaining key issues, Int. J. Hydrogen Energy 42
cells, IEEE Trans. Energy Convers. 23 (1) (2008) 179–190, http://dx.doi.org/ (38) (2017) 23966–23985, http://dx.doi.org/10.1016/j.ijhydene.2017.07.174.
10.1109/TEC.2007.914160. [170] A.S.I. Zinober, An introduction to sliding mode variable structure control, in:
[148] F. Akter, T.K. Roy, M.S. Islam, A.F. Alkhateeb, M.A. Mollah, Design of a Variable Structure and Lyapunov Control, Springer-Verlag, 2005, pp. 1–22,
nonlinear integral terminal sliding mode controller for a PEM fuel cell based http://dx.doi.org/10.1007/bfb0033676.
on a DC-DC boost converter, IEEE Access 10 (August) (2022) 97419–97428, [171] S.C. Tan, Y.M. Lai, M.K. Cheung, C.K. Tse, On the practical design of a sliding
http://dx.doi.org/10.1109/access.2022.3205733. mode voltage controlled buck converter, IEEE Trans. Power Electron. 20 (2)
[149] J.J. Hwang, L.K. Lai, W. Wu, W.R. Chang, Dynamic modeling of a photovoltaic (2005) 425–437, http://dx.doi.org/10.1109/TPEL.2004.842977.
hydrogen fuel cell hybrid system, Int. J. Hydrogen Energy 34 (23) (2009) [172] H. Komurcugil, Adaptive terminal sliding-mode control strategy for DC-DC buck
9531–9542, http://dx.doi.org/10.1016/j.ijhydene.2009.09.100. converters, ISA Trans. 51 (6) (2012) 673–681, http://dx.doi.org/10.1016/j.
[150] T. Yigit, O.F. Selamet, Mathematical modeling and dynamic Simulink simulation isatra.2012.07.005.
of high-pressure PEM electrolyzer system, Int. J. Hydrogen Energy 41 (32) [173] S.-C. Tan, Y. Lai, C.K. Tse, M.K. Cheung, Adaptive feedforward and feedback
(2016) 13901–13914, http://dx.doi.org/10.1016/j.ijhydene.2016.06.022. control schemes for sliding mode controlled power converters, IEEE Trans.
[151] W.Y. Chang, Estimating equivalent circuit parameters of proton exchange Power Electron. 21 (1) (2006) 182–192, http://dx.doi.org/10.1109/TPEL.2005.
membrane fuel cell using the current change method, Int. J. Electr. Power 861191.
Energy Syst. 53 (1) (2013) 584–591, http://dx.doi.org/10.1016/j.ijepes.2013. [174] S.C. Tan, Y.M. Lai, C.K. Tse, M.K. Cheung, Adaptive feedforward and feedback
05.031. control schemes for sliding mode controlled power converters, IEEE Trans.
[152] H. Görgün, Dynamic modelling of a proton exchange membrane (PEM) elec-
Power Electron. 21 (1) (2006) 182–192, http://dx.doi.org/10.1109/TPEL.2005.
trolyzer, Int. J. Hydrogen Energy 31 (1) (2006) 29–38, http://dx.doi.org/10.
861191.
1016/j.ijhydene.2005.04.001.
[175] S.C. Tan, Y.M. Lai, C.K. Tse, Indirect sliding mode control of power converters
[153] O. Atlam, M. Kolhe, Equivalent electrical model for a proton exchange mem-
via double integral sliding surface, IEEE Trans. Power Electron. 23 (2) (2008)
brane (PEM) electrolyser, Energy Convers. Manage. 52 (8–9) (2011) 2952–2957,
600–611, http://dx.doi.org/10.1109/TPEL.2007.915624.
http://dx.doi.org/10.1016/j.enconman.2011.04.007.
[176] S.-C. Tan, Y.-M. Lai, K.T. Chi, L. Martínez-Salamero, C.-K. Wu, A fast-response
[154] S. Haji, Analytical modeling of PEM fuel cell Is–V curve, Renew. Energy
sliding-mode controller for boost-type converters with a wide range of operating
36 (2) (2011) 451–458, http://dx.doi.org/10.1016/j.renene.2010.07.007, URL
conditions, IEEE Trans. Ind. Electron. 54 (6) (2007) 3276–3286, http://dx.doi.
https://www.sciencedirect.com/science/article/pii/S0960148110003319.
org/10.1109/TIE.2007.905969.
[155] A. Awasthi, K. Scott, S. Basu, Dynamic modeling and simulation of a proton
[177] I. Abadlia, M. Adjabi, H. Bouzeria, Sliding mode based power control of grid-
exchange membrane electrolyzer for hydrogen production, Int. J. Hydrogen
connected photovoltaic-hydrogen hybrid system, Int. J. Hydrogen Energy 42
Energy 36 (22) (2011) 14779–14786, http://dx.doi.org/10.1016/j.ijhydene.
2011.03.045.
[178] H. Armghan, M. Yang, A. Armghan, N. Ali, M.Q. Wang, I. Ahmad, Design
[156] B.W. Tuinema, E. Adabi, P.K. Ayivor, V. García Suárez, L. Liu, A. Perilla, Z.
of integral terminal sliding mode controller for the hybrid AC/DC microgrids
Ahmad, J.L. Rueda Torres, M.A. Meijden, P. Palensky, Modelling of large-sized
involving renewables and energy storage systems, Int. J. Electr. Power Energy
electrolysers for real-time simulation and study of the possibility of frequency
Syst. 119 (January) (2020) 105857, http://dx.doi.org/10.1016/j.ijepes.2020.
support by electrolysers, IET Gener. Transm. Distrib. 14 (10) (2020) 1985–
105857.
1992, http://dx.doi.org/10.1049/iet-gtd.2019.1364, URL https://onlinelibrary.
[179] B. Yang, J. Wang, Y. Sang, L. Yu, H. Shu, S. Li, T. He, L. Yang, X. Zhang,
wiley.com/doi/10.1049/iet-gtd.2019.1364.
T. Yu, Applications of supercapacitor energy storage systems in microgrid with
[157] Á. Hernández-Gómez, V. Ramirez, D. Guilbert, Investigation of PEM electrolyzer
distributed generators via passive fractional-order sliding-mode control, Energy
modeling: Electrical domain, efficiency, and specific energy consumption, Int.
187 (2019) 115905, http://dx.doi.org/10.1016/j.energy.2019.115905.
J. Hydrogen Energy 45 (29) (2020) 14625–14639, http://dx.doi.org/10.1016/
j.ijhydene.2020.03.195. [180] S.T. Venkataraman, S. Gulati, Control of nonlinear systems using terminal
[158] A.A. Rahim, A.S. Tijani, S. Kamarudin, S. Hanapi, An overview of polymer sliding modes, in: Proceedings of the American Control Conference, Vol. 1,
electrolyte membrane electrolyzer for hydrogen production: Modeling and mass Publ by American Automatic Control Council, 1992, pp. 891–893, http://dx.
transport, J. Power Sources 309 (2016) 56–65, http://dx.doi.org/10.1016/j. doi.org/10.23919/acc.1992.4792209.
jpowsour.2016.01.012. [181] X. Yu, Z. Man, Model reference adaptive control systems with terminal sliding
[159] P. Olivier, C. Bourasseau, P.B. Bouamama, Low-temperature electrolysis system modes, Internat. J. Control 64 (6) (1996) 1165–1176, http://dx.doi.org/10.
modelling: A review, Renew. Sustain. Energy Rev. 78 (2017) 280–300. 1080/00207179608921680.
[160] D.S. Gautam, A.K. Bhat, A comparison of soft-switched DC-to-DC converters [182] C. Xiu, P. Guo, Global terminal sliding mode control with the quick reaching
for electrolyzer application, IEEE Trans. Power Electron. 28 (1) (2013) 54–63, law and its application, IEEE Access 6 (2018) 49793–49800, http://dx.doi.org/
http://dx.doi.org/10.1109/TPEL.2012.2195682. 10.1109/ACCESS.2018.2868785.
[161] V. Ruuskanen, J. Koponen, K. Huoman, A. Kosonen, M. Niemelä, J. Ahola, [183] X. Yu, Y. Feng, Z. Man, Terminal sliding mode control - An overview, IEEE
PEM water electrolyzer model for a power-hardware-in-loop simulator, Int. J. Open J. Ind. Electron. Soc. 2 (November 2020) (2021) 36–52, http://dx.doi.
Hydrogen Energy 42 (16) (2017) 10775–10784, http://dx.doi.org/10.1016/j. org/10.1109/OJIES.2020.3040412.
ijhydene.2017.03.046. [184] S. Ding, W.X. Zheng, J. Sun, J. Wang, Second-order sliding-mode controller
[162] G.-R. Zhu, S.-C. Tan, Y. Chen, C.K. Tse, Mitigation of low-frequency current design and its implementation for buck converters, IEEE Trans. Ind. Inform. 14
ripple in fuel-cell inverter systems through waveform control, IEEE Trans. (5) (2018) 1990–2000, http://dx.doi.org/10.1109/TII.2017.2758263.
Power Electron. 28 (2) (2013) 779–792, http://dx.doi.org/10.1109/TPEL.2012. [185] M.A. Hossain, H.R. Pota, W. Issa, M.J. Hossain, Overview of AC microgrid
2205407. controls with inverter-interfaced generations, Energies 10 (9) (2017) 1–27,
[163] Y. Zhan, Y. Guo, J. Zhu, B. Liang, B. Yang, Comprehensive influences mea- http://dx.doi.org/10.3390/en10091300.
surement and analysis of power converter low frequency current ripple on [186] Y. Su, C. Zheng, A new nonsingular integral terminal sliding mode control
PEM fuel cell, Int. J. Hydrogen Energy 44 (59) (2019) 31352–31359, http: for robot manipulators, Internat. J. Systems Sci. 51 (8) (2020) 1418–1428,
//dx.doi.org/10.1016/j.ijhydene.2019.09.231. http://dx.doi.org/10.1080/00207721.2020.1764658.
[164] R.C. Zhu, D.Y. Zhan, Fuel cell high efficiency power converter for suppressing [187] L. Qiao, W. Zhang, Trajectory tracking control of AUVs via adaptive fast
low frequency current ripple, Procedia Comput. Sci. 166 (2020) 296–300, nonsingular integral terminal sliding mode control, IEEE Trans. Ind. Inform.
http://dx.doi.org/10.1016/j.procs.2020.02.096. 16 (2) (2020) 1248–1258, http://dx.doi.org/10.1109/TII.2019.2949007.
[165] M. Uno, K. Tanaka, Pt/C catalyst degradation in proton exchange membrane [188] M. Uzunoglu, O.C. Onar, M.S. Alam, Modeling, control and simulation of a
fuel cells due to high-frequency potential cycling induced by switching power PV/FC/UC based hybrid power generation system for stand-alone applications,
converters, J. Power Sources 196 (23) (2011) 9884–9889, http://dx.doi.org/10. Renew. Energy 34 (3) (2009) 509–520, http://dx.doi.org/10.1016/j.renene.
1016/j.jpowsour.2011.08.030. 2008.06.009.
35
[189] S. Emeryanov, S. Korovin, A. Levant, High-order sliding modes in control [213] W.R. Sultana, S.K. Sahoo, S. Sukchai, S. Yamuna, D. Venkatesh, A review on
systems, Sign (U+˜) 50 (1996) 3, http://dx.doi.org/10.1007/BF01128162. state of art development of model predictive control for renewable energy
[190] M. Derbeli, M. Farhat, O. Barambones, L. Sbita, Control of PEM fuel cell power applications, Renew. Sustain. Energy Rev. 76 (January 2016) (2017) 391–406,
system using sliding mode and super-twisting algorithms, Int. J. Hydrogen http://dx.doi.org/10.1016/j.rser.2017.03.058.
Energy 42 (13) (2017) 8833–8844, http://dx.doi.org/10.1016/j.ijhydene.2016. [214] S. Vazquez, J. Rodriguez, M. Rivera, L.G. Franquelo, M. Norambuena, Model
06.103. predictive control for power converters and drives: Advances and trends, IEEE
[191] H. Asada, J.-J. Slotine, Robot Analysis and Control, John Wiley & Sons, 1991. Trans. Ind. Electron. 64 (2) (2017) 935–947, http://dx.doi.org/10.1109/TIE.
[192] J. Zhou, C. Wen, Adaptive Backstepping Control of Uncertain Systems: Non- 2016.2625238.
smooth Nonlinearities, Interactions or Time-Variations, Springer, 2008, http: [215] Z. Gong, P. Dai, X. Yuan, X. Wu, G. Guo, Design and experimental evaluation
//dx.doi.org/10.1007/978-3-540-77807-3. of fast model predictive control for modular multilevel converters, IEEE Trans.
[193] F. Fang, L. Wei, Backstepping-based nonlinear adaptive control for coal-fired Ind. Electron. 63 (6) (2016) 3845–3856, http://dx.doi.org/10.1109/TIE.2015.
utility boiler–turbine units, Appl. Energy 88 (3) (2011) 814–824, http://dx.doi. 2497254.
org/10.1016/j.apenergy.2010.09.003. [216] S. Almer, S. Mariethoz, M. Morari, Dynamic phasor model predictive control
of switched mode power converters, IEEE Trans. Control Syst. Technol. 23 (1)
[194] M. Krstic, P.V. Kokotovic, I. Kanellakopoulos, Nonlinear and Adaptive Control
(2015) 349–356, http://dx.doi.org/10.1109/TCST.2014.2317775.
Design, John Wiley & Sons, Inc., 1995.
[217] E. Irmak, N. Güler, A model predictive control-based hybrid MPPT method for
[195] M.A. Hossain, M.R. Islam, M.Y.-Y.U. Haque, J. Hasan, T.K. Roy, M.A.H. Sadi,
boost converters, Int. J. Electron. 107 (1) (2020) 1–16, http://dx.doi.org/10.
Protecting DFIG-based multi-machine power system under transient-state by
1080/00207217.2019.1582715.
nonlinear adaptive backstepping controller-based capacitive BFCL, IET Gener.
[218] F. An, W. Song, K. Yang, N. Hou, J. Ma, Improved dynamic performance of
Transm. Distrib. 16 (22) (2022) 4528–4548, http://dx.doi.org/10.1049/gtd2.
dual active bridge DC–DC converters using MPC scheme, IET Power Electron.
12617.
11 (11) (2018) 1756–1765, http://dx.doi.org/10.1049/iet-pel.2017.0707, URL
[196] O. Kuljaca, N. Swamy, F. Lewis, C. Kwan, Design and implementation of
https://onlinelibrary.wiley.com/doi/10.1049/iet-pel.2017.0707.
industrial neural network controller using backstepping, IEEE Trans. Ind.
[219] K. Kumar, M. Alam, V. Dutta, Energy management strategy for integration of
Electron. 50 (1) (2003) 193–201, http://dx.doi.org/10.1109/TIE.2002.807675.
fuel cell-electrolyzer technologies in microgrid, Int. J. Hydrogen Energy 46 (68)
[197] S. Vaidyanathan, B.A. Idowu, A.T. Azar, Backstepping controller design for the (2021) 33738–33755, http://dx.doi.org/10.1016/j.ijhydene.2021.07.203.
global chaos synchronization of Sprott’s jerk systems, in: Chaos Modeling and [220] K.H. Cheng, C.F. Hsu, C.M. Lin, T.T. Lee, C. Li, Fuzzy-neural sliding-mode
Control Systems Design, Springer, 2015, pp. 39–58, http://dx.doi.org/10.1007/ control for DC-DC converters using asymmetric Gaussian membership functions,
978-3-319-13132-0_3. IEEE Trans. Ind. Electron. 54 (3) (2007) 1528–1536, http://dx.doi.org/10.1109/
[198] X. Hao, I. Salhi, S. Laghrouche, Y. Ait-Amirat, A. Djerdir, Backstepping TIE.2007.894717.
supertwisting control of four-phase interleaved boost converter for PEM fuel [221] A.T. Nguyen, T. Taniguchi, L. Eciolaza, V. Campos, R. Palhares, M. Sugeno,
cell, IEEE Trans. Power Electron. 37 (7) (2022) 7858–7870, http://dx.doi.org/ Fuzzy control systems: Past, present and future, IEEE Comput. Intell. Mag. 14
10.1109/TPEL.2022.3149099. (1) (2019) 56–68, http://dx.doi.org/10.1109/MCI.2018.2881644.
[199] T.K. Roy, M.A. Mahmud, A.M.T. Oo, Robust adaptive backstepping excita- [222] A. Zar, I. Ahmad, M.S. Nazir, I. Ahmed, Fuzzy optimized conditioned-barrier
tion controller design for higher-order models of synchronous generators in nonlinear control of electric vehicle for grid to vehicle & vehicle to grid
multimachine power systems, IEEE Trans. Power Syst. 34 (1) (2019) 40–51, applications, J. Energy Storage 64 (2023) 107251, http://dx.doi.org/10.1016/
http://dx.doi.org/10.1109/TPWRS.2018.2868783. j.est.2023.107251.
[200] G. Melath, M.N. Hussain, V. Agarwal, A nonlinear adaptive backstepping [223] X. Tan, Q. Li, H. Wang, Electrical Power and Energy Systems Advances and
controller for stabilizing the DC bus voltage of an islanded hybrid microgrid, trends of energy storage technology in Microgrid, Int. J. Electr. Power Energy
IEEE J. Emerg. Sel. Top. Ind. Electron. 3 (3) (2022) 538–548, http://dx.doi. Syst. 44 (1) (2013) 179–191, http://dx.doi.org/10.1016/j.ijepes.2012.07.015.
org/10.1109/JESTIE.2022.3160666. [224] L.A. Zadeh, Outline of a new approach to the analysis of complex systems and
[201] M.R. Islam, A.D. Upadhay, T.K. Roy, J. Hasan, M.A. Mahmud, Nonlinear decision processes, IEEE Trans. Syst. Man Cybern. SMC-3 (1) (1973) 28–44,
backstepping controller design for bridge-type fault current limiter to enhance http://dx.doi.org/10.1109/TSMC.1973.5408575.
the transient performance of hybrid power systems, Int. Trans. Electr. Energy [225] S. Lalouni, D. Rekioua, T. Rekioua, E. Matagne, Fuzzy logic control of stand-
Syst. 31 (11) (2021) e13097, http://dx.doi.org/10.1002/2050-7038.13097. alone photovoltaic system with battery storage, J. Power Sources 193 (2) (2009)
[202] J. Linares-Flores, C. García-Rodríguez, H. Sira-Ramírez, O.D. Ramírez-Cárdenas, 899–907, http://dx.doi.org/10.1016/j.jpowsour.2009.04.016.
Robust backstepping tracking controller for low-speed PMSM positioning sys- [226] F. Mumtaz, N. Zaihar Yahaya, S. Tanzim Meraj, B. Singh, R. Kannan, O.
tem: Design, analysis, and implementation, IEEE Trans. Ind. Inform. 11 (5) Ibrahim, Review on non-isolated DC-DC converters and their control techniques
(2015) 1130–1141, http://dx.doi.org/10.1109/TII.2015.2471814. for renewable energy applications, Ain Shams Eng. J. 12 (4) (2021) 3747–3763,
[203] A.M. Harb, B.A. Harb, Controlling chaos in Josephson-Junction using non- http://dx.doi.org/10.1016/j.asej.2021.03.022.
linear backstepping controller, IEEE Trans. Appl. Supercond. 16 (4) (2006) [227] D. Svozil, V. Kvasnicka, J. Pospichal, Introduction to multi-layer feed-forward
1988–1998, http://dx.doi.org/10.1109/TASC.2006.881811. neural networks, Chemometr. Intell. Lab. Syst. 39 (1) (1997) 43–62, http:
[204] J. Rawlings, Tutorial overview of model predictive control, IEEE Control Syst. //dx.doi.org/10.1016/S0169-7439(97)00061-0.
Mag. 20 (3) (2000) 38–52, http://dx.doi.org/10.1109/37.845037. [228] X. Luo, A. Patton, C. Singh, Real power transfer capability calculations using
multi-layer feed-forward neural networks, IEEE Trans. Power Syst. 15 (2) (2000)
[205] J.H. Lee, Model predictive control: Review of the three decades of development,
903–908, http://dx.doi.org/10.1109/59.867192.
Int. J. Control Autom. Syst. 9 (3) (2011) 415–424, http://dx.doi.org/10.1007/
[229] M.R. Habibi, H.R. Baghaee, T. Dragičević, F. Blaabjerg, False data injection
s12555-011-0300-6.
cyber-attacks mitigation in parallel DC/DC converters based on artificial
[206] E.F. Camacho, C.B. Alba, Model Predictive Control, Springer science & business
neural networks, IEEE Trans. Circuits Syst. II 68 (2) (2021) 717–721, http:
media, 2013, http://dx.doi.org/10.1007/978-0-85729-398-5.
//dx.doi.org/10.1109/TCSII.2020.3011324.
[207] P. Cortés, M.P. Kazmierkowski, R.M. Kennel, D.E. Quevedo, J. Rodriguez,
[230] W. Dong, S. Li, X. Fu, Z. Li, M. Fairbank, Y. Gao, Control of a buck
Predictive control in power electronics and drives, IEEE Trans. Ind. Electron.
DC/DC converter using approximate dynamic programming and artificial neural
55 (12) (2008) 4312–4324, http://dx.doi.org/10.1109/TIE.2008.2007480.
networks, IEEE Trans. Circuits Syst. I. Regul. Pap. 68 (4) (2021) 1760–1768,
[208] S. Kouro, P. Cortés, R. Vargas, U. Ammann, J. Rodríguez, Model predictive http://dx.doi.org/10.1109/TCSI.2021.3053468.
control - A simple and powerful Method to control power converters, IEEE [231] Z. Li, Y. Gao, X. Zhang, B. Wang, H. Ma, A model-data-hybrid-driven diagnosis
Trans. Ind. Electron. 56 (6) (2009) 1826–1838, http://dx.doi.org/10.1109/TIE. method for open-switch faults in power converters, IEEE Trans. Power Electron.
2008.2008349. 36 (5) (2021) 4965–4970, http://dx.doi.org/10.1109/TPEL.2020.3026176.
[209] G.A. Papafotiou, G.D. Demetriades, V.G. Agelidis, Technology readiness as- [232] S.V. Medina, U.P. Ajenjo, Performance improvement of artificial neural network
sessment of model predictive control in medium- and high-voltage power model in short-term forecasting of wind farm power output, J. Mod. Power
electronics, IEEE Trans. Ind. Electron. 63 (9) (2016) 5807–5815, http://dx.doi. Syst. Clean Energy 8 (3) (2020) 484–490, http://dx.doi.org/10.35833/MPCE.
org/10.1109/TIE.2016.2521350. 2018.000792.
[210] J. Holtz, Advanced PWM and predictive control-An overview, IEEE Trans. [233] S. Mohagheghi, R.G. Harley, T.G. Habetler, D. Divan, Condition monitoring of
Ind. Electron. 63 (6) (2016) 3837–3844, http://dx.doi.org/10.1109/TIE.2015. power electronic circuits using artificial neural networks, IEEE Trans. Power
2504347. Electron. 24 (10) (2009) 2363–2367, http://dx.doi.org/10.1109/TPEL.2009.
[211] S. Kouro, M.A. Perez, J. Rodriguez, A.M. Llor, H.A. Young, Model predictive 2017806.
control: MPC’s role in the evolution of power electronics, IEEE Ind. Electron. [234] K.J. Reddy, N. Sudhakar, ANFIS-MPPT control algorithm for a PEMFC system
Mag. 9 (4) (2015) 8–21, http://dx.doi.org/10.1109/MIE.2015.2478920. used in electric vehicle applications, Int. J. Hydrogen Energy 44 (29) (2019)
[212] S. Vazquez, J.I. Leon, L.G. Franquelo, J. Rodriguez, H.A. Young, A. Marquez, 15355–15369, http://dx.doi.org/10.1016/j.ijhydene.2019.04.054.
P. Zanchetta, Model predictive control: A review of its applications in power [235] L. Yang, E. Entchev, Performance prediction of a hybrid microgeneration system
electronics, IEEE Ind. Electron. Mag. 8 (1) (2014) 16–31, http://dx.doi.org/10. using adaptive neuro-fuzzy inference system (ANFIS) technique, Appl. Energy
1109/MIE.2013.2290138. 134 (2014) 197–203, http://dx.doi.org/10.1016/j.apenergy.2014.08.022.
36
[236] L.M. Halabi, S. Mekhilef, M. Hossain, Performance evaluation of hybrid adaptive [251] D. Pagnani, F. Blaabjerg, C.L. Bak, F.M. da Silva, Ł.H. Kocewiak, J. Hjerrild,
neuro-fuzzy inference system models for predicting monthly global solar radia- Offshore wind farm black start service integration: Review and outlook of ongo-
tion, Appl. Energy 213 (2018) 247–261, http://dx.doi.org/10.1016/j.apenergy. ing research, Energies 13 (23) (2020) http://dx.doi.org/10.3390/en13236286,
2018.01.035. URL https://www.mdpi.com/1996-1073/13/23/6286.
[237] J.-S. Jang, ANFIS: Adaptive-network-based fuzzy inference system, IEEE Trans. [252] T. Kerdphol, F.S. Rahman, M. Watanabe, Y. Mitani, Virtual Inertia Synthesis
Syst. Man Cybern. 23 (3) (1993) 665–685, http://dx.doi.org/10.1109/21. and Control, Springer, 2021, http://dx.doi.org/10.1007/978-3-030-57961-6.
256541. [253] Y. Zuo, Z. Yuan, F. Sossan, A. Zecchino, R. Cherkaoui, M. Paolone, Performance
[238] Y. Wang, C. Wu, S. Zhao, J. Wang, B. Zu, M. Han, Q. Du, M. Ni, K. Jiao, assessment of grid-forming and grid-following converter-interfaced battery
Coupling deep learning and multi-objective genetic algorithms to achieve high energy storage systems on frequency regulation in low-inertia power grids,
performance and durability of direct internal reforming solid oxide fuel cell, Sustain. Energy Grids Netw. 27 (2021) 100496, http://dx.doi.org/10.1016/j.
Appl. Energy 315 (2022) 119046, http://dx.doi.org/10.1016/j.apenergy.2022. segan.2021.100496.
119046. [254] M. Alam, K. Kumar, V. Dutta, Droop based control strategy for balancing the
[239] D. Min, Z. Song, H. Chen, T. Wang, T. Zhang, Genetic algorithm optimized level of hydrogen storage in direct current microgrid application, J. Energy
neural network based fuel cell hybrid electric vehicle energy management Storage 33 (2021) 102106, http://dx.doi.org/10.1016/j.est.2020.102106.
strategy under start-stop condition, Appl. Energy 306 (2022) 118036, http: [255] A. Yazdani, R. Iravani, Voltage-Sourced Converters in Power Systems: Modeling,
//dx.doi.org/10.1016/j.apenergy.2021.118036. Control, and Applications, John Wiley & Sons, 2010.
[240] A. Tapia, A. del Nozal, D. Reina, P. Millán, Three-dimensional optimization [256] M. Ndreko, S. Rüberg, W. Winter, Grid forming control scheme for power
of penstock layouts for micro-hydropower plants using genetic algorithms, systems with up to 100% power electronic interfaced generation: A case
Appl. Energy 301 (2021) 117499, http://dx.doi.org/10.1016/j.apenergy.2021. study on Great Britain test system, IET Renew. Power Gener. 14 (8) (2020)
117499. 1268–1281, http://dx.doi.org/10.1049/iet-rpg.2019.0700.
[241] S. Katoch, S.S. Chauhan, V. Kumar, A review on genetic algorithm: Past, present, [257] D.B. Rathnayake, M. Akrami, C. Phurailatpam, S.P. Me, S. Hadavi, G. Jayas-
and future, Multimedia Tools Appl. 80 (2021) 8091–8126, http://dx.doi.org/ inghe, S. Zabihi, B. Bahrani, Grid forming inverter modeling, control, and
10.1007/s11042-020-10139-6. applications, IEEE Access 9 (2021) 114781–114807, http://dx.doi.org/10.1109/
[242] J.H. Holland, Genetic algorithms, Sci. Am. 267 (1) (1992) 66–73. ACCESS.2021.3104617.
[243] M. Mohan, V.K. Sharma, E.A. Kumar, V. Gayathri, Hydrogen storage in carbon [258] Q. Li, R. Li, Y. Pu, S. Li, C. Sun, W. Chen, Coordinated control
materials—A review, Energy Storage 1 (2) (2019) e35, http://dx.doi.org/10. of electric-hydrogen hybrid energy storage for multi-microgrid with fuel
1002/est2.35.
cell/electrolyzer/PV/battery, J. Energy Storage 42 (2021) 103110, http://dx.
[244] Z. Zhang, Y. Nagasaki, D. Miyagi, M. Tsuda, T. Komagome, K. Tsukada, T.
doi.org/10.1016/j.est.2021.103110.
Hamajima, H. Ayakawa, Y. Ishii, D. Yonekura, Stored energy control for long-
[259] H. Golpîra, H. Seifi, A.R. Messina, M.R. Haghifam, Maximum penetration level
term continuous operation of an electric and hydrogen hybrid energy storage
of micro-grids in large-scale power systems: Frequency stability viewpoint,
system for emergency power supply and solar power fluctuation compensation,
Int. J. Hydrogen Energy 44 (16) (2019) 8403–8414, http://dx.doi.org/10.1016/ IEEE Trans. Power Syst. 31 (6) (2016) 5163–5171, http://dx.doi.org/10.1109/
j.ijhydene.2019.02.076. TPWRS.2016.2538083.
[245] M.Y.-Y.U. Haque, M.R. Islam, T. Ahmed, M.R.I. Sheikh, Improved voltage [260] Y. Mi, H. Zhang, Y. Fu, C. Wang, P.C. Loh, P. Wang, Intelligent power sharing
tracking of autonomous microgrid technology using a combined resonant of DC isolated microgrid based on fuzzy sliding mode droop control, IEEE
controller with lead-lag compensator adopting negative imaginary theorem, Trans. Smart Grid 10 (3) (2019) 2396–2406, http://dx.doi.org/10.1109/TSG.
Protect. Control Mod. Power Syst. 7 (1) (2022) 10, http://dx.doi.org/10.1186/ 2018.2797127.
s41601-022-00231-4. [261] D.C.D.S. Junior, J.S. Dohler, P.M. Almeida, J.G. Oliveira, Droop control for
[246] J. Fang, H. Li, Y. Tang, F. Blaabjerg, Distributed power system virtual inertia power sharing and voltage and frequency regulation in parallel distributed
implemented by grid-connected power converters, IEEE Trans. Power Electron. generations on AC microgrid, in: 2018 13th IEEE International Conference on
33 (10) (2018) 8488–8499, http://dx.doi.org/10.1109/TPEL.2017.2785218. Industry Applications, INDUSCON 2018 - Proceedings, no. Table I, IEEE, 2019,
[247] Y. Han, H. Li, P. Shen, E.A.A. Coelho, J.M. Guerrero, Review of active and pp. 1–6, http://dx.doi.org/10.1109/INDUSCON.2018.8627328.
reactive power sharing strategies in hierarchical controlled microgrids, IEEE [262] Q.C. Zhong, Y. Wang, B. Ren, Connecting the home grid to the public grid: Field
Trans. Power Electron. 32 (3) (2017) 2427–2451, http://dx.doi.org/10.1109/ demonstration of virtual synchronous machines, IEEE Power Electron. Mag. 6
TPEL.2016.2569597.
(4) (2019) 41–49, http://dx.doi.org/10.1109/MPEL.2019.2946700.
[248] A.D. Paquette, M.J. Reno, R.G. Harley, D.M. Divan, Sharing transient loads :
[263] W. Zhang, A.M. Cantarellas, J. Rocabert, A. Luna, P. Rodriguez, Synchronous
Causes of unequal transient load sharing in islanded microgrid operation, IEEE
power controller with flexible droop characteristics for renewable power
Ind. Appl. Mag. 20 (2) (2014) 23–34, http://dx.doi.org/10.1109/MIAS.2013.
generation systems, IEEE Trans. Sustain. Energy 7 (4) (2016) 1572–1582,
2288408.
[249] D. Ramasubramanian, P. Pourbeik, E. Farantatos, A. Gaikwad, Simulation of http://dx.doi.org/10.1109/TSTE.2016.2565059.
100% inverter-based resource grids with positive sequence modeling, IEEE [264] X. Zhang, W. Pei, C. Mei, W. Deng, J. Tan, Q. Zhang, Transform from
Electrif. Mag. 9 (2) (2021) 62–71, http://dx.doi.org/10.1109/MELE.2021. gasoline stations to electric-hydrogen hybrid refueling stations: An islanding
3070938. DC microgrid with electric-hydrogen hybrid energy storage system and its
[250] P. Tielens, D. Van Hertem, Grid inertia and frequency control in power systems control strategy, Int. J. Electr. Power Energy Syst. 136 (September 2021) (2022)
with high penetration of renewables, in: Young Researchers Symposium in 107684, http://dx.doi.org/10.1016/j.ijepes.2021.107684.
Electrical Power Engineering, Date: 2012/04/16-2012/04/17, Location: Delft,
the Netherlands, 2012.
37

1-s2.0-S2352152X23015670-main

Uploaded by

Copyright:

Available Formats

1-s2.0-S2352152X23015670-main

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

1-s2.0-S2352152X23015670-main

Uploaded by

Copyright:

Available Formats

Journal of Energy Storage 72 (2023) 108170

Contents lists available at ScienceDirect

Journal of Energy Storage

Control strategy review for hydrogen-renewable energy power system

ARTICLE INFO ABSTRACT

1. Introduction worldwide capacity for wind energy increased by 60.4 GW in 2019,

Fig. 2. Transition towards an inverter-dominated power system.

Fig. 3. Catalogue of storage technology.

Fig. 4. H2 production colour code.

Fig. 5. Hydrogen-based community: a modern vehicle energy network.

Project name Country Electrolyzer Hydrogen storage Fuel cell Ref.

Type Power Pressure Volume capacity Type Power

SWB Germany Alkaline 100 30 5000 PAFC 80 [118,119]

SCHATZ solar hydrogen USA Alkaline 6 5.7 30 [123,124]

SWB Project, Neunburg Germany Alkaline 100 5000 30 [119]

Freiburg Solar House Germany PEM 2 15 30 [125]

FIRST Spain PEM 1 30 _ PEM 0.42 [126]

HRI Canada Alkaline 5 120 40 PEM 5 [132]

HARI UK Alkaline 36 137 2856 PEM 7 [123,134]

Laboratory Plant IFE Norway PEM 1.5 14 16 [136]

HARI UK Alkaline 36 200 2856 PEM 7 [134]

PVFCSYS Agrate Italy Alkaline 3.4 4 10 [138]

HyDeploy UK PEM 500 20 _ [146]

control system of fuel cells and electrolyzers, intentionally excluding a

H2 ⟶ 2H+ + 2e− (1)

𝛥𝐸𝛺 = 𝐼𝑅𝑓 𝑐 (6)

where 𝑅𝑓 𝑐 is the resistance of PEM fuel cells’ electrode and electrolyte.

𝑉𝑓 𝑐 = 𝐸 − 𝛥𝐸𝑎 − 𝛥𝐸𝜌 − 𝛥𝐸𝛺 (7)

Fig. 9. Schematic of grid-connected ELZ/FC system.

Fig. 10. H2 FC/ELZ integrated DC/DC converter.

Fig. 13. PI controlled (a) ELZ and (b) FC.

Fig. 14. Small signal control loop for PI controller design.

Fig. 15. Speed of response versus system bandwidths.

Fig. 16. Cascaded PI controlled ELZ and FC.

makes the power electronic system an appropriate system where SMC

4.1.2. Sliding mode control techniques d𝑖𝑓 𝑐 𝑅𝑑 𝑢𝑓 𝑐 − 1 𝑉𝑓 𝑐

Fig. 20. Trajectory of a conventional SMC system.

Fig. 21. Standard procedure of designing SMC for a nonlinear system.

Fig. 22. The phase portraits of conventional SMC and TSMC.

Fig. 23. QRC-SMC-based H2 FC/ELZ DC/DC converter design steps.

Fig. 24. Backstepping controlled PEM FC.

Fig. 25. A traditional view of the model predictive controller.

Fig. 26. Implicit model predictive controller.

Fig. 27. General control block diagram of fuzzy logic controller.

where 𝑊𝑖 is the multiplied signals and output. Layer 3 is the normaliza-

Fig. 30. GA-ANFIS controller: sequential steps of working procedure.

Fig. 31. Architecture of the controller based on GA-ANFIS.

i. Flexible and easy i. Does not take account

Fig. 32. DC link voltage control by PI controller.

Fig. 33. Different types of hydrogen storage methods.

metal hydride unit to provide hydrogen to the fuel cell. To achieve

Fig. 36. Structure of hybrid energy storage system method.

Fig. 39. Grid connected inverter power sharing control system.

Parameters to be Direct-quadrature (dq) vector control Control of voltage magnitude

Phase-locked loop It is required It may use PLL control

Black-start Cannot black-start Can black-start a power system

System inertia Reduces system overall inertia Provides virtual inertia

Fig. 41. Classification of grid forming inverter.

Fig. 42. The 𝑃 − 𝑓 droop control.

You might also like