Review of Proton Exchange Membrane Fuel Cell-Powered Systems for Stationary Applications Using Renewable Energy Sources

Abstract

The telecommunication industry relies heavily on a reliable and continuous power supply. Traditional power sources like diesel generators have long been the backbone of telecom infrastructure. However, the growing demand for sustainable and eco-friendly solutions has spurred interest in renewable energy sources. Proton exchange membrane (PEM) fuel cell-based systems, integrated with solar and wind energy, offer a promising alternative. This review explores the potential of these hybrid systems in stationary telecom applications, providing a comprehensive overview of their architecture, energy management, and storage solutions. As the demand for telecommunication services grows, so does the need for a reliable power supply. Diesel generators are linked with high operational costs, noise pollution, and significant greenhouse gas emissions, prompting a search for more sustainable alternatives. This review analyzes the current state of PEM fuel cell systems in telecom applications, examines the architecture of microgrids incorporating renewable energy sources, and discusses optimization methods, challenges, and future directions for energy storage systems. Critical findings and recommendations are presented, highlighting objectives and constraints for future developments. Leveraging these technologies can help the telecom industry reduce fossil fuel reliance, lower operational costs, minimize environmental impact, and increase system reliability.

1. Introduction

Renewable energy sources, particularly solar and wind power, present viable options for generating the necessary electricity to produce hydrogen for PEM fuel cells. Due to the intermittent nature of renewable energy, there are spatial and temporal gaps between energy availability and consumption. Solar energy and wind turbines can be used to produce hydrogen via electrolysis, which can then be stored and utilized by PEM fuel cells to generate power on demand. To address these issues, it is necessary to develop suitable energy storage and conversion systems for the power grid [1]. Integrating these technologies into a cohesive microgrid system offers a sustainable and resilient solution for powering telecommunication infrastructure. Proton exchange membrane (PEM) fuel cells have emerged as a promising technology to meet these needs. PEM fuel cells generate electricity through an electrochemical reaction between hydrogen and oxygen, producing only water and heat as by-products. This clean energy technology offers several advantages over conventional power sources, including higher efficiency, lower emissions, and the potential for integration with renewable energy systems.

PEM fuel cell systems, combined with solar and wind energy, offer a sustainable and reliable solution for powering stationary telecom infrastructure. While challenges remain, ongoing advancements and increasing adoption of renewable energy sources are covering the way for broader implementation. This hybrid approach not only enhances energy security but also contributes to the reduction in greenhouse gas emissions, aligning with global sustainability goals.

Several projects and studies have demonstrated the feasibility and benefits of integrating PEM fuel cells with solar and wind energy for telecom applications. For instance, a solar–hydrogen system for telecom applications involves using photovoltaic (PV) panels to generate electricity from sunlight.

The following sections will delve into the literature on microgrid systems, dissecting their architecture and critically analyzing their constituent components, with a particular emphasis on energy management and storage systems. Subsequently, this text will explore optimization methods to enhance system performance and address the challenges and potential of energy storage solutions.

Research has demonstrated the potential of this system in powering remote telecom towers, especially in regions with unreliable grid access or where diesel fuel costs for backup generators are high [2].

Building upon the existing research on microgrid systems [3], the role of energy management systems (EMSs) in optimizing power flow and ensuring system stability within the hybrid framework will be explored. Additionally, an analysis of different energy storage options will be conducted, with a specific focus on comparing hydrogen storage to other methods such as batteries [4].

Through a comprehensive review of the research and advancements in technology, the aim is to assess the potential of this hybrid system to revolutionize the way we power our homes, businesses, and critical infrastructure like telecommunication towers. Existing challenges and future considerations for this promising technology, paving the way for a clean and reliable energy future, are discussed.

2. Microgrid System and Architecture

Microgrid systems, which integrate various distributed energy resources, have emerged as a viable solution for powering telecom infrastructure. Previous surveys have highlighted the effectiveness of microgrids in enhancing energy reliability and sustainability. This section delves into the architecture of microgrids, focusing on energy management and storage systems.

Microgrid systems have gained significant attention in recent years to enhance the reliability and sustainability of power supply for various applications, including telecommunication infrastructure. This chapter provides a comprehensive review of the existing literature on microgrid systems, focusing on their design, components, and integration with renewable energy sources.

A solar–hydrogen microgrid represents a cutting-edge approach to achieving sustainable and reliable energy solutions by integrating solar power with hydrogen energy storage. This hybrid system capitalizes on the strengths of photovoltaic (PV) solar panels and the flexibility of hydrogen storage, offering a resilient and environmentally friendly energy infrastructure.

Microgrids are localized energy systems that can operate independently or in conjunction with the main power grid. They consist of various distributed energy resources (DERs) such as solar panels, wind turbines, and fuel cells, along with energy storage systems [Figure 1]. The primary goal of a microgrid is to ensure a reliable and efficient power supply, particularly in remote or off-grid locations. Several studies have explored the potential of microgrid systems in different contexts.

Figure 1 illustrates the structure of the autonomous microgrid backup power system for stationary application, detailing its principle scheme for both DC and AC bus configurations in stand-alone mode. To ensure the optimal and efficient operation of the autonomous microgrid, the system’s dimensioning and structure were designed using HOMER Pro (Hybrid Optimization of Multiple Energy Resource) software. This system includes photovoltaic (PV) and wind turbine (WT) production, a battery storage system, consumers, and a hydrogen subsystem comprising fuel cells (FCs), an electrolyzer, and a hydrogen tank.

All renewable energy sources (RESs) and consumers, along with their associated energy converters, are connected in parallel to the AC(DC) bus. The energy management system (EMS) is linked to each unit through communication networks to establish proper control and management. The EMS is proposed to enhance the economy and robustness of the entire system, adding value to the developed concept. Integrating a photovoltaic energy source and wind turbines with hydrogen as an energy storage system result in a reliable energy source that is environmentally friendly and reduces overall maintenance costs.

In this hybrid system, hydrogen is produced by an electrolyzer powered by excess electricity from renewable energy sources. Hydrogen is then used to generate electricity in fuel cells, acting as a secondary energy source during periods of high demand. Simultaneously, the battery storage system is used to maintain power balance and system stability. This configuration not only ensures a consistent power supply but also contributes to a sustainable and cost-effective energy solution.

PEM fuel cells play a crucial role in this solar–hydrogen stand-alone system. Their integration into solar–hydrogen microgrids represents a significant advancement in future energy systems and the ongoing energy transition, providing a sustainable and efficient solution for energy generation and storage. In solar–hydrogen microgrids, the primary energy source is solar power, which is harnessed through photovoltaic (PV) panels. However, solar energy is intermittent, varying with weather conditions and time of day. To address this variability, solar–hydrogen microgrids incorporate energy storage systems that can store excess solar power generated during periods of high irradiance and supply power during periods of low or no solar output. PEM fuel cells play a crucial role in this context by providing a reliable and efficient means of energy storage and conversion. One of the primary functions of PEM fuel cells in solar–hydrogen microgrids is to provide a stable and continuous power supply. By converting stored hydrogen into electricity, PEM fuel cells can effectively balance the intermittent nature of solar power, ensuring that the microgrid can meet energy demands even when solar irradiance is low. This capability is particularly valuable in remote or off-grid locations where a consistent and reliable power supply is critical. As the demand for clean and reliable energy solutions continues to grow, the role of PEM fuel cells in solar–hydrogen microgrids will become increasingly important, contributing to the advancement of sustainable energy technologies and the transition to a low-carbon future. Continued research and development in PEM fuel cell technology and its integration with solar–hydrogen microgrids are essential to fully realize their potential and maximize their benefits.

PEM fuel cells operate through an electrochemical reaction in which hydrogen (H₂) and oxygen (O₂) combine to produce electricity, water (H₂O), and heat. The key reactions occurring in a PEM fuel cell are as follows:

Anode (Oxidation): H₂ → 2H⁺ + 2e⁻

(1)

Cathode (Reduction): 1/2O₂ + 2H⁺ + 2e⁻ → H₂O

(2)

Overall Cell Reaction: H₂ + 1/2O₂ → H₂O

(3)

Both electrodes generally comprise a gas diffusion layer (GDL) with a thin catalyst layer (CL) coating at the electrode–electrolyte interfaces where the reactions take place, as illustrated in Figure 2. The GDLs are porous materials, typically made of carbon fiber, located adjacent to the catalyst layers. The GDL provides structural support to the CL and contributes to removing water produced as a by-product from the catalyst layer.

Sunay Turkdogan [5] provides a detailed and well-researched examination of a renewable-based hybrid energy system for residential and transportation applications. The system is optimized using HOMER software, which simulates various configurations to find the most cost-effective and efficient setup. Manuel Castañeda et al. [6] conducted an extensive study on the architecture, optimization, and management of a hybrid energy system. The system combines photovoltaic (PV), hydrogen storage, and battery storage to create a reliable and efficient stand-alone energy solution that harnesses renewable energy sources.

Numerous case studies demonstrate the successful implementation of microgrid systems in various applications. For instance, a study conducted by the National Renewable Energy Laboratory [7] (NREL) analyzed a microgrid system for a remote telecom tower that included solar panels, a wind turbine, a PEM fuel cell, and battery storage. The study found that the hybrid system reduced fuel consumption by 70% compared to a diesel-only system and significantly lowered operational costs. The literature review indicates that microgrid systems offer a viable and sustainable solution for powering telecommunication infrastructure, especially in remote and off-grid locations.

Future research should focus on optimizing the design and operation of microgrid systems, addressing challenges related to energy storage, and exploring innovative approaches to integrating diverse energy sources. By doing so, microgrids can play a crucial role in advancing the sustainability and resilience of the telecommunication industry.

2.1. Energy Management System (EMS) in Microgrids

The EMS plays a pivotal role in ensuring the efficient and reliable operation of hybrid systems integrating solar, wind, hydrogen, and fuel cell technologies. Arshad Nawaz et al. [8], published in Applied Energy, provide an in-depth examination of the multi-microgrids (MMGs) concept, emphasizing its potential to enhance the integration of renewable energy sources (RESs) and improve power system efficiency, reliability, and stability. The study addresses the significant role of energy management systems (EMSs) and demand response programs within MMGs, highlighting their economic benefits and operational challenges.

The EMS in a microgrid performs several key functions such as Load Balancing, Peak Shaving, Renewable Integration, Battery Management, and Grid Interaction.

Several technologies and approaches are employed in the development and implementation of EMSs in microgrids [

Table 1. EMS technologies and approaches.

SN	References	Highlights
1	Guan, X. et al. [9], 2010	Algorithms such as model predictive control (MPC), genetic algorithms (GAs), and particle swarm optimization (PSO) are used to optimize energy distribution and resource allocation within the microgrid.
2	Zhou, Y. et al. [10], 2014	Real-time sensor and smart meter data are analyzed to inform energy management decisions, predict demand and supply fluctuations, and optimize energy resource operations.
3	Palensky, P. et al. [11], 2011	Demand response adjusts power demand rather than supply, using strategies like load shedding and shifting to align consumption with available supply.
4	Dongfang Chen et al. [12], 2022	A novel approach for predicting the performance degradation of proton exchange membrane (PEM) fuel cells utilizes a bidirectional Long Short-Term Memory (Bi-LSTM) neural network, optimized by a Bayesian algorithm, to enhance voltage prediction accuracy in PEM fuel cells.
5	Pavitra Sharma et al. [13], 2022	Classifies energy management systems (EMSs) by supervisory control, operating time platform, and decision-making strategies. Examines optimization techniques, including conventional programming (linear, nonlinear, mixed-integer), metaheuristic (particle swarm optimization, genetic algorithm), and AI-based methods (fuzzy logic, neural networks).
6	Mahmudul Hasan et al. [14], 2023	The article reviews microgrid control mechanisms, focusing on centralized, decentralized, and distributed strategies. It highlights the importance of effective monitoring for real-time management and discusses using IoT and advanced communication technologies to enhance microgrid monitoring and control.
7	Gugulothu et al. [15], 2023	Presents a study on the configuration and control of a DC microgrid with photovoltaic (PV) systems, fuel cells, and battery energy storage systems (BESSs). Fuel cell output is controlled based on the BESS’s state of charge (SOC) to optimize hydrogen usage under varying loads. Adjusting hydrogen and oxygen pressures in the fuel cell enhances power output, efficiently meeting additional load demands.
8	Atul S. Dahane et al. [16], 2024	Focuses on Demand Side Management (DSM) and its importance in modern energy systems, highlighting the integration of intelligent energy systems and smart loads to improve energy efficiency and grid stability.
9	Peng Wu et al. [17], 2024	The article explores the integration of Internet of Things (IoT) and digital twin technology in the management of hybrid microgrids (HMGs) with a focus on achieving net-zero energy.
10	Reagan Jean Jacques Molu et al. [18], 2023	The article presents an optimization-based energy management system (EMS) designed for grid-connected photovoltaic (PV)/battery microgrids, addressing the challenges of uncertainty in energy generation and demand. Utilizes linear programming (LP) and mixed-integer linear programming (MILP) for optimization, aiming to minimize operating costs.
11	Ali Reza Abbasi et al. [19], 2023	This review provides a comprehensive analysis of energy management strategies in MGs, focusing on advancements from 2009 to 2022. It systematically classifies various methods based on techniques, control strategies, and structures, highlighting the integration of renewable energy resources and energy storage systems. Enhancing real-time monitoring, predictive maintenance, and system optimization using IoT-enabled sensors and digital twin simulations.
12	Long Phan Van et al. [20], 2023	Analyzed different optimization methods used in EMSs, such as fuzzy logic control, model predictive control, heuristic and metaheuristic algorithms, stochastic and robust programming, and hybrid approaches. Detailed the objectives (technical, economic, environmental) and constraints (energy storage, power capacity, transmission) considered in designing EMSs for hydrogen-based microgrids.

].

Energy management systems may have numerous objective functions, some of which are suitable for specific cases. Tajjour and Chandel [21] categorize the objective functions for microgrid planning into four primary areas: social, economic, environmental, and technical. These categories encompass a wide range of factors, including but not limited to human development indicators, financial costs, environmental impact metrics, and system performance indicators. To effectively balance these often-conflicting objectives, suitable optimization algorithms are required.

The reviewed studies illustrate the dynamic and multifaceted nature of EMSs in microgrids, encompassing a wide range of optimization techniques, real-time data analysis, demand response strategies, and advanced control mechanisms. The table provides a comparative overview of various energy management systems (EMSs) for microgrids, highlighting key methodologies and their respective focuses.

The comparative analysis of EMSs in microgrid demonstrates a diverse range of methodologies and technologies aimed at enhancing energy distribution, resource allocation, and system reliability [

Table 2. List of some selected publications with their goals, aims, methods, and results.

Publication	Goals and Aims	Methods and Conditions	Some Results
Nguyen et al. [22], 2024	- Microgrid planning - Scheduling the operation of the microgrid - Minimizing the overall operating costs of microgrid and operational cost of the tank system	- Two-layer optimization model - Connected to main grid; bidirectional energy exchange is possible	The proposed model reduced the overall operating costs by 42.75% compared against single-level model
Phan-Van et al. [23], 2023	- Finding optimal size of a hydrogen-storage-based microgrid - Cost minimization of the microgrid - Contains battery and tank	- Comparison of 8 metaheuristic optimization algorithms (artificial bee colony, biogeography-based optimization, genetic algorithm, harmony search, invasive weed optimization, particle swarm optimization, shuffled complex evolution, and teaching–learning based optimization)	The particle swarm optimization algorithm was the best, reducing annual system costs by 25.3% compared to the worst energy management strategy in algorithm testing.
Ammara et al. [24], 2024	- A mathematical model for DC microgrids - Ensuring and maximizing power production	- Artificial neural network	The simulations showed that load demands were met efficiently along with the global asymptotic stability of the system
Rezaei et al. [25], 2024	- Energy management for isolated multi-energy microgrids which were used for hydrogen refueling stations	- A two-layer framework for islanding operation of a multi-energy microgrid, mixed-integer linear programming	The simulations showed that the model decreased the expected operational cost
Kilic [26], 2024	- Energy management system - Improving power quality - Reducing operational costs - Optimization of hydrogen production	- A modified version of the Symbiotic Differential Whale Optimization Algorithm	The solution maintains DC-Link voltage stability despite varying renewable energy production and provides substantial cost reduction
Yousri et al [27], 2023	- Minimize the electricity and battery degradation costs, customers’ discomfort, and peak-to-average ratio - In four-scenario testing: good, bad, and average weather scenarios and a forecasted weather profile	- A model for energy management and demand response for hybrid energy storage system consisting of batteries and hydrogen tanks - A multi-objective artificial hummingbird optimizer	The model increased customers’ savings and decreased greenhouse gas emissions in all four scenarios
Shi et al. [28], 2023	- Self-consistent microgrid with hydrogen storage for transportation	- Particle swarm optimization	The storage stabilizes varying energy production from the wind and provides adjustability
Abdelghany et al. [29], 2024	- Predictive control for islanded and grid-connected modes - Considering daily as well as real-time markets (which have different timescale)	- Advanced model predictive control strategy	Tested on lab scale The solution minimizes bidirectional exchange with grid and operational costs. It is also capable in four modes: islanded, grid-connected, and exchange of energy modes
Zhang et al. [30], 2022	- Electric–hydrogen hybrid refueling stations and DC microgrids - Voltage stability and reliable operation	- Fuzzy logic controller	Fuzzy controller provided optimal power allocation, and its use improved the lithium battery service life and hydrogen safety
Chamout et al. [31], 2024	- Requirements of the hydrogen purification - Off-grid household with availability of solar and wind electricity - Component costs and capacities	- A decision algorithm	The purification unit uses a small amount of energy but its operations affect the sizing of the components
Lyu et al. [32], 2024	- Global optimality for both sizing and scheduling of microgrid	- Combined solution methodology (Rolling Horizon Optimization and particle swarm optimization)	The method converges within 30 generations and reaches a global satisfactory solution; time-of-day tariff strategy is important
Sun et al. [33], 2024	- Operation planning for a microgrid	- A multistage stochastic mixed-integer program - A nested decomposition algorithm based on stochastic dual dynamic integer programming	The planning strategy was shown to have economic benefits and was successful for the seasonal and intraday dynamics of the system
Zhu et al. [34], 2024	- Optimal control framework containing energy management, economic optimization, and power regulation	- Distributed economic model predictive control scheme - A mixed-integer nonlinear programming algorithm	Simulations under varying irradiance and load conditions showed the model to be suitable for photovoltaic–hydrogen DC microgrid
Fang et al. [35], 2022	- Supply electricity and hydrogen and heating loads - Minimizing operational costs	- Day-ahead energy scheduling and model predictive control	Multiple-timescale energy management was shown to be suitable for multi-energy microgrid
Dong et al. [36], 2023	- Hydrogen-based microgrid fuel cell electric buses - Economic feasibility	- A two-stage robust optimization formulation with integer corrective decisions	The proposed energy management solution demonstrates significant improvements over a benchmark method; notably, it reduces mean daily operational costs by 37.08%
Shen et al. [37], 2022	- A zero-carbon microgrid	- Capacity planning and operation strategies	Operational reliability and economic feasibility are verified in a village
K/bidi et al. [38], 2022	- Solving a unit commitment problem due to different constraints of system components - A multistage power and energy management strategy	- Distributed explicit model predictive control - Mixed-integer quadratic programming	The proposed method is shown to avoid inadequate start-up of fuel cells and electrolyzers
Fang et al. [39], 2024	- The electricity–heat–hydrogen supply–demand balance and demand uncertainties	- The day-ahead scheduling stage, model predictive control (intraday rolling stage), and intraday real-time adjustment stage (markets)	Proposed methodology effectively addresses the challenges of balancing electricity, heat, and hydrogen supply and demand, especially in the face of uncertain conditions
Huangfu et al. [40], 2023	- Global optimal power distribution scheme - Rule-based judgment for reducing complexity of control	- A subsection bi-objective optimization dynamic programming strategy - A multi-objective genetic algorithm strategy	The solution improved photovoltaic utilization and fuel economy

]. Future developments in AI, IoT, and digital twin technologies are crucial for improving the real-time monitoring, predictive maintenance, and overall efficiency of microgrid operations.

Several case studies highlight the effectiveness of the EMS in microgrids. In urban settings, the EMS has been employed to enhance energy efficiency and reliability. A study by Navigant Research (2020) demonstrated that urban microgrids equipped with an advanced EMS could reduce energy costs and emissions significantly. EMSs have been crucial in the operation of microgrids powering remote telecom towers. For example, a study by NREL (2018) on a hybrid renewable energy system for a telecom tower showed that the EMS optimized the use of solar and wind energy, along with battery storage, resulting in a 70% reduction in diesel consumption. Community-based microgrids utilize EMSs to manage the collective energy resources of multiple households and businesses. This enhances the resilience and sustainability of the local energy supply [41]. Future research should focus on developing more robust and scalable EMS solutions, enhancing interoperability through standardization, and improving cybersecurity measures. By addressing these challenges, EMSs can further enhance the efficiency, reliability, and sustainability of microgrids.

2.2. Energy Storage System in Microgrids (Low- and High-Pressure Hydrogen Storage)

Microgrid systems may contain both battery and hydrogen storage to provide adequate supply without the electric grid. This has been demonstrated by many researchers (e.g., Chamout et al. [31]). Different phases of hydrogen used in storage can be naturally utilized with a battery. Jahanbin et al. [42] utilized hydrogen gas and a metal hydride battery with battery storage for a building using solar energy. Both storage solutions increased the annual renewable energy ratio which was increased further by adding battery storage.

There are already many reviews on issues related to hydrogen (e.g., Tajjour and Chandel [21], Ge et al. [43], and Khare et al. [44] studied the advantages and disadvantages of hydrogen energy and storage for a power system). This review provides detailed technical aspects of hydrogen production and storage [

Table 3. Review of selected publications on hydrogen storage systems.

Publication	Goals and Aims	Methods and Conditions	Some Results
Van et al. [45], 2023	- Energy management strategy for renewable energy microgrid with hydrogen storage system	- A state machine-based strategy combined with a hysteresis band control strategy - Connected to main grid; bidirectional energy exchange is possible	- Balanced supply and demand within the microgrid. - Extended lifespan of electrolyzer and fuel cell. - Maintained appropriate storage levels for battery and hydrogen.
Bouaouda and Sayouti [46], 2024	- A framework for microgrid size optimization and performance assessment	- Quantum-based Beluga Whale Optimization	- Simulated microgrid with renewable sources and hydrogen storage. - Demonstrated capability to supply energy to a remote off-grid site.
Karimi [47], 2024	- Sustainable scheduling of hybrid hydrogen power systems - Minimize the total costs, carbon emission, and peak load	A stochastic three-objective optimization mode and min-max approach	- Hydrogen tank enhanced system flexibility. Excess renewable energy converted to hydrogen. - Min-max approach increased load factor from 77.22% to 83.48%.
Abdelghany et al. [48], 2024	- Energy management strategy for a microgrid with wind–hydrogen strategy - Short-term and long-term operations (load demand, maximize the revenue, minimizing operation costs)	- A hierarchical model predictive control, mixed-logic dynamic framework, and a mixed-integer linear program	- Wind energy surplus converted to hydrogen. - Hydrogen stored for later use. - Hydrogen utilized in grid-islanded and connected operations.
Wu et al. [49], 2024	- Energy management for a residential microgrid - Combatting long-term and short-term uncertainty introduced by renewable sources	- Proposed hierarchical on-line energy management for hybrid hydrogen–electricity storage - Operating time: battery storage for every minute and hydrogen storage hourly; the hydrogen energy storage was for long-term, seasonal energy variation.	-Reduced overall energy cost by 21% compared to battery-only system.
Giovanniello and Wu [50], 2023	- Sizing components for a microgrid with 100% wind energy (Canada) - Long- and short-term storage	- A mixed-integer linear programming model - Lithium-ion batteries and hydrogen to solve issues in the short and long term	- Hybrid storage reduced costs significantly. - Hybrid storage is more cost-effective than single storage. - Lithium-ion battery costs dominated by energy storage capacity. - Hydrogen system impacted overall microgrid costs. - Lower electrolyzer efficiency significantly increased total microgrid cost. - Improved fuel efficiency primarily reduced total system costs. - Hydrogen system provided less energy than batteries but was crucial during peak demand periods.
Yousri et al. [27], 2023	- Minimize the electricity and battery degradation costs, customers’ discomfort, and peak-to-average ratio - In four-scenario testing: good, bad, and average weather scenarios and a forecasted weather profile	- A model for energy management and demand response for hybrid energy storage system consisting of batteries and hydrogen tanks - A multi-objective artificial hummingbird optimizer	- Model increased customer savings and reduced greenhouse gas emissions across all scenarios.
Abdelghany et al. [48], 2024	- Renewable energy was utilized in hydrogen production - Several hydrogen tanks - Autonomous operation without utility grid - Economic and operational costs, degradation aspects, physical constraints, and demands as well as smoothing variations in renewable energy production	- Model predictive control (also known as hierarchical rolling horizon control) for managing a hydrogen energy storage system in an islanded wind/solar microgrid	- Microgrid capable of independent operation. - Multiple hydrogen tanks enable long-term storage. - System complexity increased with hydrogen storage. - Hydrogen storage might not suffice for extended periods. - System performance validated through laboratory testing.
Li et al. [51], 2024	- Increasing safety and stability of large power grids because of uncertainty in energy production using wind and solar - Different storage solutions - Economic costs -Multigrid system	- Distributional robust approximation solving - A combination of the Bonferroni test and Conditional Value at Risk	- Optimal solution: independent hydrogen storage and shared battery storage. - Power transmission between microgrids favored simultaneous storage configuration. - Hydrogen storage investment cost significantly impacted overall costs.
Deng et al. [52], 2023	- A multi-microgrid system which shared electric–hydrogen storage - Combination of cooling, heating, and power systems	- Bi-layer optimization configuration model; the upper layer is energy storage capacity issues; the lower level is for multi-microgrid optimization	- Shared storage outperformed individual storage. - Reduced battery capacity by 75.94%. - Decreased daily operating costs by 11.53%. - Hybrid energy storage yielded further improvements. - Increased daily net income by 61.67%. - Reduced battery capacity by an additional 67.13%. - Decreased daily operating costs by 3.39%. - Achieved a payback period of 1.6 years.
Shi et al. [28], 2023	- Self-consistent microgrid with hydrogen storage for transportation	- Particle swarm optimization	Storage balances fluctuating wind energy output and enables grid flexibility.
Qiu et al. [53], 2024	- Optimal scheduling of microgrids with coordinated long-term and short-term energy storage - Economic optimization	- Mixed-integer linear programming which solved using the Yalmip/Gurobi commercial solver	- Combined battery and hydrogen storage for effective cross-seasonal energy management in microgrids.
Naseri et al. [54], 2022	- Controlling an islanded microgrid (stand-alone/off-grid microgrid) - Green hydrogen production, storage, and re-electrification	- Two-layer hierarchical control	Simulation results indicate that green hydrogen production is feasible using solar energy; hydrogen is stored for later electricity generation when sufficient pressure is built up.
Shao et al. [55], 2023	- Multi-energy off-grid microgrids for hydrogen - Economics and resilience	- A two-stage risk-constrained stochastic programming; the first stage was about energy resource configuration optimization, and the second stage considers long-term economics and on-emergency feasibility verification - Risk constraints via sampling approximation strategy	Hydrogen storage provides both short-term and seasonal energy storage capabilities, contributing to overall cost reduction.
Er et al. [56], 2024	Sizing microgrid for a grid–vehicle–grid approach to minimize the life cycle cost and maximize the system reliability	- Two-stage stochastic programming with a scenario-based approach	- Hybrid storage system most cost-effective with low power supply loss. - Increased power supply loss allowed for more economical solutions.
Xiang et al. [57], 2021	- Zero-emission airport operation - Techno-economic benefits	- A mixed-integer linear programming optimization method and life cycle theory	- Hydrogen system reduced annual costs by over 40% and greenhouse gas emissions by over 65%.
Kumar et al. [58], 2022	- Stand-alone microgrids (off-grid energy providers) can obtain 100% of their energy from renewables - Use of metal hydride-based hydrogen energy storage system - Optimal sizing of components	- Optimization performed via simulation software HOMER	Hybrid systems typically require smaller storage capacities compared to wind-only microgrid systems.
Li et al. [59], 2023	- Analysis of hydrogen energy storage and battery based on the levelized cost of storage, carbon emissions, and uncertainty assessments	- Monte Carlo method	- Hydrogen production using alkaline electrolyzers, pipeline delivery, and refueling demonstrated the lowest cost of 0.227 USD/kWh and CO2 emissions of 61.63 gCO2e/kWh. - Large-scale storage systems yielded comparable results.
Xu et al. [60], 2024	- Hydrogen storage for a pumping unit using renewable energy	- Simulations and control strategy	- Stand-alone operation for up to 72 h. - Energy conversion efficiency of 35%. - Proven through real-world implementation.
Pignataro et al. [61], 2024	- Modelling a power-to-gas system (synthetic methane using wind energy)	- Management strategy incl. storage - Mathematical modelling	The larger size of storage leads to better performance.
Abdolmaleki and Berardi [62], 2024	- Solar and hydrogen energy for a single-house and a midrise apartment	- HOMER software	- Simulated systems: PV–battery, PV–hydrogen, and PV–battery–hydrogen. Best economical configuration: - 522 kW photovoltaic (PV) panels; - 150 kW electrolyzers; - 20 kW fuel cell; - 200 kg hydrogen tank; - 18.6 kW converter; - 159 batteries.
Liu et al. [63], 2024	- Design of large-scale hydrogen storage pipeline networks - Costs	- A mixed-integer nonlinear optimization mode - Hybrid genetic algorithm that combines the Modified Feasible Directions Method and Genetic Algorithm Theory	- Investment of the pipeline network is affected by wellhead temperature more than ambient temperature.
Dong et al. [36], 2023	- Hydrogen-based microgrid fuel cell electric buses - Economic feasibility	- A two-stage robust optimization formulation with integer corrective decisions	- Hydrogen-based microgrids can serve as operation centers for fuel cell electric buses. - This integration offers economic feasibility and contributes to decarbonization.
K/bidi et al. [38], 2022	- Solving a unit commitment problem due different constraints of system components - A multistage power and energy management strategy	- Distributed explicit model predictive control - Mixed-integer quadratic programming	- The proposed method effectively prevents premature and inefficient start-up of fuel cells and electrolyzers.

].

The comparative analysis of recent studies on hydrogen storage in microgrids reveals a diverse range of technical solutions and methodologies, each aimed at improving the efficiency, sustainability, and cost-effectiveness of energy management systems. The primary focus across these studies is on optimizing the use of renewable energy sources through innovative storage and control strategies while addressing the inherent challenges of variability and uncertainty in renewable energy production. For instance, Van et al. [20] and Abdelghany et al. [29] aim to balance supply and demand within microgrids and convert surplus renewable energy into hydrogen. Karimi [46] and Yousri et al. [27] focus on sustainable scheduling and reducing costs and emissions, while Giovanniello and Wu [50] and Li et al. [51] explore the economic feasibility and stability of hydrogen storage systems. A variety of optimization techniques and control strategies are employed across these studies. Methods such as stochastic optimization, model predictive control, mixed-integer linear programming, and quantum-based optimization are prominently featured. For example, Karimi [46] uses a stochastic three-objective optimization model, while Bouaouda and Sayouti [47] employ quantum-based Beluga Whale Optimization. Wu et al. [49] and Giovanniello and Wu [50] leverage hierarchical and mixed-integer linear programming models, respectively.

The results consistently demonstrate improvements in system performance, cost reduction, and energy efficiency. Van et al. [20] report balanced supply and demand and extended component lifespans, while Wu et al. [49] show a 21% reduction in overall energy costs with a hybrid storage system. Deng et al. [52] highlight significant reductions in battery capacity and daily operating costs with a rapid payback period. Furthermore, hybrid storage solutions, as examined by Giovanniello and Wu [50], prove to be more cost-effective than single storage options. While the technical solutions proposed are innovative and promising, several challenges remain for real-world application. The complexity and computational demands of advanced optimization methods, such as those used by Abdelghany et al. [48] and Shi et al. [28], can be significant barriers. Additionally, the high initial investment costs for hydrogen storage systems, as noted by Li et al. [59], may limit widespread adoption. There is also the challenge of ensuring system reliability and robustness in the face of renewable energy variability, which is crucial for practical deployment.

Future research should focus on simplifying these systems, reducing costs, and validating performance under real-world conditions to facilitate the practical integration of hydrogen storage in microgrids.

2.2.1. Critical Analysis of Hydrogen Storage System Approaches

The properties of an energy management system depend on the components and processes of the system, like the power generation process. Hydrogen has been shown to be good for power generation (e.g., Bai et al. (2024) [64]), which makes storage necessary in many cases. In this subsection, some important aspects of the hydrogen storage systems are briefly evaluated.

Osman et al. [65] explore recent advancements in hydrogen storage technologies, categorized into physical-based and material-based approaches. Physical storage methods, including compressed, liquefied, and cryo-compressed hydrogen, offer innovative solutions for enhancing hydrogen density and safety. Material-based storage methods focus on the utilization of metal hydrides, complex hydrides, and carbon-based materials like activated carbon, graphene, and carbon nanotubes, which employ absorption and adsorption techniques for hydrogen retention and release. The integration of computational chemistry, high-throughput screening, and machine learning has revolutionized the development of these storage materials, enabling the optimization of structural attributes, porosity, and stability. A more detailed description of storage methods can be found, e.g., in Osman et al. [65]. The selected storage methods determine, for example, which kind of tank is needed. Sometimes, the selection is limited by the environment because underground storage is not available or cannot be utilized, e.g., for environmental reasons.

Size parameters are very important for a storage tank. Since the hydrogen can be stored in different phases (gas, liquid, and material-based hydrogen), different storage solutions are needed. For example, Ba-Alawi et al. [66] designed a storage system for reverse osmosis using a safety-oriented multi-criteria optimization with design parameters like storage size, pressure, flow rate, and temperature. The tank size is thus affected by the storage method selected.

Different phases of hydrogen are typically at different temperatures. For example, liquid hydrogen requires a temperature no higher than −252.9 °C. If liquid hydrogen storage is used, then it needs to be insulated properly for safety and efficiency (Yin et al. [67]). Insulation is still under active development (e.g., Yin et al. (2024) [68]). Hydrogen storages have different pressures, which may in some cases be selected. In these cases, the selected pressure can have economic impacts: according to Hu et al. (2024) [67], the hydrogen utilization rate may effectively be improved by the nominal working pressure of the medium-pressure and high-pressure tanks for a refueling station.

Storage often requires permission and must follow local regulations, codes, and standards. For example, high-pressure hydrogen cylinders have similar but not the same regulations, codes, and standards in different countries (Li et al. [69]). The properties of hydrogen have to be taken into account in the storage structure and material as well as safety issues. For example, Meda et al. (2023) [70] state that the susceptibility to hydrogen embrittlement increases with strength and is thus higher for high-strength steels. Their suggestion to counter this embrittlement is multi-layered coatings.

2.2.2. Existing Challenges and Future of Energy Storage Systems

Energy storage systems still have many challenges. A review of global trends and future scenarios can be found in Pleshivtseva et al. [71]. Table 4 lists the challenges and future predictions found in publications. Tajjour and Chandel (2023) [21] identified a microgrid with battery storage challenges like cyber security threats, an optimal stable power flow without any constraint violation, energy management for Peak Shaving either via adding more power generation or load shedding\shifting, and optimal network configurations. Security may become a more important topic in the future including also the hydrogen storage systems.

Hassan et al. (2023) [72] discussed issues especially related to hydrogen storage systems. The main challenges of the systems are related to large-scale hydrogen production and storage: costs, scaling up, and lack of infrastructure. Storing hydrogen also has many challenges in addition to its costs (storage material, tanks and other infrastructure, transportation). Storage materials still need development as the current ones are not cheap or efficient enough. Hydrogen storage still has low energy density and requires specific environmental conditions like high pressure or low temperature. As hydrogen is a flammable and explosive substance, safety in a hydrogen system is a critical factor.

Schiaroli et al. [73] say that flammability is one of the issues regarding using hydrogen in the transportation sector. They present an alternative storage solution utilizing the Monte Carlo method. The highest security performance was obtained using cryogenic storage in the liquid phase at ambient pressure. Safety and cost issues are also considered by Sikiru et al. [74], Ghorbani et al. (2023) [75], and Ma et al. (2023) [76]. The lack of standardization and codes are hindering the development of storage systems. The codes and standards were also discussed by Abdalla et al. (2018) [77] as well as the weight, volume, cost, and efficiency of storage. Rasul et al. (2022) [78] considered hydrogen storage to be the main challenge for systems. Novel storage systems therefore need to be developed (e.g., Bosu et al. (2024) [79] and Rasul et al. (2022) [78]).

The size of hydrogen storage affects costs, so bigger storage should be developed (Moran et al. (2024) [80]). There is also a need for better materials (Hassan et al. (2024) [81]). Hannan et al. (2022) [82] studied a hybrid system to respond to short and long time demands. They also found efficiency, costs, and security to be important. Environmental issues related to hydrogen should be addressed carefully (e.g., Saadat et al. (2024) [83] and Higgs et al. (2024) [84]), as well as operational challenges (Bosu et al. (2024) [79]).

Saberi Kamarposhti et al. (2024) [85] investigated the use of AI for energy systems utilizing hydrogen while identifying several issues like data security. The use of AI can lead to the decentralization of energy systems. Van et al. (2023) [20] discussed the need for better forecasting for renewable power production as well as load demand for energy system management. These are important factors for microgrid control.

Table 4. Challenges and future of energy storage systems.

Publication	Challenges	Future
Tajjour and Chandel [21], 2023	Battery sizing and secure operation of microgrid	AI applications, technique development, blockchains, and Reinforcement Learning techniques
Hassan et al. [72], 2023	General challenges: scaling up hydrogen storage technologies, high cost of hydrogen production and storage, need for more extensive infrastructure, and low production efficiency Storage-specific challenges: low energy density, high-pressure or low-temperature requirements, safety concerns, and storage materials	Increase in clean and sustainable energy, high-density, efficient, and cost-effective hydrogen storage materials, and reduced storage volume
Schiaroli et al. (2024) [73]	Hydrogen flammability and safety	Better risk management and control of and reduction in risks
Ghorbani et al. (2023) [75]	Safe and efficient storage, operating conditions, and application	Improvements in goals, energy safety, and efficiency
Ma et al. (2023) [76]	Cost of large-scale hydrogen storage and high energy requirements for gas compression	Decrease in costs, a novel evaluation method of the technical and economic feasibility, and new infrastructure and storage
Abdalla et al. (2018) [77]	Weight, volume, cost, efficiency, codes, and standards Safe, reliable, and cost-effective	Advances in storage technologies and infrastructure
Rasul et al. (2022) [78]	Development of hydrogen storage Storage conditions of hydrogen Container material degradation	Novel storage systems Increase in demand for hydrogen
Bosu et al. (2024) [79]	Operational challenges, cost-effective storage, technical challenges, dehumidification of hydrogen, and storage system volume problem	New storage materials, catalytic doping and structural modification, accurate lifecycle analysis, and nanocomposites
Moran et al. (2024) [80]	Storage size and cost	Increase in storage size reduces overall levelized cost
Hannan et al. (2022) [82]	A hybrid storage system (capacity, long lifespan, low cost, high efficiency, and high security)	Extended lifetime, lower cost, and higher security
Saadat et al. (2024) [83]	Hydrogen interactions with microorganisms in underground storage, in situ reactions, and leakages	Resilience and environmental safety
Higgs et al. (2024) [84]	Potential contamination and/or changes to rock properties and leakages from hydrogen storage in porous media	More research on hydrogen reactivity, mobility through seals, gas mixtures, and storage site
SaberiKamarposhti et al. (2024) [85]	Data security and privacy, interoperability, and the technical constraints of AI for hydrogen systems	Decentralization and autonomous energy management system
Van et al. (2023) [20]	Forecasting power production and load demand	Multi-microgrid systems, environmental objectives, the utilization of other hydrogen roles, and accurate microgrid modelling
Sikiru et al. [75], 2024	A safe, dependable, and cost-efficient large-scale storage system needed, as well as reliability Lack of appropriate standards and codes	Safe and dependable performance Reduced cost in large-scale hydrogen utilization Improvement in durability and efficiency

Table 4. Challenges and future of energy storage systems.

Publication	Challenges	Future
Tajjour and Chandel [21], 2023	Battery sizing and secure operation of microgrid	AI applications, technique development, blockchains, and Reinforcement Learning techniques
Hassan et al. [72], 2023	General challenges: scaling up hydrogen storage technologies, high cost of hydrogen production and storage, need for more extensive infrastructure, and low production efficiency Storage-specific challenges: low energy density, high-pressure or low-temperature requirements, safety concerns, and storage materials	Increase in clean and sustainable energy, high-density, efficient, and cost-effective hydrogen storage materials, and reduced storage volume
Schiaroli et al. (2024) [73]	Hydrogen flammability and safety	Better risk management and control of and reduction in risks
Ghorbani et al. (2023) [75]	Safe and efficient storage, operating conditions, and application	Improvements in goals, energy safety, and efficiency
Ma et al. (2023) [76]	Cost of large-scale hydrogen storage and high energy requirements for gas compression	Decrease in costs, a novel evaluation method of the technical and economic feasibility, and new infrastructure and storage
Abdalla et al. (2018) [77]	Weight, volume, cost, efficiency, codes, and standards Safe, reliable, and cost-effective	Advances in storage technologies and infrastructure
Rasul et al. (2022) [78]	Development of hydrogen storage Storage conditions of hydrogen Container material degradation	Novel storage systems Increase in demand for hydrogen
Bosu et al. (2024) [79]	Operational challenges, cost-effective storage, technical challenges, dehumidification of hydrogen, and storage system volume problem	New storage materials, catalytic doping and structural modification, accurate lifecycle analysis, and nanocomposites
Moran et al. (2024) [80]	Storage size and cost	Increase in storage size reduces overall levelized cost
Hannan et al. (2022) [82]	A hybrid storage system (capacity, long lifespan, low cost, high efficiency, and high security)	Extended lifetime, lower cost, and higher security
Saadat et al. (2024) [83]	Hydrogen interactions with microorganisms in underground storage, in situ reactions, and leakages	Resilience and environmental safety
Higgs et al. (2024) [84]	Potential contamination and/or changes to rock properties and leakages from hydrogen storage in porous media	More research on hydrogen reactivity, mobility through seals, gas mixtures, and storage site
SaberiKamarposhti et al. (2024) [85]	Data security and privacy, interoperability, and the technical constraints of AI for hydrogen systems	Decentralization and autonomous energy management system
Van et al. (2023) [20]	Forecasting power production and load demand	Multi-microgrid systems, environmental objectives, the utilization of other hydrogen roles, and accurate microgrid modelling
Sikiru et al. [75], 2024	A safe, dependable, and cost-efficient large-scale storage system needed, as well as reliability Lack of appropriate standards and codes	Safe and dependable performance Reduced cost in large-scale hydrogen utilization Improvement in durability and efficiency

The reviewed studies highlight various challenges in hydrogen storage, which can be broadly categorized into technical, economic, and safety concerns. Each study provides insights into current issues and proposes future directions to address these challenges.

Several studies focus on the technical complexities of hydrogen storage. Tajjour and Chandel [21] address the challenge of battery sizing for secure microgrid operations, suggesting advancements in AI and blockchain technologies. Hassan et al. [72] and Bosu et al. [80] discuss the low energy density and high-pressure requirements of hydrogen storage, proposing new materials and nanocomposites to improve efficiency.

High costs and low production efficiency are significant barriers identified by Hassan et al. [72] and Ma et al. [76]. Future research is directed towards reducing these costs through novel evaluation methods and the development of more efficient storage materials. Moran et al. [80] highlight the impact of storage size on reducing overall costs, advocating for increased storage capacities.

Safety is a major focus, with Schiaroli et al. [73] and Ghorbani et al. [75] emphasizing the risks associated with hydrogen flammability and safe operating conditions. Improvements in risk management and safety protocols are essential. Environmental concerns, such as hydrogen interactions with underground microorganisms, are addressed by Saadat et al. [83] and Higgs et al. [84], who call for more research into hydrogen reactivity and mobility in storage sites.

Future research aims to overcome these challenges through technological innovations and infrastructure development. Abdalla et al. [77] and Rasul et al. [78] stress the importance of advancing storage technologies and infrastructure to ensure safe, reliable, and cost-effective hydrogen storage. Van et al. [20] and Sikiru et al. [74] highlight the need for accurate microgrid modelling and the establishment of comprehensive standards and codes.

3. Optimization Methods

Numerous studies propose various control algorithms to enhance the efficiency and stability of PEM fuel cell systems. These approaches not only reduce hydrogen consumption but also ensure more reliable power output, making them a promising solution for high-power applications.

Yurdagül Bentes Yakut [86] investigates the development of a new control algorithm designed to enhance the efficiency of proton exchange membrane (PEM) fuel cells. This research is particularly relevant for applications where high efficiency and reliability are paramount, such as in the automotive industry and renewable energy systems. This system aims to optimize hydrogen fuel consumption and stabilize output voltage by employing a boost converter controlled by a Proportional–Integral (PI) controller fine-tuned using the particle swarm optimization (PSO) method. The study concludes that the proposed control algorithm, combined with PSO-optimized PI controller parameters, offers a significant improvement in the efficiency and stability of PEM fuel cell systems.

H. Rezk et al. [87] present an innovative optimization approach aimed at improving the efficiency of PEM fuel cells. The focus of their research is on the application of the Equilibrium Optimizer (EO), a recent metaheuristic algorithm inspired by physical processes, to optimize the control parameters of PEM fuel cells. The paper outlines a detailed methodology where the EO algorithm is used to fine-tune the parameters of a fuzzy logic controller (FLC) integrated into the Maximum Power Point Tracking (MPPT) system of PEM fuel cells. The optimization process involves several phases, including initialization, equilibrium pool formation, and iterative concentration updates, to achieve a balance between exploration and exploitation in the search for optimal solutions. Jiankang Wang et al. [88] explore innovative methods to enhance the performance of PEM fuel cells. The research introduces a machine learning-assisted multiphysics numerical model (MNM-ML) to optimize the parameters that mitigate NGC while maintaining high performance. The authors compare nine state-of-the-art machine learning algorithms to identify the best model for this task.

By integrating advanced optimization techniques, the study contributes to the development of sustainable and resilient energy solutions, promoting the broader adoption of hybrid renewable energy systems in remote and off-grid locations.

Abdeljelil Chammam et al. [89] explore the optimization of a novel multi-generation system designed to produce electricity, cooling, heat, and freshwater. The system utilizes a PEM fuel cell to generate electricity. To optimize this system, the study focuses on two primary objectives: exergy efficiency and the total cost rate (TCR). The optimization process employs a genetic algorithm (GA) to identify the optimal operating conditions. This method involves adjusting various design variables such as the operating temperature and pressure and the current density of the fuel cell.

Hegazy Rezk et al. [90] present a robust methodology for the accurate parameter estimation of proton exchange membrane (PEM) fuel cells using the Gradient-Based Optimizer (GBO). The optimization process treats the unknown parameters of PEM fuel cells as decision variables. The objective function, which needs to be minimized, is the sum of squared errors (SSE) between the measured and estimated data. This research focuses on a precise parameter estimation strategy for PEM fuel cells, which is crucial for ensuring an accurate emulation of the fuel cell system characteristics. The proposed methodology leverages the Gradient-Based Optimizer (GBO) to identify optimal parameters for three types of PEM fuel cells: 250 W FC stack, BCS 500 W, and SR-12 500 W. The results highlight the GBO’s capability to deliver better accuracy and reliability in PEM fuel cell parameter estimation, making it a highly effective tool for optimizing PEM fuel cells’ performance.

Chengjun Guoa et al. [91] delve into various optimization strategies for improving the performance of proton exchange membrane (PEM) fuel cells. The focus is primarily on the application of advanced optimization algorithms to fine-tune critical parameters of PEM fuel cells, ensuring enhanced efficiency and operational reliability. The study evaluates several state-of-the-art optimization techniques, including particle swarm optimization (PSO), genetic algorithms (GAs), Differential Evolution (DE), and Gradient-Based Optimizer (GBO). These methods are applied to a comprehensive model of the PEM fuel cell to identify the optimal parameter settings that minimize errors between experimental data and theoretical predictions. Key optimization variables considered in this study include the membrane thickness, electrode properties, and operational conditions such as temperature and pressure. The optimization process aims to achieve a balance between maximizing the fuel cell’s power output and minimizing the total cost of operation, including the maintenance and replacement of components. The Gradient-Based Optimizer, in particular, shows promise as a leading tool for achieving these optimization goals, paving the way for more efficient and sustainable energy solutions.

Wei Zhao et al. [92] highlight the critical role of effective water management in the performance and durability of proton exchange membrane fuel cells (PEMFCs). The paper provides a comprehensive review of common water management issues such as flooding and membrane dehydration, which significantly impact the efficiency and lifespan of PEMFCs. The paper concludes that while significant progress has been made, there are still challenges and opportunities for further development in water management for PEMFCs. Future research should focus on improving diagnostic accuracy, developing cross-scale and multiphysics simulation studies, and advancing data-driven diagnostic methods for efficient and stable PEMFC operation. These efforts are crucial for the commercialization and broader application of PEMFCs in achieving carbon neutrality and sustainable energy solutions.

Abdullah G. Alharbi et al. [93] present a comprehensive approach to optimizing proton exchange membrane (PEM) fuel cells using a variety of advanced algorithms, with the primary aim of enhancing efficiency and reducing energy losses. The optimization strategies discussed include the Gradient-Based Optimizer (GBO), genetic algorithm (GA), hybrid optimization techniques, multi-objective optimization, and particle swarm optimization (PSO).

The Gradient-Based Optimizer (GBO) is highlighted for its ability to quickly converge on optimal solutions by leveraging Newton’s gradient-based principles. This method is particularly effective for complex engineering problems, showing superior performance in optimizing PEM fuel cell parameters compared to other algorithms. It asserts that these advanced optimization methods significantly improve the performance and efficiency of PEM fuel cells. The comprehensive comparison of different algorithms highlights the effectiveness of these strategies in achieving optimal system performance, making PEM fuel cells more viable for various applications, including off-grid and hybrid energy systems.

Dong fang Chen et al. [94] present a novel approach for predicting the performance degradation of proton exchange membrane (PEM) fuel cells. The authors utilize a bidirectional Long Short-Term Memory (Bi-LSTM) neural network optimized by a Bayesian algorithm to enhance the accuracy of voltage prediction in PEM fuel cells. The study addresses the challenges of the long-term operation of fuel cells, where performance degradation is a critical issue due to the aging of components such as the membrane electrode assembly. The study concludes that the proposed Bayesian-optimized Bi-LSTM neural network model offers a robust and accurate method for predicting the short-term performance degradation of PEM fuel cells, with potential applications in enhancing the reliability and efficiency of fuel cell systems in real-world scenarios.

4. Critical Findings and Future Recommendations Including Objectives and Constraints

The findings and recommendations can be roughly divided into the following categories: regulatory and policies, economics, storage and its properties, environmental issues, and applications, as well as management systems.

Hassan et al. [81] recommend policy and regulatory support which includes incentives and financial support from governments, changes in regulations, and international collaboration. Hannan et al. [82] found it necessary to change the energy market. The financial and economic factors are critical, as one of the issues in hydrogen usage in power production is the price. The price of hydrogen may drop to an economically suitable level of 1–2 USD/kg within a couple of decades (Rasul et al. [78]). Capital costs should decrease so that hydrogen storage systems would be an economic solution (de la Cruz-Soto et al. [95] and Diaz et al. [96]). Kilic [26] noted that hydrogen production has intrinsically slow dynamics from electricity; thus, the optimization of hydrogen production is needed from solar and wind.

Diaz et al. [96] found that the hydrogen approach reduced the total annual cost of the microgrid by 14.1%. The selection of technologies was influenced by market factors (e.g., investment and tariff structure costs) and customer characteristics (the load profile and the availability of renewable energy sources). Aba et al. [97] recommend utilizing hydrogen as an energy storage medium rather than a carrier. It is also suitable for regions where grid extensions are difficult as well as in industries which cannot otherwise reduce their emissions (hard to abate).

Le et al. [98] describe critical future research objectives. The very important parameters are efficiency, sustainability, safety, and economic feasibility. Hydrogen has significant and numerous obstacles; one of them is storage. Yang et al. [99] studied battery and hydrogen energy systems critically. The important factors which are lacking include the following: large storage capacity in limited space, frequent storage with rapid response, and continuous storage without loss. Batteries are good short-term storage but have serious limits for longer use like a self-discharge rate (>1%) and capacity loss (~20%). Hydrogen is better for longer use, but the problem is with low efficiency. Yang et al. (2024) [99] recommend using a hybrid energy system using batteries and hydrogen storage.

Er et al. [56] used two-stage stochastic programming with a scenario-based approach for sizing the microgrid (a grid–vehicle–grid approach). A hybrid storage system was found to be the most cost-effective option with a low loss of power supply probability. When a higher loss of power supply probability was allowed, it was a more economical solution.

Ba-Alawi et al. [66] identified the constraints of different phase storage systems: for gas storage systems, the long-term financials are a concern, e.g., due to high investment cost. In liquid storage systems, the constraints are liquefaction and boil-off operations and the necessary risk management. Hydrogen material-based systems require precise heat management and high investments.

Mehr et al. [100] highlight the critical role of hydrogen storage in the overall hydrogen value chain. Their research provides a comprehensive analysis of hydrogen storage technologies across various scales, considering technical maturity and economic viability. The life cycle cost analysis, feasibility, and safety are very important for systems. Giovanniello and Wu [50] also noted that high proportions of renewable sources increase supply–demand mismatches due to their variability. These mismatches can happen in multiple timescales, thus indicating the importance of storage. Hren et al. [101] evaluate several aspects for hydrogen: greenhouse gas and energy footprints, acidification, eutrophication, human toxicity potential, and eco-cost. They utilize the Life Cycle Assessment. One of their important findings was that storage and transport causes around 35.5% of the greenhouse gas emissions of the whole hydrogen chain.

Sikiru et al. [74] state that hydrogen still has adaptability constraints. Accidents related to hydrogen are causing negative public opinion which affects policies. Safety issues are thus one of the critical factors. Tariq et al. [102] provide an evaluation of fuel cells with a hydrogen tank for an emergency or backup power scenario. This evaluation found many factors like the possibility of an extra hydrogen tank influencing the system. Sadeq et al. [103] analyzed a hydrogen energy system and found it suitable for transportation, industry, and residential heating to reduce greenhouse gas emissions. To improve the economy of hydrogen systems, they recommend more research on improving hydrogen production as well as regulatory measures and incentives. It is also important to make rigorous safety regulations. Mohammad and Iqbal [104] developed a successful hybrid energy system for housing complexes in Srinagar, India, which indicates again the suitability of hydrogen systems for residential buildings. They recommended, e.g., flexible energy policies and grid extension for hydrogen-supported systems. The most cost-effective system produced, however, more greenhouse gas than the configuration with fuel cells.

Qiu et al. [105] identify high capital expenditure and low energy conversion efficiency as primary obstacles for power-to-gas technology. To address these challenges, they introduce green hydrogen-based energy storage as a service model. By simulating this model using Shanghai’s electricity market and weather data, the authors demonstrate a complete recovery of excess renewable energy and a substantial increase in renewable energy integration within microgrids, from 59% to 83%. To increase the use of hydrogen, they recommend governmental subsidies and support policies as well as co-operation between microgrids.

Van et al. [20] provide a comprehensive overview of energy management techniques applicable to microgrids, including fuzzy logic, model predictive control, heuristic and metaheuristic algorithms, and stochastic and robust programming methods. While each approach offers distinct advantages, such as the flexibility of fuzzy logic, they also come with challenges like high implementation costs. The authors further emphasize the multifaceted nature of microgrid control by outlining a range of critical factors, encompassing technical, economic, and environmental objectives, as well as operational and system configuration considerations.

Simulations for hydrogen refueling stations made by Hu et al. [67] indicate that continuous filling is necessary to make simulated and experimental values close to each other.

In future backup power energy systems, PEM fuel cells are recognized as a promising technology for microgrid applications due to their high efficiency, low emissions, and capability to provide reliable power. However, their widespread adoption is vulnerable due to degradation issues that affect their performance and lifespan. Xingwang Tang et al. [106] provide a comprehensive analysis of a degradation-adaptive energy management strategy for fuel cell hybrid electric vehicles (FCHEVs). This strategy focuses on understanding and managing the degradation processes of proton exchange membrane fuel cells (PEMFCs) to enhance their durability and performance over time. The authors conduct accelerated durability tests and temperature sensitivity tests to determine the temperature sensitivity characteristics of PEMFCs at different state-of-health (SOH) levels.

The utilization of PEM fuel cells in microgrid applications brings to light several critical issues concerning the raw and critical materials required for their production. PEM fuel cells rely on several key materials, including platinum for the catalysts, perfluorinated sulfonic acid (PFSA) for the membrane, and various carbon-based materials for the electrodes and gas diffusion layers. Platinum, in particular, is a crucial material due to its excellent catalytic properties, which facilitate the necessary electrochemical reactions. However, platinum is an expensive and scarce resource, making its cost a significant barrier to the widespread adoption of PEM fuel cells. Erik Eikeng et al. [107] provide an in-depth review of the critical and strategic raw materials (CRMs) essential for the development and implementation of hydrogen technologies. Critical raw materials (CRMs) such as platinum, palladium, rare earth elements, and others play a pivotal role in the hydrogen economy. Developing alternative materials or enhancing the efficiency of existing ones can significantly decrease the dependency on CRMs. For instance, advancements in nanotechnology and catalyst development might provide new pathways to achieve similar or better performance with less or no use of traditional CRMs. Critical raw materials are the backbone of the emerging hydrogen economy, enabling the technological advancements necessary for large-scale adoption. However, their strategic importance also brings challenges that need to be addressed through sustainable practices, innovation, and strategic policy-making. By understanding and mitigating these challenges, we can ensure the continued growth and sustainability of the hydrogen sector, paving the way for a cleaner energy future. Veeresh Patil [108] provides an in-depth review of the various degradation mechanisms (chemical, mechanical, catalyst, thermal, and contamination) that affect Polymer Electrolyte Membrane (PEM) fuel cells, which are crucial for improving their durability and commercial viability. Understanding and mitigating these degradation mechanisms are essential for enhancing the performance and longevity of PEM fuel cells.

L. Vichard et al. [109] explore the long-term durability and degradation mechanisms of proton exchange membrane fuel cells (PEMFCs), focusing on their application in transportation and stationary systems. The study highlights the necessity of enhancing fuel cell durability to facilitate widespread adoption. The researchers developed a degradation model using an Echo State Neural Network (ESN) to predict performance over extended periods. The study concludes that the ESN-based model effectively predicts fuel cell degradation, with potential applications in real-time performance monitoring and prognostics for fuel cell systems. The findings underscore the impact of ambient temperature on fuel cell performance and suggest that improved humidification strategies could significantly enhance durability.

Another critical issue in the operation of PEM fuel cells is water management, particularly within the context of microgrid applications. Effective water management is essential for maintaining optimal performance, ensuring durability, and preventing operational issues such as the flooding or drying of the fuel cell components. M. Ait Ziane et al. [110] present a new diagnostic approach for water management in Polymer Electrolyte Membrane Fuel Cells (PEMFCs). The focus is on addressing faults related to water management from a control perspective. The study emphasizes the importance of controlling operational conditions and early fault detection to enhance the durability and performance of PEM fuel cells. The study highlighted the significant impact of temperature sensor faults on PEMFC water management and the effectiveness of the proposed control and diagnostic strategies in enhancing fuel cell durability and performance.

The water content in PEMFCs significantly affects the transport of reactants and the conductivity of the membrane. Effective water management strategies can enhance the performance and extend the lifespan of the fuel cell. Wenxin Wan et al. [111] outline the development of an internal resistance–operating condition model that considers the coupling effect of temperature and humidity to determine the variation trend in total resistance and stack humidity with single-factor operating conditions. The proposed method includes optimizing operating conditions such as the working temperature, anode gas pressure, cathode gas pressure, anode gas humidity, and cathode gas humidity to achieve the optimal water management state within the fuel cell stack.

M. Rahimi- Esbo et al. [112] investigate the effects of water accumulation in Polymer Electrolyte Membrane Fuel Cells (PEMFCs) operated in dead-end mode. The study utilizes a transparent PEMFC stack to visualize water flow within the cells, providing direct insights into water management challenges. Water accumulation within the fuel cell channels leads to the non-uniform distribution of reactant flow, causing voltage instability and the degradation of the electrodes. Water tends to accumulate more in the lower half of the flow channels, necessitating a design adjustment to increase the flow velocity and improve water removal. At higher current densities (e.g., 300 mA/cm²), water accumulation can transition from droplet form to film and slug flow, significantly impacting performance by blocking the channels and reducing the active area. The study emphasizes the importance of effective water management through proper purge timing and design optimization to prevent performance degradation and improve the efficiency and longevity of PEM fuel cells.

5. Conclusions

This article comprehensively reviews the potential of proton exchange membrane (PEM) fuel cell-based systems integrated with renewable energy sources, specifically solar and wind, for stationary applications in the telecommunications industry. It highlights the critical role these hybrid systems play in addressing the increasing demand for reliable and continuous power supply while mitigating the environmental and operational drawbacks associated with traditional diesel generators.

PEM fuel cell systems, when integrated with renewable energy sources, offer significant reductions in greenhouse gas emissions and operational costs compared to diesel generators. This shift supports the industry’s move towards more sustainable and eco-friendly power solutions. Effective energy management and advanced storage solutions are vital for optimizing the performance and reliability of PEM fuel cell systems. The integration of sophisticated control algorithms and energy management strategies enhances the efficiency and stability of these systems, making them suitable for high-power applications in telecommunications. Despite the progress, several challenges remain, including high costs, storage efficiency, and the need for more extensive infrastructure. This article recommends further research into improving hydrogen production and storage technologies, alongside developing rigorous safety regulations to enhance public acceptance and policy support. Future research should focus on improving diagnostic accuracy, developing cross-scale and multiphysics simulation studies, and advancing data-driven diagnostic methods for efficient and stable PEM fuel cell operation. These efforts are crucial for the commercialization and broader application of PEM fuel cells in achieving carbon neutrality and sustainable energy solutions. In conclusion, leveraging PEM fuel cell systems integrated with renewable energy can significantly reduce the telecommunications industry’s reliance on fossil fuels, lower the operational costs, and minimize the environmental impact, thus enhancing the reliability and sustainability of power supply infrastructure.