The increasing penetration of renewable energy is shifting water electrolysis for hydrogen production from steady-state operation to dynamic conditions involving frequent load changes, low-load operation, overload, and repeated start–stop cycles. Proton exchange membrane (PEM) and alkaline (ALK) water electrolyzers are the two most widely deployed low-temperature electrolysis technologies, but they differ substantially in ion-conducting media, reactant transport pathways, gas evolution behavior, thermal management, and intrinsic time constants. These differences lead to distinct transient responses, performance evolution, and degradation risks under fluctuating power inputs. This review provides a critical and comparative synthesis of dynamic modeling studies on PEM and ALK electrolyzers under renewable-energy-driven operation. Existing models are examined from three interrelated perspectives: spatial non-uniformity of state variables, dynamic evolution and degradation of key performance parameters, and bidirectional multiphysics coupling with feedback closure. The analysis shows that PEM models should retain local current density, membrane water content, temperature gradients, membrane conductivity, and interfacial impedance, whereas ALK models should emphasize KOH concentration, electrolyte circulation, bubble coverage, gas void fraction, effective reaction area, and separator resistance. High-fidelity models provide mechanistic insight into local transport, thermal, and electrochemical processes, but their system-level application is limited by extensive parameter requirements and high computational cost. Reduced-order models are more suitable for control and operational optimization, yet may omit critical feedback pathways. Future research should prioritize dominant-state retention, dynamic parameter updating, feedback-loop closure, engineering deployability, and standardized validation under fluctuating operating conditions.

Low seawater temperatures constrain the operation of open rack vaporizers (ORVs) and intermediate fluid vaporizers (IFVs), while also increasing pumping-related emissions at LNG terminals. This study establishes a carbon-oriented framework for an expanded ORV–IFV regasification system sharing a fixed-speed seawater pump network and evaluates thermal discharge from an adjacent power plant as a supplementary heat source. Using measured LNG composition, we developed an Aspen HYSYS model based on the Peng–Robinson equation of state and steady-state energy balances, which was validated against field data. Electricity-related CO2 emissions from seawater pumps and auxiliaries were quantified using the regional grid emission factor, while pump scheduling was formulated as a mixed-integer nonlinear programming (MINLP) problem. Model predictions differed from measurements by approximately 2%. Lower seawater temperatures increased emissions and restricted maximum regasification capacity to 80% and 57% of the design value at 3–4 °C and 2–3 °C, respectively. For LNG throughputs of 300, 500, and 700 t/h, CO2 reduction increased with warm-seawater flow and inlet temperature; maximum reductions reached approximately 50–55% under 3–7 °C ambient seawater conditions and 40% under 6–20 °C conditions, with a 95% confidence interval of ±3.9 percentage points. Monthly discharge data indicated reductions of approximately 20% in winter and 45% in summer. Integrating power-plant waste heat with load-dependent pump scheduling can improve the carbon performance of LNG regasification.

Accurate fault diagnosis of high-voltage circuit breakers is essential for improving power-system reliability. However, coil-current signals are often affected by environmental interference and random fluctuations in the measurement circuit, which may reduce diagnostic accuracy and lead to unstable recognition results. To address this problem, this paper proposes a fault diagnosis method based on opening and closing coil-current signals of high-voltage circuit breakers.First, an improved wavelet-threshold variational mode decomposition method is used to suppress noise while preserving fault-sensitive waveform characteristics. Then, extrema, temporal features, and waveform statistical features are extracted from the denoised signals. A CNN-BiGRU-Attention model is constructed to learn local waveform features and temporal dependencies. The model parameters are further optimized using a multi-strategy improved dung beetle optimization algorithm to improve parameter search performance and reduce the risk of local optima.Experiments were conducted on a physical ZN63–12 high-voltage vacuum circuit breaker platform under controlled laboratory fault-simulation conditions. The proposed method achieved an overall diagnostic accuracy of 96.5% across five operating states. The results indicate that the method can effectively distinguish normal operation, inter-turn short circuit, poor contact, overvoltage, and undervoltage conditions within the tested experimental scope. This study provides a potential data-driven approach for coil-current-based fault diagnosis of high-voltage circuit breakers, while further validation on additional breaker types and field operating conditions is still required.

Multiple virtual synchronous generators (VSGs) encounter the challenge of weak damping during parallel operation, which increases the risk of low-frequency oscillations. To address this challenge, this paper proposes a transient power feedback (TPF) branch to suppress low-frequency oscillations in the coordinated operation of multiple VSGs. The TPF branch consists of a transient link and an inertial link, establishing feedback from the output frequency to the input active power to enhance system damping. Simulation results show that the proposed method can reduce the active power overshoot by more than 13

Voltage regulation accuracy and high-efficiency operation are critical objectives for wireless power transfer (WPT) systems. Conventional phase-shifted voltage control strategies suffer from significant efficiency degradation under light-load conditions. To address this limitation, a novel hybrid frequency-phase adaptive control (HFPAC) strategy is proposed to minimize switching losses in both the inverter and active rectifier (AR), thereby optimizing system efficiency during light-load conditions. The system’s performance is enhanced by implementing AR synchronous control and system tuning through current-phase closed-loop control. Furthermore, a Newton-Raphson-based optimizer (NRBO) is integrated with a backpropagation neural network (BPNN) to enhance the BPNN’s prediction accuracy and convergence speed, enabling rapid and precise inverter frequency switching. Resonant network parameter optimization is also performed to suppress current fluctuations and ensure stability. Finally, experimental validation using a 150 W prototype demonstrates a peak system efficiency of 90.5% at 60% load. The novelty of this research lies in the synergistic integration of adaptive control algorithms and hybrid optimization methodologies, providing a systematic approach to efficiency enhancement in WPT systems.

Hybrid hydrogen production systems are crucial for green hydrogen energy development. However, they face challenges like low power conversion efficiency and high costs, which limit large-scale use. To address the shortcomings of current scheduling strategies, this paper constructs and validates an integrated optimization framework. Firstly, we elucidate the coupling relationship between scheduling time scales and power fluctuations, establishing a self-adaptive dynamic time-scale scheduling mechanism. This mechanism achieves a 90.7% fluctuation suppression rate, representing an 11.6% improvement over conventional fixed-interval strategies. Secondly, we design a multi-objective coordinated power allocation mechanism with a price-sensitive dynamic weighting adaptation. This mechanism increases the net profit by an average of 3.18% compared to fixed-weight models under hydrogen price fluctuations. Finally, we design an intelligent optimization algorithm leveraging multi-scenario memory for precise pressure and flow rate control. The algorithm elevates the average hydrogen production efficiency to 68.22% for AEL and 83.64% for PEMEL. The proposed strategy demonstrates superior performance through a closed-loop workflow of fluctuation suppression, economic dispatch, and precise control, significantly enhancing the system's adaptability, economy, and efficiency for time-varying operations.

ABSTRACT In offshore wind farm systems, the submarine‐cable topology mainly consists of collector and transmission submarine cables, which are responsible for power collection and transmission, respectively. For the 35 kV collector submarine cable, a COMSOL electromagnetic–thermal coupling model was established to calculate layer losses and ampacity under direct‐buried conditions. The simulated ampacity differed from the IEC 60287 result by 3.90%, indicating acceptable consistency for engineering comparison. For the 220 kV transmission submarine cable, electromagnetic–thermal–fluid multiphysics coupling models were developed for both direct‐buried and J‐tube laying conditions. Ocean‐current cooling changed the direct‐buried ampacity by only 0.11%, whereas the J‐tube ampacity was 679.5 A, 14.26% lower than that of the direct‐buried section. The effects of wind speed, solar radiation intensity, and ambient temperature on J‐tube ampacity were further quantified, and a quadratic regression model was established. The model achieved an R 2 of 0.998564 with a maximum fitting error of 1.5%, and the annual ampacity at the project site ranged from 611.122 to 896.771 A.

The integration of numerous new energy generators into the AC grid will diminish the inertia characteristics traditionally provided by synchronous generators, leading to frequency instability. The inertia and damping constants of the more widely used grid-forming converter (GFC) control methods are fixed. This paper proposes a multi-intelligent deep learning (MADRL) based optimization strategy to enhance the inertia and damping characteristics of GFCs. We begin by analyzing the constraints of the inertia that GFC source measurements can provide and propose a method for acquiring the inertia time constant (ITC) based on node active power and frequency. This method facilitates the partitioning of the distribution grid according to ITC values. Furthermore, we implement characteristic-enhanced branching to optimize inertia and damping characteristics through MADRL decision-making. Utilizing an IEEE-129 network with multiple new energy field stations, the experimental results demonstrate that this approach can increase the ITC of most nodes by over five times. In addition, we maintain frequency fluctuations within 0.01pu, representing a reduction of more than 95% compared to virtual synchronous generator control, while the maximum frequency transient fluctuation decreases by 81%. Furthermore, through comparative experiments and transfer learning, it can be demonstrated that this method possesses sufficient superiority and plug-and-play capability.

To enhance the control performance and anti-interference ability of permanent magnet synchronous motor, a speed control strategy that integrates integral terminal sliding mode control (ITSMC) with an internal model control (IMC) observer is proposed. ITSMC is designed by combining integral sliding mode control with terminal sliding mode control, which can effectively reduce the steady state error of the system as well as enhance its reaching speed. The sliding mode reaching law has been improved. The newly improved sliding mode reaching law can adaptively adjust its own size according to the state of the system. It shortens the system reaching time and further enhances overall control performance. The load disturbance observer is constructed based on the IMC approach. The observed load is converted into an electric current and fed forward, effectively suppressing disturbances and enhancing the system control performance and robustness. Finally, simulations and experiments were conducted on the proposed method. Analysis of the experimental results shows that, compared to traditional SMC, the improved integral terminal sliding mode control (IITSMC) reduces convergence time by 48.6

To address the critical industry challenge of sharply reduced heat transfer efficiency and increased energy consumption of Open Rack Vaporizers (ORV) in Liquefied Natural Gas receiving terminals under low-temperature seawater conditions, this paper proposes an optimized co-operational strategy for ORV and Intermediate Fluid Vaporizers (IFV) based on the utilization of waste heat from warm water discharge. Focusing on a domestic LNG terminal equipped with ORVs, Submerged Combustion Vaporizers (SCV), and IFVs, a process and mathematical model for the combined ORV-IFV operation was developed using simulation software. The study specifically investigates the influence mechanism of introducing warm water discharge from an adjacent power plant on the energy efficiency of the system. The research reveals the direct impact of seawater temperature on the energy consumption of the combined system: low seawater inlet temperatures severely constrain the heat exchange efficiency of ORVs, preventing them from reaching their rated design capacity. The integration of warm water discharge fundamentally improves heat transfer efficiency by increasing the temperature differential within the vaporizers. Furthermore, the study quantifies the coupled relationships between the system's minimum theoretical energy consumption and key operational parameters—including thermal discharge input volume, seawater inlet temperature difference, and LNG processing capacity—under both winter and summer conditions. This study reveals that low seawater temperatures (5 °C) severely constrain ORV performance, limiting its throughput to only 57% of rated capacity at 10 MPa. The integration of warm water discharge fundamentally overcomes this limitation. Quantitative analysis demonstrates that the proposed ORV-IFV co-operation strategy, utilizing waste heat, reduces total system energy consumption by approximately 20% in winter and up to 45% in summer. A subsequent economic analysis, based on actual industrial data, reveals annual cost savings of approximately 152 million CNY with a simple payback period of 2.0 years. The results demonstrate that the warm water integration strategy significantly reduces the total energy consumption of the combined system and effectively compensates for the performance deficiency of ORVs at low temperatures, achieving efficient cascade utilization of industrial waste heat. This innovative approach not only provides a feasible pathway to enhance the full-condition energy efficiency and economic viability of LNG receiving terminals but also offers a theoretical foundation and reference for energy-saving retrofitting projects utilizing low-grade heat sources in similar terminals.

Weak grid (WG) is an interface power constraint which weakens the voltage and frequency stability of nodes. The virtual synchronous generators (VSGs) operating in parallel in a WG system face frequency instability, and potential low-frequency oscillations may lead to grid collapse. To address this challenge, this paper proposes a time delay correction (TDC) scheme to strengthen the stability of VSGs in WG. First, this paper proposes a power small-signal decoupling method for parallel VSGs. Secondly, a voltage source model of weak grid dominated by synchronous generators is constructed based on the renewable energy station short circuit ratio (RSCR) of new energy field stations. Then, to improve the synchronization stability of the system, we enhance the control strategy of the VSG by connecting a time-delay correction algorithm in series at the active power input. Through hardware-in-the-loop experiments, the TDC scheme significantly reduces the frequency fluctuation of the three parallel VSG system by 57.45% to 82.16% compared with the conventional strategy. In addition, the stability of TDC outperforms the conventional strategy as well as the existing improved strategies under lower short-circuit ratios.

Due to worsening environmental pollution and severe energy shortages, the optimal operation of low-carbon park integrated energy systems has become a research hotspot. This paper constructs a park integrated energy system including wind-solar power generation, electrolytic hydrogen production, combined heat and power generation, and other equipment. A VMD-BOA-LSTM combined prediction model based on Variational Mode Decomposition (VMD), Butterfly Optimization Algorithm (BOA), and Long Short-Term Memory network (LSTM) is proposed to achieve high-precision prediction of wind and solar power. On this basis, a multi-time-scale optimization model for different seasons, including day-ahead, intraday, and real-time scales, is established, aiming at minimizing the economic cost and environmental cost and maximizing wind-solar energy consumption rate. Three schemes are proposed based on the impact of seasonal differences on the system. Finally, taking the Zhangjiakou Chongli wind-solar coupled hydrogen production demonstration project as a case study, it effectively proves that the proposed optimization scheme has economic efficiency and environmental friendliness. Through energy integration, resource waste has been significantly reduced.

Aiming at the low-carbon economic operation of integrated energy systems (IES), this paper proposes a collaborative optimization model that addresses multi-source renewable energy uncertainty by integrating refined power-to-gas (P2G) technology with a reward-penalty stepped carbon trading mechanism. First, the original wind-PV historical data are transformed into marginal distributions for wind and PV power using kernel density estimation (KDE). Subsequently, based on these marginal distributions, a multi-time-scale joint probability model for wind-PV output is established through Copula theory. By combining Latin hypercube sampling (LHS) combined with K-means clustering is then applied to generate a set of typical scenarios that retain the spatiotemporal coupling characteristics of renewable generation. On this basis, P2G technology is introduced to realize bidirectional conversion between electricity and natural gas as well as carbon cycle utilization, and a dynamic carbon trading cost function with reward and penalty factors is constructed. Through sensitivity analysis of carbon trading parameters, the influence mechanism of stepped carbon price thresholds and reward- penalty coefficients on system economy is revealed. Finally, comparative experiments based on four different scenarios demonstrate that the proposed method can substantially reduce carbon emissions and total costs, providing an innovative solution for the low-carbon economic operation of IES under the background of the new power system.

The integration of grid-forming energy storage systems (GFM-ESSs) provides essential support for the stable operation of grid-connected converters in renewable energy systems. However, GFM-ESSs may exhibit low-frequency oscillations in response to grid state variations, posing a threat to power system stability. To address this challenge, this paper proposes a fast continuous optimization method for the active power-frequency control loop of multi-VSG-based GFM-ESSs. First, a parameter coupling model for multiple VSGs is established, and an internal parameter decoupling control strategy is proposed. Subsequently, an iterative optimization model based on a gradient-based master-slave game is developed, in which the minimization of converter frequency deviation serves as the leader's objective, while the minimization of system frequency deviation acts as the follower's objective. Frequency fluctuations are further mitigated through tracking differentiator-based active power compensation. The effectiveness of the proposed method is validated through simulation with six GFM-ESS units integrated into a modified IEEE 33-node system featuring six renewable energy stations. Simulation results demonstrate that the proposed approach significantly suppresses frequency fluctuations while also reducing the response time and the rate of frequency change under grid disturbance conditions.

Alkaline Water Electrolysis (AWE) has become the mainstream technology for current green hydrogen production due to its high technical maturity, low equipment cost, and scale advantage. However, it still faces challenges such as bottlenecks in key material performance, insufficient adaptability to dynamic operating conditions, and resource dependence. Thus, there is an urgent need to promote efficiency improvement and cost reduction through material innovation. Although AWE technology has developed relatively maturely, existing review literatures lack considerations on the systematic synergistic optimization of AWE materials. To address this issue, this paper systematically reviews the action mechanisms and material properties of various hydrogen production materials from three core dimensions: electrodes (catalysts), membranes, and electrolytes. It conducts an in-depth analysis of the optimization mechanisms of material properties and proposes targeted optimization strategies. Meanwhile, it clarifies the key challenges and future development directions faced by various materials, and lists the directions for engineering optimization. Aiming to help readers gain a comprehensive understanding of electrolysis systems, broaden research ideas, and provide theoretical support and technical route guidance for constructing high-efficiency electrolysis systems with material synergistic optimization. Among them, carbon-based composite materials with excellent structural designability, self-healing membranes with a unique dynamic self-healing design, and hydrogen production via seawater electrolysis with significant cost-effectiveness and deep coupling with renewable energy all show great application potential in the future.

The global shift towards sustainable energy underscores the critical role of green hydrogen. Hybrid hydrogen production systems, which integrate alkaline (AEL) and proton exchange membrane (PEMEL) electrolyzers, offer a promising solution for utilizing intermittent renewable power. However, their complex structure and dynamic operation pose significant challenges for fault detection and diagnosis. This paper proposes a Dynamic Deep Coupled Dictionary Learning (DDCDL) method to address these issues. The approach constructs a Unified Joint Dictionary to fuse multimodal AEL and PEMEL data for integrated diagnostics, a Dynamic Adaptive Dictionary updated via online learning to minimize false alarms during operational transitions, and a Deep Sparse Dictionary combined with a temporal convolutional network to enhance sensitivity to incipient faults. Validated with data from a 100 MW wind-solar hydrogen demonstration project, the method demonstrates superior detection timeliness, accuracy, and robustness across single faults, coupled faults, mode-switching faults, and early weak faults compared to established techniques. This work provides an effective technical pathway to improve the operational reliability and economic efficiency of hybrid hydrogen production systems.

The integration of renewable energy sources, such as wind and solar power, at high proportions has become an inevitable trend in the development of power systems under the new power system framework. The construction of a microgrid system incorporating hydrogen energy storage and battery energy storage can leverage the complementary advantages of long-term and short-term hybrid storage, achieving power and energy balance across multiple time scales in the power system. To prevent frequent start-stop cycles of hydrogen storage devices and lithium battery storage under overcharge and overdischarge conditions, a coordinated control strategy for power distribution in a microgrid with hydrogen storage is proposed. First, a fuzzy control algorithm is used for power distribution between hydrogen storage and lithium battery storage. Then, the hydrogen storage tank’s state of health (SOH) and the lithium battery’s state of charge (SOC) are compared, with the goal of selecting a multi-stack fuel cell system operating at its optimal efficiency point, where each fuel cell stack outputs 10 kW. This further ensures that the SOC and SOH remain within reasonable ranges. Finally, simulations are conducted in MATLAB/Simulink R2018b to verify that the proposed strategy maintains stability in the DC bus and alleviates issues of overcharge and overdischarge, ensuring that both the system’s SOC and SOH remain within a reasonable range, thereby enhancing equipment lifespan and system stability.

Aiming at the characteristics of weak vibration signal, strong interference, unevenness and nonlinearity in rolling bearing faults, this paper proposes a bearing fault intelligent diagnosis method based on variational modal decomposition (VMD) and Multisynchrosqueezing transform (MSST).First, the original vibration signal of the bearing is decomposed into multiple intrinsic mode functions (IMF) through the optimal parameter VMD, and then an effective IMF is selected for signal reconstruction according to the kurtosis value and mutual information.Secondly, MSST is applied to the reconstructed signal to obtain a time-frequency (TF) images with high energy concentration, and then extract the time-frequency and timeamplitude signals of the vibration signal in the TF images according to the ridge detection algorithm.Finally, these time series are used as input, using Long short-term memory (LSTM) network is trained to complete the intelligent classification and diagnosis of bearing failure.This method is compared with other fault diagnosis methods through test data sets.The results show that the method proposed in this paper is superior to other methods in recognition accuracy and recognition type, can accurately identify and classify bearing faults, and has strong generalization.Ability to better meet the needs of actual engineering.

The high penetration of renewable energy and the application of power electronic equipment have exacerbated harmonic problems in new power systems, requiring effective mitigation strategy. In addition, static synchronous compensators (STATCOMs) and multifunctional grid-connected inverters (MFGCIs) with definite installation sites in distribution networks both have reactive power compensation and harmonic mitigation functions. These can assist active power filters (APFs) in harmonic mitigation. Therefore, an allocation strategy for APFs is proposed in this study to optimise the installation sites and capacities of APFs, as well as the capacities of STATCOMs and MFGCIs. Firstly, harmonic sensitivity is constructed to quantify the sensitivity of the total harmonic voltage distortion (THDv) at each node. Based on harmonic sensitivity, a node classification method is proposed to identify candidate nodes for APFs installation. Secondly, a two-layer K-means clustering algorithm is established to deal with uncertain operation scenarios. In the first-layer of clustering, typical days are clustered from the entire year and these typical days form representative operation scenarios in the second-layer of clustering. Then, an optimal allocation model with the objective of minimal total cost is constructed and solved with the dynamic adaptive mutation particle swarm optimisation (DAM-PSO) algorithm to optimise the capacities of APFs, STATCOMs, and MFGCIs. Finally, simulations on a modified IEEE 33 bus system verify that the proposed strategy achieved effective harmonic mitigation and reactive power compensation with economic advantage compared to the conventional strategy.

Renewable energy-based water electrolysis for hydrogen production is an effective pathway to achieve green energy transition. However, the intermittency and randomness of renewable energy pose numerous challenges to the safe and stable operation of hydrogen production systems, with the wide power fluctuation adaptability and economic efficiency of electrolyzers being prominent issues. Hybrid electrolyzers combine the operational characteristics of proton exchange membrane (PEM) and alkaline electrolyzers, leveraging the advantages of both to improve adaptability to wide power fluctuations and economic efficiency, thereby enhancing the overall system efficiency. To ensure coordinated operation of hybrid electrolyzers, it is essential to consider their start-stop characteristics and the impact of hydrogen to oxygen (HTO) concentration on the hydrogen production system. To achieve this, we first discuss the operating characteristics of both types of electrolyzers and the influence of system parameters on HTO concentration. A control scheme for hybrid electrolyzer systems considering HTO content is proposed. By analyzing the electrolyzer efficiency curve, the optimal efficiency point under low power operation is identified, enabling the electrolyzers to operate at this optimal efficiency, thus enhancing the efficiency of the hybrid electrolyzer system. The implementation of a dual-layer rotation control strategy effectively balances the lifecycle loss of the electrolyzers. Additionally, reducing the pressure during startup broadens the startup range of the hybrid electrolyzer.