The rapid development of fuel cell electric vehicles (FCEVs) has highlighted the critical importance of optimizing energy management strategies to improve vehicle performance, energy efficiency, durability, and reduce hydrogen consumption and operational costs. However, existing approaches often face limitations in real-time applicability, adaptability to varying driving conditions, and computational efficiency. This paper aims to provide a comprehensive review of the current state of FCEV energy management strategies, systematically classifying methods and evaluating their technical principles, advantages, and practical limitations. Key techniques, including optimization-based methods (dynamic programming, model predictive control) and machine learning-based approaches (reinforcement learning, deep neural networks), are analyzed and compared in terms of energy distribution efficiency, computational demand, system complexity, and real-time performance. The review also addresses emerging technologies such as artificial intelligence, vehicle-to-everything (V2X) communication, and multi-energy collaborative control. The outcomes highlight the main bottlenecks in current strategies, their engineering applicability, and potential for improvement. This study provides theoretical guidance and practical reference for the design, implementation, and advancement of intelligent and adaptive energy management systems in FCEVs, contributing to the broader goal of efficient and low-carbon vehicle operation.

There are several drawbacks for lithium-ion batteries at low temperatures, including weak electrolyte conductivity, low chemical reaction rate and greatly increased impedance. Thus, it is inefficient to charge lithium-ion batteries at low temperatures. This work proposes an AC incentive fast charging strategy at low-temperatures for lithium-ion batteries based on the analysis and comparison of the existing charging and heating methods. The charging speed, temperature variation, the capacity loss of the constant current constant voltage (CCCV) charging strategy and the proposed method with different current and frequency conditions are compared and analyzed. The results show that it takes about 1400 s for the proposed method to fully charge a lithium-ion battery in the case of 2.2 A current beginning at 25% state of charge (SOC). In addition, the temperature rises about 8 °C. In contrast, the charging time of the CCCV method is 400 s slower than the proposed method and the temperature of the CCCV method increases only about 2 °C. In the case of 1.5 A current beginning at 0% SOC, the charging time of the proposed method is 500 s faster than the CCCV method. The results indicate that the proposed charging method can significantly improve the charging efficiency of lithium-ion batteries at low temperatures.

The large-scale production and utilization of nitrogen-containing compounds have greatly promoted the development of modern agriculture and the chemical industry, but have also resulted in increasingly severe nitrate contamination and an imbalance of the nitrogen cycle. The efficient conversion of nitrate into value-added ammonia not only contributes to pollutant remediation but also provides a promising route for green ammonia synthesis. Owing to its mild reaction conditions, potentially lower environmental impact, and compatibility with renewable electricity, electrocatalytic nitrate reduction to ammonia has attracted considerable attention in recent years. This process involves a multielectron transfer process involving numerous intermediate transformations, and its catalytic performance largely depends on the adsorption and conversion of key intermediates on the catalyst surface, as well as the suppression of the competing hydrogen evolution reaction. This review systematically summarizes recent advances in electrocatalytic nitrate reduction to ammonia, with emphasis on the reaction mechanisms and major reaction pathways, as well as the design strategies, structure–activity relationships, and performance enhancement mechanisms of metal-based, carbon-based, and composite catalysts. In addition, the main challenges in this field, including product selectivity, mass transport, in-situ mechanistic characterization, and long-term stability, are discussed. Finally, the construction of highly efficient catalytic systems and key directions for future research are outlined, with particular emphasis on nitrate valorization and sustainable ammonia synthesis.

Accurate state-of-charge (SOC) estimation is indispensable for the reliable operation of lithium-ion battery management systems. Nevertheless, the performance of existing deep learning models is often compromised when deployed under practical operating conditions, where temperature variations, unseen driving patterns, sensor faults, and battery-to-battery discrepancies induce significant distribution shifts. To address this challenge, this study develops a mask-enhanced deep learning framework for robust SOC estimation. A temporal masking strategy is incorporated into an LSTM-based network to randomly mask a subset of input observations during training. This strategy encourages the model to learn robust battery representations that generalize effectively to unseen operating conditions without requiring additional retraining or fine-tuning. Extensive evaluations are conducted under unknown temperatures, unseen drive cycles, sensor anomalies, varying input sequence lengths, and different battery chemistries. The proposed framework consistently outperforms conventional machine learning and deep learning methods across all evaluation scenarios. Under unknown-temperature conditions, the root mean square error (RMSE) and mean absolute error (MAE) are reduced from 18.43% and 16.20% to 10.13% and 9.13%, respectively, while under an unseen driving cycle they further decrease from 30.43% and 24.60% to 7.16% and 5.82%, respectively. Moreover, the proposed framework demonstrates strong transferability across different battery types and maintains stable estimation performance under sensor perturbations, highlighting its robustness to practical deployment challenges. Overall, this work provides a simple yet effective solution for reliable data-driven SOC estimation and facilitates the practical deployment of trustworthy battery management systems under complex real-world operating conditions.