To investigate hydrogen behaviors in the high-entropy alloy AlCrTiNiV, density functional theory and transition state theory were used to explore the molecular $H_2$ absorption and dissociation and the atomic H adsorption, diffusion, and penetration progress. The $H_2$ molecule, where the H-H band is parallel to the surface layer, is more inclined to absorb on the top site of the Ti atom site of first atomic layer on the AlCrTiNiV surface, then diffuse into the hollow sites, through the bridge site, after dissociating into two H atoms. Atomic H is more likely to be absorbed on the hollow site. The absorption capacity for atomic H on the surface tends to decline with the increase in H coverage. By calculating the energy barriers of atomic H penetration in AlCrTiNiV, it was indicated that lattice distortion may be one important factor that impacts the permeation rate of hydrogen. Our theory research suggests that high-entropy alloys have potential for use as a hydrogen resistant coating material.

The service life and fuel consumption of fuel cell system (FCS) are the main factors limiting the commercialization of fuel cell electric vehicles (FCEV). Effective energy management strategies (EMS) can reduce fuel consumption during the cycle and prolong the service life of FCS. This paper proposes an energy management strategy based on the deep reinforcement learning (DRL) algorithm, deep Q-learning (DQL). Considering the unstable performance of conventional DQL during the training process, a new algorithm called Double Deep Q Learning (DDQL) is introduced. The DDQL uses a target evaluation network to evaluate output actions and a delayed update strategy to improve the convergence and stability of DRL. This article trains the strategy using UDDS cycle, tests it using combined cycles UDDS-WLTC-NEDC, and compares it with traditional ECM-based EMS. The results demonstrate that under the combined cycle, the strategy effectively reduced FCS voltage degradation by 50%, maintained fuel economy, and ensured consistency between the initial and final state of charge (SOC) of LIB.

Rechargeable aqueous Zn-MnO₂ batteries are positioned as a highly promising candidate for next-generation energy storage, owing to their compelling combination of economic viability, inherent safety, exceptional capacity (with a theoretical value of ≈308 mAh·g⁻¹), and eco-sustainability. However, this system still faces multiple critical challenges that hinder its practical application, primarily including the ambiguous energy storage reaction mechanism (e.g., unresolved debates on core issues such as ion transport pathways and phase transition kinetics), dendrite growth and side reactions (e.g., the hydrogen evolution reaction and corrosion reaction) on the metallic Zn anode, inadequate intrinsic electrical conductivity of MnO₂ cathodes (≈10⁻⁵ S·cm⁻¹), active material dissolution, and structural collapse. This review begins by systematically summarizing the prevailing theoretical models that describe the energy storage reactions in Zn-Mn batteries, categorizing them into the Zn²⁺ insertion/extraction model, the conversion reaction involving MnO_x dissolution–deposition, and the hybrid mechanism of H⁺/Zn²⁺ co-intercalation. Subsequently, we present a comprehensive discussion on Zn anode protection strategies, such as surface protective layer construction, 3D structure design, and electrolyte additive regulation. Furthermore, we focus on analyzing the performance optimization strategies for MnO₂ cathodes, covering key pathways including metal ion doping (e.g., introduction of heteroions such as Al³⁺ and Ni²⁺), defect engineering (oxygen vacancy/cation vacancy regulation), structural topology optimization (layered/tunnel-type structure design), and composite modification with high-conductivity substrates (e.g., carbon nanotubes and graphene). Therefore, this review aims to establish a theoretical foundation and offer practical guidance for advancing both fundamental research and practical engineering of Zn-manganese oxide secondary batteries.

While the combustion vehicle promote the economic development, they also bring environmental pollution and energy crisis issues. As a new type of new energy vehicle, Fuel Cell Hybrid Electric Vehicles (FCHEV) attracted widespread attentions due to its characteristics of zero-emission and high energy density. At present, fuel economy and fuel cell aging are the important factors affecting the development of FCHEV. To this end, a deep reinforcement learning based energy management strategy (DQN) is proposed in this paper for energy management system of FCHEV, and improve the lifetime of fuel cell by reducing frequent fluctuation in fuel cell output power. The standard CycleRun are used to train the energy management strategy, and compared the training result with two benchmark energy management strategies. The result shows that the proposed energy management strategy can achieve lower computation cost, and reduce the power fluctuations.