The thermal management system of Proton Exchange Membrane Fuel Cell (PEMFC) needs to keep the stack running at normal operating temperature. It plays an important role in the normal operation of fuel cell system. This paper reviews the related researches in recent years. Firstly, the structure of thermal management system for different power systems is summarized. Secondly, the heat generation and heat transfer in the system are summarized, especially the heat generation and heat transfer characteristics in the fuel cell stack. Thirdly, different control strategies applied to the system are summarized, and the characteristics of different control strategies are compared. Finally, the relevant research methods are summarized. Through the summary of relevant research work, the emphases of work are analyzed and the future research direction is pointed out.

As the power source of pure electric vehicles, the thermal management of lithium-ion batteries can not be ignored. This paper comprehensively introduces the thermal safety problems caused by the heat change of lithium-ion batteries in the work, and summarizes the thermal management methods of lithium-ion batteries. Specifically, air cooling, liquid cooling, phase-change cooling and their improvement and optimization are reviewed. The analysis shows that the common cooling methods include air cooling, liquid cooling, phase-change material cooling and heat pipe cooling. The coupling of multiple cooling methods can improve the cooling efficiency of the cooling system, which can continue to cool after a certain cooling method fails, and make the temperature uniformity of the battery better.

The aerodynamic characteristics of automobiles have important influences on the power, economy, operation stability and comforts of automobiles. In order to summarize the research status and development trends of automobile aerodynamic characteristics, the CiteSpace, a visualization bibliometrics tool, is applied to analyze the spectrums of automobile aerodynamic characteristics.The Chinese literatures on the aerodynamic characteristics of automobiles included in the China Knowledge Network (CNKI) database from 2000 to 2022 were utilized as data sample to visualize and analyze the chronological publication volumes, journal source categories, research institutions, scholars and key words to generate relevant network graphs.The results show that research on the topic of “Automotive Aerodynamics” has been continue to receive academic attention, research in this area has been depending on the relevant key laboratories, future research will be still focusing on aerodynamic drags, crosswind stabilities and aerodynamic noises. 3 important methods to reduce aerodynamic drag are proposed. It is shown that the importance of studying crosswind stability and active control method and air jet method are new ways, which are worth of exploring for reducing aerodynamic noises.

Lightweight is one of the mainstreams of the future development of automotive. The bumper anti-collision beam system plays an important role in energy transfer and absorption during vehicle collision, which has always been the focus of lightweight technology application. This paper analyzes the research of application status of new materials and new technologies in the development of bumper anti-collision beam in recent years. It is concluded that the multi-form mixed material is the development trend of the future bumper anti-collision beam in the material lightweight application, and the new processes corresponding to the new materials will also be diversified. Finally, it is pointed out that the lower cost high strength steel is the main part of the bumper anti-collision beam, and the proportion of aluminum alloy is gradually increasing, in the middle and low end cars. 3D hot roll bending anti-collision beam of ultra-high strength steel will be gradually popularized and applied.

Illegal operations by pedestrians and motorcycle drivers are the primary cause of accidents. Increasing supervision of helmet wear and pedestrian violations by motorcycle drivers is critical to motorcycle driving safety. This paper discusses the main tasks and characteristics of driver helmet and pedestrian detection, introduces the principle and advantages and disadvantages of the current deep learning-based detection algorithms, summarizes the research status of safety helmets and pedestrians in recent years, and analyzes and prospects in this field in combination with detection difficulties.

The hybrid A* algorithm of automated driving vehicles runs quickly, but it is difficult to ensure the continuity of curvature and comfort of vehicles, and can not guarantee the optimized results without collision. Aiming at the above problems, this paper aims to improve the comfort of vehicles, and optimize A* algorithm based on minimum impact. Algorithm pre-study in MATLAB and raster map is conducted to solve the problem of insufficient comfort during the motion of automated driving vehicles in low-speed scenarios. Then, the optimization algorithm is improved to ensure the safety of vehicles for the problem that the optimization algorithm may lead to trajectory collision, and the strategy of automatic adjustment of the “corridor” is given. The simulation results show that the optimized trajectory is collision-free and the curvature of the trajectory is smaller, and the acceleration does not exceed the normal range, which can meet the normal driving requirements.

The energy recovery of electric vehicle involves motor positive and negative torque switch and zero crossing, and both front and rear axles of dual motor for 4-wheel drive electric vehicle will have zero crossing. Poor torque control will cause driving jitter at zero crossing. In order to optimize the driving jitter caused by double motor zero crossing of pure electric 4-wheel-drive vehicle, alternate axle zero crossing control strategy for a 4-wheel-drive vehicleis designed. In this strategy, a torque zero crossing platform is designed for positive and negative switching of motor torque, and the zero-crossing reversing process is controlled, in the meantime, the front and rear motor axles have complementary zero crossing, to reduce the number of zero crossing. Through the subjective and objective evaluation method, it is verified that the scheme can effectively optimize the drivability of electric 4-wheel-drive vehicle when the torque crosses zero.

Aimed at the urgent need of high precision and high efficiency numerical simulation in the automobile industry, the applicability of the industrial simulation software INTESIM 5.0 supporting distributed parallel computing for large-scale fine simulation is studied. The research method is as follows: 3 cylinder models with different grid sizes are established, INTESIM and a commercial software are used to simulate 1 million grid models respectively. The results of 2 models are compared to verify the correctness of INTESIM calculations. INTESIM is used to simulate 100 million grid models on the commercial computing system of Sunway TaihuLight, it is found that the parallel speed up ratio and parallel efficiency conform to the general law. The advantages of fine modeling are demonstrated by comparing the results of 3 models solved by INTESIM. The results show that INTESIM is fully competent for large-scale parallel computing, to obtain high-precision simulation results with low time cost, which is helpful to improve the optimization efficiency of the reliability and durability of automobile engine cylinder block.