关键词: 电池包热管理, 变密度拓扑优化, 树状流道设计, 响应面优化, 非支配遗传算法

Abstract:

The thermal management of electric vehicles predominantly relies on liquid cooling. Recognizing the limitations of traditional serpentine liquid cold plate, characterized by poor temperature uniformity and high voltage drop, this study explores the application of topology optimization technology to design the flow channels. The objective is to develop a cooling system that effectively addresses the critical requirements of high-temperature safety and uniform temperature distribution within the battery pack. First, based on the 2D simulation of Comsol variable density topology optimization, by taking the lowest average temperature in the design domain as the objective function and the flow channel volume fraction as the constraint condition, the flow channel distribution law in the design domain was obtained using a variable control method. Sensitivity filtering was implemented using the Helmholtz filter, resulting in the design of a novel tree-like topology optimization channel. Furthermore, the 2D topology simulation results were transformed into the actual channel geometry, and the tree-like topology runner heat sink was prepared using 3D printing technology. The experimental design of the response surface was carried out using the simulation technology of heat flux coupling, and the interaction effect of the volume fraction A, inlet temperature B, and flow rate C on the heat dissipation performance of the runner was studied. The actual temperature control ability of the topological flow channel was verified by repeated group experiments, and this confirmed the high prediction accuracy of the simulation. The optimal Pareto front solution was obtained using the optimization iterative analysis of the non-dominant genetic algorithm: A = 0.3, B = 20 ℃, and C = 10 L/min had the best heat dissipation performance. Compared with the serpentine channel, the optimal topological flow channel reduces the inlet and outlet pressure drop from 4863 Pa to 822 Pa, a decrease of 490%. The maximum temperature of the battery module decreased from 27.88℃ to 27.21 ℃, a decrease of 2.4%; The temperature difference decreased from 5.7 ℃ to 4.95 ℃, a decrease of 13.2%; The above results meet the experimental test requirements under the driven-durability condition of the battery module. In this study, the advantages of tree-topology optimized flow channels for battery module thermal management are proposed and verified, and an effective scheme is provided for the design of a battery thermal management system.

Key words: battery pack thermal management, variable density topology optimization, tree-like flow channel design, response surface optimization, non-dominated genetic algorithms