Thermal optimization of Li-ion battery pack using genetic algorithm integrated with machine learning

This research work is focused on the "Air-Cooled" Battery Thermal Management System (BTMS) through the optimization of cell spacing of the battery pack. The Computational Fluid Dynamics (CFD) model has been developed to analyze the temperature field of the battery pack. The cell structure optimization is carried out using variable spacing obtained by numerical optimization and genetic algorithm (GA) integrated with the machine learning (ML) approach "Support Vector Machine", to enhance the temperature uniformity across the cells. The temperature distribution of the geometries obtained by numerical optimization and GA-SVM is analyzed using CFD. The maximum cell temperature of GA battery pack is reduced by approximately 3.5 K and the temperature uniformity across the cell is increased by more than 70 %. Moreover, the effects of the inlet air flow rate on the thermal behavior of the battery pack are also analyzed. The results obtained in this study suggest that the proposed optimization method is an effective tool to design the cell spacing for improvement of the cooling efficiency of the battery pack.