To enhance the mechanical performance and energy density (ED) of the Battery Pack System (BPS), this study proposes a multi-objective optimization (MOO) method that automates cell array layout design. Using cell arrangement parameters as design variables, the approach optimizes ED, crush deformation (Def), root mean square equivalent stress (RMSES), and energy absorption rate (REA). A surrogate model based on Support Vector Regression (SVR) is established and combined with an improved Reference Vector Guided Evolutionary Algorithm (RVEA) to generate the Pareto front. The optimal solution is selected via the Entropy Weight-VIKOR method. Results show a 4.063% increase in ED, along with reductions of 8.131% in RMSES, 21.646% in Def, and 10.204% in REA. This method offers an innovative strategy for lightweight and safety-oriented BPS design with strong practical applicability.
Lian et al. (Thu,) studied this question.