Analysis and optimization of injection molding for the part of EV charging equipment with NSGA-II and machine learning-enhanced RSM
摘要
This paper presents an analysis and optimization method for the injection molding of electric vehicle (EV) charging equipment components, with the goal of reducing warpage and enhancing surface quality. Initially, the injection molding process of the side bright strip plastic part on the side of the charging pile is studied through mold flow analysis, revealing a maximum total warpage deformation of 8.42E−03 m, which exceeds the design specification of being less than 5.0E−03 m. Subsequently, the NSGA-II genetic algorithm is employed to innovatively optimize the number and positioning of gates, which preliminarily reduced the warpage deformation of the injection-molded part. The outcomes indicated that with three gates, the maximum total warpage deformation is decreased to 5.39E−03 m. Thirdly, the process parameters for the 3-gates injection are further optimized using the NSGA-II genetic algorithm. It is found that at a mold temperature of 64.7 °C, a melt temperature of 276 °C, and an injection time of 2.99 s, the maximum total warpage deformation of the part is reduced to 2.30 E−03 m, marking a 72.68% reduction and meeting the assembly requirements of the side bright strip part design. Finally, in conjunction with DOE design and the application of machine learning to obtain an accurate RSM response surface, the influence of process parameters on warpage deformation is further investigated, providing systematic data for mold design and process adjustment. Injection molding proofing validates that the maximum total warpage deformation of the plastic part meets the requirements utilizing the optimized injection gate and process parameters.