Multi-Objective Optimization of Injection Molding Process Parameters for Junction Boxes Based on BP Neural Network...

Area of Science:

• Materials Science
• Manufacturing Engineering
• Computational Science

Background:

• Injection molding quality is influenced by numerous factors, including process parameters, mold materials, and part geometry.
• Product defects often arise from suboptimal process parameter settings, necessitating advanced optimization techniques.

Purpose of the Study:

• To develop and validate a multi-objective optimization method for injection molding process parameters.
• To minimize volume shrinkage rate and warpage deformation in plastic components.

Main Methods:

• Numerical simulation using Moldflow software for a junction box shell.
• Design of a six-factor, five-level orthogonal experiment to study process parameter effects.
• Application of a BP neural network combined with the NSGA-II algorithm for multi-objective optimization.

Main Results:

• Melt temperature was identified as the most critical factor affecting junction box quality.
• Optimized parameters achieved a 33.2% reduction in volume shrinkage and a 3.8% reduction in warpage deformation.
• The BP-NSGA-II optimization method demonstrated high reliability with prediction errors of 1.9% and 3.4% compared to simulation.

Conclusions:

• The proposed BP neural network and NSGA-II algorithm-based optimization method is effective and reliable for improving injection molding quality.
• Optimized process parameters significantly enhance product quality by reducing critical defects.
• This approach offers a robust solution for addressing quality control challenges in plastic injection molding.