Multi-Objective Optimization of Thin-Walled Connectors in Injection Molding Process Based on Integrated Algorithms.

Area of Science:

• Manufacturing Engineering
• Materials Science
• Computational Science

Background:

• Warpage is a significant challenge in thin-walled connector manufacturing via injection molding, impacting dimensional accuracy.
• Existing methods struggle to consistently achieve desired shape consistency and quality.

Purpose of the Study:

• To develop and validate an integrated optimization methodology for injection molding process parameters.
• To minimize warpage and volume shrinkage while managing clamping force in thin-walled connectors.

Main Methods:

• Utilized Latin Hypercube Sampling (LHS) for experimental design.
• Employed numerical simulations to assess process parameter influences.
• Developed a DBO-BP neural network for enhanced prediction accuracy.
• Applied NSGA-II for multi-objective optimization to identify optimal process parameters.

Main Results:

• The DBO-BP neural network achieved prediction accuracy within 5% for warpage and volume shrinkage.
• Identified optimal process parameters yielding a warpage of 0.173 mm and volume shrinkage of 7.5%.
• Maintained a relatively high clamping force of 15.83 tons.

Conclusions:

• The proposed methodology effectively optimizes injection molding parameters for thin-walled connectors.
• Achieved significant improvements in dimensional accuracy and shape consistency.
• Enhances production efficiency and reduces manufacturing defects.