Undeterministic analysis and process optimization for short-fiber composite injection molding

When designing vehicle parts, reliability is required to be insensitive to the change of design parameters, so as to improve product safety and robustness. The traditional design method mainly considers the deterministic model and ignores the influence of noise factors. Based on deterministic optimization, a multi-objective robust design model with multi-quality characteristics is proposed in this paper. The response surface model (RSM) of controllable factors, noise factors and quality objectives is established. Then based on the RSM model, it is applied to do robust optimization design with undeterministic analysis method, 6 Sigma optimization design and NSGA-II optimization algorithm. Taking the optimization design of the vehicle navigation fixed base as an example, the feasibility and effectiveness of the optimization method are verified by numerical simulation and actual injection molding experiment respectively.The results show that the 6 Sigma robust design achieves the minimum mean square error and improves the robustness of the optimized solution, compared to the deterministic optimization. • The traditional design method ignores the influence of noise factors. • A multi-objective robust design model with multi-quality characteristics is proposed. • The response surface model of controllable factors, noise factors and quality objectives is established. • Do robust optimization design with undeterministic analysis method and NSGA-II optimization algorithm.