基于响应面法和广泛学习粒子群算法的注塑件体积收缩优化

Volume shrinkage optimization of injection parts based on RSM and CLPSO

以柜式空调面板为研究对象,将体积收缩率作为优化目标。利用实验设计分析出对体积收缩率影响较大的4个工艺参数为熔体温度、模具温度、保压压力及保压时间。以这4个工艺参数为实验变量,通过响应面法(RSM)构建出它们与体积收缩率之间的响应面模型,使用残差分析检验响应面模型的拟合质量,并对响应面法优化结果进行CAE模拟验证,得出响应面模型具有较高精度。最后,运用广泛学习粒子群算法(CLPSO)对响应面模型进行迭代寻优,并验证优化结果。结果表明,将实验设计、响应面法及广泛学习粒子群算法相结合的优化效果好,能够显著降低塑件体积收缩率。

The cabinet air conditioning panel was taken as the research object,and the volume shrinkage rate was taken as the optimization objective. It was analyzed by design of experiments that the four process parameters which had great influence on volume shrinkage rate were melting temperature,mold temperature,holding pressure and holding time. Then,taking the four parameters as the experimental variables,the response surface model between the four parameters and volume shrinkage rate was constructed by response surface methodology( RSM). Residual analysis was used to test the fitting quality of response surface model,and the optimization results of response surface method were verified by CAE simulation. It is concluded that the response surface model has high accuracy. Finally,the response surface model was iteratively optimized by using the comprehensive learning particle swarm optimization( CLPSO),and the optimization results were verified. The results show that the combination of experimental design,RSM and CLPSO has good optimization effect,and can significantly reduce the volume shrinkage rate of plastic parts.