基于RBF神经网络断路器注塑成型工艺优化

Injection Molding Process Optimization of Circuit Breaker Based on RBF Neural Network

塑壳断路器一般通过注塑成型工艺制得。在注塑成型过程中,模具温度、熔体温度、保压压力以及保压时间均对制件的翘曲变形产生一定的影响。以模具温度、熔体温度、保压压力以及冷却时间作为研究参数,以翘曲变形量作为研究目标,采用最优拉丁超立方抽样法抽取合适的样本,建立RBF神经网络模型,结合遗传算法对制件的翘曲变形量进行优化,得到最佳的成型工艺参数组合。结果表明:四个因素的影响程度大小为模具温度>冷却时间>保压压力>熔体温度。当模具温度为50℃、熔体温度为250℃、保压压力为60 MPa以及冷却时间为10 s时,制件的翘曲变形量最小为2.3077 mm,相较未优化前降低1.2942 mm,制件成型质量得到明显改善。

Molded case circuit breaker is generally made by injection molding process.In the process of injection molding,the mold temperature,melt temperature,holding pressure and cooling time all have a certain influence on the warpage of the part.The mold temperature,melt temperature,holding pressure and holding time were taken as the research parameters,and the warpage deformation was taken as the research target.The optimal Latin hypercube sampling method was used to extract the appropriate samples,and the RBF neural network model was established.The warpage deformation of the part was optimized by genetic algorithm to obtain the best combination of molding process parameters.The results show that the influence degree of the four factors is:mold temperature>cooling time>holding pressure>melt temperature.When the mold temperature is 50℃,the melt temperature is 250℃,the holding pressure is 60 MPa and the cooling time is 10 s,the minimum warpage deformation of the part is 2.3077 mm,which is 1.2942 mm lower than that before optimization,and the forming quality of the part is obviously improved.