基于粒子群算法的汽车仪表框架注塑工艺参数优化

Optimization of Injection Process Parameters for Automotive Instrument Framework Based on Particle Swarm Optimization Algorithm

以某汽车仪表框架为例,对塑件注塑后的翘曲量进行研究。选取模具温度、熔体温度、冷却时间、保压压力、保压时间和注射时间作为自变量,以翘曲量作为优化对象。通过正交试验得到各个工艺参数范围内翘曲量较小的工艺参数组合,利用BP神经网络建立塑件翘曲量和各工艺参数之间的非线性模型。将BP神经网络模型作为粒子群算法的适应度函数,通过粒子群算法的优化能力对工艺参数进行优化,并得到最小的翘曲量0.9363 mm和最优的工艺参数组合,即模具温度55.813℃,熔体温度259.568℃,冷却时间29.650 s,保压压力85.02 MPa,保压时间29.187 s,注射时间1.23 s。利用Moldflow模流分析软件对最优的工艺参数组合进行翘曲分析,得到翘曲量为0.9489 mm。与初始默认值和正交试验优化的工艺参数结果对比,翘曲量分别降低了64.48%和19.17%,达到优化塑件质量的目的。

Taking a dashboard frame as an example,the warpage of plastic parts after injection molding was studied.Selecting mold temperature,melt temperature,cooling time,holding pressure,holding time,and injection time as independent variables,and warpage was treated as the optimization object.Through orthogonal experiments,process parameter combinations with smaller warpage within each process parameter range were obtained.BP neural network was used to establish a nonlinear model between the warpage of plastic parts and various process parameters.Using the BP neural network model as the fitness function of the particle swarm algorithm,the process parameters are optimized through the optimization ability of the particle swarm algorithm,and the minimum warpage of 0.9363 mm and the optimal combination of process parameters are obtained,namely mold temperature 55.813℃,melt temperature 259.568℃,cooling time 29.650 s,holding pressure 85.02 MPa,holding time 29.187 s,and injection time 1.23 s.Using Moldflow mold flow analysis software,the optimal combination of process parameters is analyzed for warpage,and the simulated warpage is 0.9489 mm.Comparing with the initial default values and the process parameter results are optimized by orthogonal experiments,the warpage is reduced by 64.48%and 19.17%,respectively,achieving the goal of optimizing the quality of plastic parts.