利用正交试验方法,借助Mold Flow软件对车灯副反射器进行翘曲变形分析。利用BP神经网络建立注塑工艺参数(熔体温度、模具温度、注射时间、保压时间、冷却时间)与产品翘曲变形之间的关系。利用遗传算法对BP神经网络进行全局寻优,得到产品的最小翘曲变形,以此实现对注塑成型工艺参数的优化,进而减小产品翘曲变形,提高产品质量。 Using orthogonal test method, warping deformation analysis of headlamp sub-reflector is carried out with Mold Flow software. Using BP neural network to establish the relationship between the injection molding process parameters (melt temperature, mold temperature, injection time, dwell time, cooling time) and product warpage. The genetic algorithm is used to optimize the BP neural network globally to obtain the minimum warpage of the product, so as to optimize the parameters of the injection molding process, and further reduce the warpage of the product and improve the product quality.