以激光器支架为例,运用Moldflow软件进行模流分析,并设置了正交试验,以得到各因素水平的最佳组合,从而减小翘曲变形量,提高塑件质量,使其达到装配要求。然后根据所得数据建立了BP神经网络预测模型,再利用测试样本验证模型的准确性,结果发现仿真值与预测值的误差均在±3%以内。 Taking the laser bracket as an example, the Moldflow software was used to analyze the mold flow, and orthogonal test was set up to get the best combination of all factors so as to reduce warpage and improve the quality of the plastic parts to meet the assembly requirements. Then the BP neural network prediction model is established based on the obtained data, and then the test sample is used to verify the accuracy of the model. It is found that the error between the simulation value and the prediction value is within ± 3%.