摘要
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 mold flow analysis was carried out by Moldflow soft ware. In order to reduce the warping deformation, improve the quality of plastic parts, and to meet the requirements of assembly, the orthogonal test was set and the optimal combination of factor levels was obtained. According to the data, the BP neural network forecast model was established. Then the accuracy of the model was verified by using the test sample. The results show that the errors of the simulation value and the forecast value are both within ±3%.
出处
《塑料科技》
CAS
北大核心
2016年第4期76-80,共5页
Plastics Science and Technology