基于偏最小二乘回归的冲压件回弹预测研究

Prediction on Springback of Stamping Based on Partial Least-squares Regression

为了研究工艺参数对汽车储物箱底板冲压件回弹的影响,首先根据正交试验安排进行了3因数3水平的有限元数值仿真,得到18组模拟值(9组作为模型建立数据,9组作为模型检验数据),然后利用偏最小二乘回归方法建立了冲压件回弹的预测模型。运用该模型,预测回弹量的相对误差均在8%以下,充分验证了该预测模型的合理性及适用性。

In order to study the influence of process parameters on springback of stamping for automobile storage tank bottom plate, first,finite element numerical simulation with 3 factors 3 level were done based on the orthogonal experiment. 18 groups of simulation value were got （9 group as a model to establish data, 9 group as model to test data）; then, the prediction model of springback of stamping was established by the partial least-squares regression method. The relative errors of predicting springback are less than 8% by using this model and the rationality and applicability of the predicting model can be fully verified.