基于数值模拟和回归正交设计的AA5754铝合金温成形回弹特性

来源期刊：中南大学学报(自然科学版)2017年第10期

论文作者：赵坤民 申国哲 葛永鹏 刘立忠

文章页码：2590 - 2597

关键词：数值模拟；回归正交试验；铝合金；温成形；回弹；回归模型

Key words：numerical simulation; regression orthogonal test; aluminum alloy; warm forming; springback; regression model

摘    要：以圆筒温拉深切环实验为基础，采用数值模拟与回归正交试验相结合的方法，建立铝合金温成形过程中板料成形初始温度、摩擦因数、压边力、凹模圆角半径和凸凹模间隙与回弹之间的回归模型，得出各工艺参数对制件回弹的影响规律，确定最优工艺参数组合。通过回归模型进行回弹预测，并与相同工艺条件下4种温度的实验结果和数值模拟结果进行比较。研究结果表明：回归模型显著性明显，拟合度高，预测结果具有较高的可信度。

Abstract: Based on the split-ring experiment for cylindrical drawing cups, a regression model of relationship between springback and process parameters, including initial forming temperature of blank, friction coefficient, blank holder force, die radius and clearance between punch and die, was established for warm forming of aluminum alloy by numerical simulation method in conjunction with the regression orthogonal test. The effect of process parameters on springback was investigated, and the optimal combination of process parameters was determined. The regression model was applied to predict the springback, whose results were compared with those of test and numerical simulation at four kinds of temperatures. The results show that with high significance and fitting degree, the regression model is reliable.