基于NSGA-Ⅱ算法的汽车后背门内板拉延成形质量控制与预测

Quality control and prediction of drawing forming of automotive back door inner panel based on NSGA-Ⅱ algorithm

为避免汽车覆盖件冲压时出现起皱、开裂等成形缺陷,提出基于带精英策略的非排序遗传算法(NSGA-Ⅱ)对冲压成形工艺参数进行优化,实现对冲压成形质量的控制与预测。以某型汽车后背门内板作为研究对象,以压边力、摩擦系数、冲压速度和模具间隙作为试验因素,拉延变形时的最大增厚率和最大减薄率作为质量控制目标。应用拉丁超立方抽样方法并结合CAE分析技术建立试验样本,构建冲压工艺参数同成形质量控制目标之间的响应面模型,基于NSGA-Ⅱ算法在构建的响应面模型内计算获得了满足成形质量的一组最优工艺参数组合,同时给出了成形质量的预测值。实际冲压试模结果证明了文章所提方法的有效性,为汽车覆盖件的冲压成形质量控制与预测提供了一套可借鉴的方法。

In order to avoid forming defects such as wrinkling and cracking during the stamping of automobile cover parts,a non-ranking genetic algorithm(NSGA-Ⅱ)based on elite strategy was proposed to optimize the stamping process parameters and realize the control and prediction of stamping forming quality.The forming process parameters such as blank holder force,friction coefficient,forming force and die clearance were taken as test factors,and the maximum thickening rate and the maximum thinning rate during drawing deformation were taken as quality control objectives.Latin hypercube sampling method and combining with CAE analysis technology to establish the test sample,build between stamping process parameters and forming quality control target of the response surface model,based on the NSGA-Ⅱ algorithm was obtained in constructing the response surface model of a set of optimal process parameters meet the forming quality combination,at the same time,the forming quality of forecast is given.The actual stamping die test results prove the effectiveness of the method proposed in this paper,which provides a set of reference methods for the quality control and prediction of automobile cover parts.