CAE-Based Robust Optimization for Deep-Drawing Process of Sheet Metal Forming Using Grey Relational Analysis Method

The theory of grey systems is a new technique for performing prediction, relational analysis and decision making in many areas. The grey relational analysis was used to optimize the deep-drawing process parameters with considerations of the multiple response (the wrinkle, crack and thinning variation). The deep-drawing parameters, such as the blank holding force (F_ bh ), the radii of punch and die (R_1,R_2), the coefficients of friction (μ_1,μ_2,μ_3) are considered. An orthogonal array is used for the experimental design. The multiple response values are obtained making use of finite element analysis (FEA). Optimal process parameters are determined by the grey relational grade obtained from the grey relational analysis for multi-performance characteristics (the wrinkle, crack and the thinning). The analysis of variance (ANOVA) for the grey relational grade is implemented. The results show that the quality of stamped parts can be improved effectively through the new approach. The grey relational analysis can be applied in sheet metal forming.     

基于RBF神经网络模型的板料成形变压边力优化

为了解决变压边力优化过程中RBF(radial basis function )神经网络隐层节点训练难的问题,利用人工智能算法的优越性,建立了基于人工免疫算法的RBF神经网络,并将其用于非线性函数的逼近中.结合分块压边圈与改变压边力控制技术,通过Dynaform软件进行数值模拟获得成形数据,建立了变压边力与成形质量之间的RBF神经网络近似模型.利用人工免疫智能算法对该近似模型进行优化,获得最优压边力参数.将该方法应用于S形梁冲压成形中,与优化前的结果进行比较,采用优化后最优变压边力可以抑制起皱,最大起皱量减少了89.53%.