中国机械工程 ›› 2007, Vol. 18 ›› Issue (21): 2614-2617.

• 先进材料加工工程 • 上一篇    下一篇

基于微型多目标遗传算法的薄板冲压成形变压边力优化

刘桂萍;韩旭;姜潮   

  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-11-10 发布日期:2007-11-10

Optimization of Variable Binder Force in Sheet Metal Forming Using the Micro Multi-objective Genetic Algorithm

Liu Guiping;Han Xu;Jiang Chao   

  • Received:1900-01-01 Revised:1900-01-01 Online:2007-11-10 Published:2007-11-10

摘要:

提出一种高效的薄板冲压成形变压边力多目标优化方法,该方法以减少冲压件的成形缺陷为优化目标,以变压边力曲线的特征参数为优化变量,采用自主开发的微型多目标遗传算法作为优化算法,并在优化过程中引入神经网络近似模型以减少数值模拟的次数,提高优化效率。通过NUMISHEET’93的U形弯曲标准模型和某车型前地板角支撑板冲压成形模型两个变压边力优化实例对该方法进行了验证。结果表明,该方法既能高效率地解决薄板冲压成形变压边力优化问题,又能仅通过一次计算就提供多组方案以满足对冲压件成形质量控制的不同需要。

关键词: 多目标优化, 变压边力, 薄板冲压成形, 微型多目标遗传算法

Abstract:

A multi-objective optimization method for the optimization of variable binder force in sheet metal forming was suggested. The objectives were to minimize the forming shortcomings and the characteristic parameters of the variable binder force curve were selected as the design variables. The micro multi-objective genetic algorithm with high efficiency was adopted. An approximation model based on the neural network was also used to improve the efficiency. As an examples, a 2D draw bending problem of NUMISHEET’93 and a shoe plate of the front floor of a car stamping problem were solved using the presented method. Eventually, several variable binder force curves which could fulfill different needs are acquired efficiency.

Key words: multi-objective optimization, variable binder force;sheet metal forming, micro multi-objective genetic algorithm

中图分类号: