摘要
针对渐进成形中板料严重减薄、易开裂和尺寸精度差及成形效率低等问题,采用BP神经网络建立了优化目标和工艺参数之间的非线性映射关系。然后利用多目标遗传算法(NSGA-II)实现方锥件渐进成形工艺参数的优化。结果表明,采用BP神经网络与多目标遗传算法能很好地优化渐进成形工艺参数,提高零件成形质量。
Aiming at the problems of serious thinning, easy cracking of sheet, poor dimensional precision and low forming efficiency in incremental forming, a nonlinear mapping relation between forming process parameters and optimization objective was established based on BP neural network. Then, the optimization of incremental forming process parameters for square cone-shaped part was realized by using the multi-objective genetic algorithm (NSGA-Ⅱ). The results show that using BP neural network and the multi-objective genetic algorithm can preferably optimize incremental forming process parameters and improve the forming quality of parts.
出处
《热加工工艺》
CSCD
北大核心
2014年第23期108-112,共5页
Hot Working Technology
基金
福建省自然科学基金资助项目(2008J0153)
福建省大学生创新创业项目(201313470019)
关键词
方锥件
渐进成形
工艺优化
数值模拟
神经网络
square cone-shaped part
incremental forming
technology optimization
numerical simulation
genetic neuralnetwork