PREDICTION MODEL CONSTRUCTION FOR RESIDUAL STRESS DISTRIBUTION IN HIGH SPEED MACHINING

摘要：

This paper presents a residual stress prediction model for high-speed machining using the finite element method in conjunction with neural network. The finite element method is utilized to simulate a chip formation process, which is constituted step by step from the workpiece removal under the conditions of high-speed machining. The residual stress distributions underneath the machined surface of the workpiece after the cutting tool edge plows the workpiece are determined. The artificial neural network is in turn applied to synthesize the data calculated from the finite element method and a prediction model for residual stress distributions within the machined subsurface of the workpiece is thus constructed. The model can predict the residual stress component distributions at different locations beneath the machined surface of the workpiece for various workpiece materials under different combinations of cutting conditions such as cutting speed, feed rate, rake angle and edge radius of the tool more effectively.

关键词： residual stress high-speed machining finite element method neural network

作者: S.Y. Lin S. H. Cheng

作者单位: Institute of Mechanical and Electro-Mechanical Engineering, National Formosa University, 64 Wunhua R Institute of Mechanical and Electro-Mechanical Engineering,National Formosa University, 64 Wunhua Rd

会议类型: 国际会议

会议名称: 第二十届国际生产研究大会

会议地点: 上海

会议语种:英文

页码: 1-6

在线出版日期: 2009-08-02（万方平台首次上网日期，不代表论文的发表时间）