摘要
研究了电火花加工(EDM)技术的加工机理.以峰值电流、脉冲宽度、脉冲间隔、抬刀时间和加工时间为输入参数,并以加工速度和表面粗糙度为输出参数,分别用神经网络技术与非线性回归技术建立了EDM工艺模型.经过与实验数据的比较,认为这两种模型均能较精确地预测出给定条件下的加工速度和表面粗糙度,反映了该机床的加工工艺规律.其中。
This paper presented an attempt at modeling the process through nonlinear regression and artificial neural networks (ANN). The peak current, pulse width, pulse interval, lifting time and machining time were selected as the input parameters, while the material removal rate and the surface roughness were the output parameters. Verification experiments were carried out to check the validity of the developed models. It is concluded that both models provide accurate results for process, and the ANN model fits the experiments result better.
出处
《上海交通大学学报》
EI
CAS
CSCD
北大核心
2000年第3期299-302,共4页
Journal of Shanghai Jiaotong University
基金
上海市青年科技启明星计划资助项目! ( 96 QF140 0 6 )
关键词
电火花加工
神经网络
非线性回归
工艺效果
建模
electrical discharge machining
neural networks
nonlinear regression
process modeling
