基于BP神经网络的空间轮廓误差自适应补偿

J4 ›› 201016, Vol. 21 ›› Issue (16): 1902-1906.

基于BP神经网络的空间轮廓误差自适应补偿

李小力;陈威;闫蓉

华中科技大学国家数控系统工程技术研究中心,武汉,430074

出版日期:2010-08-25 发布日期:2010-08-25

Adaptive Compensation of Contour Errors Based on BP Neural Networks

Li Xiaoli;Chen Wei;Yan Rong

National Numerical Control System Engineering Research Center,Huazhong University of Science and Technology,Wuhan,430074

Online:2010-08-25 Published:2010-08-25

摘要/Abstract

摘要：

提出了矢量化的空间轮廓误差建模方法,并基于BP神经网络提出了交叉耦合控制器参数的自适应调整方法,使用三层神经元,根据进给速率以及曲率的变化进行控制器参数的自适应调整,而在补偿量的分配问题上则充分利用已有数据,

以减小系统的计算量。实例仿真表明,所提出的自适应控制方法在进给速率设定值变化以及加工轮廓的半径发生变化时,都可以很好地控制轮廓误差的大小,提高轮廓加工精度。

关键词:

轮廓误差;交叉耦合控制器;神经网络;自适应控制

Abstract:

A vector modeling method of contour errors that simplified the error calculation was developed to improve the exiting methods.After that, an adaptive tuning method of parameters for cross coupling controller was presented to deal with the changing feedrate and path curvature based on BP neural network with simple three layer nerve cell. Moreover, the calculated amount of the system can be reduced effectively using exiting

data to distribute compensation proportion. A simulation study shows that the adaptive control method presented herein can improve the contour machining precision according to the control of contour errors under conditions of changeable feedrate and path curvature. 

Key words: contour error, cross coupling controller, neural network, adaptive control

中图分类号:

TP242.2

李小力, 陈威, 闫蓉.

基于BP神经网络的空间轮廓误差自适应补偿

[J]. J4, 201016, 21(16): 1902-1906.

LI Xiao-Li, CHEN Wei, YAN Rong.

Adaptive Compensation of Contour Errors Based on BP Neural Networks

[J]. J4, 201016, 21(16): 1902-1906.