基于GA-BP网络的多轴机床热误差建模

基金项目:

“高档数控机床与基础制造装备”重大科技专项项目(2010ZX04015-011)资助

Modeling of the Multi-axis NC Machine Tool Thermal Error Based on GA-BP Neural Network

1.
Institute of Advanced Design and Manufacturing,School of Mechanical Engineering, Southwest Jiaotong University,Chengdu 610031

摘要: 为最大限度减少热误差对多轴联动机床加工精度的影响,综合遗传算法全局收敛性和人工神经网络局部搜索快速性的优点,提出一种基于遗传算法优化BP网络隐层节点数及初始值的机床热误差建模方法。运用Matlab-GUI工具开发了具有通用性的交互式多轴机床热误差建模仿真系统,通过与传统的BP神经网络进行对比分析及试验论证,证明该模型预测精度更高、通用性强。
- 数控机床 /
- 热误差 /
- 遗传算法 /
- 神经网络 /
- Matlab-GUI
Abstract: In order to eliminate the influence of thermal error on the machining precision, a novel model for thermalerror in machine tools is proposed by using the BP via GA to optimize BP neural network structure and parameters.A versatile interactive thermal error modeling simulation system interface is designed by using Matlab-GUI, the results show that the present model performs better than the traditional BP model.It is good in fitting and precise inprediction for a great number of working conditions.
- CNC machine tool /
- thermal error /
- genetic algorithms /
- neural networks /
- Matlab-GUI

[1]	Ni J. CNC machine accuracy enhancement through real-time error compensation[J]. Journal of Manufacturing Science and Engineering, 1997, 119(11)
[2]	李永祥, 童恒超, 曹洪涛等. 数控机床热误差的时序分析法建模及其应用[J]. 四川大学学报, 2006, 38(2): 74～78
[3]	杨建国, 任永强, 朱卫斌等. 数控机床热误差补偿模型在线修正方法研究[J]. 机械工程学报, 2003, 39(3): 81～84
[4]	闫嘉钰, 杨建国. 灰色 GM(X, N) 模型在数控机床热误差建模中的应用[J]. 中国机械工程, 2009, 20(11), 1297～1300
[5]	林伟青, 傅建中, 陈子辰等. 数控机床热误差的动态自适应加权最小二乘支持矢量机建模方法[J]. 机械工程学报, 2009，45(3): 178～182
[6]	傅建中, 陈子辰. 精密机械热动态误差模糊神经网络建模研究[J]. 浙江大学学报(工学版), 2004, 38(6)
[7]	杜正春, 杨建国, 窦小龙等. 基于 RBF 神经网络的数控机床热误差建模[J]. 上海交通大学学报, 2003, 37(1): 26～29
[8]	Mize C D, Ziegert J C. Neural network thermal error compensation of a maching center[J]. Precision Engineering, 2000, 24(4): 338～346
[9]	Yang H, Ni J. Dynamic neural network modeling for nonlineer、nonstationary machine tool thermally induced error[J]. International Journal of Machine Tools and Manufacture, 2005, 45:445～465