贝叶斯证据框架下的LS-SVM多工况数控机床热误差建模

摘要/Abstract

摘要：

最小二乘支持向量机(LS-SVM)模型是表征数控机床热误差特性的有效工具，但该模型中的参数设置直接影响建模的精度。传统的基于交叉验证法或网格法的参数获取方法存在计算量大、精度低的缺点，且同一组模型常数往往不能准确表征机床多种工况条件下所产生的热误差。为解决这一问题，提出了一种基于贝叶斯证据框架理论的LS-SVM多工况参数优化方法。通过测量不同工况下数控机床温度值与主轴热变形量，采用贝叶斯证据框架的3个推断对LS-SVM模型进行训练并对参数进行辨识和优化，推导出了不同工况所对应的最优模型和参数。热误差建模实验验证了该参数优化方法的有效性，结果显示，经优化的模型具有泛化能力强、预测精度高、计算速度快的特点，能够较准确地描述多种典型工况条件下的实际热误差特性。

关键词: 贝叶斯证据框架, 最小二乘支持向量机(LS-SVM), 热误差建模, 多工况, 参数优化

Abstract:

LS-SVM is an effective tool for machine error modeling. The traditional methods to set the parameters of LS-SVM which determined the modeling accuracy included ten-fold cross validation and grid method. However, using these methods, parameter calculation was complex and prone to low accuracy. Moreover, the LS-SVM model with only one set of parameters was hard to precisely describe the thermal error behaviors under the different work conditions. In order to solve these problems, based on Bayesian evidence framework a novel method was proposed to identify and optimize LS-SVM parameters under multiple working conditions. Three inferring levels of Bayesian evidence framework were used to derive the optimal model parameters corresponding to the different operating conditions. A series of experiments for thermal error modeling verified the validity of this method. LS-SVM model based on Bayesian evidence framework has good generalization ability, accurate prediction, and rapid calculation speed, so it can describe the actual thermal error characteristics more accurately under multiple work conditions.

Key words: Bayesian evidence framework, least square support vector machine(LS-SVM), thermal error modeling, multiple operating condition, parameter optimization

中图分类号:

TH161

余文利, 姚鑫骅, 傅建中, 孙磊. 贝叶斯证据框架下的LS-SVM多工况数控机床热误差建模[J]. 中国机械工程, 2014, 25(17): 2361-2368.

Yu Wenli, Yao Xinhua, Fu Jianzhong, Sun Lei. Modeling of CNC Machine Tool Thermal Errors Based on LS-SVM within Bayesian Evidence Framework[J]. China Mechanical Engineering, 2014, 25(17): 2361-2368.

参考文献

[1]Mayr J, Jedrzejewski J, Uhlmann E. Thermal Issues in Machine Tools[J]. CIRP Annals-Manufacturing Technology, 2012, 61(2): 771-791.
[2]Pawel T, Jerzy J, Wojciech M. Methods of Machine Tool Error Compensation[J]. Journal of Machine Engineering, 2010,10(4):5-25.
[3]杨建国, 姚晓栋. 数控机床误差补偿技术与展望[J]. 世界制造技术与装备市场, 2012(3): 64-71.
Yang Jianguo,Yao Xiaodon. Present Situation and Prospect of Error Compensation Technology for CNC Machine Tools[J]. World Manufacturing Engineering & Market, 2012(3): 64-71.
[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.
Yan Jiayu, Yang Jianguo. Application of Grey GM(X, N) Model on CNC Machine Thermal Error Modeling[J]. China Mechanical Engineering, 2009, 20(11): 1297-1300.
[5]吴雄彪, 姚鑫骅, 傅建中. 基于贝叶斯网络的数控机床热误差建模[J]. 中国机械工程, 2009, 20(3): 293-296.
Wu Xiongbiao,Yao Xinhua, Fu Jianzhong. Thermal Error Modeling of NC Machine Tools Based on Bayesian Networks[J]. China Mechanical Engineering, 2009, 20(3): 293-296.
[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): 742-746.
Fu Jianzhong, Chen Zichen. Research on Modeling Thermal Dynamic Errors of Precision Machine Based on Fuzzy Logic and Artifical Neural Network[J]. Journal of Zhejiang University(Engineering Science), 2004, 38(6): 742-746.
[7]王福吉, 贾振元, 阳江源, 等. 基于动态模糊神经网络的机床时变定位误差补偿[J]. 机械工程学报, 2011, 47(13): 175-179.
Wang Fuji, Jia Zhenyuan, Yang Jiangyuan, et al. Time-varying Position Error Compensation of Machine Tools Based on Dynamic Fuzzy Neural Networks[J]. Journal of Mechanical Engineering, 2011, 47(13): 175-179.
[8]Zhang Y, Yang J G, Jiang H. Machine Tool Thermal Error Modeling and Prediction by Grey Neural Network[J]. International Journal of Advanced Manufacturing Technology, 2012, 59(9/12): 1065-1072.
[9]张毅, 杨建国. 基于灰色理论预处理的神经网络机床热误差建模[J]. 机械工程学报, 2011, 47(7): 134-139.
Zhang Yi, Yang Jianguo.Modeling for Machine Tool Thermal Error Based on Grey Model Preprocessing Neural Network[J]. Journal of Mechanical Engineering, 2011, 47(7): 134-139.
[10]Vapnik V. Nature of Statistical Learning Theory[M]. New York: Springer Verlag, 1999.
[11]Suykens J A K, Vandewalle J. Least Squares Support Vector Machine Classifiers[J]. Neural Processing Letters, 1999, 9(3): 293-300.
[12]林伟青,傅建中,许亚洲,等. 基于在线最小二乘支持向量机的数控机床热误差建模与补偿[J]. 计算机集成制造系统, 2008, 14(2): 295-299.
Lin Weiqing, Fu Jianzhong, Xu Yazhou, et al. Thermal Error Modeling & Compensation of Numerical Control Machine Tools Based on On-line Least Squares Support Vector Machine[J]. Computer Integrated Manufacturing Systems, 2008, 14(2): 295-299.
[13]林伟青,傅建中,许亚洲,等. 基于最小二乘支持向量机的数控机床热误差预测[J]. 浙江大学学报(工学版), 2008, 42(6): 905-908.
Lin Weiqing, Fu Jianzhong, Xu Yazhou, et al. Thermal Error Prediction of Numerical Control Machine Tools Based on Least Squares Support Vector Machines[J]. Journal of Zhejiang University(Engineering Science), 2008, 42(6): 905-908.
[14]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.
Lin Weiqing, Fu Jianzhong, Chen Zichen, et al. Modeling of NC Machine Tool Thermal Error Based on Adaptive Best-fitting WLS-SVM[J]. Journal of Mechanical Engineering, 2009, 45(3): 178-182.
[15]Mackay D J C. Probable Network and Plausible Predictions-a Review of Practical Bayesian Methods for Supervised Neural Networks[J]. Network Computation in Neural Systems, 1995, 6(3): 469-505.
[16]Boser B E, Guyon I M, Vapnik V N. A Training Algorithm for Optimal Margin Classifiers[C]//Proceedings of 5th Annual Workshop on Computational Learning Theory. Pittsburgh: ACM, 1992: 144-152.
[17]荣海娜, 张葛祥, 金炜东. 系统辨识中支持向量机核函数及其参数的研究[J]. 系统仿真学报, 2006,18(11): 3204-3208.
Rong Haina, Zhang Gexiang, Jin Weidong. Selection of Kernel Functions and Parameters for Support Vector Machines in System Identification[J]. Journal of System Simulation, 2006,18(11): 3204-3208.
[18]Bishop C M. Neural Networks for Pattern Recognition[M]. Oxford：Oxford University, Press, 1995.
[19]Mackay D J C. Bayesian Interpolation[J]. Neural Computation, 1992, 4(3): 415-447.
[20]林伟青, 傅建中. 数控机床热误差建模中的温度传感器优化研究[J]. 现代设计技术, 2007, 24(3): 5-8, 28-29.
Lin Weiqing, Fu Jianzhong. Research on Optimization of Temperature Sensors in the Modeling of Thermally Induced Error of NC Machine Tool[J]. Group Technology & Production Modernization, 2007, 24(3): 5-8, 28-29.