为了提高数控机床几何误差建模精度，改进补偿效果，先用测力环等仪器模拟施加并测量机床主切削力，再用激光干涉仪同步测量机床俯仰角和偏摆角误差. 根据粒子群优化算法(particle swarm optimization,PSO)优化支持向量机(support vector machine, SVM)的相应参数，并以实际测量数据进行训练，从而建立
了PSO-SVM力-几何误差预测模型. 实际试验表明，PSO-SVM误差预测模型输出的偏摆角误差预测值与实测数据的最大差值仅为0.6 μrad，俯仰角误差预测值与实测数据的最大差值仅为0.21 μrad，远小于利用BP神经网络以及常规方法优化的SVM所建立的力-几何误差预测模型的误差，因此该模型可用于数控机床几何误差的高精度实时补偿.
To improve modeling precision of CNC geometric error and error compensation, the main cutting force is simulated and measured using a dynamometer. Errors in the pitching and deflection angles are measured with a laser interferometer. Trained with practically measured error data, the force-geometric error
predicting model based on PSO-SVM is established with the key parameters optimized using the particle swarm optimization algorithm (PSO). Verification experiments show that difference between the measured error and the maximum deflection angle error using the PSO-SVM model is 0.6 μrad, and that with the maximum error of pitching angle error is 0.21 μrad. Compared with the force-geometric error predicting model based on BP neural networks and SVM whose parameters is optimized using a conventional method, the prediction precision of the PSO-SVM error model is greatly improved. Therefore the proposed model can compensate geometric error of CNC machine tools in real-time with high-precision.
[1] 费业泰，赵静，王宏涛等. 三坐标测量机动态误差研究分析[J].仪器仪表学报，2004, 25(4)：773-776.
Fei Yetai, Zhao Jing, Wang Hongtao, A Review of Research on Dynamic Errors of Coordinate Measuring Machine[J]. Chinese Journal of Scientific Instrument 2004, 25(4)：773-776.（in Chinese）
[2] F.Y. Peng, J. Y. Ma, W. Wang. Total differential methods based universal post processing algorithm considering geometric error for multi-axis NC machine tool[J]. International Journal of Machine Tools & Manufacture 70, 2013:53–62.
[3] Shaowei Zhu, Guofu Ding, Shengfeng Qin. Integrated geometric error modeling, identification and compensation of CNC machine tools[J]. International Journal of Machine Tools & Manufacture 52, 2012:24–29.
[4] J.G. Yang, Y.Q. Ren, Z.C. Du. An application of real-time error compensation on an NC twin-spindle lathe[J]. Journal of Materials Processing Technology 129, 2002: 474-479.
[5] 王维，杨建国，姚晓栋等.数控机床几何误差与热误差综合建模及其实时补偿[J].机械工程学报，2012, 48(7)：165-170.
Wang Wei, Yang Jianguo, Yao Xiaodong. Synthesis Modeling and Real-time Compensation of Geometric Error and Thermal Error for CNC Machine Tools[J]. Journal of Mechanical Engineering 2012, 48(7)：165-170. （in Chinese）
[6] 伍迪.数控机床误差补偿技术及研究现状[J].煤矿机械，2011, 32(06)：8-10.
Wu Di. Error Compensation Technology and Present Research Situation for CNC Machine Tools [J]. Coal Mine Machinery，2011, 32(06)：8-10. （in Chinese）
[7] Lei Ni, JunChen Jiang, Yong Pan. Leak location of pipelines based on transient model and PSO-SVM[J]. Journal of Loss Prevention in the Process Industries,2013：1-9.
[8] 苗恩铭，龚亚运，成天驹等. 支持向量回归机在数控加工中心热误差建模中的应用[J].光学精密工程，2013, 21(4)：980-985.
Miao Enming, Gong Yayun, Cheng Tianju. Application of Support Vector Regression Machine to Thermal Error Modeling of Machine Tools[J]. Optics and Precision Engineering, 2013, 21(4)：980-985. （in Chinese）
[9] 郑大腾，费业泰. 柔性坐标测量机空间误差模型研究[J].机械工程学报，2010, 46(10)：19-24.
Zheng Dateng, Fei Yetai. Research on Spatial Error Model of Flexible Coordinate Measuring Machine[J]. Journal of Mechanical Engineering，2010,46(10)：19-24. （in Chinese）
[10] JuiHsi Fu, SingLing Lee. A multi-class SVM classification system based on learning methods from indistinguishable Chinese official documents[J]. Expert Systems with Applications 39, 2012： 3127–3134.
[11] 杨洪涛，刘勇，费业泰等. 三坐标测量机动态误差混合建模方法[J].仪器仪表学报，2010,31(8)：1861-1866.
Yang Hongatao, Fei Yetai. Hybrid modeling method for CMM dynamic error[J]. Chinese Journal of Scientific Instrument，2010, 31(8)：1861-1866. （in Chinese）
[12] 孙林,杨世元.基于LS-SVM的温度传感器非线性关系拟合及参考端温度补偿[J].应用科学学报，2009, 27(6)： 616-622.
Sun Lin, Yang Shiyuan. Fitting of Non-linear Relation of Temperature Sensor and Reference Temperature Compensation Based on LS-SVM[J].Journal of Applied Sciences-Electronics and Information Engineering，2009,27(6)：616-622. （in Chinese）
[13] 白鹏，张喜斌，张斌等. 支持向量机理论及工程应用实例[M].西安：西安电子科技大学出版社，2008,18(7):2033-2036.
Bai Peng, Zhang Xibin, Zhang Bin. The support vector machine theory and engineering application[M]. Xi’an: Xidian university press，2008, 18(7):2033-2036. （in Chinese）
[14] 钟伟红，马修水，关宏伟等. 基于RBF神经网络的三坐标测量机动态测量误差预测[J].中国科技论文，2012, 7(7)：560-562.
Zhong Weihong, Ma Xiushui, Guan Hongwei. Dynamic measurement errors estimation of CMMs based on RBF neural network[J].China Science paper, 2012,7（7）：560-562. （in Chinese）
[15] 许志军. 基于粒子群算法优化支持向量机的数控机床状态预测[J].现代制造工程, 2011（7）：46-49.
Xu Zhijun. State prediction for CNC machine based on PSO-SVM[J]. Modern Manufacturing Engineering, 2011(7)：46-49. （in Chinese）
