基于GS-SVM的数控机床热误差预测研究.docxVIP

MECHANICALENGINEERME

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

机械工程师

MECHANICALENGINEER

MECHANICALENGINEERMECHANICALENGINEERMECHANICALENGINEERMECHANICALENGINEERMECHANICALENGINEERMECHANICALENGINEERMECHANICAL

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

MECHANICALENGINEER

机械工程师

MECHANICALENGINEER

PAGE36圆园员

PAGE36

圆园员9年第11期网址：电邮：hrbengineer@163.com

37网址：

37

网址：电邮：hrbengineer@163.com圆园员9年第11期

基于GS-SVM的数控机床热误差预测研究

张腰，杨庆东

（北京信息科技大学机电工程学院，北京100192）

摘要：为了提高立式加工中心热误差的预测精度，考虑影响立式机床热误差的主要因素，提出了支持向量机预测热误差

摘要：为了提高立式加工中心热误差的预测精度，考虑影响立式机床热误差的主要因素，提出了支持向量机预测热误差预测模型，并用网格搜索算法对支持向量机的参数惩罚因子和核参数进行搜索。通过实验数据分析得出结论，基于GS-SVM模型对数控机床主轴热变形能够进行准确预测，并与BP算法模型进行对比。结果表明GS-SVM主轴热变形预测模型效果优于BP算法预测算法，具有较高的预测精度，为数控机床热误差研究提供参考。

关键词：数控机床；热误差建模；网格搜索算法；支持向量机

中图分类号:TG659；TG502.15 文献标志码:粤 文章编号：员园园圆原圆猿猿猿（圆园员9）11原园036原园4

（SchoolofMechanicalandElectricalEngineering,BeijingInformationScienceandTechnologyUniversity,Beijing100192,China）Abstract:Inordertoimprovethepredictionaccuracyofthermalerrorinverticalmachiningcenterandconsiderthemainfactorsaffectingthethermalerrorofverticalmachinetools,apredictivethermalerrorpredictionmodelbasedonsupportvectormachineisproposed.Thegridpenaltyalgorithmisusedtosearchtheparameterpenaltyfactorandkernelparameterofthesupportvectormachine.Throughtheanalysisofexperimentaldata,itisconcludedthatthe