Application of support vector regression machine to thermal error modelling of machine tools
| Autor: | 龚亚运 Gong Ya-yun, 苗恩铭 Miao En-ming, 陈海东 Chen Hai-dong, 成天驹 Cheng Tian-ju |
|---|---|
| Rok vydání: | 2013 |
| Předmět: |
Engineering
Support vector regression machine business.product_category business.industry Atomic and Molecular Physics and Optics Standard deviation Electronic Optical and Magnetic Materials Machine tool Support vector machine Machining Robustness (computer science) Linear regression Thermal business Algorithm Simulation |
| Zdroj: | Optics and Precision Engineering. 21:980-986 |
| ISSN: | 1004-924X |
| Popis: | This paper explored and selected an optimal thermal error model of the Computer Numberical Control(CNC) machining center to compensate the main error source,the thermal error of spindle,in the machine processing and to improve the machining accuracy.In experiments,the leaderway-V450 machining center was taken as a compensation object,and the Support Vector Regression(SVR)model and Multiple Regression(MLR) model were analyzed and compared.Firstly,the MLR model and the SVR model were established according to the first batch of data of the CNC center gained in summer.Then,by substituting the second batch of data measured in summer into two kinds of models respectively,the compensation accuracy of each model was calculated.Furthermore,by substituting the third batch of data measured in autumn into two kinds of models respectively,the compensation accuracy of each model was calculated again.Finally,the robustness between both models was compared according to the precision variation regulation.The experiment shows that the compensation standard deviations of SVR model both in summer and autumn are less than 2 μm,and that of MLR model in summer is less than 2 μm,while less than 8 μm in autumn.These data show that the SVR model not only has high accuracy,but also has higher robustness for the thermal error modeling of CNC center. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|