Thermal error modeling and compensation for a high-speed motorized spindle

To improve the precision of CNC machine tools, a motorized spindle thermal error model based on least square support vector machine (LS-SVM) was proposed. A thermal error compensation method was implemented, which takes the length of cutting tools and thermal tilt angles into account. A five-point method was applied to measure radial thermal declinations and axial expansion of the spindle with eddy-current sensors. This resolves a problem arising out of the three-point thermal error measurement, where the radial thermal-induced angle errors cannot be obtained. Variables sensitive to thermal error were selected by grouping and optimizing temperature variables using a combined fuzzy cluster and correlation analysis. LS-SVM models were established for axial elongation and radial thermal yaw and pitch angle errors. Moreover, a method to test the goodness of prediction for the results based on the model is discussed. The results indicated that the LS-SVM has high predictive ability based on fuzzy cluster grouping, and prediction accuracy reached up to 90 %. In addition, the axial accuracy was improved by 82.6 % after error compensation, and the axial maximum error decreased from 39 to 8 mu m. Moreover, the X/Y direction accuracy can reach up to 77.4 and 86 %, respectively, which demonstrated that the proposed methodology of measurement, modeling, and compensation was effective.