Grey Theory and Radial Basis Function Neural Network Applied to Thermal Error Compensation in a CNC Lathe

Abstract

The thermal effect on machine tools has become a well-recognized problem in response to the increasing requirement of product quality. The performance of a thermal error compensation system strongly depends on the accuracy of the thermal error model. To establish the compensation model of the thermal error of a CNC two-turret lathe, the methods of the grey theory (GT), feed-forward neural network (FNN), radial basis function neural network (RBFNN), and generalized regression neural network (GRNN) were used. Results found by the grey theory showed that the characteristic temperature rise at the spindle nose is the most important factor influencing the thermal deformation. Comparisons among all mentioned neural network models showed that the RBFNN model has the best ability to map the thermal drift to temperature ascent of the machine structure.