Time Series Prediction for Machining Errors Using Support Vector Regression

Abstract

A time series prediction method using support vector regression (SVR) for machining errors is presented in this paper. The design steps and learning algorithm are also addressed. Since SVR have greater generalization ability and guarantee global minima for given training data, it is believed that SVR will perform well for time series for machining errors. A typical machining process of cutting bearing outer race is carried out and the real measured data are used to contrast experiment. The experimental results demonstrate the feasibility of applying SVR in machining errors prediction and prove that SVR is applicable and performs well for small-batch machining process analysis.