Application of time series to thermal error compensation of machine tools

Abstract

The thermal error compensation of CNC machine tool is of great value to improving the accuracy, and the modeling method is a proximate factor of thermal error compensation and its robustness. Currently, internationally adopted modeling methods include multiple linear regression, neural network method etc. And the most commonly used modeling method is multiple linear regression, what is simple and quick. But forecast accuracy which needs to be improved limits to the application of thermal error modeling of precision CNC machine. When we model a time series modeling, we study variables and its extrapolate mechanism to forecast changes of time series, give heavier weight to the data near by the prediction, increase short-term parameters' impact of model, so as to achieve improving forecasting precision of model. And time series models have been used widely in economic, sociology and medicine, but few in thermal error modeling of CNC machine. Adopt autoregressive distributed lag model of time series analysis, and contract of results among time series model, multiple linear regression model, demonstrate that forecast accuracy can be improved using time series model, and time series analysis has a bright future.