Remaining Useful Life Prediction of Computer Numerical Control Machine Tool Components Considering Operating Condition Information

To improve the accuracy of predicting the remaining useful life (RUL) of computer numerical control (CNC) machine tool components, this study proposes a novel method. In the method, a condition monitoring platform for components is built to obtain component operation information. The collected information is processed to acquire signal features with better trend. The Weibull model is modified via the fusion of internal signal features and external operating information. Accordingly, a Weibull regression model that fully considers the operating condition information of components is established. The fminsearch function is applied to solve the WRM with complete parameterization, and the optimal parameter estimates of the model are obtained. The RUL prediction method is demonstrated using a specific example. Multiple indexes are used to evaluate the model performance. Furthermore, the validity of the model is verified by comparison. The proposed method can obtain more accurate RUL prediction results of the components. It is of great significance to the health operation of CNC machine tools.