Reliability prediction of CNC machine tools based on GM(1,1)-SVM combined model

Abstract

Aiming at the characteristics of low cost and complex influencing factors of CNC machine tool failure data sample collection, a combination prediction model based on gray prediction and support vector machine is proposed, use Python and MATLAB software to predict the time between failures of CNC machine tools in the SVM model and the GM(1,1) model, respectively, and obtain the predicted time between failures, use the standard deviation method to determine the weight of the combined model, and establish GM(1,1)-SVM combined prediction model, the accuracy and practicability of the combined forecasting model are verified through examples. The results show that the prediction of the combined model is higher than that of the single model and the results are reliable, the combined model provides a theoretical basis for the CNC machine tool users and maintenance personnel to maintain it in advance, and has guiding value for the reliability prediction.