Fault prediction of the CNC machine tool servo system based on the BRB

In order to achieve optimal maintenance and Prediction and Health Management (PHM) of CNC (Computer Numerical Control) machine tool, a belief rule base (BRB) is proposed to predict the fault of the servo system by studying the key techniques of fault prediction. Based on the analysis of fault mechanism in CNC machine tool servo system, a BRB based fault prognosis model of the servo system is established by combing expert knowledge with quantitative information. Moreover, the evidential reasoning (ER) algorithm is used to achieve fault on-line prediction. Experimental results show that the proposed method can accurately reflect the behavior of the system and take full use of uncertain information to improve the accuracy of fault prediction.