Fault diagnosis method based on gray correlation and evidence theory

Based on the evidence theory, combing with gray correlation and information entropy theory, a new method is proposed for machinery fault diagnosis. Firstly, based on information entropy feature of machinery fault, it builds the standard feature vectors of fault diagnosis. Secondly, the Basic Probability Assignment Function (BPAF) of evidence is built by gray correlation theory, and then a space-time second-level fusion algorithm based on evidence theory is provided, which includes the time domain fusion of single sensor with multi-measuring period and the space domain fusion of multi-sensor. Finally, a decision-making method based on the basic probability number is used for the fault model recognition. The typical instance of rotational machinery indicates that the new machinery fault diagnosis method is valid and feasible for recognizing fault pattern.