Compound Fault Diagnosis of Rolling Bearings Based on Fastica and Granger Causality Analysis

Aiming at the problem that it is difficult to separate the multi-channel compound fault signals of rolling bearing and it is difficult to determine the type of compound fault through the separated signals, a compound fault diagnosis method of rolling bearing based on Fast Independent Component Analysis (FastICA) and Granger Causality Analysis was proposed. Firstly, the multi-channel compound fault signal is obtained by mixing the single fault signal. Secondly, the FastICA algorithm is used to separate the compound fault signals, and the best estimated signal of the source signal is separated. After that, the fault type of each separated estimated signal is inferred by envelope analysis. Finally, Granger causality analysis was introduced to establish the causal relationship model between the compound fault signal and each estimated signal, and the composition of the compound fault signal was inferred to realize the compound fault diagnosis. The effectiveness of the compound fault diagnosis method is verified by the compound fault diagnosis experiment on the standard data set. The results show that the proposed method can correctly diagnose the compound fault type of rolling bearing, and has high diagnostic accuracy, which provides a new method for the compound fault diagnosis of rolling bearing.