Initial Fault Feature Extraction for Rolling Bearings Based on Piecewise Matching Pursuit

In the initial fault of rolling bearings, impulse signal reflecting the fault feature is too weak to extract from strong noise and harmonic components. For this reason, a fault feature extraction method based on piecewise matching pursuit is proposed. Based on impulse structural characteristics, a Gabor dictionary was formed. In sparse decomposition, original signal was divided into several dyadic pieces, and the best atom was search by artificial fish swarm algorithm to match the impulse in each interval signal. An enhanced threshold method is put forward to enlarge the weak impulses in simulating experiments. The analysis results of the experiments show that the proposed method can successfully extract the impulse from noise of a certain intensity and harmonics.