A new fault feature extraction method of rotating machinery based on finite sample function

Abstract

Feature extraction plays a crucial role in the diagnosis of rotating machinery’s faults. In order to separate different fault vibration signals from measured mixtures and diagnose the fault features of the machine effectively according to the separated signals, a blind source separation (BSS) method using kernel function based on finite support samples was proposed. The method is stronger adaptability to the score functions estimated according to finite support observed signal samples. The simulation results prove that the proposed BSS algorithm is able to separate mixed signals that contain both sub-Gaussian and super-Gaussian sources. It is shown that the algorithm has better separation performance when compared with other BSS ones. The results of an experiment under the rotor’s composite fault states with rub-impact fault and unbalance fault show that this method has higher efficiency and accuracy.