噪声干扰下基于PCA-SF的轴承故障诊断方法OA北大核心CSTPCD

Mechanical fault diagnosis is very important for reducing maintenance cost and preventing accidents.Vibration signal monitoring is an effective and feasible method for mechanical fault diagnosis.However,the collected fault signals can be interfered easily by the external noise from the other equipment.Therefore,it is the key as well as the difficulty of fault diagnosis to extract the periodic pulse related to the faults from the noise interfered monitoring signals.To solve this problem,this paper proposed a mechanical fault feature extraction method based on Principal Component Analysis(PCA)and Sparse Filtering(SF).Specifically,PCA is used to extract the principal components of noise interfered signal segments.And then SF is used to extract effective features from the principal components.In order to reduce the over-fitting problem of SF model,the L1/2 norm is used to regularize the objective function.Finally,the extracted features are input into Softmax classifier for fault identification.The effectiveness of the proposed PCA-SF method is verified by a set of simulation and experimental cases.Experimental results show that the proposed method can achieve accurate fault classification,and also outperform other traditional methods.