Research on Fault Diagnosis of Rotating Machinery based on Entropy Weight Method
Under the background of big data,the research on fault diagnosis of rotating machinery based on machine learning is developing rapidly.For nonlinear and non-stationary vibration signals,this paper proposes a PSO-SVM fault diagnosis model based on entropy weight method to strengthen the feature extraction ability.The data in this paper are from the CWRU data set and measured data of pumped storage power station.Firstly,the four kinds of fault sliding sampling,smoothing and noise reduction of every data set are preprocessed respectively.Secondly,the fault sample VMD is decomposed,and the sample entropy,energy entropy,fuzzy entropy and power spectrum entropy are used to construct the feature vector.The entropy weight method is used to select the feature vector with the largest weight as the input of the EWM-PSO-SVM model,and the diagnosis results are obtained.At the same time,the validity and accuracy of the method are verified by comparison with other methods.
NETL
NSTL
万方数据
