Bearing fault feature extraction method based on RIME optimised VMD-HHT
In order to solve the current problems of difficulty in rolling bearing fault feature extraction and blind selection of mode number and penalty factor when performing Variational Modal Decomposition(VMD),as well as the problem that the Fourier analysis spectrum only reflects the possibility of the existence of a certain frequency in the signal compared with the HHT marginal spectrum,this paper proposes a bearing fault feature extraction method based on the RIME optimised VMD-HHT.Firstly,the rolling bearing signal is analysed by using the Rime Optimization Algorithm(RIME),and the optimal number of decomposition layers and penalty factor are calculated by using the sample entropy as the fitness function;then,based on the optimal decomposition parameters obtained,the bearing signal is decomposed to obtain the modal components,and then the validity is verified based on the centre frequency,and it is compared with the Northern Goshawk Optimization Algorithm(NGO)optimization VMD method,the spectral characteristics of each modal component are obtained by using the Hilbert transform;finally,the eigenparameters of each modal component are calculated to form a set of eigenquantities,which are used to identify the bearing fault signal.The experimental results show that the parameters obtained by this method are reasonable,effective and optimal,and the proposed feature extraction method can effectively decompose the faulty signals of rolling bearings and construct the corresponding feature set.
万方数据
维普
