Title: Sparse representation of vibration signals of rolling bearing based on K-SVD combined with DCT
Abstract:To achieve a fast and effective fault diagnosis of rolling bearing, this paper proposed a dictionary based on k- singular value decomposition (K-SVD) combined with discrete cosine transform (DCT) for ...To achieve a fast and effective fault diagnosis of rolling bearing, this paper proposed a dictionary based on k- singular value decomposition (K-SVD) combined with discrete cosine transform (DCT) for sparse representation of vibration signals. Firstly, two dictionaries are separately got. One of them is directly composed of signal samples, and the other is obtained by DCT. Then orthogonal matching pursuit (OMP) is used to sparsely decompose the first dictionary based on the dictionary by DCT. Next, the sparse coefficient is constantly updated by K-SVD. Finally, the updated dictionary is obtained using the sparse coefficients of the first dictionary. After the final dictionary is obtained, the Gaussian random matrix and OMP are respectively used to compress and reconstruct. The proposed method is verified by the vibration signal of the rolling bearing. The results show that it can effectively reduce the sparse time base on ensuring the reconstruction quality. It provides a reference value for the real-time diagnosis of rolling bearing faults.Read More
Publication Year: 2021
Publication Date: 2021-07-26
Language: en
Type: article
Indexed In: ['crossref']
Access and Citation
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot