A clustering K-SVD-based sparse representation method for rolling bearing fault diagnosis

Abstract

It is challenging to extract weak impulse features from vibration signals corrupted by strong noise in mechanical fault diagnosis. Due to its simple calculation, fast convergence and easy implementation, K-singular value decomposition (K-SVD) has been widely used in rolling bearing fault diagnosis. However, it fails to consider the influence of noise and harmonics on atoms learning from impulse characteristics, which results in many irrelevant atoms, and then increases the difficulty of extracting the impulse features in bearing fault signals. Therefore, a clustering K-SVD-based sparse representation method is proposed in this paper and it is combined with the particle swarm optimisation (PSO)-based time-varying filter empirical mode decomposition (TVF-EMD) for rolling bearing fault diagnosis. The PSO-based TVF-EMD is developed to automatically decompose the original signal to eliminate the impact of noise and harmonics on the impulses in the signal. Then, the clustering K-SVD method is applied to perform dictionary learning on the sensitive component containing impulses to obtain a redundant dictionary of atoms with obvious impulse patterns. Finally, the orthogonal matching pursuit (OMP) algorithm is introduced to extract the fault features from rolling bearing vibration signals. The experimental results show that the proposed method can improve the robustness of the dictionary atoms to noise and achieve the extraction of rolling bearing fault features.

Clustering Application for Condition-Based Maintenance in Time-Varying Processes: A Review Using Latent Dirichlet Allocation

“…In these contributions, with an unusual application, a k-means clustering helps to evaluate the efficacy of the proposed model (e.g., a denoising model) [39]. Another contribution, again with the help of kmeans, suggests the application of K-SVD-based sparse representation method clustering to implement dictionary learning and extract information from vibration impulses [40]. Another contribution proposes the application of clustering based on spectrum overlap coefficient and kurtosis index to implement the extraction of periodic pulses from bearing vibration signals [41].…”

Clustering Application for Condition-based Maintenance in Time-varying Processes: A Review

“…a denoising model) [39]. Another contribution, again with the help of k-means, suggests the application of K-SVD-based sparse representation method clustering to implement dictionary learning and extract information from vibration impulses [40]. Finally, on the subject of feature extraction, another contribution proposes the application of clustering based on spectrum overlap coefficient and kurtosis index to implement the extraction of periodic pulses from bearing vibration signals [41].…”

Improved EEMD and overlapping group sparse second-order total variation

“…The running conditions of motor rolling bearing are complicated and its vibration signal is easily affected by noise and other excitation sources, thus resulting in non-stationary and nonlinear characteristics of vibration signals [1]. If fault features can be efficiently extracted in the early failure stage, the damaged bearing can be replaced or repaired in time so that the economic loss caused by the fault can be largely avoided [2].…”