Bearing failure diagnosis at time-varying speed based on adaptive clustered fractional Gabor transform

Abstract

For bearing fault diagnosis at time-varying speed with tachometer-free and non-resampling, the crucial process is to obtain a high-resolution time-frequency representation and extract fault features. However, current multi-component non-stationary signal feature extraction methods based on time-frequency transform suffer from fixed parameter settings and insufficient resolution for low signal-to-noise ratio signals. To address these issues, a novel adaptive clustered fractional Gabor transform is proposed and applied to extract bearing fault features at time-varying speed. Firstly, the Grey Wolf optimization is utilized to adaptively search for the optimal fractional order and Gauss window length based on the maximum spectral kurtosis and the generalized time-bandwidth product to achieve the most adequate fractional Gabor spectrum. Then, the Clustering by Fast Search and Find of Density Peaks algorithm reconstructs the sparse representation of the fractional Gabor spectrum, remapping multi-component signals into single-component clusters. Bearing fault diagnosis is achieved by matching the relative order of each cluster with the bearing fault characteristic coefficients. Simulation signals validate the superiority of the feature extraction method, and experimental signals validate the feasibility of the bearing fault diagnosis method.

Intelligent fault diagnosis of double-aligned domain adaptation based on multi-structure fusion and multi-attention mechanism

“…The setting of transfer diagnosis task: the datasets of different speeds of 500, 1000 and 1500 rpm were recorded as D 2 1 , D 2 2 and D 2 3 , respectively, where {NO,IR1,IR2,OR1,OR2} ∈ D 2 i ⊆ D 2 , i = 1, 2, 3. From there, six groups of transfer tasks were constructed:…”

“…The specific experimental comparison results . The reason for this phenomenon is that the dataset D 2 3 collected at 1500 rpm is very different from the dataset D 2 1 collected at 500 rpm, and it is difficult to transfer the knowledge form the transfer task D 2 3 to the transfer task D 2 1 . On the contrary, the data distribution D 2 3 is closer to D 2 2 .…”

“…The reason for this phenomenon is that the dataset D 2 3 collected at 1500 rpm is very different from the dataset D 2 1 collected at 500 rpm, and it is difficult to transfer the knowledge form the transfer task D 2 3 to the transfer task D 2 1 . On the contrary, the data distribution D 2 3 is closer to D 2 2 . Although the diagnostic accuracy is generally low due to the large difference in working conditions in various transfer tasks, the average accuracy of the proposed method reaches 97.75%, with a maximum increase of 6.24%, and the optimal diagnostic accuracy and a small standard deviation are obtained, as shown figure 13.…”

“…In order to visually display the cross-domain features extracted by different depth transfer methods, we use the t-SNE algorithm to reduce the dimensionality of the high-dimensional features extracted by the feature extractor to obtain figure 14. Figure 14 is a variety of highdimensional features extracted under the transfer task D 2 3 → D 2 2 . Figures 14(a On the contrary, the distance of the samples of different classes is obvious and the same kind of samples are densely compacted in figure 14(g).…”

Bearings fault diagnosis under variable speed conditions by hypothesis-based FRFT technique

“…Due to the complex environmental noises and structural deformation, bearings are prone to failures. In order to reduce economic losses and prevent accidents, bearing fault diagnosis based on vibration analysis has become a crucial subject of study in academic and industrial fields for many years [1][2][3][4].…”