Fault diagnosis of rolling element bearing weak fault based on sparse decomposition and broad learning network

Abstract

Rolling element bearings are widely used in rotating machinery and, at the same time, they are easily damaged due to harsh operating environments and conditions. As a result, rolling element bearings are critical to the safe operation of the mechanical devices. The incipient fault information extraction of rolling bearings mainly faces the following difficulties: (1) The fault signal is too weak. (2) The fault mechanism and the dynamic model of the rolling bearing system are complex. (3) The oscillations caused by the fault shocks are overlapped due to the smaller impact between two adjacent faults. (4) The impact interval of the fault will change randomly. To overcome the aforementioned difficulties, a connection network constructed by resonance-based sparse signal decomposition (RSSD) and broad learning system (BLS) without the need for deep architecture, namely RSSD-BLS, is proposed for intelligent fault diagnosis. We construct RSSD-BLS by input layer, RSSD decomposition layer, feature layer and output layer. So, when the observed vibration signals are the input layer, the network first uses RSSD to decompose the raw vibration signal into high resonance components and low resonance components. Then, the network obtains energy spectrum features of high resonance components which decomposed by RSSD to extract the unique features in the feature. Finally, the network recognizes different fault conditions in the output layer. Through comparing with commonly used intelligent network diagnosis method, the superiority of the proposed RSSD-BLS is verified.

A modular fault diagnosis method for rolling bearing based on mask kernel and multi-head self-attention mechanism

“…Therefore, we randomly divide the data set into a training set and a test set to fit the data collected in the actual project. As the input of the deep network, the features of 1D data are relatively simple and insufficient to feed the network to learn all the features of the fault samples, and they are considered to be transformed into image data with time-frequency features to enrich the information contained in the vibration signals and to release the data deficiency problem to some extent (Li et al, 2020). These images are obtained from time-series data by time-frequency transformation, and the common transformation method include the Fourier transform, the Hilbert yellow transform, and the wavelet transform as well as their respective derived transformations.…”

“…With the growing demand for the operational stability of machinery and equipment, reliable condition monitoring and intelligent maintenance strategies have gradually become a current research hotspot (Cheng et al, 2022; Li et al, 2020; Zhu et al, 2019). As an important transmission component, the health status of rolling bearings is closely related to the operating performance of the whole machine.…”

Rolling bearing fault feature extraction based on maximum correlated kurtosis deconvolution and improved autocorrelation spectral kurtograph

“…Due to the influence of external noise, impact response of other components and sensor working conditions ,the early fault characteristics of rolling bearing are buried in strong background noise [2] , which make it difficult to effectively extract its fault features [3] .MED is a time-domain blind convolution technique [4] , it has been applied to the fault diagnosis of rolling bearing [5] .But the filter it solves is not necessarily the global optimal filter, and often only a few pulse components can be extracted. In view of the limitations of MED, some scholars proposed the MCKD based on the correlation kurtosis [6] , Song [7] proposed an adaptive bearing fault feature extraction method combining the advantages of UPEMD and MCKD, Chen [8] proposed a bearing ball weak fault detection method combining EEWT and EMCKD, Ren [9] proposed a rolling bearing fault diagnosis method combining SSA, VMD and MCKD.…”

Rolling bearing fault diagnosis under variable working conditions using deep convolutional fuzzy system

“…To date, the vibration signal is considered to be an important information source, and the vibration analysis is one of the most effective techniques for rolling bearing fault diagnosis (Li et al, 2020a). Many intelligent approaches have been put forward and used to realize the identification of various working conditions of rolling bearings (Li et al, 2020c; Zheng et al, 2014; Zhong et al, 2021).…”