An effective multi-channel fault diagnosis approach for rotating machinery based on multivariate generalized refined composite multi-scale sample entropy

Abstract

Fault diagnosis of critical rotating machinery components is necessary to ensure safe operation. However, the commonly used fault diagnosis methods for rotating machinery are generally based on the single-channel signal processing method, which is not suitable for processing multi-channel signals. Thus, to extract features and carry out the intelligent diagnosis of multi-channel signals, a novel method for rotating machinery fault diagnosis is proposed.Firstly, a novel non-linear dynamics technique, named the multi-variate generalized refined composite multi-scale sample entropy, is presented and applied to extract fusion entropy features of multi-channel signals. Secondly, a practical manifold learning known as supervised isometric mapping is introduced to map the high-dimensional fusion entropy features in a low-dimensional space. In a third step, the Harris hawks optimization-based support vector machine is utilized to achieve intelligent fault recognition. Finally, aiming to verify the effectiveness of the proposed method and present its advantages, it was applied to analyze the rotating machinery system bearing and gear data. The experimental results have shown that the method at hand can accurately identify various bearing and gear faults. Furthermore, in addition to being suitable for multi-channel signal fault diagnosis, it displayed higher recognition accuracy when compared with other multi-channel or single-channel methods.

Multivariate multi-scale cross-fuzzy entropy and SSA-SVM-based fault diagnosis method of gearbox

“…Hence, the fault information hidden in the other channels will be ignored. Drawing inspiration from the multi-channel analysis methods [34][35][36][37] and building upon the MCFE along with multivariate embedding theory [38,39], this study proposes MvMCFE. The MvMCFE method is employed to assess the synchrony, similarity, and mutual predictability of two multivariate time series, thereby elucidating the dynamic changes of vibration signals across different channels.…”

Investigation on the fusion reliability and cluster consistency of multivariable entropy method

“…Wei et al [23] combined composite processes with multivariate multiscale symbolic dynamics entropy for gearbox fault identification. Wang et al [24] improved the multivariate multiscale sample entropy by taking advantage of the generalized refined composite process and applied the improved multivariate multiscale sample entropy to the fault diagnosis of rotating machinery. Ma et al [25] proposed a multivariate multiscale fuzzy distribution entropy by changing the coarse-grained and calculation process of the multiscale fuzzy distribution entropy and applied the method to the fault diagnosis of rotating machinery.…”

Intelligent diagnosis method of bearing fault based on ICEEMDAN and Ghost-IRCNN

“…17 However, the fault feature information extracted by the SE at a single scale is not comprehensive, so the multi-scale sample entropy (MSE) of the reconstructed signals are used to construct the feature vector as the input of the diagnosis model. 18 Since the intelligent algorithm based on machine learning was proposed, it has been widely used in the field of mechanical fault diagnosis. Li et al 19 used back propagation neural network (BPNN) to learn the local characteristics of bearing signals from different scales, and achieved excellent fault diagnosis results.…”

“…17 However, the fault feature information extracted by the SE at a single scale is not comprehensive, so the multi-scale sample entropy (MSE) of the reconstructed signals are used to construct the feature vector as the input of the diagnosis model. 18…”