A novel complex network community clustering method for fault diagnosis

Abstract

Complex network, as a method for analyzing nonlinear and non-stationary signals, overcomes the shortcomings of traditional time-frequency analysis methods and proves its effectiveness in mechanical fault diagnosis. Community clustering, as one of them, has made great progress in recent years. However, the existing community clustering algorithms have the disadvantages of lacking significant global extreme value and huge search space. Therefore, a Fast Newman algorithm based on reliability judgment is proposed. Starting from the community structure characteristics of the complex network, with the fault sample as a network node, the relationship between the samples as a connected edge, and a complex network model of fault data is established. Clusters in troubleshooting are transformed into community structure discovery in the network. Firstly, the initial division of the community is obtained by measuring the distance between the samples. Then, the modularity index of the network is used as a standard function of the community division, and the bottom-up community merger is performed. The local edge density index is used for reliability determination before each combination to achieve global optimization, and the network block structure is the most obvious. Finally, with all data being merged into one community, the optimal division of the community structure is obtained while accurate community clustering and fault diagnosis is realized. The benchmark graphs for testing community detection (LFR standard test network) and different fault data of rolling bearings under multiple operating conditions are applied to verify the effectiveness of this method, and the experiment results prove that the modified Fast Newman algorithm has better clustering effect and higher correct rate than the original one. Compared with K-means clustering and fuzzy clustering, the modified Fast Newman algorithm achieves higher performance for fault diagnosis of rolling bearing under multiple operating conditions.

A lightweight complex-domain acoustic feature extraction method for rotating machinery fault detection

“…Conventional methods such as MFCC and VGGish structure or the DL methods, as addressed in the aforementioned studies, ignore phase information, which only focus on amplitude while training or extracting the features of the acoustic signals. Although scholars have proposed deep complex network architectures to extract features in the complex domain [50,51] and have demonstrated their noise robustness and lightweight in speech enhancement [52], these methods of extracting complex-domain features…”

Feature selection and interpretability analysis of compound faults in rolling bearings based on the causal feature weighted network

“…Another research direction involved applying complex network modeling to rolling bearing faults [25]. Early studies predominantly used this approach for fault diagnosis [26], transforming fault diagnosis into detection of network community structures.…”

Special features on continuous learning based condition monitoring for crucial components

“…The ensemble weights of selected ELM subnetworks are also calculated adaptively, which avoids the complex process of manually selecting subnetworks and calculating ensemble weights. Article [9] introduces an improved Fast Newman algorithm based on reliability judgment for complex network community clustering in mechanical fault diagnosis. This method has high clustering accuracy and stability under multiload and multispeed conditions.…”