Robust fault diagnosis of rolling bearings via entropy-weighted nuisance attribute projection and neural network under various operating conditions

Abstract

Rolling bearings are crucial components in the fields of mechanical, civil, and aerospace engineering. They sometimes work under various operating conditions, which makes it harder to distinguish faults from normal signals. Nuisance attribute projection (NAP) is a technique that has been widely used in audio and image recognition to eliminate interference information in the extracted feature space. In constructing the weighted matrix of NAP, the setting of the weighted value represents the degree of interference between the feature vectors. The interference is either taken into consideration in whole, or not considered at all, which will inevitably lead to information loss. In our work, an entropy-weighted NAP (EWNAP) is proposed to deal with such “bipolar problem” in constructing the weighted matrix. The eigenvalues of covariance matrix of collected signals contain dynamical information, and the fuzzy entropy is adopted to evaluate the dispersion degree of these eigenvalues. After normalization, these entropy values are used to express the weight relationship in the weighted matrix of EWNAP. The features processed by EWNAP can be used as samples and combined with neural network to achieve fault diagnosis of rolling bearings. Furthermore, a fault diagnosis approach with insufficient data is demonstrated to validate the effectiveness of the proposed scheme. In the case studies, Case Western Reserve University bearing database and data collected from the bearing fault simulation bench are used. These case studies show that the proposed EWNAP alleviates the interference caused by various operating conditions, and the comparative analysis confirms that the proposed method works better than the conventional methods.

A novel in-situ tool wear monitoring approach using multivariate signal processing and intrinsic multiscale entropy analysis

“…Entropy theory was proposed to measure the regularity and complexity of time series quantitatively. The commonly used entropies consist of permutation entropy (PE) [28], sample entropy (SE) [29], and fuzzy entropy (FE) [30]. However, the above algorithms only analyze the time series on a single scale, which cannot describe the complexity of time series accurately, and generates inaccurate information.…”

A Novel Multivariate Cutting Force-Based Tool Wear Monitoring Method Using One-Dimensional Convolutional Neural Network

“…Deep learning has been highly successful in many fields, such as computer vision and speech recognition, of which the convolutional neural network (CNN) is the most widely applied. Compared with traditional machine learning models [ 30 ], the advantages of CNN are the combination of feature extraction and classification and the ability to perform adaptive screening of features. Therefore, CNN has been widely used in tool wear condition monitoring [ 31 , 32 ].…”

Fault Feature Extraction Method of Reciprocating Compressor Valve Based on SPA-MF

“…Reciprocating compressor valve in the work will be affected by friction, impact and other factors, resulting in its vibration signal has strong nonlinear and non-stationary characteristics [2]. At present, wavelet analysis, empirical mode decomposition, variational mode decomposition, entropy method and multifractal method are mostly used to analyze and study nonlinear and non-stationary signals [3][4][5][6][7]. For example, Cai used wavelet threshold denoising (WTD) and ensemble empirical mode decomposition to analyze and process the vibration signal of diesel engine, and combined rule-based algorithm and BNs/BPNNs to achieve accurate detection and identification of diesel engine faults [8].…”