Deep PCA-Based Incipient Fault Diagnosis and Diagnosability Analysis of High-Speed Railway Traction System via FNR Enhancement

To address the problem of low fault diagnosis accuracy caused by insufficient fault samples of rolling bearings, a dual-input deep anomaly detection method with zero fault samples is proposed for early fault warning of rolling bearings. First, the main framework of dual-input feature extraction based on a convolutional neural network (CNN) is established, and the two outputs of the main frame are subjected to the autoencoder structure. Then, the secondary feature extraction is performed. At the same time, the experience pool structure is introduced to improve the feature learning ability of the network. A new objective loss function is also proposed to learn the network parameters. Then, the vibration acceleration signal is preprocessed by wavelet to obtain multiple signals in different frequency bands, and the two signals in the high-frequency band are two-dimensionally encoded and used as the network input. Finally, the unsupervised learning of the model is completed on five sets of actual full-life rolling bearing fault data sets relying only on some samples in a normal state. The verification results show that the proposed method can realize earlier than the RMS, Kurtosis, and other features. The early fault warning and the accuracy rate of more than 98% show that the method is highly capable of early fault warning and anomaly detection.

Section: Introductionmentioning

confidence: 99%

A New Dual-Input Deep Anomaly Detection Method for Early Faults Warning of Rolling Bearings

Kang,

Chen,

Wang

et al. 2023

“…A method based on wavelet transform (WT) and an improved residual neural network was proposed by [ 9 ], but its use of SVD to improve the pooling layer made it less efficient. Reference [ 22 ] proposed combining PCA theory with deep learning technology and applying it to a high-speed rail system for early fault diagnosis.Reference [ 23 ] proposed a deep learning method for bearing fault diagnosis by superimposing residual null convolution, but it only considered a single load situation and a high signal-to-noise ratio, which cannot meet the requirements of practical complex environments. Reference [ 24 ] proposed a model with high fault diagnosis accuracy based on a working mechanism of soft thresholding and global context, but its sharing of a threshold value for all channels led to ignoring the possibility of different amounts of noise features in different channels.Reference [ 25 ] proposed a deep transient feature learning method that forms a training dataset by simulating the underlying signals of different pulse wavelet bases and learns to anticipate repetitive transient pulse features during the training process, which in turn constructs a mapping of noisy TFDs to clean TFDs; however, the method is only used to remove noise, and there is a lack of fault diagnosis.…”

Section: Introductionmentioning

confidence: 99%

MAB-DrNet: Bearing Fault Diagnosis Method Based on an Improved Dilated Convolutional Neural Network

Zhang

Yin

et al. 2023

Rolling bearing fault diagnosis is of great significance to the safe and reliable operation of manufacturing equipment. In the actual complex environment, the collected bearing signals usually contain a large amount of noises from the resonances of the environment and other components, resulting in the nonlinear characteristics of the collected data. Existing deep-learning-based solutions for bearing fault diagnosis perform poorly in classification performance under noises. To address the above problems, this paper proposes an improved dilated-convolutional-neural network-based bearing fault diagnosis method in noisy environments named MAB-DrNet. First, a basic model called the dilated residual network (DrNet) was designed based on the residual block to enlarge the model’s perceptual field to better capture the features from bearing fault signals. Then, a max-average block (MAB) module was designed to improve the feature extraction capability of the model. In addition, the global residual block (GRB) module was introduced into MAB-DrNet to further improve the performance of the proposed model, enabling the model to better handle the global information of the input data and improve the classification accuracy of the model in noisy environments. Finally, the proposed method was tested on the CWRU dataset, and the results showed that the proposed method had good noise immunity; the accuracy was 95.57% when adding Gaussian white noises with a signal-to-noise ratio of −6 dB. The proposed method was also compared with existing advanced methods to further prove its high accuracy.

“…Existing research in fault detection for induction machines has explored several diagnostic techniques. Studies have focused on detecting faults in both stator and rotor windings of doubly fed induction machines (DFIMs) [ 18 , 19 , 20 , 21 , 22 , 23 ]. Some researchers have utilized neural-network-based modeling of vibration spectra to address electrical faults in the stator winding [ 24 ].…”

Section: Introductionmentioning

confidence: 99%

Efficient Fault Detection of Rotor Minor Inter-Turn Short Circuit in Induction Machines Using Wavelet Transform and Empirical Mode Decomposition

Rehman

Jiao

Sun

et al. 2023

This paper introduces a novel approach for detecting inter-turn short-circuit faults in rotor windings using wavelet transformation and empirical mode decomposition. A MATLAB/Simulink model is developed based on electrical parameters to simulate the inter-turn short circuit by adding a resistor parallel to phase “a” of the rotor. The resulting high current in the new phase indicates the presence of the short circuit. By measuring the rotor and stator three-phase currents, the fault can be detected as the currents exhibit asymmetric behavior. Fluctuations in the electromagnetic torque also occur during the fault. The wavelet transform is applied to the rotor current, revealing an effective analysis of sideband frequency components. Specifically, changes in amplitude and frequency, particularly in d7 and a7, indicate the presence of harmonics generated by the inter-turn short circuit. The simulation results demonstrate the effectiveness of wavelet transformation in analyzing these frequency components. Additionally, this study explores the use of empirical mode decomposition to detect faults in their early stages, observing substantial changes in the instantaneous amplitudes of the first three intrinsic mode functions during fault onset. The proposed technique is straightforward and reliable, making it suitable for application in wind turbines with simple electrical inputs.