Enhanced deep residual network with multilevel correlation information for fault diagnosis of rotating machinery

Abstract

Modern machinery becomes more precious with the advance of science, and fault diagnosis is vital for avoiding economical losses or casualties. Among massive diagnosis methods, deep learning algorithms stand out to open an era of intelligent fault diagnosis. Deep residual networks are the state-of-the-art deep learning models which can continuously improve performance by deepening the network structures. However, in vibration-based fault diagnosis, the transient property instability of vibration signal usually calls for time–frequency analysis methods, and the characters of time–frequency matrices are distinct from standard images, which brings some natural limitations for the diagnosis performance of deep learning algorithms. To handle this issue, an enhanced deep residual network named the multilevel correlation stack-deep residual network is proposed in this article. Wavelet packet transform is used to preprocess the sensor signal, and then the proposed multilevel correlation stack-deep residual network uses kernels with different shapes to fully dig various kinds of useful information from any local regions of the processed input. Experiments on two rolling bearing datasets are carried out. Test results show that the multilevel correlation stack-deep residual network exhibits a more satisfactory classification performance than original deep residual networks and other similar methods, revealing significant potentials for realistic fault diagnosis applications.

A novel dense residual network based on Adam-S optimizer for fault diagnosis of bearings under different working conditions

“…The ResNet is composed of many residual blocks, each of which is composed of an identity mapping part and residual part [35,36]. For a ResNet architecture, as shown in figure 2, this is obtained by introducing identity mapping.…”

Power Equipment Fault Diagnosis Method Based on Energy Spectrogram and Deep Learning

“…Li et al [ 29 ] designed a multi-scale multi-sensor feature fusion convolutional neural network (MSMFCNN), which fused the rich information provided by multiple sensors and conducted fault diagnosis based on CNN, achieving good diagnosis results. Xiong et al [ 30 ] proposed an enhanced deep residual network with multilevel correlation information for fault diagnosis of rotating machinery, which is used to process the feature information obtained by wavelet packet transformation. Shi et al [ 31 ] proposed a fault diagnosis method based on IMFs and WDenseNets, in which the components of vibration signals obtained through empirical mode decomposition were weighted and input into WDenseNets for fault identification and classification [ 32 ].…”

Fault diagnosis method of multi-rotor UAV based on one-dimensional convolutional neural network with adaptive batch normalization algorithm

“…Feature extraction plays an important role in fault diagnosis, and the powerful feature extraction capability of deep learning makes it highly popular among experts and scholars [12]. Classical deep learning methods for fault diagnosis include autoencoder, convolutional neural network (CNN), deep neural network (DNN), and so on [13][14][15][16][17][18][19][20]. Combining deep learning with transfer learning in fault diagnosis is a new research direction [21].…”