Rolling bearing fault diagnosis based on 2D time-frequency images and data augmentation technique

Rolling bearings play a vital role in ensuring the safe operation of rotating machinery. However, in many application scenarios, the collected data has a low signal-to-noise ratio and the samples with faults are rare, which affects the generalization capability of the model, making it impossible to achieve accurate diagnosis. To solve this problem, the selection of time-frequency (TF) maps was considered in this paper through reinforcement learning. The time-frequency maps are built by four classical time-frequency characterization methods such as short-time Fourier transform (STFT) and synchro squeezing transform (SST). And the match-reinforcement learning time frequency selection (MRLTFS) fault diagnosis model is proposed to extract the fault-related features. Experiments show that the proposed MRLTFS method is superior to existing methods in robustness, generalization and feature selection capability.

Section: Data Representationmentioning

confidence: 99%

Match-reinforcement learning with time frequency selection for bearing fault diagnosis

Wang

Gao

Zhu

et al. 2023

“…In recent years, deep learning has attracted widespread attention in various fields due to its strong feature mining capability, providing a new perspective for fault diagnosis [10][11][12][13][14]. Compared with traditional fault diagnosis methods, deep learning can adaptively extract the fault information from vibration signals, avoiding the information loss caused by manual processing.…”

Section: Introductionmentioning

confidence: 99%

Multi-scale residual neural network with enhanced gated recurrent unit for fault diagnosis of rolling bearing

Liao,

Fu,

Yang

et al. 2024

Self Cite

As the key component of rotating machinery, effective and reliable fault diagnosis of rolling bearing is particularly critical for promoting production safety and economic benefits. The powerful representation learning ability of convolutional neural network (CNN) enables it to effectively extract fault information from vibration signals of rolling bearing. Nevertheless, challenges are faced by CNN in extracting features at multi-scale and capturing temporal features. With regard to this issue, a hybrid deep learning model that incorporates the multi-scale residual neural network (MSRN) with the enhanced gated recurrent unit (EGRU), namely MSRN-EGRU, is proposed in this paper. To begin with, MSRN is designed by introducing multi-scale structure and residual connections into CNN for extracting local features effectively and improving the feature representation of the model. Then, the extracted local features are input into EGRU to further extract temporal features, where EGRU is proposed by improving gated recurrent unit (GRU) structure and embeding scaled exponential liner unit (SELU), which enhances the nonlinear modeling and memory ability. Eventually, the obtained features are processed by α-Dropout and global average pooling (GAP) before being inputted into the softmax layer for fault diagnosis. To validate the effectiveness of the proposed model, three baseline models and two ablation models were employed for comparative experiments with two bearing datasets. The experimental results reveal that the proposed model achieves commendable performance in terms of accuracy, robustness, and convergence for fault diagnosis of rolling bearing.

“…The vibration signals are one-dimensional data, one-dimensional CNN has been applied to the field of fault diagnosis with a large number of excellent results. However, the input of CNN can be two-dimensional, some scholars have proposed the transformation of one-dimensional vibration data into two-dimensional images before extracting features through CNN, which can effectively enhance diagnostic accuracy [35][36][37]. Li et al [38] utilized STFT to convert one-dimensional signals into time-frequency images and successfully diagnosed bearing faults with CNN.…”

Section: Introductionmentioning

confidence: 99%

Bearing fault diagnosis based on CNN-BiLSTM and residual module

Fu,

Wei,

Yang

et al. 2023

Bearings are key components of rotating machinery, and their fault diagnosis is essential for machinery operation. Bearing vibration signals belong to time series data, but traditional convolutional neural networks or recurrent neural networks cannot fully extract the fault features from these data. To address the insufficient feature extraction and poor noise resistance, this paper proposes a fault diagnosis model based on CWT, CNN with channel attention, BiLSTM and residual module. Firstly, a parallel dual-path feature extraction mechanism is constructed which takes time-domain signals and time-frequency images transformed via CWT as the input respectively. Then BiLSTM extracts the time features of the signal as one path, and the CNN with efficient channel attention extracts the spatial features as the other path. This parallel neural network contributes to better feature extraction. Then, the residual module is applied to extract the global features to further improve the feature extraction ability and noise immunity. The experimental results demonstrate that the proposed model on the CWRU dataset has better diagnostic accuracy under different working conditions and different SNRs than other comparative methods. In addition, the model shows good generalization performance on JNU dataset.