An Intelligent Fault Diagnosis Method of Rolling Bearings Based on Short-Time Fourier Transform and Convolutional Neural Network

Zhang, Qi; Deng, Liya

doi:10.1007/s11668-023-01616-9

Cited by 45 publications

(22 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In terms of STFT, Zhang and Deng [55] converted onedimensional vibration signals into time-frequency images through STFT. Then, the time-frequency image was input into the CNN for fault diagnosis, which solved the loss in extracting one-dimensional vibration signal features and improved the accuracy of feature extraction.…”

Section: Time-frequency Algorithmmentioning

confidence: 99%

A review on convolutional neural network in rolling bearing fault diagnosis

Li,

Ma,

Yuan

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

The health condition of rolling bearings has a direct impact on the safe operation of high-speed trains. And their working environment is harsh and the working condition is complex, which brings challenges to fault diagnosis. With the development of computer technology, deep learning has been applied in the field of fault diagnosis and has rapidly developed. Among them, convolutional neural network (CNN) has received great attention from researchers due to its powerful data mining ability and feature adaptive learning ability. Based on recent research hotspots, the development history and trend of CNN is summarized and analyzed. Firstly, the basic structure of CNN is introduced and the important progress of classical CNN models for rolling bearing fault diagnosis in recent years is studied. The problems with the classic CNN algorithm have been pointed out. Secondly, to solve the above problems, combined with recent research achievements, various methods and principles for optimizing CNN are introduced and compared from the perspectives of deep feature extraction, hyperparameter optimization, network structure optimization. Although significant progress has been made in the research of fault diagnosis of rolling bearings based on CNN, there is still room for improvement and development in addressing issues such as low accuracy of imbalanced data, weak model generalization, and poor network interpretability. Therefore, the future development trend of CNN networks is discussed finally. And transfer learning models are introduced to improve the generalization ability of CNN and interpretable CNN is used to increase the interpretability of CNN networks.

show abstract

Section: Time-frequency Algorithmmentioning

confidence: 99%

A review on convolutional neural network in rolling bearing fault diagnosis

Li,

Ma,

Yuan

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…These architectures provide much better performance in detecting the rare sound event in the audios of the industrial dataset of DCASE 2022 in case of extremely low SNRs than the systems without the denoising frontends. The models in [216] and [217] use U-Net and U-Net++ respectively for denoising the spectrograms of the sound from the planetary gearboxes which are widely used in many industrial categories such as mining, wind power generation, metal forming, etc. These TF spectrograms are later used for fault diagnoses.…”

Section: C) Industrial Soundsmentioning

confidence: 99%

“…U-Net has outperformed the classical methods in reducing the number of false positives and the processing time, making it a better choice for meeting the real-time requirements of fault diagnosis. In [217], an improved version of U-Net++ using Tversky loss [218] as an optimization objective is utilized for further improving the segmentation F1 score from 0.942 (in [216]) to 0.949 (in [217]).…”

Section: C) Industrial Soundsmentioning

confidence: 99%

A Survey of Audio Enhancement Algorithms for Music, Speech, Bioacoustics, Biomedical, Industrial, and Environmental Sounds by Image U-Net

Gul,

Khan

2023

IEEE Access

View full text Add to dashboard Cite

The recent surge in the use of Deep Neural Networks (DNNs) has also made its mark in the field of Audio Enhancement (AE), providing much better quality than the classical methods. Although, there are dedicated audio processing DNNs, yet, many recent models of AE have utilized U-Net: a DNN based on Convolutional Neural Network (CNN), fundamentally developed for image segmentation. It is found that the useful features hidden in the time domain are highlighted when the audio signal is converted to a spectrogram, which can be treated as an image. In this article, we will review the recent work, utilizing U-Nets for different AE applications. Different than other published reviews, this review focuses entirely on AE techniques based on image U-Nets. We will discuss the need for AE, U-Net comparison to other DNNs, the benefits of converting the audio to 2D, input representations that are useful for different AE applications, the architecture of vanilla U-Net and the pre-trained models, variations in vanilla architecture incorporated in different E models, and the state-of-the-art AE algorithms based on U-Net in various applications. Apart from speech and music, this article discusses a wide range of audio signals e.g. environmental, biomedical, bioacoustics, and industrial sounds, not covered collectively in a single article in previously published studies. The article ends with the discussion of colored spectrograms in future AE applications.INDEX TERMS CNNs; image processing deep neural networks; pre-trained networks; spectrogram.;U-Net.

show abstract

“…The traditional fault diagnosis methods of rolling bearings mainly consist of feature extraction and fault identification. In feature extraction, time-domain features are usually extracted by using waveform factors, peaks, etc [6,7], frequencydomain features are usually extracted by using Fourier transforms, spectral cliff, etc [8,9], and time-frequency-domain features are usually extracted by using wavelet transform, empirical modal decomposition, etc [10][11][12][13], then the obtained data are input into machine learning algorithms for fault identification [14,15]. Liu et al [16] established a new spalling propagation assessment algorithm relying on spectral amplitude ratio and statistical features, and effectively diagnosed the location and degree of spalling damage.…”

Section: Introductionmentioning

confidence: 99%

Rolling bearing fault diagnosis model based on DSCB-NFAM

Zhao,

Guo

2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

Machine learning techniques have had great success in fault diagnosis. However, the traditional machine learning methods rely heavily on manual priori knowledge leading to poor fault diagnosis results in rolling bearing fault diagnosis. Deep learning techniques can improve the accuracy of fault intelligent diagnosis with the help of automatic extraction of fault features. In this article, a method of smart fault diagnosis for rolling bearings based on depth-separable convolutional block (DSCB)-non-local feature awareness module (NFAM) is proposed. Firstly, a DSCB is constructed, which can reduce the number of parameters and effectively alleviate the gradient disappearance phenomenon. Then a NFAM is constructed to enhance the global sensing ability while adaptively adjusting the weight of each channel. Further, a feature extraction layer is constructed to extend the fault feature of the input information. Finally, a feature pyramid approach is used to fuse the extracted information for output. The validity of the suggested method is tested on both of the bearing datasets. By comparing with several mainstream intelligent diagnosis algorithms, the results of the experiments show that the proposed method has better advancement under complex conditions such as variable load and variable noise.

show abstract

An Intelligent Fault Diagnosis Method of Rolling Bearings Based on Short-Time Fourier Transform and Convolutional Neural Network

Cited by 45 publications

References 39 publications

A review on convolutional neural network in rolling bearing fault diagnosis

A review on convolutional neural network in rolling bearing fault diagnosis

A Survey of Audio Enhancement Algorithms for Music, Speech, Bioacoustics, Biomedical, Industrial, and Environmental Sounds by Image U-Net

Rolling bearing fault diagnosis model based on DSCB-NFAM

Contact Info

Product

Resources

About