Rolling Bearing Fault Severity Recognition via Data Mining Integrated With Convolutional Neural Network

Liu, Dongdong; Cui, Lingli; Chen, Weidong; Zhao, Dongfeng; Wen, Weigang

doi:10.1109/jsen.2022.3146151

Cited by 38 publications

(13 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, Ince et al [17] proposed an adaptive 1-D CNN for motor fault detection that extracted features from the raw signals. Liu et al [18] utilized the matrix profile to extract the motifs in noisy 1-D signals, which were then fed into CNN for bearing fault classification. Sun and Li [19] proposed a novel intelligent fault diagnostic method based on symmetrized dot pattern and CNN.…”

Section: Introductionmentioning

confidence: 99%

Rotating machinery fault diagnosis based on optimized Hilbert curve images and a novel bi-channel CNN with attention mechanism

Sun

Liu

Cui

2023

Meas. Sci. Technol.

Self Cite

View full text Add to dashboard Cite

Deep learning methods have been widely investigated in machinery fault diagnosis owing to their powerful feature learning capability. However, high accuracy is hard to achieve due to the limited fault information in a single domain when the data volume is small. In this paper, an optimized Hilbert Curve (OHC) method is developed, which can generate a novel domain to highlight the fault impulses of vibration signals. To fully mine the fault information, a bidirectional-channel convolutional neural network with an attention mechanism (BCCNN-AM) is further proposed, in which two channels are constructed and a transmission channel selection is conducted by a novel improved convolutional block attention module (ICBAM).First, the OHC images and the time-frequency representations are obtained by OHC and wavelet transform respectively. Second, the two types of representations are fed into the channels respectively for feature learning. Finally, the learned features are allocated to different attention mechanism for feature fusion and classification. The proposed method is evaluated via the datasets of rolling bearings and planetary gearboxes, and results show that it outperforms the comparison methods.

show abstract

Section: Introductionmentioning

confidence: 99%

Rotating machinery fault diagnosis based on optimized Hilbert curve images and a novel bi-channel CNN with attention mechanism

Sun

Liu

Cui

2023

Meas. Sci. Technol.

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, bearings often work in complex and harsh environments, which makes the collected vibration signals contain a lot of environmental noise. Especially in the fault state, the fault characters are covered by noise, which seriously affects the accuracy of state monitoring (Liu et al, 2022). In order to effectively reduce noise interference and improve diagnosis accuracy, the denoise procedure is necessary.…”

Section: Introductionmentioning

confidence: 99%

LN-MRSCAE: A novel deep learning based denoising method for mechanical vibration signals

Yang

Wang

et al. 2023

Journal of Vibration and Control

View full text Add to dashboard Cite

Vibration signals are used to monitor the running state of mechanical equipment, but always suffer from a lot of noise in the acquisition process. In order to eliminate noise interference as much as possible, a multilevel residual convolution autoencoder network based noise learning method (LN-MRSCAE) is proposed in this paper. The LN-MRSCAE consists of a deep convolutional autoencoder network and multilevel residual, in which the learning noise module encodes and decodes the noise components, combining with a residual module to obtain the denoised signal. The multilevel residual structure is designed to address the gradient disappearance in network learning and improve the training speed. In specific, the noise learning replaces the traditional clean signal learning. Instead of learning the characteristics of the original clean signal, just learn the characteristics of the noise signal. The effectiveness of LN-MRSCAE network denoising is verified by classic analog signal and actual bearing vibration signals with different faults. The results show that the LN-MRSCAE model is superior to the latest models, and the noise component of the signal is obviously suppressed.

show abstract

“…In some real applications, the equipment often operates under time-varying conditions. For example, the speeds of wind turbines often fluctuate due to the variations in wind power and directions; 15 the speeds of aircraft drivetrains also vary due to the intricate flight tasks. 16 Under variable speeds, the frequencies caused by the faults change following the speed variation profiles.…”

Section: Introductionmentioning

confidence: 99%

“…4. The final demodulated signal is obtained per equation(15), and the demodulated spectrum is obtained by applying FFT to the demodulated signal.…”

mentioning

confidence: 99%

Flexible iterative generalized demodulation filtering for the fault diagnosis of rotating machinery under nonstationary conditions

Liu

Cui

Chen

2022

Structural Health Monitoring

Self Cite

View full text Add to dashboard Cite

The health monitoring and diagnosis of rotating machinery under nonstationary conditions are still challenging due to the complex modulation characteristics and the interfering noises. In this paper, a novel flexible iterative generalized demodulation filtering method is proposed for the machinery fault diagnosis. First, the Hilbert transform is applied to the vibration signals to highlight the characteristic frequencies as well as their harmonics. Second, the phase functions used for mapping the interest frequencies are designed, and no matter how the speed varies, the time-varying frequencies of different signal segments with the same physical meaning are transformed into the same constant frequencies. Then, the filters are designed based on the introduced base frequency and the characteristic coefficients, and then the modulation rotating frequency, fault characteristic frequencies, and their harmonics are filtered. Finally, the demodulated signals are reconstructed and the health conditions are determined by the demodulated spectrums. The method is evaluated by the vibration signals of faulty rolling bearings and planetary gearboxes. The results demonstrate that the method can well reveal the fault-related frequencies and that the demodulated frequency values are not subject to the speed fluctuation profiles.

show abstract

Rolling Bearing Fault Severity Recognition via Data Mining Integrated With Convolutional Neural Network

Cited by 38 publications

References 44 publications

Rotating machinery fault diagnosis based on optimized Hilbert curve images and a novel bi-channel CNN with attention mechanism

Rotating machinery fault diagnosis based on optimized Hilbert curve images and a novel bi-channel CNN with attention mechanism

LN-MRSCAE: A novel deep learning based denoising method for mechanical vibration signals

Flexible iterative generalized demodulation filtering for the fault diagnosis of rotating machinery under nonstationary conditions

Contact Info

Product

Resources

About