Developing and applying OEGOA-VMD algorithm for feature extraction for early fault detection in cryogenic rolling bearing

Wang, Bin; Guo, Yanbao; Zhang, Zheng; Wang, Deguo; Wang, Junqiang; Zhang, Yuansheng

doi:10.1016/j.measurement.2023.112908

Cited by 12 publications

(3 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Considering that the bearing fault signals contain complex frequency components, signal decomposition techniques are employed to enhance the fault features and suppress the influence of irrelevant parts. The techniques can adaptively decompose the signal into different modal functions with various frequency bands based on the signal characteristics, commonly used include the empirical mode decomposition (EMD), 7,8 the variational mode decomposition (VMD), [9][10][11] the empirical wavelet decomposition (EWT) 12,13 and the empirical Fourier decomposition (EFD). 14 Though the above approaches are verified to achieve ideal performance, some limitations still exist.…”

Section: Introductionmentioning

confidence: 99%

“…15 Moreover, some new parameter selection strategies are also utilized which improves the performance of the above techniques. 3,9 As the interference components are distributed in different frequency bands in the actual bearing fault signals, the selection of the appropriate analysis band has received extensive attention. Resonance demodulation is most commonly applied which core is to determine the appropriate center frequency and bandwidth, and then the optimal band signal can be acquired.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Ensemble average of enhanced generalized envelope spectrum for fault detection of rolling element bearings

Zhu,

Liu,

Zhu

2024

Adaptive Control & Signal

View full text Add to dashboard Cite

SummarySince bearing defects usually occur which threaten the stable operation of the machinery, bearing fault detection is of great importance. However, the bearing fault signals inevitably exhibit strong interference components due to the complex structures of the real equipment, which leads to difficulty in fault feature detection. To address the problem, a fault diagnosis method based on the ensemble average of enhanced generalized envelope spectrum (EAEGES) was constructed. First, to suppress the irrelevant components, the optimal analysis frequency band was determined based on the variable frequency band division criteria. Second, the optimal signal was divided into numerous sub‐signals, the EAEGES was established based on the principles of the generalized envelope spectrum to strengthen the capability in detecting bearing fault features, while the improved pulse extraction operator was utilized as the target. Finally, abundant fault information can be distinguished in the feature map acquired using the proposed technique. This technique shows high effectiveness in extracting the defect signatures of the rolling element bearing, which is demonstrated using the simulation and experimental signals and can be applied to real applications.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Ensemble average of enhanced generalized envelope spectrum for fault detection of rolling element bearings

Zhu,

Liu,

Zhu

2024

Adaptive Control & Signal

View full text Add to dashboard Cite

show abstract

“…Rolling bearings usually operate in harsh and variable operating environments. This invariably increases the difficulty of predicting remaining useful life (RUL) in applications related to engineering [7]. Even while rolling bearings are utilized in the same working conditions as product components, they might have very different useful lives [8].…”

Section: Introductionmentioning

confidence: 99%

Rolling Bearing Residual Useful Life Prediction Model Based on the Particle Swarm Optimization-Optimized Fusion of Convolutional Neural Network and Bidirectional Long–Short-Term Memory–Multihead Self-Attention

Yang,

Zhang,

Liu

et al. 2024

Electronics

View full text Add to dashboard Cite

In the context of predicting the remaining useful life (RUL) of rolling bearings, many models often encounter challenges in identifying the starting point of the degradation stage, and the accuracy of predictions is not high. Accordingly, this paper proposes a technique that utilizes particle swarm optimization (PSO) in combination with the fusing of a one-dimensional convolutional neural network (CNN) and a multihead self-attention (MHSA) bidirectional long short-term memory (BiLSTM) network called PSO-CNN-BiLSTM-MHSA. Initially, the original signals undergo correlation signal processing to calculate the features, such as standard deviation, variance, and kurtosis, to help identify the beginning location of the rolling bearing degradation stage. A new dataset is constructed with similar degradation trend features. Subsequently, the particle swarm optimization (PSO) algorithm is employed to find the optimal values of important hyperparameters in the model. Then, a convolutional neural network (CNN) is utilized to extract the deterioration features of rolling bearings in order to predict their remaining lifespan. The degradation features are inputted into the BiLSTM-MHSA network to facilitate the learning process and estimate the remaining lifespan of rolling bearings. Finally, the degradation features are converted to the remaining usable life (RUL) via the fully connected layer. The XJTU-SY rolling bearing accelerated life experimental dataset was used to verify the effectiveness of the proposed method by k-fold cross-validation. After comparing our model to the CNN-LSTM network model and other models, we found that our model can achieve reductions in mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE) of 9.27%, 6.76%, and 2.35%, respectively. Therefore, the experimental results demonstrate the model’s accuracy in forecasting remaining lifetime and support its ability to forecast breakdowns.

show abstract

Fault diagnosis of high-speed rolling bearing in the whole life cycle based on improved grey wolf optimizer-least squares support vector machines

Li,

Luo,

Bai

et al. 2024

Digital Signal Processing

View full text Add to dashboard Cite

Developing and applying OEGOA-VMD algorithm for feature extraction for early fault detection in cryogenic rolling bearing

Cited by 12 publications

References 31 publications

Ensemble average of enhanced generalized envelope spectrum for fault detection of rolling element bearings

Ensemble average of enhanced generalized envelope spectrum for fault detection of rolling element bearings

Rolling Bearing Residual Useful Life Prediction Model Based on the Particle Swarm Optimization-Optimized Fusion of Convolutional Neural Network and Bidirectional Long–Short-Term Memory–Multihead Self-Attention

Fault diagnosis of high-speed rolling bearing in the whole life cycle based on improved grey wolf optimizer-least squares support vector machines

Contact Info

Product

Resources

About