An enhanced rolling bearing fault detection method combining sparse code shrinkage denoising with fast spectral correlation

Li, Jimeng; Yu, Qingwen; Wang, Xiangdong; Zhang, Yungang

doi:10.1016/j.isatra.2020.02.031

Cited by 50 publications

(27 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As one of attractive signal processing methods, cyclic spectral analysis presents the special strengths in analyzing cyclostationary signals in comparison to other methods, such as spectral kurtosis, envelope analysis, and short-time FT [89,90,91,92,93 ]. Cyclic spectral correlation can be the considered as double FT of the signal, which depicts the relationship of the spectral frequency f and the cyclic frequency.…”

Section: Cyclic Spectral Coherencementioning

confidence: 99%

Data Preprocessing Techniques in Convolutional Neural Network Based on Fault Diagnosis Towards Rotating Machinery

2020

View full text Add to dashboard Cite

Rotating machinery plays a critical role in many significant fields. However, the unpredictable machinery faults may lead to the severe damage and losses. Hence, it is of great value to explore the precise approaches for fault diagnosis. With the development of the intelligent fault diagnosis methods based on deep learning, convolutional neural network (CNN) has aroused the attention of researchers in machinery fault diagnosis. In the light of the reduction of difficulty in feature learning and the improvement of final diagnosis accuracy, data preprocessing is necessary and crucial in CNN-based fault diagnosis methods. This review focuses on CNN-based fault diagnosis approaches in rotating machinery. Firstly, data preprocessing methods are overviewed. Then, we emphatically analyze and discuss several main techniques applied in CNN-based intelligent diagnosis, principally including the fast Fourier transform, wavelet transform, data augmentation, S-transform, and cyclic spectral analysis. Finally, the potential challenges and research objects are prospected on data preprocessing in intelligent fault diagnosis of rotary machinery. INDEX TERMS Data preprocessing, convolutional neural network, intelligent fault diagnosis, rotating machinery.

show abstract

Section: Cyclic Spectral Coherencementioning

confidence: 99%

Data Preprocessing Techniques in Convolutional Neural Network Based on Fault Diagnosis Towards Rotating Machinery

2020

View full text Add to dashboard Cite

show abstract

“…e specific cyclostationary components could be extracted from other interferences by applying SC on the vibration signal of faulty rotating machinery. However, the traditional SC has the defects of great computational complexity while dealing with nonstationary vibration signals as stated in the paper [23], which hinders its wide engineering application. To solve the above defect of the above problem of traditional SC, the FSC proposed in the paper [24] is used in the paper, and steps of FSC are stated as follows [24]:…”

Section: Fscmentioning

confidence: 99%

“…SC was used for vibration analysis in a wind turbine generator successfully, and the effectiveness of SC with corresponding precautionary measures for protecting bearing in generators is verified and suggested [22]. However, as stated in paper [23], since the transformation from one-dimensional time domain to two-dimensional spectrogram involves a large number of discrete Fourier transform operations and correlations, the traditional SC is very complicated and has a very low efficiency, so the engineering application of SC is very limited. Accordingly, a new method calculating SC named FSC [24] was proposed to solve the above shortcomings of traditional SC.…”

Section: Introductionmentioning

confidence: 99%

“…Kurtosis weighting was introduced to FSC to improve the performance of enhanced envelope spectrum (EES) [25]. However, FSC still has the defect of low robustness to strong background noise as being verified in the paper [23]. e sparse representation self-learning dictionary method such as KSVD could capture the characteristic components hidden in the vibration signal without prior knowledge of the analyzed signal such as wavelet transform.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fast Spectral Correlation Based on Sparse Representation Self-Learning Dictionary and Its Application in Fault Diagnosis of Rotating Machinery

Wang

2020

Complexity

View full text Add to dashboard Cite

Rolling element bearing and gear are the typical supporting or rotating parts in mechanical equipment, and it has important economy and security to realize their quick and accurate fault detection. As one kind of powerful cyclostationarity signal analyzing method, spectral correlation (SC) could identify the impulsive characteristic component buried in the vibration signals of rotating machinery effectively. However, the fault feature such as impulsive characteristic component is often interfered by other background noise, and the situation is serious especially in early weak fault stage. Besides, the traditional SC method has a drawback of low computation efficiency which hinders its wide application to some extent. To address the above problems, an impulsive feature-enhanced method which combines fast spectral correlation (FSC) with sparse representation self-learning dictionary is proposed in the paper. Firstly, the sparse representation self-learning dictionary method-K-means singular value decomposition (KSVD) is improved and the improved KSVD (IKSVD) method is used to denoise the original signal, and the periodic impulses are highlighted. Then, the FSC algorithm is applied on the denoised signal and spectral correlation image could be obtained. Finally, the calculated enhanced envelope spectrum (EES) of the denoised signal is obtained by using the spectral correlation image to identify the accurate fault position. The feasibility and superiority of the proposed method is verified through simulation, experiment, and engineering application.

show abstract

“…This index can suppress shock interference and harmonic interference in the time domain, making periodic fault components more specific. In order to solve the problem that it is difficult to extract the weak pulses masked by strong background noise in the fault diagnosis of rolling bearings, Li et al [16] proposed an enhanced fault detection method based on the cyclic statistical characteristics of defective bearing vibration signals, combined with sparse code shrinkage denoising and fast spectral correlation. Furthermore, Antoni [17] introduced entropy into the improvement of Kurtogram.…”

Section: Introductionmentioning

confidence: 99%

Bearing Fault Diagnosis Based on Iterative 1.5-Dimensional Spectral Kurtosis

Zhang

Wan

et al. 2020

IEEE Access

View full text Add to dashboard Cite

The key step of bearing fault diagnosis is to select a suitable resonance frequency band, so as to filter out interference components to the maximum extent and retain fault information in the resonance band. Kurtogram algorithm can locate the resonance frequency band well, which has been widely researched and applied in recent years, and has produced many derivative algorithms. The statistical indicators used by these methods to identify frequency band features are divided into time domain indicators and frequency domain indicators. Time domain indicators are more sensitive to a single accidental impact components, while frequency domain indicators are easily affected by harmonics in the time domain, that is, single or several frequency extremes in the frequency domain. In order to overcome the impact of non-periodic transient impulse components and modulation harmonic components, this paper proposes a new method. This method uses wavelet packet transform (WPT) to divide the frequency band plane, and adopts 3 iterations 1.5-dimensional spectrum (1.5D spectrum) method, which can eliminate the impulse interference that has no coupling relationship in the time domain and frequency domain. Based on the above process, the K I-1.5D gram method is constructed, which can realize more accurate positioning of the fault information. Finally, through simulated and experimental analysis, the effectiveness of the proposed method is verified.

show abstract

An enhanced rolling bearing fault detection method combining sparse code shrinkage denoising with fast spectral correlation

Cited by 50 publications

References 34 publications

Data Preprocessing Techniques in Convolutional Neural Network Based on Fault Diagnosis Towards Rotating Machinery

Data Preprocessing Techniques in Convolutional Neural Network Based on Fault Diagnosis Towards Rotating Machinery

Fast Spectral Correlation Based on Sparse Representation Self-Learning Dictionary and Its Application in Fault Diagnosis of Rotating Machinery

Bearing Fault Diagnosis Based on Iterative 1.5-Dimensional Spectral Kurtosis

Contact Info

Product

Resources

About