Sparse representation based on parametric impulsive dictionary design for bearing fault diagnosis

Sun, Ruo-Bin; Yang, Zhibo; Zhai, Zhi; Chen, Xuefeng

doi:10.1016/j.ymssp.2018.12.054

Cited by 107 publications

(41 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…6 Among these methods, the vibration method is the most widely used one. It was found in many previous studies that fault features in the vibration signals were widely used as the basis for fault diagnosis [7][8][9] not only for bearings, but also for some other common rotors, such as gears. [10][11][12] There are also some publications dealing not only the detection of the incipient fault, but also the diagnose of various sizes of fault.…”

Section: Introductionmentioning

confidence: 99%

Contact characteristic and vibration mechanism of rolling element bearing in the process of fault evolution

Yang

et al. 2021

Proceedings of the Institution of Mechanical Engineers, Part K:

View full text Add to dashboard Cite

The vibration response of rolling element bearing has a close relation with its fault. An accurate evaluation of the bearing vibration response is essential to the bearing fault diagnosis. At present, most bearing dynamics models are built based on rigid assumptions, which may not faithfully reveal the dynamic characteristics of bearing in the presence of fault. Moreover, previous similar works mainly focus on the fault with a specified size without considering the varying contact characteristics as the fault evolves. This paper developed an explicit dynamics finite element model for the bearing with three types of raceway faults considering the flexibility of each bearing component in order to accurately study the contact characteristic and vibration mechanism of defective bearings in the process of fault evolution. The developed model is validated by comparing its simulation results with both analytical and experimental results. The dynamic contact patterns between the rolling elements and the fault, the additional displacement due to the fault and the faulty characteristics within the bearing vibration signal during the fault evolution process are investigated. The analysis results from this work can provide practitioners an in-depth understanding towards the internal contact characteristics with the existence of raceway fault and theoretical basis for rolling bearing fault diagnosis.

show abstract

Section: Introductionmentioning

confidence: 99%

Contact characteristic and vibration mechanism of rolling element bearing in the process of fault evolution

Yang

et al. 2021

Proceedings of the Institution of Mechanical Engineers, Part K:

View full text Add to dashboard Cite

show abstract

“…Yang et al [14] developed a sliding-window dictionary learning denoising method with time-domain signals. Zhou and Sun et al [15][16] established an overcomplete dictionary (or overcomplete atomic library) of attended cosines basis and a parameter pulse dictionary that highly match the bearing fault waveform. Jiao et al [17] presented a hierarchical discrimination sparse coding (HDSC) method, which successfully extracted weak fault features under strong noise and environmental interference.…”

Section: Introductionmentioning

confidence: 99%

Dictionary Learning via a Mixed Noise Model for Sparse Representation Classification of Rolling Bearings

Zhang

2020

IEEE Access

View full text Add to dashboard Cite

Rotating machinery contains a great number of rolling bearings, which play an indispensable role. However, bearing vibration signals in complex environments are often mixed with various noises, which makes it difficult to extract fault characteristics from original signals. It is still challenging to identify the fault types of rolling bearings. To address this issue, a dictionary learning method based on a mixed noise model for the sparse representation classification of rolling bearings (DLMN-SRC) is proposed. Our framework constructs a dictionary learning method based on mixed noise, which has better robustness to complex noise pollution than a single noise model. Then, an alternating direction method of multipliers (ADMM) algorithm is used to solve the optimization problem of the proposed model. Eventually, the redundant errors between the detected and reconstructed signals in the dictionary learning model are calculated for sparse representation classification. The results of two examples prove that faults of the rolling bearing are successfully extracted and classified by DLMN-SRC. Compared with traditional diagnosis methods, the performance of this method has obvious superiority and good application prospects.

show abstract

“…The diagnosis of rolling bearing early weak fault is not only difficult, but also is very hot, and the fault diagnosis of rolling bearing early weak fault has been attracted a lot of attention in recent years. Sparse representation [9,10] is a promising tool for early weak fault feature extraction of the rolling bearing. However, there exist two problems: Firstly, the prior knowledge is needed to construct the predefined dictionary to match the objected signal.…”

Section: Introductionmentioning

confidence: 99%

A frequency slice wavelet transform based on wavelet de-noising using neighboring coefficients method and its application in feature extraction of rolling bearing’ early weak fault

Wang¹

2020

J. vibroeng.

View full text Add to dashboard Cite

Extracting the characteristics of rolling bearings in early weak failure stage before the occurring of complete failure has important safety and economic significance in engineering application. The wavelet transform (WT) is the commonly used and effective time-frequency method for fault feature extraction of rotating machinery due to that it could reflect the fault feature in time and frequency domains synchronously. However, WT would not work effectively when the impulsive fault signal is buried by strong background noise, and the situation is particularly serious in the early weak fault stage of rolling bearing. A frequency slice wavelet transform based on wavelet de-noising using neighboring coefficients method is proposed in the paper by combing frequency slice wavelet transform with wavelet de-noising using neighboring coefficient to solve the above problem: Firstly, the vibration signal of rolling bearing is de-noised by wavelet de-noising using the neighboring coefficients method. Then the frequency slice wavelet transform is applied on the de-noised signal, and satisfactory analysis results could be obtained. The effectiveness of the proposed method is verified by the vibration data of rolling bearing accelerated fatigue test. Besides, the analysis result of the same vibration data of rolling bearing accelerated fatigue test using Kurtogram method is also presented in the paper to verify the advantage of the proposed method.

show abstract

Sparse representation based on parametric impulsive dictionary design for bearing fault diagnosis

Cited by 107 publications

References 27 publications

Contact characteristic and vibration mechanism of rolling element bearing in the process of fault evolution

Contact characteristic and vibration mechanism of rolling element bearing in the process of fault evolution

Dictionary Learning via a Mixed Noise Model for Sparse Representation Classification of Rolling Bearings

A frequency slice wavelet transform based on wavelet de-noising using neighboring coefficients method and its application in feature extraction of rolling bearing’ early weak fault

Contact Info

Product

Resources

About