Understanding and improving deep learning-based rolling bearing fault diagnosis with attention mechanism

Li, Xiang; Zhang, Wei; Ding, Qian

doi:10.1016/j.sigpro.2019.03.019

Cited by 300 publications

(114 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The process of the OSF is illustrated by a data sequence [3,4,3,5,7,5,3,4,8,10,6,3,5,4,2], and the formation of the upper envelope is seen in Figure 1. The formation process is described as follows: First, the distance of adjacent maxima points in the data sequence is calculated.…”

Section: Order-statistic Filtering Fourier Decomposition Methodsmentioning

confidence: 99%

“…points is respectively supplemented at both ends of the original data sequence by mirroring, that is, the new data sequence is [4,3,4,3,5,7,5,3,4,8,10,6,3,5,4,2,4]. Finally, the data is filtered from left to right using a window with a window width of 3, and the maximum value in each window is used to form a filtered result of [4,4,5,7,7,7,5,8,10,10,10,6,5,5,4], the results obtained are shown in Figure 2.…”

Section: Order-statistic Filtering Fourier Decomposition Methodsmentioning

confidence: 99%

“…Rolling bearing is an indispensable element of rotating equipment, and its failure will lead to abnormal operation of the equipment and even cause huge economic losses. Therefore, the correct detection of rolling bearing failure is the hot spot but also the difficulty of current research [1][2][3][4][5]. The empirical mode decomposition (EMD) formally proposed by Huang et al [6,7] has been commonly used in dealing with mechanical signals.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Order-statistic filtering Fourier decomposition and its applications to rolling bearing fault diagnosis

Huang

Zheng

Pan

et al. 2020

Preprint

View full text Add to dashboard Cite

Inspired by the empirical wavelet transform (EWT) method, a new method for nonstationary signal analysis termed order-statistic filtering Fourier decomposition (OSFFD) is proposed in this paper. The OSFFD method uses order-statistic filtering and smoothing to preprocesses the Fourier spectrum of original signal, which improves the problem of sometimes unreasonable boundaries obtained by EWT directly segmenting the Fourier spectrum. Then, the mono-components with physical significance are obtained by adaptively reconstructing the coefficient of fast Fourier transform in each interval, which improves the problem of too many false components obtained by Fourier decomposition (FDM). The OSFFD method also is compared with the existing nonstationary signal decomposition methods including empirical mode decomposition（EMD）, EWT, FDM and variational mode decomposition（VMD） through analyzing simulation signals and the result indicates that OSFFD is less affected by noise and is much more accurate and reasonable in obtaining mono-components. After that, the OSFFD method is compared with the mentioned methods in diagnostic accuracy through analyzing the tested faulty bearing vibration signals and the effectiveness and superiority of OSFFD to the comparative methods in bearing fault identification are verified.

show abstract

Section: Order-statistic Filtering Fourier Decomposition Methodsmentioning

confidence: 99%

Section: Order-statistic Filtering Fourier Decomposition Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Order-statistic filtering Fourier decomposition and its applications to rolling bearing fault diagnosis

Huang

Zheng

Pan

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…The raw signal of sliding bearing is contaminated with intensive background noises due to the characteristic of the abnormal state in the early stage, and it is difficult to ensure the diagnostic efficiency and accuracy. Although deep learning has a strong ability to extract useful information through multiple non-linear transformations and approximate complex non-linear functions with little error [25], it is necessary to filter the signal with the statistical filter according to the quality of the practical signal. The whole process can be described as Figure 8.…”

Section: Data Preprocess With Statistical Filtermentioning

confidence: 99%

Stepwise Intelligent Diagnosis Method for Rotor System with Sliding Bearing Based on Statistical Filter and Stacked Auto-Encoder

et al. 2020

View full text Add to dashboard Cite

Since the raw signal collected from the sliding bearing is contaminated with background noise, and it is difficult to obtain high-precision results for the traditional methods due to the low signal-to-noise ratio (SNR). Therefore, a stepwise intelligent diagnosis method based on statistical filter and stacked auto-encoder (SAE) that is established with several auto-encoders is proposed to identify several faults of sliding bearing in a rotor system. Firstly, the statistical filter is utilized to reduce the interference information for the different abnormal states and to increase the SNR. Secondly, the stepwise intelligent diagnosis based on SAE is performed to learn the useful fault features, and it can automatically complete the fault diagnosis which is contributed with the superiority binary classification to fully mine the relationship between the fault characteristics and the health condition of bearing. Finally, the diagnosis of the oil whirl and structural faults in a rotor system is cited as an example to demonstrate the effectiveness of proposed method. It can effectively illustrate the advantages of the stepwise diagnosis method to obtain the maximum diagnostic accuracy.

show abstract

“…However, the model of equipment may not be modeled well when the equipment is complicated. The data-driven technique does not require the creation of a physical model of the device; use the monitored data during the operation of the equipment to diagnose the fault type of the equipment, for example, machine learning, [19][20][21][22] multivariate statistical analysis, 23 signal processing, 24 rough set, 25,26 fuzzy set, 27 and multi-sensors or multi-sources information fusion method. [28][29][30][31] In the multi-sensors information fusion based method, in which the data of multiple sensors (or sources) are fused, reflects the diversity, redundancy, and complementarity of multiple information.…”

Section: Introductionmentioning

confidence: 99%

Fault diagnosis method based on time domain weighted data aggregation and information fusion

Zhang

Jiang

Deng

2019

International Journal of Distributed Sensor Networks

View full text Add to dashboard Cite

Fault diagnosis of equipment is a key issue in the industrial field, and it is essential to keep abreast of equipment status. However, previous studies either considered fault data at a single moment or gave the same weight to data over a period of time. In view of the problems above, fault diagnosis method based on time domain weighted data aggregation and information fusion is proposed in this article. First, the monitored data of sensors loaded by the equipment are aggregated utilizing the linear decaying weights. Then, Gaussian models of each fault type under different fault features are established based on aggregated data. And the basic probability assignments are generated by matching aggregated testing samples with the constructed Gaussian model. At last, the basic probability assignments generated under each fault feature are fused by Dempster combination rule. The proposed method is verified and the results show that the total fault recognition rate can reach 97.5%, which increased by 1.9% compared with the method that Gaussian model constructed by original data.

show abstract

Understanding and improving deep learning-based rolling bearing fault diagnosis with attention mechanism

Cited by 300 publications

References 60 publications

Order-statistic filtering Fourier decomposition and its applications to rolling bearing fault diagnosis

Order-statistic filtering Fourier decomposition and its applications to rolling bearing fault diagnosis

Stepwise Intelligent Diagnosis Method for Rotor System with Sliding Bearing Based on Statistical Filter and Stacked Auto-Encoder

Fault diagnosis method based on time domain weighted data aggregation and information fusion

Contact Info

Product

Resources

About