New domain adaptation method in shallow and deep layers of the CNN for bearing fault diagnosis under different working conditions

Jin, Tongtong; Yan, Chuliang; Chen, Chuanhai; Yang, Zhaojun; Tian, Hailong; Guo, Jinyan

doi:10.1007/s00170-021-07385-9

Cited by 41 publications

(16 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At this time, the corresponding spectrum distribution and amplitude will also change. Bearing fault diagnosis often uses statistical parameters in the time domain and frequency domain to characterize fault characteristics [26][27][28][29]. e statistical parameters selected in the paper are shown in Table 1.…”

Section: Det Methodmentioning

confidence: 99%

Fault Identification of Low-Speed Hub Bearing of Crane Based on MBMD and BP Neural Network

Guo

Yang

Cheng

et al. 2022

Shock and Vibration

View full text Add to dashboard Cite

As the key bearing part of the crane, the low-speed hub bearing of the crane exists in special working conditions of low-speed and alternating heavy load. It is difficult to extract its fault characteristics accurately by existing analysis methods. The main idea of the broadband mode decomposition (BMD) method previously proposed is to search in the association dictionary library containing broadband and narrowband signals. However, when it is applied to the broadband signals interfered by strong noise, the decomposition is easy to produce modal confusion, so the modulated broadband mode decomposition (MBMD) method is proposed. The fault signal just can be analyzed by MBMD, so it is applied to the fault diagnosis of low-speed hub bearing of the crane. To realize the fault identification of low-speed hub bearing of the crane, firstly, the original signal is decomposed by MBMD. Secondly, the eigenvalues of the first three-component signals are calculated, the eigenvalue matrix is constructed, and the marked features are selected by the distance evaluation technique (DET). Finally, the marked features are input into BP neural network for training and testing to identify the types of bearing fault. Compared with EEMD, VMD, and BMD, the MBMD method combined with BP neural network has good performance in feature extraction and fault identification.

show abstract

Section: Det Methodmentioning

confidence: 99%

Fault Identification of Low-Speed Hub Bearing of Crane Based on MBMD and BP Neural Network

Guo

Yang

Cheng

et al. 2022

Shock and Vibration

View full text Add to dashboard Cite

show abstract

“…Cheng [26] and Hu [27] et al studied the ball screw accuracy loss characteristics caused by sliding motion under time-varying axial loads and feed rates. Zhang [28], Niu [29] and Jin [30] et al studied the machining quality evaluation and machining accuracy improvement of multi-axis CNC machine tools, through analyzing the geometric errors of key functional components.…”

Section: Introductionmentioning

confidence: 99%

Accuracy decay mechanism of ball screw in CNC machine tools for mixed sliding-rolling motion under non-constant operating conditions

Zhao

Chen

et al. 2022

Int J Adv Manuf Technol

Self Cite

View full text Add to dashboard Cite

The ball screw (BS) has become an indispensable key functional component in many fields, such as CNC machine tools, precision transmission system and special service requirements. Its accuracy reliability is a key performance in many application fields. There is a phenomenon of mixed slip-roll motion during the BS operation, however, both non-constant working conditions and mixed sliding-rolling motion mode affect the its accuracy decay. In this paper, the modeling method of accuracy decay for mixed sliding-rolling motion behavior is established under non-constant operating conditions. The accuracy reliability of the BS is analyzed under single non-constant operation condition or multiple non-constant operating conditions, and the accuracy decay is predicted though modeling method proposed. In addition, the experimental analysis of the BS accuracy degradation is carried out under three different amplitudes of AL (non-constant axial load), FR (non-constant feed rate) and AL+FR combination conditions. The average value of the accuracy loss relative error between theoretical model and experimental test results is smaller than that under pure sliding motion. It shows that the accuracy decay analysis model established is beneficial to predict the BS accuracy reliability.

show abstract

“…Considering the randomness of the performance degradation process, Zhou et al 32 established a performance degradation model including two stages of stable and accelerated degradation. In addition, Zhang, 33 Niu, 34 and Jin 35 studied the machining quality evaluation and machining accuracy improvement of multi-axis CNC equipment, by analyzing the geometric errors of key functional components.…”

Section: Introductionmentioning

confidence: 99%

Numerical analysis and experimental research on the accuracy decay of a ball screw with geometric errors under non-constant operating conditions

Cheng

Liu

et al. 2022

Proceedings of the Institution of Mechanical Engineers, Part C:

Self Cite

View full text Add to dashboard Cite

The ball screw (BS) has become an indispensable key functional component in many fields, such as precision manufacturing equipment and intelligent manufacturing production lines. Inevitable geometric errors and non-constant operating conditions affect the accuracy decay of the BS component. In this paper, a numerical analysis method of this accuracy decay with geometric parameters errors, under non-constant operating conditions, is established. The accuracy degradation of the BS, under a single non-constant operation condition or multiple non-constant operating conditions, is analyzed, using a proposed numerical analysis method. The average value of the relative error between numerical analysis and theoretical model results was 5.52%, 5.66%, and 5.40%, under the three operating conditions of non-constant rotation rate (RR), non-constant contact load (CL), and non-constant rotation rate and contact load (RR + CL), respectively. And the maximum relative error value was 11.11%, 11.11%, and 10.98%. In addition, the numerical analysis method of BS accuracy degradation was compared to experimental tests. The average value of the relative error between numerical analysis outcome and experimental tests was 7.07%, 6.08%, 6.56%, and the maximum relative error value was 11.11%, 14.29%, and 13.04% under the three operating conditions of RR, CL, and RR + CL, respectively.

show abstract

New domain adaptation method in shallow and deep layers of the CNN for bearing fault diagnosis under different working conditions

Cited by 41 publications

References 22 publications

Fault Identification of Low-Speed Hub Bearing of Crane Based on MBMD and BP Neural Network

Fault Identification of Low-Speed Hub Bearing of Crane Based on MBMD and BP Neural Network

Accuracy decay mechanism of ball screw in CNC machine tools for mixed sliding-rolling motion under non-constant operating conditions

Numerical analysis and experimental research on the accuracy decay of a ball screw with geometric errors under non-constant operating conditions

Contact Info

Product

Resources

About