A new dynamic model and transfer learning based intelligent fault diagnosis framework for rolling element bearings race faults: Solving the small sample problem

Dong, Yunjia; Li, Yuqing; Zheng, Haitao; Wang, Rixin; Xu, Min

doi:10.1016/j.isatra.2021.03.042

Cited by 114 publications

(53 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such a problem has been widely studied in the field of machine learning. As we can observe, its appearance usually accompanies by transfer learning [79].…”

Section: B Challenges 1) Fault Diagnosismentioning

confidence: 99%

“…Furthermore, transfer learning was extended to the application of prognostic and health management tasks in [78]. For a dynamic rolling system, [79] proposed an improved FD method based on parameter transfer, where the small sample problem was also taken into account. Knowledge transfer is usually adopted to address multiple tasks; for instance, [80] proposed a transfer learning-based FD strategy by incorporating higherorder spectral analysis.…”

Section: Developments Of Transfer Learning-based Fdmentioning

confidence: 99%

See 1 more Smart Citation

Transfer Learning-motivated Intelligent Fault Diagnosis Designs: A Survey, Insights, and Perspectives

Chen¹,

Luo²,

Huang³

et al. 2022

Preprint

View full text Add to dashboard Cite

<p>Over the last decade, transfer learning has attracted a great deal of attention as a new learning paradigm, based on which fault diagnosis (FD) approaches have been intensively developed to improve the safety and reliability of modern automation systems. Because of inevitable factors such as the varying work environment, performance degradation of components, and heterogeneity among similar automation systems, the FD method having long-term applicabilities becomes attractive. Motivated by these facts, transfer learning has been an indispensable tool that endows the FD methods with self-learning and adaptive abilities. On the presentation of basic knowledge in this field, a comprehensive review of transfer learning-motivated FD methods, whose two subclasses are developed based on knowledge calibration and knowledge compromise, is carried out in this survey paper. Finally, some open problems, potential research directions, and conclusions are highlighted. Different from the existing reviews of transfer learning, this survey focuses on how to utilize knowledge specifically for the FD tasks, based on which three principles and a new classification strategy of transfer learning-motivated FD techniques are also presented. We hope this work will constitute a timely contribution to transfer learning-motivated techniques regarding the FD topic.</p>

show abstract

“…Such a problem has been widely studied in the field of machine learning. As we can observe, its appearance usually accompanies by transfer learning [79].…”

Section: B Challenges 1) Fault Diagnosismentioning

confidence: 99%

Section: Developments Of Transfer Learning-based Fdmentioning

confidence: 99%

Transfer Learning-motivated Intelligent Fault Diagnosis Designs: A Survey, Insights, and Perspectives

Chen¹,

Luo²,

Huang³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Cui et al [ 7 ] proposed a coupled multistable stochastic resonance method with two first-order multistable stochastic resonance systems, which has higher output signal-to-noise ratio than multistable stochastic resonance. Dong et al [ 8 ] used a bearing dynamics model to generate a large amount of simulated data for the small sample problem, and then implemented transfer learning between the diagnostic knowledge in the simulated data to the real scenario based on convolutional neural network and parameter transfer strategies. Considering the compensation balance excitation caused by the rotor mass eccentricity, Liu et al [ 9 ] developed a dynamic model for the bearing rotor system of a high-speed train under variable speed conditions.…”

Section: Introductionmentioning

confidence: 99%

Coupled Electromagnetic-Dynamic Modeling and Bearing Fault Characteristics of Induction Motors considering Unbalanced Magnetic Pull

Huang

Shen

et al. 2022

Entropy

View full text Add to dashboard Cite

Induction motors are complex energy conversion systems across the domains of dynamics, electricity, and magnetism. Most existing models mainly consider unidirectional coupling, such as the effect of dynamics on electromagnetic properties, or the effect of unbalanced magnetic pull on dynamics, while in practice it should be a bidirectional coupling effect. The bidirectionally coupled electromagnetic-dynamics model is beneficial to the analysis of induction motor fault mechanisms and characteristics. This paper proposes a coupled electromagnetic-dynamic modeling method that introduces unbalanced magnetic pull. By using the rotor velocity, air gap length, and unbalanced magnetic pull as the coupling parameters, the coupled simulation of the dynamic and electromagnetic models can be effectively realized. Simulation results for bearing faults show that the introduction of magnetic pull induces a more complex dynamic behavior of the rotor, which in turn leads to modulation in the vibration spectrum. The fault characteristics can be found in the frequency domain of the vibration and current signals. Through the comparison between simulation and experimental results, the effectiveness of the coupled modeling approach and the frequency domain characteristics caused by the unbalanced magnetic pull are verified. The proposed model can help to obtain a variety of information that is difficult to measure in reality and can also serve as a technical basis for further research on nonlinear characteristics and chaos in induction motors.

show abstract

“…This method improved the robustness and generalization of ADCNN, and avoid over-fitting with limited number of labeled samples. Dong et al [ 32 ] proposed a dynamic model of bearing to generate massive and various simulation data, and diagnosis for real scenario are based on transfer strategies and CNN. Pan et al [ 33 ] proposed a semi-supervised multi-scale convolutional generative adversarial network for bearing fault identification when the labeled data are insufficient.…”

Section: Introductionmentioning

confidence: 99%

A Novel Anti-Noise Fault Diagnosis Approach for Rolling Bearings Based on Convolutional Neural Network Fusing Frequency Domain Feature Matching Algorithm

Zhou

Mao

2021

Sensors

View full text Add to dashboard Cite

The development of deep learning provides a new research method for fault diagnosis. However, in the industrial field, the labeled samples are insufficient and the noise interference is strong so that raw data obtained by the sensor are occupied with noise signal. It is difficult to recognize time-domain fault signals under the severe noise environment. In order to solve these problems, the convolutional neural network (CNN) fusing frequency domain feature matching algorithm (FDFM), called CNN-FDFM, is proposed in this paper. FDFM extracts key frequency features from signals in the frequency domain, which can maintain high accuracy in the case of strong noise and limited samples. CNN automatically extracts features from time-domain signals, and by using dropout to simulate noise input and increasing the size of the first-layer convolutional kernel, the anti-noise ability of the network is improved. Softmax with temperature parameter T and D-S evidence theory are used to fuse the two models. As FDFM and CNN can provide different diagnostic information in frequency domain, and time domain, respectively, the fused model CNN-FDFM achieves higher accuracy under severe noise environment. In the experiment, when a signal-to-noise ratio (SNR) drops to -10 dB, the diagnosis accuracy of CNN-FDFM still reaches 93.33%, higher than CNN’s accuracy of 45.43%. Besides, when SNR is greater than -6 dB, the accuracy of CNN-FDFM is higher than 99%.

show abstract

A new dynamic model and transfer learning based intelligent fault diagnosis framework for rolling element bearings race faults: Solving the small sample problem

Cited by 114 publications

References 59 publications

Transfer Learning-motivated Intelligent Fault Diagnosis Designs: A Survey, Insights, and Perspectives

Transfer Learning-motivated Intelligent Fault Diagnosis Designs: A Survey, Insights, and Perspectives

Coupled Electromagnetic-Dynamic Modeling and Bearing Fault Characteristics of Induction Motors considering Unbalanced Magnetic Pull

A Novel Anti-Noise Fault Diagnosis Approach for Rolling Bearings Based on Convolutional Neural Network Fusing Frequency Domain Feature Matching Algorithm

Contact Info

Product

Resources

About