Fault diagnosis method of rolling bearings under varying working conditions based on deep feature transfer

Kang, Shouqiang; Qiao, Chunyang; Wang, Yujing; Wang, Qingyan; Hu, Mingwu; Микулович, В. И.

doi:10.1007/s12206-020-1003-9

Cited by 19 publications

(16 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Qian et al [12] employed a sparse filtering method with divergence to learn the shared identification characteristics of the source domain and target domain. Kang et al [13] put forward a transfer fault diagnosis method based on variable mode decomposition (VMD) and multi-feature construction. Lee et al [14] associated the output of the first convolution layer with the structural meaning of the original data, to locate the variables and time information representing the process fault.…”

Section: Jinst 15 P06002mentioning

confidence: 99%

Wasserstein distance based deep multi-feature adversarial transfer diagnosis approach under variable working conditions

et al. 2020

View full text Add to dashboard Cite

Intelligent mechanical fault diagnosis is a crucial measure to ensure the safe operation of equipment. To solve the problem that network features is not fully utilized in the adversarial transfer learning, this paper develops a Wasserstein distance based deep multi-feature adversarial (WDDMA) transfer diagnosis approach under variable working conditions. Domain adaptation is realized by adapting multi-feature in adversarial training and reducing the Wasserstein distance between the two domains after the discriminator network. The multi-feature are adapted at the same time to improve the recognition ability of the transfer network to the target domain. Experimental datasets under variable working conditions support the value of our approach.

show abstract

Section: Jinst 15 P06002mentioning

confidence: 99%

Wasserstein distance based deep multi-feature adversarial transfer diagnosis approach under variable working conditions

et al. 2020

View full text Add to dashboard Cite

show abstract

“…TCA, which is proposed by Pan et al [30] has been employed to address cross-domain fault diagnosis of gear by Xie et al [60], [61], rolling element bearing by Chen et al [72], and delta 3D printer by Guo et al [123]. Similarly, Kang et al [67] utilized the Semi-supervised TCA (SSTCA) to diagnose bearing fault under variations of operating conditions. A multi-kernel kernel function is constructed for SSTCA by combining Polynomial kernel K poly and Radial Basis Function (RBF) kernel K rbf , that is K i,j = αK poly + (1 − α) K rbf where 0 ≤ α ≤ 1 is a multikernel coefficient.…”

Section: ) Traditional Transfer Approachesmentioning

confidence: 99%

“…Mostly, the inputs of the cross-domain diagnosis approaches based on traditional transfer learning were handcrafted features, such as statistical parameters of vibration signal [60], [61], [67], [71], [80], [96], [123], SVD eigenvalues [81], [83] etc. The features used in these approaches were usually elaborately extracted and selected from massive candidates, and in this procedure the expert knowledge that which features are more discriminative has been considered.…”

Section: ) Inputs Of Cross-domain Diagnosis Approachesmentioning

confidence: 99%

“…In their research, different domains were simulated by the data from different operating conditions (including rotating speed and load), and TrAdaboost method was applied to achieve knowledge transfer. After that, Zhang and Peng et al [73], [74], [116], Li, Zhang, and Ding et al [82], [84], [112], Zhang, Li and Tong et al [62], [75]- [78], Wen and Gao et al [88], [105], [117], [118], Qian and Li et al [63]- [65], [124], Han et al [85], Xu et al [87], Cheng et al [66], M.J.Hasan and Kim [69], [70], and others [67], [86], [110], [127] also focused on the cross-domain diagnosis problem of bearing under variation of operating conditions. Aiming at the data discrepancy caused by different fault severities, references [90], [91] and [93] achieved the cross-domain diagnosis between different bearing fault sizes using corresponding transfer learning methods.…”

Section: Applicationsmentioning

confidence: 99%

See 1 more Smart Citation

Cross-Domain Fault Diagnosis Using Knowledge Transfer Strategy: A Review

et al. 2019

View full text Add to dashboard Cite

Data-driven fault diagnosis has been a hot topic in recent years with the development of machine learning techniques. However, the prerequisite that the training data and the test data should follow an identical distribution prevents the conventional data-driven diagnosis methods from being applied to the engineering diagnosis problems. To tackle this dilemma, cross-domain fault diagnosis using knowledge transfer strategy is becoming popular in the past five years. The diagnosis methods based on transfer learning aim to build models that can perform well on target tasks by leveraging knowledge from semantic related but distribution different source domains. This paper for the first time summarizes the state-of-art cross-domain fault diagnosis research works. The literatures are introduced from three different viewpoints: research motivations, cross-domain strategies, and application objects. In addition, the corresponding open-source fault datasets and several future directions are also presented. The survey provides readers a framework for better understanding and identifying the research status, challenges and future directions of cross-domain fault diagnosis. INDEX TERMS Cross-domain, domain adaptation, fault diagnosis, review, transfer learning.

show abstract

“…9,10 In the area of fault diagnosis, some research on transfer diagnosis methods by using traditional methods have been carried out such as transfer component analysis, 11 singular value decomposition and TrAdaBoost. 12 Kang et al 13 introduced a transfer fault diagnosis method by multifeature construction and variable mode decomposition.…”

Section: Introductionmentioning

confidence: 99%

Diversity maximization-based transfer diagnosis approach of rotating machinery

She

Chen

Yan

et al. 2023

Structural Health Monitoring

View full text Add to dashboard Cite

The existing transfer diagnosis methods based on entropy minimization are easy to lead to trivial solution. To solve this problem, a deep diversity maximization-based adversarial transfer diagnosis approach for rotating machinery is presented in this paper. Firstly, the deep convolution neural network is utilized as the feature encoder to learn the characteristics of vibration signals in different working conditions. The diversity maximization strategy is taken to balance the entropy minimization, so as to avoid trivial local minimum. The categories predicted by nontrivial domain adaptation method are more diverse. Moreover, the entropy is conducted to evaluate the uncertainty of the predicted result of the classifier. Using this deterministic strategy based on entropy to adjust the domain discriminator. The experimental study demonstrates the effectiveness of the developed method.

show abstract

Fault diagnosis method of rolling bearings under varying working conditions based on deep feature transfer

Cited by 19 publications

References 15 publications

Wasserstein distance based deep multi-feature adversarial transfer diagnosis approach under variable working conditions

Wasserstein distance based deep multi-feature adversarial transfer diagnosis approach under variable working conditions

Cross-Domain Fault Diagnosis Using Knowledge Transfer Strategy: A Review

Diversity maximization-based transfer diagnosis approach of rotating machinery

Contact Info

Product

Resources

About