Cross-Domain Fault Diagnosis of Rolling Element Bearings Using Deep Generative Neural Networks

Li, Xiang; Zhang, Wei; Ding, Qian

doi:10.1109/tie.2018.2868023

Cited by 333 publications

(145 citation statements)

References 29 publications

Supporting

Mentioning

143

Contrasting

Unclassified

Order By: Relevance

“…We choose the architecture in Fig 2 because it composes a rather strong baseline for domain adaptation tasks. The effectiveness of the architecture is proved in [14], and also validated by our re-produced results.…”

Section: A Baseline Architecturementioning

confidence: 53%

See 1 more Smart Citation

Domain Adaptive Transfer Learning for Fault Diagnosis

Wang

Michau

Fink

2019

2019 Prognostics and System Health Management Conference (PHM-Paris)

113

View full text Add to dashboard Cite

Thanks to digitization of industrial assets in fleets, the ambitious goal of transferring fault diagnosis models from one machine to the other has raised great interest. Solving these domain adaptive transfer learning tasks has the potential to save large efforts on manually labeling data and modifying models for new machines in the same fleet. Although datadriven methods have shown great potential in fault diagnosis applications, their ability to generalize on new machines and new working conditions are limited because of their tendency to overfit to the training set in reality. One promising solution to this problem is to use domain adaptation techniques. It aims to improve model performance on the target new machine. Inspired by its successful implementation in computer vision, we introduced Domain-Adversarial Neural Networks (DANN) to our context, along with two other popular methods existing in previous fault diagnosis research. We then carefully justify the applicability of these methods in realistic fault diagnosis settings, and offer a unified experimental protocol for a fair comparison between domain adaptation methods for fault diagnosis problems.

show abstract

Section: A Baseline Architecturementioning

confidence: 53%

“…The CNN backbone from [14] is used as shown in Fig 2. The basic architecture is composed of two parts: a feature extractor, and a basic classifier.…”

Section: A Baseline Architecturementioning

confidence: 99%

Domain Adaptive Transfer Learning for Fault Diagnosis

Wang

Michau

Fink

2019

2019 Prognostics and System Health Management Conference (PHM-Paris)

113

View full text Add to dashboard Cite

show abstract

“…We follow the setup used by [35] whenever possible. Thus drive-end accelerometer data are used as our input.…”

Section: ) Cwrumentioning

confidence: 99%

“…For the CWRU dataset, we follow the same pre-processing steps as in [29], [35]. As shown in Figure 4, first, we downsample and truncate each raw recording.…”

Section: B Pre-processingmentioning

confidence: 99%

See 1 more Smart Citation

Missing-Class-Robust Domain Adaptation by Unilateral Alignment

Wang

Michau

Fink

2021

IEEE Trans. Ind. Electron.

View full text Add to dashboard Cite

Domain adaptation aims at improving model performance by leveraging the learned knowledge in the source domain and transferring it to the target domain. Recently, domain adversarial methods have been particularly successful in alleviating the distribution shift between the source and the target domains. However, these methods assume an identical label space between the two domains. This assumption imposes a significant limitation for real applications since the target training set may not contain the complete set of classes. We demonstrate in this paper that the performance of domain adversarial methods can be vulnerable to an incomplete target label space during training. To overcome this issue, we propose a two-stage unilateral alignment approach. The proposed methodology makes use of the inter-class relationships of the source domain and aligns unilaterally the target to the source domain. The benefits of the proposed methodology are first evaluated on the MNIST→MNIST-M adaptation task. The proposed methodology is also evaluated on a fault diagnosis task, where the problem of missing fault types in the target training dataset is common in practice. Both experiments demonstrate the effectiveness of the proposed methodology.Index Terms-fault diagnosis, domain adaptation, feature alignment

show abstract

Self‐optimization for smelting process of fused magnesium furnace based on operation status assessment

2021

View full text Add to dashboard Cite

Based on operation status assessment and self‐optimizing regulation, an optimization strategy for the smelting process of a fused magnesium furnace is proposed to improve the benefits of smelting magnesia. In the online evaluation stage, according to the qualitative information, the data is divided by the fuzzy model recognition method, and then the posterior probability of online data in different Gaussian mixture models is calculated to obtain the evaluation results. When the evaluation results are not optimal, the non‐optimal variables are obtained by calculating the contribution rate, and then the self‐optimizing adjustment is carried out by case search. In order to improve the recovery rate of self‐optimizing regulation, a time series prediction method is used to predict the operation status, and the corresponding operation schemes for different prediction results are proposed. Experimental results show that the proposed method is accurate and effective.

show abstract

Cross-Domain Fault Diagnosis of Rolling Element Bearings Using Deep Generative Neural Networks

Cited by 333 publications

References 29 publications

Domain Adaptive Transfer Learning for Fault Diagnosis

Domain Adaptive Transfer Learning for Fault Diagnosis

Missing-Class-Robust Domain Adaptation by Unilateral Alignment

Self‐optimization for smelting process of fused magnesium furnace based on operation status assessment

Contact Info

Product

Resources

About