A dual-view alignment-based domain adaptation network for fault diagnosis

Zhao, Chi; Liu, Guokai; Shen, Weiming

doi:10.1088/1361-6501/ac100e

Cited by 13 publications

(7 citation statements)

References 34 publications

(17 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Liu et al [221] also used the Wasserstein distance to calculate the distribution discrepancy of bearing condition diagnosis datasets. Zhao et al [222] used both the MMD and Wasserstein distance for the condition diagnosis of bearings. Another similarity measure utilized in transfer learning is Kullback-Leibler divergence.…”

Section: B Feature Alignment By Other Similarity Measuresmentioning

confidence: 99%

See 1 more Smart Citation

Using Data From Similar Systems for Data-Driven Condition Diagnosis and Prognosis of Engineering Systems: A Review and an Outline of Future Research Challenges

Braig

Zeiler

2023

IEEE Access

View full text Add to dashboard Cite

Prognostics and health management (PHM) is an engineering approach dealing with the diagnosis, prognosis, and management of the health state of engineering systems. Methods that can analyze system behavior, fault conditions, and degradation are crucial for PHM applications, as they create the basis for determining, predicting, and monitoring the health of engineering systems. Data-driven methods have been proven to be suitable for automated diagnosis or prognosis due to their pattern recognition and anomaly detection abilities. Moreover, they do not require knowledge of the underlying degradation process. However, training data-driven methods usually requires a large amount of data, whose collection, cleansing, organization, and preparation are generally very time-consuming and costly. There are usually little or no run-to-failure data available at market launch, especially for new systems such as new machine generations. Nevertheless, related systems, hereinafter referred to as similar systems, often already exist, differing only in some technical characteristics. In this paper, the similar system problem is defined, and explanations of the difficulties that arise when using data from similar systems are presented. Furthermore, it is discussed why the usage of these data offers great potential for condition diagnosis and prognosis of engineering systems. An overview of data-driven methods that can be used to utilize data from similar systems is provided, and the methods that such systems already consider are highlighted. Two related research areas are identified, namely, fleet learning and transfer learning. In the paper, it is shown that similar system approaches will become an important branch of research in PHM. However, some difficulties must be overcome.INDEX TERMS Condition diagnosis, condition prognosis, data-driven methods, fleet learning, prognostics and health management (PHM), similar system approach, similar system problem, transfer learning.

show abstract

Section: B Feature Alignment By Other Similarity Measuresmentioning

confidence: 99%

“…In some approaches, multiple similarity measures were used simultaneously or in combination. As already mentioned, Zhao et al [222] applied the MMD and Wasserstein distance for condition diagnosis. In addition, Cao et al [231] used these two measures for RUL prognosis of bearings.…”

Section: B Feature Alignment By Other Similarity Measuresmentioning

confidence: 99%

Using Data From Similar Systems for Data-Driven Condition Diagnosis and Prognosis of Engineering Systems: A Review and an Outline of Future Research Challenges

Braig

Zeiler

2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…In order to cope with the challenges brought by global domain adaptation, more and more researchers have begun to pay attention to subdomain adaptation methods [22][23][24][25][26]. Compared with the global domain adaptive method, the subdomain adaptive method pays more attention to the matching of local data distribution and achieves more accurate domain adaptation by aligning the distribution of the same type of data in the source domain and target domain.…”

Section: Definition Of Subdomain Adaptive Problemmentioning

confidence: 99%

Scale adaptive subdomain matching network for bearing fault diagnosis

Zhang¹,

Yuan²,

Wang³

et al. 2021

Meas. Sci. Technol.

View full text Add to dashboard Cite

The wide application of transfer learning technology can effectively solves the problem of differences between data collection and actual application equipment faced by traditional intelligent fault diagnosis methods in the practical application process. However, the difference in subdomain space and serious imbalance of data samples in the process of simultaneous transfer restrict the deep transfer learning technology to the engineering application of high-precision diagnosis. Aiming at the problem of subdomain matching with different subspaces and unbalanced data samples, this paper studies the subdomain adaptive method and proposes a Scale Adaptive Subdomain Matching (SASM) method. The SASM method divides the global feature space according to the sample labels, and the features with the same label will be divided into the same sub-feature space. Using the edge distribution of the sample and the category weight of the label, the SASM method can effectively optimize the feature distribution of the same subdomain and the weight distribution of different subdomains. Based on the establishment of a clearer internal structure of features, the field adaptation effect is improved, and the matching ability is enhanced when the sample is unevenly distributed. At the same time, the Scale Adaptive Subdomain Matching Network (SASMN) method for unsupervised bearing fault diagnosis is constructed and validated by experiments. The results indicate that SASMN can effectively optimize the subdomain adaptive effect, and the diagnostic accuracy on the target domain dataset is significantly higher than the other three currently popular domain adaptive fault diagnosis methods.

show abstract

“…the daily operation of the entire working system but may also cause serious accidents resulting in personnel injury and significant economic losses. Therefore, the research and development of rotating machinery fault diagnosis technology are of great significance [4][5][6].…”

Section: Introductionmentioning

confidence: 99%

A feature extraction method for rotating machinery fault diagnosis based on a target detection index and successive variational mode decomposition

Cao,

Zhang,

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

Rotating machinery feature extraction is critical for the subsequent fault diagnosis and ensuring safe and stable operation. However, the commonly used methods often have limitations, such as the extracted features being redundant or insufficient and the method parameters generally being set based on experience. Although some feature extraction methods apply optimization algorithms for parameter setting, their objective functions are often too simple to obtain favorable fault diagnosis results. To address these issues, a target detection index (TDI) is constructed which can consider the discrimination among features more comprehensively and make the extracted features more sensitive. Furthermore, a fault feature extraction method is proposed based on TDI and successive variational mode decomposition (SVMD). Taking TDI as the objective function, genetic algorithm (GA) is used for the feature selection process and SVMD parameter optimization. The obtained features are then fused and visualized using t-distributed stochastic neighbor embedding and are classified using support vector machines. The Case Western Reserve University data and hydropower generating unit data are employed for method verification. When compared with other decomposition algorithms, the proposed method exhibits great ability at extracting highly sensitive features.

show abstract

A dual-view alignment-based domain adaptation network for fault diagnosis

Cited by 13 publications

References 34 publications

Using Data From Similar Systems for Data-Driven Condition Diagnosis and Prognosis of Engineering Systems: A Review and an Outline of Future Research Challenges

Using Data From Similar Systems for Data-Driven Condition Diagnosis and Prognosis of Engineering Systems: A Review and an Outline of Future Research Challenges

Scale adaptive subdomain matching network for bearing fault diagnosis

A feature extraction method for rotating machinery fault diagnosis based on a target detection index and successive variational mode decomposition

Contact Info

Product

Resources

About