Cross-domain fault diagnosis of bearing using improved semi-supervised meta-learning towards interference of out-of-distribution samples

Lin, Jian; Shao, Haidong; Min, Zhishan; Luo, Jingjie; Xiao, Yiming; Yan, Shen; Zhou, Jian

doi:10.1016/j.knosys.2022.109493

Cited by 43 publications

(6 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This method leveraged the meta-learning ability to rapidly adapt to new tasks, enabling complex condition few-shot fault diagnosis. Lin et al [157] considered the problem of out-of-distribution unlabeled samples, proposed an improved semi-supervised meta-learning model. They initially made full use of the information in unlabeled samples through a label allocation strategy, alleviating the out-ofdistribution problem.…”

Section: Meta Learningmentioning

confidence: 99%

Application of deep learning to fault diagnosis of rotating machineries

Su,

Xiang,

2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

Deep learning has attained remarkable achievements in diagnosing faults for rotary machineries. Capitalizing on the formidable learning capacity of deep learning, it has the potential to automate human labor and augment the efficiency of fault diagnosis in rotary machinery. These advantages have engendered escalating interest over the past decade. Although recent reviews of the literature have encapsulated the utilization of deep learning in diagnosing faults in rotating machinery, they no longer encompass the introduction of novel methodologies and emerging directions as deep learning methodologies continually evolve. Moreover, in practical application, novel issues and trajectories perpetually manifest, demanding a comprehensive exegesis. To rectify this lacuna, this article amalgamates current research trends and avant-garde methodologies while systematizing the utilization of anterior deep learning techniques. The evolution and extant status of deep learning in diagnosing faults for rotary machinery were delineated, with the intent of providing orientation for prospective research. Over the bygone decade, archetypal deep learning theory has empowered the diagnosis of faults in rotating machinery by directly establishing the nexus between mechanical data and fault conditions. In recent years, meta learning methods aimed at solving small sample scenarios and large model transformers aimed at mining big data features have both received widespread attention and development in the field of fault diagnosis of rotating machinery equipment. Although excellent results have been achieved in these two directions, there is no review and summary article yet, so it is necessary to update the review literature in the field of fault diagnosis of rotating machinery equipment. Lastly, predicated on a survey of the literature and the current developmental landscape, the challenges and prospective orientations of deep learning in rotary machinery fault diagnosis are presented.

show abstract

Section: Meta Learningmentioning

confidence: 99%

Application of deep learning to fault diagnosis of rotating machineries

Su,

Xiang,

2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…Zhang et al proposed a domain adaptation meta-learning network with feature-oriented discard-supplement module for few-shot cross-domain rotating machinery fault diagnosis [26]. Lin et al implemented semi-supervised fault diagnosis based on meta-learning [27]. Li et al proposed a novel few-shot method using attention and meta-transfer learning [28].…”

Section: Introductionmentioning

confidence: 99%

A novel multi-task transfer model to realize unsupervised fault diagnosis of newly constructed wind turbines under variable conditions

liu,

Sha,

Zhang

et al. 2024

Fourth International Conference on Mechanical Engineering, Intelligent Manufacturing, and Automation Technology (MEMAT 2023)

View full text Add to dashboard Cite

Fault diagnosis can effectively improve the power generation of the wind turbines. Deep learning has promoted the intelligent development of wind turbine fault diagnosis. Traditional deep learning usually requires a sufficient amount of labeled data. However, for newly constructed wind turbines, there are problems such as insufficient samples, limited labels, variable operating conditions. Transfer learning provides a new way to solve these problems. Establishing appropriate models to reduce the distribution differences between existing and newly built units is the key to improving unsupervised fault diagnosis accuracy for newly built units. To address these challenges, a novel multi-task transfer model based on improved model agnostic meta learning (MT-TL-MAML) was proposed to realize unsupervised fault diagnosis of newly constructed wind turbines. The proposed model utilizes the advantages of MAML in generalization of new tasks and small sample diagnosis, and can adapt to randomly changing working conditions. By means of iterative learning, the gap between source domain and target domain is further narrowed, and the classifier can realize more accurate diagnosis of target domain data. This article takes the SCADA and CMS data of two wind farms as case studies to conduct unsupervised fault diagnosis and compare it with other literatures. The results validate the advantages of the proposed model in unsupervised fault diagnosis of newly constructed wind turbines.

show abstract

“…It then applies the knowledge learned from a small number of samples to a new field. Since most few-shot tasks provide algorithms that transfer knowledge from a large collection of source datasets to a sparsely annotated collection of target categories, this method essentially falls under the category of meta-learning [ 7 ]. Therefore, this paper focuses primarily on the transfer learning algorithm based on meta-learning as its main research content.…”

Section: Introductionmentioning

confidence: 99%

Research on a Cross-Domain Few-Shot Adaptive Classification Algorithm Based on Knowledge Distillation Technology

Gao,

Li,

Xia

et al. 2024

Sensors

View full text Add to dashboard Cite

With the development of deep learning and sensors and sensor collection methods, computer vision inspection technology has developed rapidly. The deep-learning-based classification algorithm requires the acquisition of a model with superior generalization capabilities through the utilization of a substantial quantity of training samples. However, due to issues such as privacy, annotation costs, and sensor-captured images, how to make full use of limited samples has become a major challenge for practical training and deployment. Furthermore, when simulating models and transferring them to actual image scenarios, discrepancies often arise between the common training sets and the target domain (domain offset). Currently, meta-learning offers a promising solution for few-shot learning problems. However, the quantity of supporting set data on the target domain remains limited, leading to limited cross-domain learning effectiveness. To address this challenge, we have developed a self-distillation and mixing (SDM) method utilizing a Teacher–Student framework. This method effectively transfers knowledge from the source domain to the target domain by applying self-distillation techniques and mixed data augmentation, learning better image representations from relatively abundant datasets, and achieving fine-tuning in the target domain. In comparison with nine classical models, the experimental results demonstrate that the SDM method excels in terms of training time and accuracy. Furthermore, SDM effectively transfers knowledge from the source domain to the target domain, even with a limited number of target domain samples.

show abstract

Cross-domain fault diagnosis of bearing using improved semi-supervised meta-learning towards interference of out-of-distribution samples

Cited by 43 publications

References 30 publications

Application of deep learning to fault diagnosis of rotating machineries

Application of deep learning to fault diagnosis of rotating machineries

A novel multi-task transfer model to realize unsupervised fault diagnosis of newly constructed wind turbines under variable conditions

Research on a Cross-Domain Few-Shot Adaptive Classification Algorithm Based on Knowledge Distillation Technology

Contact Info

Product

Resources

About