TechRxiv

Blade Fault Diagnosis Based on Hybrid Physics and Domain Adaptation: A Case Study of Variable-Speed Marine Current Turbines

Download (9.36 MB)
preprint
posted on 2023-11-22, 15:19 authored by Tao XieTao Xie, Zhihuang Hu, Tianzhen Wang, weidong zhang, Hongtian ChenHongtian Chen

As a machine learning approach, domain adaptation methods are widely applied in cross-scenario fault diagnosis. However, the target domain may need more annotated data, posing challenges to the performance of domain adaptation methods. This paper proposes a fault diagnosis method based on hybrid physics and domain adaptation (HPDA) with its application to marine current turbines (MCTs) scenarios. Specifically, this method first establishes a rotational feature alignment model based on physical variables. Then, it aligns the feature of the target domain data with physical parameters. Finally, an augmented domain adversarial model is trained using pre-alignment samples. Data from MCT prototypes are collected to validate the effectiveness of the proposed method. Experimental results demonstrate the proposed method's superior stability and data transferability compared with the state-of-the-art.

Funding

National Natural Science Foundation of China 62303305

National Natural Science Foundation of China 62303308

National Natural Science Foundation of China U2141234

Shanghai Pujiang Program 23PJ1404700

National Key R and D Program of China 2022ZD0119900

Shanghai Science and Technology program 22015810300

Hainan Province Science and Technology Special Fund ZDYF2021GXJS041

History

Email Address of Submitting Author

xietao0906@sjtu.edu.cn

ORCID of Submitting Author

0000-0001-7911-0517

Submitting Author's Institution

Shanghai Jiaotong University

Submitting Author's Country

  • China

Usage metrics

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC