Asynchronous Decentralized Federated Learning for Collaborative Fault Diagnosis of PV Stations

被引:32
作者
Liu, Qi [1 ,2 ,3 ]
Yang, Bo [1 ,2 ,3 ]
Wang, Zhaojian [1 ,2 ,3 ]
Zhu, Dafeng [1 ,2 ,3 ]
Wang, Xinyi [1 ,2 ,3 ]
Ma, Kai [4 ]
Guan, Xinping [1 ,2 ,3 ]
机构
[1] Shanghai Jiao Tong Univ, Dept Automat, Shanghai 200240, Peoples R China
[2] Minist Educ China, Key Lab Syst Control & Informat Proc, Shanghai 200240, Peoples R China
[3] Minist Educ China, Shanghai Engn Res Ctr Intelligent Control & Manag, Shanghai 200240, Peoples R China
[4] Yanshan Univ, Sch Elect Engn, Qinhuangdao 066004, Hebei, Peoples R China
来源
IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING | 2022年 / 9卷 / 03期
关键词
Data models; Computational modeling; Fault diagnosis; Collaboration; Robustness; Photovoltaic systems; Heuristic algorithms; Asynchronous decentralized learning; collaborative fault diagnosis; federated learning; multiple PV stations; LOCALIZATION;
D O I
10.1109/TNSE.2022.3150182
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
Due to the different losses caused by various photovoltaic (PV) array faults, accurate diagnosis of fault types is becoming increasingly important. Compared with a single one, multiple PV stations collect sufficient fault samples, but their data is not allowed to be shared directly due to potential conflicts of interest. Therefore, federated learning can be exploited to train a collaborative fault diagnosis model. However, the modeling efficiency is seriously affected by the model update mechanism since each PV station has a different computing capability and amount of data. Moreover, for the safe and stable operation of the PV system, the robustness of collaborative modeling must be guaranteed rather than simply being processed on a central server. To address these challenges, a novel asynchronous decentralized federated learning (ADFL) framework is proposed. Each PV station not only trains its local model but also participates in collaborative fault diagnosis by exchanging model parameters to improve the generalization without losing accuracy. The global model is aggregated distributedly to avoid central node failure. By designing the asynchronous update scheme, the communication overhead and training time are greatly reduced. Both the experiments and numerical simulations are carried out to verify the effectiveness of the proposed method.
引用
收藏
页码:1680 / 1696
页数:17
相关论文
共 40 条
  • [1] A Comprehensive Review of Catastrophic Faults in PV Arrays: Types, Detection, and Mitigation Techniques
    Alam, Mohammed Khorshed
    Khan, Faisal
    Johnson, Jay
    Flicker, Jack
    [J]. IEEE JOURNAL OF PHOTOVOLTAICS, 2015, 5 (03): : 982 - 997
  • [2] Convex Optimization: Algorithms and Complexity
    不详
    [J]. FOUNDATIONS AND TRENDS IN MACHINE LEARNING, 2015, 8 (3-4): : 232 - +
  • [3] PV array power output maximization under partial shading using new shifted PV array arrangements
    Belhaouas, N.
    Cheikh, M. -S. Ait
    Agathoklis, P.
    Oularbi, M. -R.
    Amrouche, B.
    Sedraoui, K.
    Djilali, N.
    [J]. APPLIED ENERGY, 2017, 187 : 326 - 337
  • [4] Adaptive Fault Localization in Photovoltaic Systems
    Chen, Leian
    Wang, Xiaodong
    [J]. IEEE TRANSACTIONS ON SMART GRID, 2018, 9 (06) : 6752 - 6763
  • [5] Internet of Things Based Smart Grids Supported by Intelligent Edge Computing
    Chen, Songlin
    Wen, Hong
    Wu, Jinsong
    Lei, Wenxin
    Hou, Wenjing
    Liu, Wenjie
    Xu, Aidong
    Jiang, Yixin
    [J]. IEEE ACCESS, 2019, 7 : 74089 - 74102
  • [6] Rapid and accurate modeling of PV modules based on extreme learning machine and large datasets of I-V curves
    Chen, Zhicong
    Yu, Hui
    Luo, Linlu
    Wu, Lijun
    Zheng, Qiao
    Wu, Zhenhui
    Cheng, Shuying
    Lin, Peijie
    [J]. APPLIED ENERGY, 2021, 292
  • [7] Deep residual network based fault detection and diagnosis of photovoltaic arrays using current-voltage curves and ambient conditions
    Chen, Zhicong
    Chen, Yixiang
    Wu, Lijun
    Cheng, Shuying
    Lin, Peijie
    [J]. ENERGY CONVERSION AND MANAGEMENT, 2019, 198
  • [8] Intelligent fault diagnosis of photovoltaic arrays based on optimized kernel extreme learning machine and I-V characteristics
    Chen, Zhicong
    Wu, Lijun
    Cheng, Shuying
    Lin, Peijie
    Wu, Yue
    Lin, Wencheng
    [J]. APPLIED ENERGY, 2017, 204 : 912 - 931
  • [9] Optimal Contract Design for Efficient Federated Learning With Multi-Dimensional Private Information
    Ding, Ningning
    Fang, Zhixuan
    Huang, Jianwei
    [J]. IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, 2021, 39 (01) : 186 - 200
  • [10] Domain Knowledge-Based Deep-Broad Learning Framework for Fault Diagnosis
    Feng, Jian
    Yao, Yu
    Lu, Senxiang
    Liu, Yue
    [J]. IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2021, 68 (04) : 3454 - 3464