A clustered federated learning framework for collaborative fault diagnosis of wind turbines

被引:5
作者
Zhou, Rui [1 ]
Li, Yanting [1 ]
Lin, Xinhua [2 ]
机构
[1] Shanghai Jiao Tong Univ, Dept Ind Engn & Management, Shanghai, Peoples R China
[2] Shanghai Jiao Tong Univ, High Performance Comp Ctr, Shanghai, Peoples R China
基金
中国国家自然科学基金;
关键词
Wind turbine; Fault diagnosis; Federated learning; Data heterogeneity; Model similarity; CONVOLUTIONAL NEURAL-NETWORK; ROTATING MACHINERY;
D O I
10.1016/j.apenergy.2024.124532
中图分类号
TE [石油、天然气工业]; TK [能源与动力工程];
学科分类号
0807 ; 0820 ;
摘要
Data-driven approaches demonstrate significant potential in accurately diagnosing faults in wind turbines. To enhance diagnostic performance and reduce communication costs in federated learning with data heterogeneity among different clients, we introduce a clustered federated learning framework to wind turbine fault diagnosis. Initially, a lightweight multiscale separable residual network (LMSRN) model is proposed for each local client. The LMSRN model integrates a multiscale spatial feature derivation unit and a depthwise separable feature extraction unit. Subsequently, to tackle data heterogeneity among clients, canonical correlation coefficients of representations are extracted from the intermediate layers of local LMSRN models, and a representational canonical correlation clustering (RCCC) method is proposed to assess the similarity of local LMSRN models and group them into clusters. Finally, a global model is trained for each cluster. Real-world wind turbine data experiments showcase the superior performance of the proposed clustered federated learning framework over traditional methods in terms of diagnostic accuracy and computational speed. Additionally, the optimal choice of the number of clusters is also discussed.
引用
收藏
页数:21
相关论文
共 39 条
[1]  
Ali Sher Muhammad, 2021, 2021 IEEE International Conference on Artificial Intelligence and Computer Applications (ICAICA), P102, DOI 10.1109/ICAICA52286.2021.9498027
[2]   Deep learning for automated drivetrain fault detection [J].
Bach-Andersen, Martin ;
Romer-Odgaard, Bo ;
Winther, Ole .
WIND ENERGY, 2018, 21 (01) :29-41
[3]   Federated learning with hierarchical clustering of local updates to improve training on non-IID data [J].
Briggs, Christopher ;
Fan, Zhong ;
Andras, Peter .
2020 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2020,
[4]   SMOTE: Synthetic minority over-sampling technique [J].
Chawla, Nitesh V. ;
Bowyer, Kevin W. ;
Hall, Lawrence O. ;
Kegelmeyer, W. Philip .
2002, American Association for Artificial Intelligence (16)
[5]   Diagnosis of wind turbine faults with transfer learning algorithms [J].
Chen, Wanqiu ;
Qiu, Yingning ;
Feng, Yanhui ;
Li, Ye ;
Kusiak, Andrew .
RENEWABLE ENERGY, 2021, 163 :2053-2067
[6]   Wind turbine blade icing detection: a federated learning approach [J].
Cheng, Xu ;
Shi, Fan ;
Liu, Yongping ;
Liu, Xiufeng ;
Huang, Lizhen .
ENERGY, 2022, 254
[7]   Xception: Deep Learning with Depthwise Separable Convolutions [J].
Chollet, Francois .
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, :1800-1807
[8]  
Council GWE, 2023, Global Wind Report 2023
[9]  
Duong LR, 2023, Arxiv, DOI [arXiv:2211.11665, DOI 10.48550/ARXIV.2211.11665]
[10]  
Glorot Xavier., 2011, Proceedings of the 14th International Conference on Artificial Intelligence and Statistics. JMLR WCP, V15, P315