Malfunction Diagnosis of Microgrid Devices Based on Optimized Graph Neural Network
DOI:
https://doi.org/10.4108/ew.12180Keywords:
Graph neural network, Malfunction diagnosis, Data repairing, Multi-head attentionAbstract
Malfunction diagnosis based on deep learning ensures the stable operation of microgrid. However, the microgrid system with new energy access appears to be increasingly complex and changeable, and its frequently changing topology brings challenges to the training of traditional neural network models. A method for malfunction diagnosis of microgrid devic-es based on optimized Graph Neural Network is proposed in the paper. Firstly, data reconstruction is completed through the mapping of nodes, edges and graphs in order to standardize changeable data. Secondly, abnormal data is processed to improve the quality of database. In that way, device misoperation caused by extreme data loss can be avoided, thus im-proving the training efficiency. Then, aiming at fully capturing the characteristics of time and space dimensions of data at different latitudes, the graph-convolution network under Multi-head Attention mechanism is adopted in the process. Finally, a microgrid simulation model is built to prove that the proposed scheme has the advantages of good robustness, strong adaptability, high reliability and accurate diagnosis.
Downloads
References
[1] Mo Haojie,Peng Yonggang,Wei Wei,Xi Wei,Cai Tiantian. SR-GNN Based Fault Classification and Location in Power Distribution Network[J]. Energies,2022,16(1).
[2] Gao Hao,Zhang Tong,Chen Songqiang,Wang Lina,Yu Fajiang. FUSION: Measuring Binary Function Similarity with Code-Specific Embedding and Order-Sensitive GNN[J]. Symmetry,2022,14(12).
[3] Veyrin-Forrer Luca,Kamal Ataollah,Duffner Stef-an,Plantevit Marc,Robardet Céline. In pursuit of the hidden features of GNN’s internal representations[J]. Data & Knowledge Engineering,2022,142.
[4] Zhou Xiangyu,Zhang Yuhui,Wei Qianru. Few-Shot Fine-Grained Image Classification via GNN[J]. Sen-sors,2022,22(19).
[5] Wei Quanmin,Wang Jinyan,Hu Jun,Li Xianxian,Yi Tong. OGT: optimize graph then training GNNs for node classifi-cation[J]. Neural Computing and Applications,2022,34(24).
[6] Sathana V.,Mathumathi M.,Makanyadevi K.. Prediction of material property using optimized augmented graph-attention layer in GNN[J]. Materials Today: Proceed-ings,2022,69(P3).
[7] Dash, Tirtharaj,Srinivasan, Ashwin,Baskar, A.. Inclusion of domain-knowledge into GNNs using mode-directed inverse entailment[J]. Machine Learning,2021(prepublish).
[8] Li Tianyu,Xu Huiqi,Zeng Weigui. Ship Classification Method for Massive AIS Trajectories Based on GNN[J]. Journal of Physics: Conference Series,2021,2025(1).
[9] Predrag S. Stanimirović,Marko D. Petković. Improved GNN Models for Constant Matrix Inversion[J]. Neural Processing Letters,2019,50(1).
[10] Priyanka Chaudhary,M. Rizwan. Short term solar energy forecasting using GNN integrated wavelet-based ap-proach[J]. International Journal of Renewable Energy Technology,2019,10(3).
[11] Chen Renxiang, Huang Xing, Yang Lixia,etc. Rolling bear-ing fault diagnosis based on convolution neural network and discrete wavelet transform[J]. Journal of Vibration En-gineering,2018,31(05):883-891.
[12] Liu Ruonan，Yang Boyuan，Zio Enrico，et al． Artificial intelligence for fault diagnosis of rotating machinery: A re-view [J]. Mechanical Systems and Signal Processing, 2018, 10(8): 8-16.
[13] Ye Zhuang, Yu Jianbo. Gearbox Fault Diagnosis Method Based on Multi-channel One-dimensional Convolution Neural Network Feature Learning [J]. Vibration and shock, 2020, 39(20): 55-66.
[14] Guillaume Jaume,An-phi Nguyen,María Rodríguez Mar-tínez,Jean-Philippe Thiran,Maria Gabrani. edGNN: a Sim-ple and Powerful GNN for Directed Labeled Graphs.[J]. CoRR,2019,abs/1904.08745.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Qiyue Huang, Qi Song, Hao Lei
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
This is an open-access article distributed under the terms of the Creative Commons Attribution CC BY 4.0 license, which permits unlimited use, distribution, and reproduction in any medium so long as the original work is properly cited.
