Fault diagnosis of the HVDC system based on the CatBoost algorithm using knowledge graphs

• 出版年: Frontiers in Energy Research
• 主要な著者: Jiyang Wu, Qiang Li, Qian Chen, Nan Zhang, Chizu Mao, Litai Yang, Jinyu Wang
• フォーマット: 論文
• 言語: 英語
• 出版事項: Frontiers Media S.A. 2023-03-01
• 主題: HVDC; CatBoost; fault diagnosis; knowledge graph; BP
• オンライン･アクセス: https://www.frontiersin.org/articles/10.3389/fenrg.2023.1144785/full

In order to overcome the difficulty of fault diagnosis in the high-voltage direct current (HVDC) transmission system, a fault diagnosis method based on the categorical boosting (CatBoost) algorithm is proposed in this work. To make the research conform to the actual situation, three kinds of measured fault data in the HVDC system of the Southern Power Grid are selected as the original data set. First, the core role and significance of fault diagnosis in knowledge graphs (KGs) are given, and the characteristics and specific causes of the four fault types are explained in detail. Second, the fault dates are preprocessed and divided into the training data set and the test data set, and the CatBoost algorithm is employed to train and test fault data to realize fault diagnosis. Finally, to verify the progressiveness and effectiveness of the proposed method, the diagnostic results obtained by CatBoost are compared with those obtained by the BP neural network algorithm. The results show that the diagnostic accuracy of the CatBoost algorithm in the three test sets is always higher than that of the BP neural network algorithm; the accuracy rates in the three case studies of the CatBoost algorithm are 94.74%, 100.00%, and 98.21%, respectively, which fully proves that the CatBoost algorithm has a very good fault diagnosis effect on the HVDC system.