AUTHOR=Xiao Dajun , Xu Xialing , Zhang Bo , Zhang Yue , Shan Lianfei , Liu Tao , Li Xin , Qiao Yongtian , Jiang Tao , Wang Yu 

TITLE=Power grid fault handling plan matching method based on a hybrid neural network

JOURNAL=Frontiers in Energy Research

VOLUME=Volume 12 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2024.1468651

DOI=10.3389/fenrg.2024.1468651

ISSN=2296-598X

ABSTRACT=To improve the accuracy of online matching and pushing of power grid fault handling plan, a matching method of fault handling plan based on hybrid neural network is proposed. Firstly, the ERNIE 3.0 encoding and double-pointer decoding module are used to replace the generative model in the universal information extraction (UIE) framework, and the mapping relationship between entities and entity labels of fault handling plan is trained by adjusting the hyper-parameters of the UIE framework. Then, the semantic distance between the fault equipment, fault type, fault phenomenon and the entity of fault handling plan is calculated based on the residual vector-embedding vector-encoded vector (RE2). The hybrid neural network model for power grid fault handling plan matching is established. Finally, through the verification of fault related data of a regional power grid, the proposed fault handling plan matching method has higher matching accuracy and stronger generalization ability compared with other algorithms. The average precision rate, recall rate and F1 value of the built fault handling plan matching model are 97.61%, 98.24% and 97.91%, respectively, which can provide auxiliary decision for timely and rapid treatment of power grid faults.