Author:
An Zhiguo,Yi Mancheng,Liu Jing,Li Ying,Peng Zheng,Yu Sifan,Liu Jianxin,Huang Weirong,Fang Chunhua
Abstract
The traditional power grid ticket filling method has a large workload, low efficiency, and cannot achieve comprehensive and effective reference of historical tickets. This paper proposes a method of intelligent filling in a power grid working ticket based on a historical ticket knowledge base. Firstly, the historical ticket data are preprocessed, then the historical ticket information is mined by the association rule algorithm, and the method of establishing the historical ticket knowledge base is proposed. Based on the improved word bag model, an intelligent grid work ticket filling model is established based on the historical ticket knowledge base, and the correctness of the method is verified by an example. The results show that the accuracy of the proposed method is at least 18% higher than that of the traditional model, and the matching efficiency is 50% higher than the evaluation results of the three models based on semantic expressions. The method enables the identification and extraction of similar and associated work tickets, improves the efficiency of filling work tickets for power grids, and promotes the intelligence of the safety procedures for power grid operations.
Subject
Economics and Econometrics,Energy Engineering and Power Technology,Fuel Technology,Renewable Energy, Sustainability and the Environment
Reference19 articles.
1. Semantic Framework-Based Text Mining Technique for Grid Defects and its Application;Cao;Power Grid Technol.,2017
2. Text Mining of Typical Fault Cases of Power Equipment;Du;High Voltage Technol.,2018
3. Data Mining-Based Cyber Security Posture Assessment of Electric Power Information Systems;Gao;Electr. Meas. Instrumentation,2019
4. Intelligent Error Prevention Technology of Maintenance Work Ticket Based on Layered Finite State Machine;Gui;J. Hefei Univ. Technol. (Natural Sci. Edition),2021
5. A Method for Similarity Analysis of Power System State;Han,2016
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献