首页 | 本学科首页   官方微博 | 高级检索  
     

基于混合神经网络的电力调度文本事件抽取方法
引用本文:刘赫,皮俊波,宋鹏程,赵翰林,张越,刘显壮. 基于混合神经网络的电力调度文本事件抽取方法[J]. 中国电力, 2022, 55(9): 105-110+120. DOI: 10.11930/j.issn.1004-9649.202111096
作者姓名:刘赫  皮俊波  宋鹏程  赵翰林  张越  刘显壮
作者单位:1. 国家电网有限公司 国家电力调度控制中心,北京 100031;2. 南瑞集团有限公司(国网电力科学研究院有限公司),江苏 南京 211106;3. 北京科东电力控制系统有限责任公司,北京 100192
基金项目:国家电网有限公司科技项目(5108-202040024A-0-0-00)。
摘    要:为了提升电力调度文本在业务场景的应用效果,提出基于混合神经网络的电力调度文本事件抽取方法。以电力调度文本故障处置预案为研究对象,建立以预案触发词为中心的事件抽取模型。通过仿真案例验证了所提方法具有较好的事件抽取效果,能够提升实体和实体关系识别的准确率。

关 键 词:电网调度  混合神经网络  触发词  事件抽取
收稿时间:2021-11-21

An Event Extraction Method for Power Dispatching Text Based on Hybrid Neural Network
LIU He,PI Junbo,SONG Pengcheng,ZHAO Hanlin,ZHANG Yue,LIU Xianzhuang. An Event Extraction Method for Power Dispatching Text Based on Hybrid Neural Network[J]. Electric Power, 2022, 55(9): 105-110+120. DOI: 10.11930/j.issn.1004-9649.202111096
Authors:LIU He  PI Junbo  SONG Pengcheng  ZHAO Hanlin  ZHANG Yue  LIU Xianzhuang
Affiliation:1. National Electric Power Dispatching and Control Center of State Grid Corporation of China, Beijing 100031, China;2. NARI Group Corporation (State Grid Electric Power Research Institute), Nanjing 211106, China;3. Beijing Kedong Electric Power Control System Co., Ltd., Beijing 100192, China
Abstract:In order to improve the application effect of power dispatching text in business scenarios, a method for extracting power dispatching text events is proposed based on hybrid neural network. Taking the power dispatching text fault handling plan as the research object, an event extraction model is established centering on the trigger words of the plan. It is proved through case simulation that the proposed method can better extract the entity elements in the plan event. Compared with other methods, the accuracy of the entity recognition and entity relation recognition is improved.
Keywords:power grid dispatching  hybrid neural network  trigger word  event extraction  
点击此处可从《中国电力》浏览原始摘要信息
点击此处可从《中国电力》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号