极端天气条件下风电机组事故演化知识图谱的构建及分析  

Construction and analysis of knowledge graph for wind turbine accident evolution under extreme weather conditions

作　　者：刘浩[1,2] 徐晴晴 张来斌 李云涛[1,2] 周涛涛 LIU Hao;XU Qingqing;ZHANG Laibin;LI Yuntao;ZHOU Taotao(College of Safety and Ocean Engineering,China University of Petroleum(Beijing),Beijing 102249,China;Key Laboratory of Oil and Gas Production Safety and Emergency Technology,Ministry of Emergency Management,Beijing 102249,China)

机构地区：[1]中国石油大学(北京)安全与海洋工程学院,北京102249 [2]应急管理部油气生产安全与应急技术重点实验室,北京102249

出　　处：《中国安全生产科学技术》2024年第10期5-11,共7页Journal of Safety Science and Technology

基　　金：中国石油科技创新基金研究项目(2021DQ02-0801)。

摘　　要：极端天气(强风、雷电、盐雾、冰冻)频发,对风电机组的安全稳定运行带来极大的挑战。传统的事故演化和风险预警方法对事故运行数据有较强的依赖性,而极端天气条件下难以提供足够的可靠数据进行分析,这大大增加对风电机组进行监测预警的难度。针对上述问题,本文利用专家知识和风电机组事故的文本数据,提出一种基于知识图谱的风电机组事故演化分析模型。该模型首先借助深度学习算法进行知识抽取,以实现对于事故风险的感知,随后进行知识图谱的构建以及事故的演化推理。本文以雷电和高温灾害为例,借助所提模型进行风电机组事故演化推理和事故风险的定性推理分析。研究结果表明:该模型能够实现极端天气条件下的风电机组事故信息的有效感知,并进行事故的演化推理,可为风电机组的风险管理提供一个有效的工具。The increasing frequency of extreme weather events(including strong winds,lightning,salt fog and freezing conditions)poses significant challenges to the safe and stable operation of wind turbines.The traditional accident evolution and risk early-warning methods have strong dependence on the accident operation data,and it is difficult to provide sufficient reliable data for analysis under extreme weather conditions,which greatly increases the difficulty of monitoring and early-warning of wind turbines.In view of the above problems,an analysis model of wind turbine accident evolution based on knowledge graph was proposed by using the expert knowledge and text data of wind turbine accidents.The knowledge extraction was carried out by means of deep learning algorithm to realize the perception of accident risk,and then the construction of knowledge graph and the evolutionary reasoning of accident were conducted.Taking the lightning and high temperature disasters as examples,the proposed model was used to carry out the qualitative reasoning analysis of wind turbine accident evolution reasoning and accident risk.The results show that the model can realize the effective perception of wind turbine accident information under extreme weather conditions,and carry out the evolution reasoning of accident,which can provide an effective tool for the risk management of wind turbines.

关 键 词：极端天气条件 风电机组 深度学习 知识图谱 事故演化 

分 类 号：X913[环境科学与工程—安全科学]