基于数据挖掘的输电线路覆冰舞动预警方法  

Data Mining Based Early Warning Method of Transmission Line Icing Galloping

作　　者：刘善峰 卢明 王超 王津宇 孟高军[2] LIU Shanfeng;LU Ming;WANG Chao;WANG Jinyu;MENG Gaojun(Electric Power Research Institute,Henan Electric Power Company,Zhengzhou He’nan 450052,China;School of Electric Power Engineering,Nanjing Institute of Technology,Nanjing Jiangsu 211167,China)

机构地区：[1]国网河南省电力公司电力科学研究院,河南郑州450052 [2]南京工程学院电力工程学院,江苏南京211167

出　　处：《电子器件》2024年第3期764-771,共8页Chinese Journal of Electron Devices

基　　金：江苏省高等学校自然科学研究项目(20KJB480006)。

摘　　要：近年来输电线路舞动事故频发,严重威胁着电网的安全运行。为了达到提前预警、及早防范的目的,降低输电线路舞动对电网安全运行的影响,提出了一种基于数据挖掘的输电线路覆冰舞动预警方法,首先建立了输电线路覆冰增长模型,利用BP神经网络分析导线覆冰的各气象因素所占权重,并在气象数值预报结果与地形信息的基础上,采用Bayes-Adaboost方法对输电线路的舞动概率进行建模,实现输电线路的舞动概率预警,最终,依据河南省电网历史气象资料进行了验证,证明了模型的有效性。In recent years,transmission line galloping accidents occur frequently,seriously threatening the safe operation of the power grid.In order to achieve the purpose of early warning and early prevention,and reduce the impact of transmission line galloping on the safe operation of the power grid,a transmission line galloping early warning method is proposed based on data mining.Firstly,the transmission line icing growth model is established,and the weight of each meteorological factor of conductor icing is analyzed by using BP neural network.Based on the meteorological numerical prediction results and topographic information,Bayes-Adaboost method is used to model the galloping probability of transmission lines and realize the galloping probability early warning of transmission lines.Finally,the validity of the model is verified according to the historical meteorological data of Henan power grid.

关 键 词：输电线路 舞动 数据挖掘 气象 预警 

分 类 号：TM863[电气工程—高电压与绝缘技术]