风暴灾害下电力断线倒塔概率预测  

Probabilistic Forecasting of Transmission Line Damage and Tower Collapse from Storm Disaster

作　　者：徐文军 戴宇 建艳龙 易俊[4] 

机构地区：[1]浙江省丽水电业局,浙江丽水323000 [2]金华东阳市供电局,浙江金华322100 [3]江苏省南京河海南自水电自动化有限公司,江苏南京210032 [4]长沙理工大学电气与信息工程学院,湖南长沙410114

出　　处：《华中电力》2012年第3期77-81,84,共4页Central China Electric Power

摘　　要：提出了基于极端学习机(Extreme Learning Machine,ELM)和广义极值(Generalized Extreme Value,GEV)分布的风暴灾害下输电线路断线概率预测模型。该模型首先针对极端风速,通过ELM网络的训练学习,预测出实时变化的风速;随后从概率的角度考虑实时变化的电力线风荷载极值,提出电力线风荷载的实时广义极值分布;从而实现风暴灾害下输电线路的实时断线概率预测。结合输电线路的历史数据展开算例分析,验证了该预测方法的有效性和准确性。A probability forecasting model of transmission line broken and collapsed from storm disaster based on the Extreme Learning Machine(ELM) and Generalized Extreme Value(GEV) distribution is proposed.Facing the extreme wind speed,the ELM is first used to forecast the variational real-time wind speed.Subsequently,considering the variational real-time wind loads are the extreme probability events,the GEV is then employed to fit the wind loads on both transmission lines and towers.Based on the proposed model,probabilities of transmission line damage and tower collapse can be calculated.The validity and accuracy of the proposed model is verified by comparing the numerical results given by the model to the historical data of a real transmission line.

关 键 词：极端学习机 广义极值分布 断线倒塔概率 输电线路 风暴 

分 类 号：TM75[电气工程—电力系统及自动化]