基于随机风场概率加权的台风灾害下输电线路损毁预警

为解决台风灾害下输电线路损毁预测精度和数据利用率低的问题,建立了基于随机风场概率加权的输电线路损毁概率预测混合模型。该模型综合考虑了气象、微地形、杆塔运行和实时损毁等空间多源异构信息的融合。首先,使用极值Ⅰ型概率分布进行阵风的风值模拟,通过柯尔莫哥洛夫-斯米尔诺夫假设检验证明了其具有较高的精度;其次,利用蒙特卡洛法实现了随机风场的概率生成;最后,利用随机森林法预测每个风场下的线路损毁概率,实现了基于随机风场概率加权的输电线路损毁概率预测,该预测结果可为电力部门提供有效的预警决策支持。以台风"山竹"为实例,验证了所提基于随机风场概率加权的输电线路损毁概率预测混合模型的科学性及有效性。

Early Warning for Transmission Line Damage Under Typhoon Disaster Based on Random Wind Field Probability Weighting

To address the problem of low prediction accuracy and data utilization rate of transmission line damage under typhoon disaster,a hybrid model for the probability prediction of transmission line damage based on random wind field probability weighting is built.This model comprehensively considers the fusion of spatial multi-source heterogeneous information such as weather,microtopography,tower operation,and real-time damage.Firstly,the Gumbel distribution is used to simulate the wind value of gusts,and Kolmogorov-Smirnov test proves its high accuracy.Secondly,the Monte Carlo method is used to realize the probability generation of random wind fields.Finally,the random forest method is used to predict the probability of line damage in each wind field,realizing the probability prediction of transmission line damage based on random wind field probability weighting.The prediction result can provide effective early warning decision support for the sectors of power industry.An example of the typhoon"Mangkhut"verifies the scientific nature and validity of the proposed hybrid model for the probability prediction of transmission line damage based on random wind field probability weighting.