基于GA-BP算法的输电线路弧垂预测模型  

Transmission line sag prediction model based on GA-BP algorithm

作　　者：籍海亮 夏林 迟长春[2] JI Hailiang;XIA Lin;CHI Changchun(East China Power Transmission and Transformation Engineering Co.,Ltd.,Shanghai 200335,China;School of Electrical Engineering,Shanghai Dianji University,Shanghai 201306,China)

机构地区：[1]华东送变电工程有限公司,上海200335 [2]上海电机学院电气学院,上海201306

出　　处：《黑龙江电力》2023年第3期189-192,共4页Heilongjiang Electric Power

基　　金：上海电机学院校企合作产教融合项目(项目编号:E3-22-6302-2-238,Z-22B0187)。

摘　　要：针对实时获取输电线路的弧垂工作量大、操作繁琐的问题,构建GA-BP神经网络模型对弧垂进行计算。将风速、辐照度、导线温度和导线载流量、弧垂作为训练神经网络的参数,并比较了GA-BP模型与传统BP模型之间的差异。仿真结果表明,GA-BP模型的误差率小于2%,相比BP模型,其收敛性能较好、计算精度较高。In response to the large workload and tedious operation to obtain the arc sag of transmission lines in real time,a GA-BP neural network model is constructed to calculate the sag.Wind speed,irradiance,conductor temperature and conductor load capacity,and sag were used as parameters for training the neural network,and the differences between the GA-BP model and the conventional BP model were compared.The simulation results show that the error rate of the GA-BP model is less than 2%,which has better convergence performance and higher computational accuracy compared with the BP model.

关 键 词：架空输电线路 导线弧垂 神经网络 遗传算法 

分 类 号：TM752[电气工程—电力系统及自动化]