摘要
针对现有输电线路覆冰厚度预测方法上存在的不足,采用狼群算法(wolf pack algorithm,WPA)对最小二乘支持向量机(least squares support vector machine,LSSVM)的惩罚因子和核函数参数进行优化,建立基于WPA优化LSSVM的输电线路覆冰厚度预测模型。采用某500 kV输电线路实际覆冰数据进行算例分析,结果表明,WPA-LSSVM覆冰厚度预测模型的均方根误差、平均相对误差和全局最大相对误差分别为0.634、2.61%和3.27%,优于其他覆冰厚度预测模型,验证了该模型的正确性和实用性。
In order to overcome the shortcomings of existing methods for ice thickness prediction of transmission lines,the penalty factor and kernel function parameters of least squares support vector machine(LSSVM)are optimized by wolf colony algorithm(WPA).The prediction model of ice thickness of transmission lines based on WPA optimized LSSVM is established.The actual icing data of a 500 kV transmission line are used to analyze the calculation example.The results show that the root mean square error,average relative error and global maximum relative error of WPA-LSSVM icing prediction model are 0.634,2.61%and 3.27%respectively,which are better than other icing prediction models and verify the correctness and practicability of the model.
作者
汪晗
李启迪
黄治翰
刘闯
WANG Han;LI Qidi;HUANG Zhihan;LIU Chuang(State Grid Ezhou Power Supply Company,Ezhou,Hubei 436000,China;State Grid Jingmen Power Supply Company,Jingmen,Hubei 448000,China)
出处
《东北电力技术》
2022年第2期42-46,共5页
Northeast Electric Power Technology
关键词
输电线路
覆冰厚度
狼群算法
最小二乘支持向量机
transmission line
icing thickness
wolf pack algorithm
least squares support vector machine
