摘要
为了提高输电线路覆冰厚度预测精度,利用灰色关联分析确定覆冰影响因素对输电线路覆冰增长量的影响权重,采用PSO算法对LSSVM的参数优化,建立了考虑灰色关联权重的PSO-LSSVM输电线路覆冰厚度预测模型。采用实际运行线路的覆冰增长数据进行仿真分析,并与其他覆冰预测模型对比,考虑灰色关联权重的PSO-LSSVM输电线路等值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误差分别为0.575、3.124%和4.015%,均小于其他三种预测模型,验证了模型的正确性和实用性。
In order to improve the prediction accuracy of transmission line icing thickness,the influence weight of icing influencing factors on transmission line icing growth is determined by grey correlation analysis.The parameters of LSSVM are optimized by PSO algorithm,and the PSO-LSSVM transmission line icing thickness prediction model considering grey correlation weight is established.The ice growth data of actual operation lines are used for simulation analysis and compared with other icing prediction models.The root mean square error,average relative error and global maximum error of PSO-LSSVM transmission line equivalent icing thickness prediction model considering grey correlation weight are 0.575%,3.124%and 4.015%respectively,which are less than the other three prediction models,The correctness and practicability of the model are verified.
作者
郭开春
王波
GUO Kaichun;WANG Bo(College of Electrical and New Energy,Three Gorges University,Hubei Yichang 443000,China;School of Electrical and Automation,Wuhan University,Wuhan 430072,China)
出处
《电工材料》
CAS
2022年第1期15-19,24,共6页
Electrical Engineering Materials
基金
国家自然科学基金资助项目(61876097,51777142)。
关键词
灰色关联权重
粒子群优化算法
最小二乘支持向量机
输电线路
覆冰厚度
grey correlation weight
particle swarm optimization algorithm
least squares support vector machine
transmission line
icing thickness
