摘要
针对目前输电线路覆冰预测误差大、效率低等问题,提出了一种基于优化SVRM的输电线路覆冰厚度预测模型.首先运用主成分分析法(PCA)提取出影响线路覆冰的主要特征,其次利用粒子群算法(PSO)对支持向量回归机(SVRM)中的主要参数进行迭代最优化,同时搭建输电线路覆冰预测模型.最后结合四川电力勘探设计院某观冰站自动监测系统采集的现场2 234组覆冰气象数据集,对所提出的预测模型进行训练与测试,验证预测模型的有效性与实用性.与未优化及同类型预测方法相比,其预测平均均方误差分别减少了约28%、21%、3%,预测准确度有一定的提高.
Aiming at the problems of large error and low efficiency of transmission line icing prediction,a prediction model based on optimized SVRM is proposed.Firstly,principal component analysis(PCA)is used to extract the main features affecting line icing.Secondly,particle swarm optimization(PSO)is used to iteratively optimize the main parameters of support vector regression machine(SVRM),and the icing prediction model is established at the same time.Finally,combined with the 2234 sets of field icing data set collected by the automatic monitoring system of an ice observation station of Sichuan Electric Power Exploration and Design Institute,the effectiveness of the prediction model is verified.Compared with the non optimized and the same type of prediction methods,the average mean square error of prediction is reduced by about 28%,21%and 3%respectively,and the prediction accuracy is improved to a certain extent.
作者
汤伟
桑旬
刘家兵
武健
TANG Wei;SANG Xun;LIU Jia-bing;WU Jian(School of Electrical and Control Engineering,Shaanxi University of Science&Technology,Xi′an 710021,China;Xi′an Jinpower Electric Co.,Ltd.,Xi′an 710075,China)
出处
《陕西科技大学学报》
北大核心
2023年第4期151-157,共7页
Journal of Shaanxi University of Science & Technology
基金
国家自然科学基金项目(62073206)
陕西省科技厅科学技术研究发展计划项目(2018GY-031)。
关键词
输电线路
自动监测
覆冰预测
主成分分析
支持向量回归机
power transmission line
automatic monitoring
icing prediction
principal component analysis
support vector regression machine
