摘要
污区等级是评价绝缘子以及输电网安全性能的重要指标,为了研究所处地区的气象数据与污区等级的关系,建立基于气象数据的污区等级评估方案,本文利用现有云南省气象数据及污区等级分布数据,提出了基于CSO-LSSVM的污区等级评估方法,首先对现有气象数据进行整理分析,采用了LSSVM算法对数据模型进行学习和评估,针对LSSVM参数确定较为困难的问题,引入了CSO算法对LSSVM的参数进行寻优,实验结果表明,相较于传统BP神经网络的评估模型,CSO-LSSVM算法所构建的气象数据与污区等级评估模型评估结果正确率较高,具有一定的实际应用意义。
The pollution degree is an important index to evaluate the safety performance of insulators and the transmission grid.To research the relationship between the meteorological data and the pollution degree of an area,establish the evaluation scheme of pollution degree based on meteorological data.Using the existing meteorological data of Yunnan Province and the distributed data of the polluted area,this paper proposed an appraisal procedure of the pollution degree based on CSO-LSSVM.First of all,analyze the existing meteorological data,and then evaluate the data model by LSSVM algorithm.Aiming at the relatively di cult problem of determining the LSSVM parameter,introduce the CSO algorithm to optimize the parameters of LSSVM.The experimental results show that the accuracy of the model evaluation results on meteorological data and pollution degree by CSO-LSSVM algorithm is higher,compared with the traditional BP neural network evaluation model,which has certain practical application of signi cance.
作者
黄绪勇
沈志
王昕
Huang Xuyong;Shen Zhi;Wang Xin(Yunnan Power Grid Company Limited,Electric Power Research Institute,Kunming 100080,China;Center of Electrical&Electronic Technology,Shanghai Jiao Tong University,Shanghai 200240,China)
出处
《云南电力技术》
2019年第6期75-78,共4页
Yunnan Electric Power
关键词
污区等级评估
复杂气象条件
数据分析
最小二乘支持向量机
鸡群优化算法
assessment of pollution degree
complex weather conditions
data analysis
Least Squares support vector machine
chicken swarm optimization algorithm
