基于CSO-LSSVM的复杂气象条件下污区等级评估方法
|
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。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。试验结果表明,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的评估结果正确率较高,具有一定的实际应用意义。
Pollution area level is an important index for the evaluation of insulators and safety performance of the transmission grid. To study the relationship between the meteorological data and pollution area level in the region, this paper set up a scheme for assessment of pollution area level based on meteorological data. Using existing meteorological data for Yunnan Province as well as the data on distribution of pollution area level, this paper proposed a pollution area level assessment method based on CSO-LSSVM. Firstly, it sorted out and analyzed existing meteorological data. Then, the data model was leaned and evaluated in the LSSVM algorithm. In view of the difficulty in determining LSSVM parameters, CSO algorithm was introduced to optimize LSSVM parameters. Experimental results showed that, compared with traditional BP neural network evaluation model, the evaluation model for meteorological data and pollution area level assessment constructed by CSO-LSSVM algorithm had a higher accuracy and a certain value for practical application.
Pollution area level is an important index for the evaluation of insulators and safety performance of the transmission grid. To study the relationship between the meteorological data and pollution area level in the region, this paper set up a scheme for assessment of pollution area level based on meteorological data. Using existing meteorological data for Yunnan Province as well as the data on distribution of pollution area level, this paper proposed a pollution area level assessment method based on CSO-LSSVM. Firstly, it sorted out and analyzed existing meteorological data. Then, the data model was leaned and evaluated in the LSSVM algorithm. In view of the difficulty in determining LSSVM parameters, CSO algorithm was introduced to optimize LSSVM parameters. Experimental results showed that, compared with traditional BP neural network evaluation model, the evaluation model for meteorological data and pollution area level assessment constructed by CSO-LSSVM algorithm had a higher accuracy and a certain value for practical application.
引文
[1] 红伟,曹彬,王耿耿,等.绝缘子表面局部等值盐密测量方法[J].电网技术,2016,40(4):1289-1294.
[2] SHUAI H,GONG Q,WU J.Insulators ESDD predicting under various meteorological conditions based on least squares support vector machines[C]// Asia-Pacific Power and Energy Engineering Conference.IEEE,2011:1-4.
[3] REN A,LI Q,XIAO H.Influence analysis and prediction of ESDD and NSDD based on random forests[J].Energies,2017,10(7):878.
[4] 蒋兴良,赵世华,毕茂强,等.污秽绝缘子闪络特性与泄漏电流特性研究[J].中国电机工程学报,2013,31(3):220-226;27.
[5] 张晋,汲胜昌,欧小波,等.基于泄漏电流基波阻性分量的绝缘子污秽度表示方法的研究[J].高压电器,2010,46(6):29-33;39.
[6] CHRZAN K L.Leakage currents on naturally contaminated porcelain and silicone insulators[J].IEEE Transactions on Power Delivery,2012,25(2):904-910.
[7] 谭国梁,王思华.基于熵权模糊神经网络的绝缘子污闪状态评估[J].电瓷避雷器,2015,12(4):5-9.
[8] M'HAMDI B,TEGUAR M,MEKHALDI A.Optimal design of corona ring on HV composite insulator using PSO approach with dynamic population size[J].IEEE Transactions on Dielectrics & Electrical Insulation,2016,23(2):1048-1057.
[9] 熊宇,陈孝明,阮羚,等.卫星AOD数据驱动的区域电力污秽等级评估方法[J].测绘通报,2016,6(8):39-43.
[10] 何洪英,姚建刚,蒋正龙,等.基于支持向量机的高压绝缘子污秽等级红外热像检测[J].电力系统自动化,2005,29(24):70-74;82.
[11] 吴军,刘飞.基于最小二乘支持向量机的复杂气象条件下绝缘子等值附盐密度的预测[J].陕西电力,2011,39(2):9-12.
[12] 何相佑,向凤红,忽建蕊.基于BP人工神经网络的绝缘子等值附盐密度预测[J].绝缘材料,2008,41(4):58-61.
[13] 李哲,姜昀芃,卢明,等.基于AMPSO和GRNN的绝缘子污秽度预测模型[J].广东电力,2017,30(7):76-82.
[14] 白国振,俞洁皓.基于改进模糊神经网络的PID参数自整定[J].计算机应用研究,2016,33(11):3358-3363;3368.
[15] 史丽萍,祝艳华,朱宁坦.应用最小二乘支持向量机在线预测绝缘子等值附盐密度[J].高压电器,2013,49(4):82-85.
[16] 周绍磊,廖剑,史贤俊.RBF-SVM的核参数选择方法及其在故障诊断中的应用[J].电子测量与仪器学报,2014,28(3):240-246.
[17] 王玲玲,王昕,郑益慧,等.计及多个风电机组出力相关性的配电网无功优化[J].电网技术,2017,41(11):3463-3469.
[18] DL/T 374—2010,电力系统污区分布图绘制方法[S].
[2] SHUAI H,GONG Q,WU J.Insulators ESDD predicting under various meteorological conditions based on least squares support vector machines[C]// Asia-Pacific Power and Energy Engineering Conference.IEEE,2011:1-4.
[3] REN A,LI Q,XIAO H.Influence analysis and prediction of ESDD and NSDD based on random forests[J].Energies,2017,10(7):878.
[4] 蒋兴良,赵世华,毕茂强,等.污秽绝缘子闪络特性与泄漏电流特性研究[J].中国电机工程学报,2013,31(3):220-226;27.
[5] 张晋,汲胜昌,欧小波,等.基于泄漏电流基波阻性分量的绝缘子污秽度表示方法的研究[J].高压电器,2010,46(6):29-33;39.
[6] CHRZAN K L.Leakage currents on naturally contaminated porcelain and silicone insulators[J].IEEE Transactions on Power Delivery,2012,25(2):904-910.
[7] 谭国梁,王思华.基于熵权模糊神经网络的绝缘子污闪状态评估[J].电瓷避雷器,2015,12(4):5-9.
[8] M'HAMDI B,TEGUAR M,MEKHALDI A.Optimal design of corona ring on HV composite insulator using PSO approach with dynamic population size[J].IEEE Transactions on Dielectrics & Electrical Insulation,2016,23(2):1048-1057.
[9] 熊宇,陈孝明,阮羚,等.卫星AOD数据驱动的区域电力污秽等级评估方法[J].测绘通报,2016,6(8):39-43.
[10] 何洪英,姚建刚,蒋正龙,等.基于支持向量机的高压绝缘子污秽等级红外热像检测[J].电力系统自动化,2005,29(24):70-74;82.
[11] 吴军,刘飞.基于最小二乘支持向量机的复杂气象条件下绝缘子等值附盐密度的预测[J].陕西电力,2011,39(2):9-12.
[12] 何相佑,向凤红,忽建蕊.基于BP人工神经网络的绝缘子等值附盐密度预测[J].绝缘材料,2008,41(4):58-61.
[13] 李哲,姜昀芃,卢明,等.基于AMPSO和GRNN的绝缘子污秽度预测模型[J].广东电力,2017,30(7):76-82.
[14] 白国振,俞洁皓.基于改进模糊神经网络的PID参数自整定[J].计算机应用研究,2016,33(11):3358-3363;3368.
[15] 史丽萍,祝艳华,朱宁坦.应用最小二乘支持向量机在线预测绝缘子等值附盐密度[J].高压电器,2013,49(4):82-85.
[16] 周绍磊,廖剑,史贤俊.RBF-SVM的核参数选择方法及其在故障诊断中的应用[J].电子测量与仪器学报,2014,28(3):240-246.
[17] 王玲玲,王昕,郑益慧,等.计及多个风电机组出力相关性的配电网无功优化[J].电网技术,2017,41(11):3463-3469.
[18] DL/T 374—2010,电力系统污区分布图绘制方法[S].