摘要
针对电力金属设施在土壤中的腐蚀预测问题,分析现有腐蚀预测方法的不足,考虑金属腐蚀影响因素,研究提出了一种采用改进粒子群优化LSSVM的金属腐蚀速率预测方法。在传统粒子群算法中引入收缩因子,以控制粒子速度,增强粒子的搜索能力,从而解决粒子群早熟问题。采用实验数据进行仿真分析,改进PSO-LSSVM预测模型的平均相对误差仅为2.24%,与其他几种方法相比,改进粒子群优化的LSSVM算法具有更高的预测精度。
In view of the corrosion prediction of the electric metal facilities in the soil,the deficiencies of the existing corrosion prediction methods are analyzed,and the factors affecting the metal corrosion are considered,and a prediction method of metal corrosion rate using improved particle swarm optimizing LSSVM is proposed.The shrinkage factor is introduced into the traditional particle swarm optimization algorithm to control particle velocity and enhance search ability,so as to solve the premature problem of particle swarm optimization.The average relative error of the improved PSO-LSSVM prediction model is only 2.24%.Compared with several other methods,the improved particle swarm optimizing LSSVM algorithm has higher prediction accuracy than other methods.
引文
[1]耿进锋,时洪飞.变电站接地网腐蚀与防护技术进展[J].腐蚀与防护,2009,30(08):523-525.GENG Jin-feng,SHI Hong-fei.Corrosion and Protection of Grounding Grid for Substation[J].Corrosion&Protection,2009,30(08):523-525.
[2]祝郦伟,潘晨,钱洲亥,等.输电线路杆塔接地装置的腐蚀与防护[J].全面腐蚀控制,2015,29(04):44-47+64.ZHU Li-wei,PAN Chen,QIAN Zhou-hai,et al.Corrosion and Prevention of the Grounding Device of Power Transition Lines[J].Total Corrosion Control,2015,29(04):44-47+64.
[3]左旭,高波.500kV输电线路拉线塔拉线棒防护措施[J].吉林电力,2010,38(01):50-51+56.ZUO Xu,GAO Bo.Guyed Poles Protection Measurements of Guyed Tower in Transmission Lines[J].Jilin Electric Power,2010,38(01):50-51+56.
[4]The Substation Committee of the IEEE Power Engineering Society.IEEE Guide for Safety in AC Substation Grounding:IEEE Std 80-2000[S].New York:The Institute of Electrical and Electronics Engineers,Inc,2000.
[5]DAWALIBI F P,SELBL A.Electromagnetic Fields of Energized Conductors[J].IEEE Transactions on Power Deliver,1993,08(03):1275-1284.
[6]徐松,杨漫兮,刘凯,等.500kV变电站接地网腐蚀分析及防护措施[J].湖南电力,2017,37(z2):71-75.XU Song,YANG Man-xi,LIU Kai,et al.Analysis on Corrosion Failure of 500kV Substation Grounding Grid[J].Hunan Electric Power,2017,37(z2):71-75.
[7]李丽,李晓刚,邢士波,等.BP人工神经网络对国内典型地区碳钢土壤腐蚀的预测研究[J].腐蚀科学与防护技术,2013,25(05):372-376.LI Li,LI Xiao-gang,XING Shi-bo,et al.Research on Soil Corrosion Rate Prediction of Carbon Steel in Typical Chinese Cities Based on BP Artificial Neural Network[J].Corrosion Science and Protection Technology,2013,25(05):372-376.
[8]花广如,李文浩,郭阳阳.基于神经网络模型的海南变电站接地网Q235钢腐蚀率预测[J].腐蚀与防护,2017,38(08):573-577+588.HUA Guang-ru,LI Wen-hao,GUO Yang-yang.Corrosion Rate Prediction of Q235Steel in Hainan Substation Grounding Grid Based on Neural Network Models[J].Corrosion&Protection,2017,38(08):573-577+588.
[9]张公永,李伟.基于灰色最小二乘支持向量机的变压器油溶解气体预测[J].电力学报,2012,27(02):111-115.ZHANG Gong-yong,LI Wei.Study on Transformer Oil Dissolved Gas Prediction Based on Gray Least Square Support Vector Machine[J].Journal of Electric Power,2012,27(02):111-115.
[10]VAPNIK V N.The Nature of Statistical Learning Theory[M].New York:Springer,1995.
[11]SUYKENS J A K,VANDEWALLE J.Least Squares Support Vector Machine Classifiers[J].Neural Processing Letters,1999,09(03):293-300.
[12]汪庆年,饶利强,龚文军,等.基于PSO优化的BP神经网络在电动机绝缘剩余寿命预测中的应用[J].水电能源科学,2015,33(12):161-164.WANG Qing-nian,RAO Li-qiang,GONG Wen-jun,et al.Application of BP Neural Network Based on PSO in Motor Insulation Residual Life Prediction[J].Water Resources and Power,2015,33(12):161-164.
[13]CLERC M,KENNEDY J.The Particle Swarm-explosion,Stability,and Convergence in a Multidimensional Complex Space[J].IEEE Transations on Evolutionary Computation,2002,06(01):58-73.
[14]郭阳阳.基于神经网络的海南变电站土壤对Q235钢的腐蚀预测研究[D].北京:华北电力大学,2016.GUO Yang-yang.Study on Corrosion Prediction of Q235Steel from Soil in Hainan Substation Based on Artificial Neural Network[D].Beijing:North China Electric Power University,2016.