混合布谷鸟算法在高压断路器故障诊断上的应用
|
摘要
为了更准确、快速地对高压断路器故障进行分类、诊断,提出一种基于混合布谷鸟算法优化的最小二乘支持向量机(LSSVM)的故障诊断方法。首先提取分合闸线圈的时间和电流特征量得到特征向量,再利用模拟退火算法(SA)与布谷鸟算法(CS)结合形成的混合布谷鸟算法(CS-SA),对支持向量机进行寻优,旨在得到具有最优参数支持向量机分类模型,提高诊断结果的准确性。最后,利用收集到的数据对该算法进行诊断验证,结果表明利用混合布谷鸟算法优化后的LS-SVM得到的分类模型比常用的粒子群算法、遗传算法、标准布谷鸟算法优化得到的模型准确率更高。
A diagnostic method based on least squares support vector machine(SVM)optimized by Hybrid cuckoo algorithm is proposed to accurately and quickly classification,diagnose faults in high voltage circuit breakers. First,the eigenvectors are obtained by extracting the time and current characteristic of the tripping and closing coil. Second,optimizing the support vector machine(SVM)use Hybrid cuckoo algorithm(CS-SA)which synthetized by Simulated annealing algorithm(SA)and cuckoo algorithm(CS)to get the support vector machine classification model with optimal parameter and improve the accuracy of diagnostic results. Finally,using the collected data to verify this algorithm,and,the experimentation shows that the accuracy rate of classification model obtained by LS-SVM optimized by CS-SA will be higher than optimized by particle swarm optimization,genetic algorithm or normal cuckoo algorithm.
A diagnostic method based on least squares support vector machine(SVM)optimized by Hybrid cuckoo algorithm is proposed to accurately and quickly classification,diagnose faults in high voltage circuit breakers. First,the eigenvectors are obtained by extracting the time and current characteristic of the tripping and closing coil. Second,optimizing the support vector machine(SVM)use Hybrid cuckoo algorithm(CS-SA)which synthetized by Simulated annealing algorithm(SA)and cuckoo algorithm(CS)to get the support vector machine classification model with optimal parameter and improve the accuracy of diagnostic results. Finally,using the collected data to verify this algorithm,and,the experimentation shows that the accuracy rate of classification model obtained by LS-SVM optimized by CS-SA will be higher than optimized by particle swarm optimization,genetic algorithm or normal cuckoo algorithm.
引文
[1]王志华,郭明叙.断路器本体非全相保护装置故障分析及防范措施[J].南方电网技术,2013,7(2):124-126.WANG Zhihua,GUO Mingxu.The fault analysis of circuit breaker open-phase protection device and related countermeasures[J].Southern Power System Technology,2013,7(2):124-126.
[2]艾绍贵,高峰,黄永宁,等.断路器型零损耗故障限流装置的研制及试验[J].南方电网技术,2015,9(4):103-108.AI Shaogui,GAO Feng,HUANG Yongning,et al.Development and experiment of breaker-type zero-loss fault current limiter[J].Southern Power System Technology,2015,9(4):103-108.
[3]张卫正,李永丽,姚创.基于最小二乘支持向量机的高压断路器故障诊断[J].高压电器,2015,51(12):79-83.ZHANG Weizheng,LI Yongli,YAO Chuang.Fault diagnosis for high voltage circuit breaker based on least square support vector machine[J].High Voltage Apparatus,2015,51(12):79-83.
[4]胡晓光,孙来军,纪延超.基于线圈电流和触点状态的断路器故障分析[J].电力自动化设备,2006,26(8):5-7.HU Xiaoguang,SUN Laijun,JI Yanchao.Circuit breakerfault analysis based on coil currents and contact states[J].Electric Power Automation Equipment,2006,26(8):5-7.
[5]宋锦刚,许长青,朱统亮.基于波形辨识技术的SF6断路器分/合闸线圈电流监测[J].电力建设,2011,32(3):65-68.SONG Jingang,XU Changqing,ZHU Tongliang.Coilcurrent monitoring of SF6circuit breakers based on wave-form identification[J].Electric Power Construction,2011,32(3):65-68.
[6]罗小安,黄时细,程勇.基于正交算法的RBF网络在高压断路器故障诊断中的应用研究[J].电力科学与工程,2008,24(3):13-15.LUO Xiao’an,HUANG Shixi,CHENG Yong.Study on high-voltage circuit breaker fault diagnosis based on RBFnetwork of orthogonal learning algorithm[J].Electric Power Science and Engineering,2008,24(3):13-15.
[7]薛浩然,张珂珩,李斌,等.基于布谷鸟算法和支持向量机的变压器故障诊断[J].电力系统保护与控制,2015,43(8):8-13.XUE Haoran,ZHANG Keheng,LI Bin,et al.Fault diagnosis of transformer based on the cuckoo search and support vector machine[J].Power System Protection and Control,2015,43(8):8-13.
[8]兰少峰,刘升.布谷鸟搜索算法研究综述[J].计算机工程与设计,2015,36(4):1063-1067.LAN Shaofeng,LIU Sheng.Overview of research on Cuckoo search algorithm[J].Computer Engineering and Design,2015,36(4):1063-1067.
[9]高叶军,连志刚,曹宇.基于改进布谷鸟算法的火电厂机组组合优化[J].电气自动化,2015,37(4):64-66.GAO Yejun,LIAN Zhigang,CAO Yu.Optimization of unit commitment of thermal power plant based on improved cuckoo algorithm[J].Electrical Automation,2015,37(4):64-66.
[10]刘瑞瑞.基于遗传模拟退火算法的三维离线装箱优化问题研究[D].吉林:吉林大学,2014.LIU Ruirui.Research on three-dimensional offline bin packing problem on genetic simulated annealing algorithm[D].Jilin:Jilin University,2014.
[11]陆平静,李宝,易任娇,等.一种基于改进模拟退火算法的程序性能优化参数搜索算法[J].计算机工程与科学,2015,37(7):1227-1232.LU Pingjing,LI Bao,YI Renjiao,et al.An improved simulated annealing algorithm for program optimization parameters search[J].Computer Engineering and Science,2015,37(7):1227-1232.
[12]郑巧燕,莫愿斌,刘付永,等.一种小规模多种群布谷鸟算法[J].计算机应用与软件,2014(10):278-280.ZHENG Qiaoyan,MO Yuanbin,LIU Fuyong,et al.Asmall-scale and multi-population cuckoo search[J].Computer Applications and Software,2014(10):278-280.
[13]刘颖.基于半监督集成支持向量机的土地覆盖遥感分类方法研究[D].北京:中国科学院研究生院(东北地理与农业生态研究所),2013.LIU Ying.The study of semisupervised ensembledsupport vector[D].Beijing:Graduate University of Chi-nese Academy of Sciences(Northeast Institute of Geog-raphy and Agroecology,Chinese Academy of Sciences),2013.
[14]白志艳.高压断路器故障诊断与状态监测研究[D].北京:华中科技大学,2013.BAI Zhiyan.Research of fault diagnosis and condi-tion monitoring for high-voltage circuit breaker[D].Bei-jing:Huazhong University of Science and Technology,2013.
[15]孙银山,张文涛,张一茗,等.高压断路器分合闸线圈电流信号特征提取与故障判别方法研究[J].高压电器,2015,51(9):134-139.SUN Yinshan,ZHANG Wentao,ZHANG Yiming,et al.Research on feature value extraction and fault recogni-tion of coil current signal in high-voltage circuit breaker[J].High Voltage Apparatus,2015,51(9):134-139.
[16]严信.基于云模型的模糊支持向量机分类方法研究[D].太原:太原理工大学,2013.YAN Xin.FSVM classification based on cloud model[D].Taiyuan:Taiyuan University of Technology,2013.
[17]彭搏.高压断路器在线监测与故障诊断系统的研究[D].上海:上海交通大学,2013.PENG Bo.Research about the on-lien monitoring and fault diagnosis system of high voltage circuit breaker[D].Shanghai:Shanghai Jiao Tong University,2013.
[18]杨红,孙卓,刘夏平,等.基于多最小二乘支持向量机的桥梁温度挠度效应的分离[J].振动与冲击,2014(1):71-76.YANG Hong,SUN Zhuo,LIU Xiaping,et al.Separation of bridge temperature deflection effect based on MLS-SVM[J].Journal of Vibration and Shock,2014(1):71-76.
[19]戴臻.布谷鸟算法优化支持向量机的网络热点话题预测[J].计算机应用与软件,2014(4):330-333.DAI Zhen.Internet hot topic prediction based on support vector machine optimised by cuckoo search algorithm[J].Computer Applications and Software,2014(4):330-333.
[20]罗小安.基于径向基神经网络的高压断路器故障诊断应用研究[D].成都:西南交通大学,2007.LUO Xiao’an.Study high-voltage circuit breaker fault diagnosis based on RBFNN[D].Chengdu:Southwest Jiaotong University,2007.
[2]艾绍贵,高峰,黄永宁,等.断路器型零损耗故障限流装置的研制及试验[J].南方电网技术,2015,9(4):103-108.AI Shaogui,GAO Feng,HUANG Yongning,et al.Development and experiment of breaker-type zero-loss fault current limiter[J].Southern Power System Technology,2015,9(4):103-108.
[3]张卫正,李永丽,姚创.基于最小二乘支持向量机的高压断路器故障诊断[J].高压电器,2015,51(12):79-83.ZHANG Weizheng,LI Yongli,YAO Chuang.Fault diagnosis for high voltage circuit breaker based on least square support vector machine[J].High Voltage Apparatus,2015,51(12):79-83.
[4]胡晓光,孙来军,纪延超.基于线圈电流和触点状态的断路器故障分析[J].电力自动化设备,2006,26(8):5-7.HU Xiaoguang,SUN Laijun,JI Yanchao.Circuit breakerfault analysis based on coil currents and contact states[J].Electric Power Automation Equipment,2006,26(8):5-7.
[5]宋锦刚,许长青,朱统亮.基于波形辨识技术的SF6断路器分/合闸线圈电流监测[J].电力建设,2011,32(3):65-68.SONG Jingang,XU Changqing,ZHU Tongliang.Coilcurrent monitoring of SF6circuit breakers based on wave-form identification[J].Electric Power Construction,2011,32(3):65-68.
[6]罗小安,黄时细,程勇.基于正交算法的RBF网络在高压断路器故障诊断中的应用研究[J].电力科学与工程,2008,24(3):13-15.LUO Xiao’an,HUANG Shixi,CHENG Yong.Study on high-voltage circuit breaker fault diagnosis based on RBFnetwork of orthogonal learning algorithm[J].Electric Power Science and Engineering,2008,24(3):13-15.
[7]薛浩然,张珂珩,李斌,等.基于布谷鸟算法和支持向量机的变压器故障诊断[J].电力系统保护与控制,2015,43(8):8-13.XUE Haoran,ZHANG Keheng,LI Bin,et al.Fault diagnosis of transformer based on the cuckoo search and support vector machine[J].Power System Protection and Control,2015,43(8):8-13.
[8]兰少峰,刘升.布谷鸟搜索算法研究综述[J].计算机工程与设计,2015,36(4):1063-1067.LAN Shaofeng,LIU Sheng.Overview of research on Cuckoo search algorithm[J].Computer Engineering and Design,2015,36(4):1063-1067.
[9]高叶军,连志刚,曹宇.基于改进布谷鸟算法的火电厂机组组合优化[J].电气自动化,2015,37(4):64-66.GAO Yejun,LIAN Zhigang,CAO Yu.Optimization of unit commitment of thermal power plant based on improved cuckoo algorithm[J].Electrical Automation,2015,37(4):64-66.
[10]刘瑞瑞.基于遗传模拟退火算法的三维离线装箱优化问题研究[D].吉林:吉林大学,2014.LIU Ruirui.Research on three-dimensional offline bin packing problem on genetic simulated annealing algorithm[D].Jilin:Jilin University,2014.
[11]陆平静,李宝,易任娇,等.一种基于改进模拟退火算法的程序性能优化参数搜索算法[J].计算机工程与科学,2015,37(7):1227-1232.LU Pingjing,LI Bao,YI Renjiao,et al.An improved simulated annealing algorithm for program optimization parameters search[J].Computer Engineering and Science,2015,37(7):1227-1232.
[12]郑巧燕,莫愿斌,刘付永,等.一种小规模多种群布谷鸟算法[J].计算机应用与软件,2014(10):278-280.ZHENG Qiaoyan,MO Yuanbin,LIU Fuyong,et al.Asmall-scale and multi-population cuckoo search[J].Computer Applications and Software,2014(10):278-280.
[13]刘颖.基于半监督集成支持向量机的土地覆盖遥感分类方法研究[D].北京:中国科学院研究生院(东北地理与农业生态研究所),2013.LIU Ying.The study of semisupervised ensembledsupport vector[D].Beijing:Graduate University of Chi-nese Academy of Sciences(Northeast Institute of Geog-raphy and Agroecology,Chinese Academy of Sciences),2013.
[14]白志艳.高压断路器故障诊断与状态监测研究[D].北京:华中科技大学,2013.BAI Zhiyan.Research of fault diagnosis and condi-tion monitoring for high-voltage circuit breaker[D].Bei-jing:Huazhong University of Science and Technology,2013.
[15]孙银山,张文涛,张一茗,等.高压断路器分合闸线圈电流信号特征提取与故障判别方法研究[J].高压电器,2015,51(9):134-139.SUN Yinshan,ZHANG Wentao,ZHANG Yiming,et al.Research on feature value extraction and fault recogni-tion of coil current signal in high-voltage circuit breaker[J].High Voltage Apparatus,2015,51(9):134-139.
[16]严信.基于云模型的模糊支持向量机分类方法研究[D].太原:太原理工大学,2013.YAN Xin.FSVM classification based on cloud model[D].Taiyuan:Taiyuan University of Technology,2013.
[17]彭搏.高压断路器在线监测与故障诊断系统的研究[D].上海:上海交通大学,2013.PENG Bo.Research about the on-lien monitoring and fault diagnosis system of high voltage circuit breaker[D].Shanghai:Shanghai Jiao Tong University,2013.
[18]杨红,孙卓,刘夏平,等.基于多最小二乘支持向量机的桥梁温度挠度效应的分离[J].振动与冲击,2014(1):71-76.YANG Hong,SUN Zhuo,LIU Xiaping,et al.Separation of bridge temperature deflection effect based on MLS-SVM[J].Journal of Vibration and Shock,2014(1):71-76.
[19]戴臻.布谷鸟算法优化支持向量机的网络热点话题预测[J].计算机应用与软件,2014(4):330-333.DAI Zhen.Internet hot topic prediction based on support vector machine optimised by cuckoo search algorithm[J].Computer Applications and Software,2014(4):330-333.
[20]罗小安.基于径向基神经网络的高压断路器故障诊断应用研究[D].成都:西南交通大学,2007.LUO Xiao’an.Study high-voltage circuit breaker fault diagnosis based on RBFNN[D].Chengdu:Southwest Jiaotong University,2007.