IFS-LSSVM及其在时延序列预测中的应用
|
摘要
针对最小二乘支持向量机预测模型中最优参数难以确定的问题,提出一种基于改进的自由搜索算法确定最小二乘支持向量机最优参数的方法(IFS-LSSVM)。对标准自由搜索算法进行改进,使之可应用于最小二乘支持向量机的参数优化,改进之后的算法具有更好的优化性能。将具有时间序列性质的网络时延作为预测对象,利用本文的IFS-LSVM算法进行预测。在仿真中与遗传算法优化的最小二乘支持向量机(GA-LSSVM)、粒子群优化算法优化的最小二乘支持向量机(PSOLSSVM)、标准最小二乘支持向量机工具箱中的网格搜索算法(Grid-LSSVM)进行了对比。仿真对比结果表明本文的方法具有更高的预测精度与更小的预测误差。
It is difficult to determine the optimal parameters of least squares support vector machine prediction model,so a prediction method based on improved free search algorithm( IFS-LSSVM) was proposed to determine the optimal parameters of least squares support vector machines. First,the standard free search algorithm was improved so that it can be applied to the parameter optimization of least squares support vector machines,the improved harmony search algorithm has better optimization performance.Then the least squares support vector machines was applied to predict the time-delay series of the network based on improved free search optimization algorithm. Finally,time-delay series was used as prediction simulation object,genetic algorithm optimized least squares support vector machines( GA-LSSVM),particle swarm optimization algorithm optimized least squares support vector machines( PSO-LSSVM),standard grid search method of least squares support vector machines( Grid-LSSVM) toolbox were compared. Simulation comparison results show that the proposed method has higher prediction accuracy and smaller prediction error.
It is difficult to determine the optimal parameters of least squares support vector machine prediction model,so a prediction method based on improved free search algorithm( IFS-LSSVM) was proposed to determine the optimal parameters of least squares support vector machines. First,the standard free search algorithm was improved so that it can be applied to the parameter optimization of least squares support vector machines,the improved harmony search algorithm has better optimization performance.Then the least squares support vector machines was applied to predict the time-delay series of the network based on improved free search optimization algorithm. Finally,time-delay series was used as prediction simulation object,genetic algorithm optimized least squares support vector machines( GA-LSSVM),particle swarm optimization algorithm optimized least squares support vector machines( PSO-LSSVM),standard grid search method of least squares support vector machines( Grid-LSSVM) toolbox were compared. Simulation comparison results show that the proposed method has higher prediction accuracy and smaller prediction error.
引文
[1]DING S F,HUA X P,YU Z J.An overview on nonparallel hyperplane support vector machine algorithms[J].Neural Computing and Applications,2013,25(5):975-982.
[2]SUYKENS J A K,VANDEWALLE J.Least squares support vector machines classifiers[J].Neural Processing Letters,1999,9(3):293-300.
[3]田中大,高宪文,李琨.基于EMD与LS-SVM的网络控制系统时延预测方法[J].电子学报,2014,42(5):868-874.TIAN Zhongda,GAO Xianwen,LI Kun.Time-delay prediction method of networked control system based on EMD and LS-SVM[J].Acta Electronica Sinica,2014,42(5):868-874.
[4]田中大,高宪文,史美华,等.资源受限网络控制系统的模糊反馈调度[J].电机与控制学报,2013,17(1):94-101.TIAN Zhongda,GAO Xianwen,SHI Meihua,et al.Fuzzy feedback scheduling of resource constrained networked control system[J].Electric Machines and Control,2013,17(1):94-101.
[5]HU T J,HUANG X X,TAN Q.Time delay prediction for space teleoperation based on non-Gaussian auto-regressive model[C]//2012 Proceedings of International Conference on Modelling,Identification&Control(ICMIC),June 24-26,2012,Wuhan,China.2012:567-572.
[6]时为国,邵成,孙正阳.基于AR模型时延预测的改进GPC网络控制算法[J].控制与决策,2012,27(3):477-480.SHI Weiguo,SHAO Cheng,SUN Zhengyang.Improved GPC network-control algorithm based on AR model time-delay prediction[J].Control and Decision,2012,27(3):477-480.
[7]YANG M,RU J,LI X R,et al.Predicting Internet end-to-end delay:a multiple-model approach[C]//24th Annual Joint Conference of the IEEE Computer and Communications Societies,March 13-17,2005,Miami,USA.2005,4:2815-2819.
[8]TABIB S R S,JALALI A A.Modelling and prediction of internet time-delay by feed-forward multi-layer perceptron neural network[C]//Second UKSIM European Symposium on Computer Modeling and Simulation,Sept 8-10,2008,Liverpool,England.2008:611-616.
[9]RAHMANI B,MARKAZI A H D,MOZAYANI N.Real time prediction of time delays in a networked control system[C]//Proceedings of the 3rd International Symposium on Communications,Control and Signal Processing,March 12-14,2008,St Julians,Malta.2008:1242-1245.
[10]SADEGHZADEH N,AFSHAR A,MENHAJ M B.An MLP neural network for time delay prediction in networked control systems[C]//Control and Decision Conference,July 2-4,2008,Yantai,China.2008:5314-5318.
[11]田中大,高宪文,李琨.基于KPCA与LSSVM的网络控制系统时延预测方法[J].系统工程与电子技术,2013,35(6):1281-1285.TIAN Zhongda,GAO Xianwen,LI Kun.Networked control system timne-delay prediction method based on KPCA and LSSVM[J].Systems Engineering and Electronics,2013,35(6):1281-1285.
[12]李春茂,肖建,张玥.基于LS-SVM的网络化控制系统自适应预测控制[J].系统仿真学报,2007,19(155):3494-3498.LI Chunmao,XIAO Jian,ZHANG Yue.Approach of adaptive prediction control on networked control systems based on leastsquares support vector machines[J].Journal of System Simulation,2007,19(155):3494-3498.
[13]杨晓冬,王崇林,史丽萍.H桥逆变器IGBT开路故障诊断方法研究[J].电机与控制学报,2014,18(5):112-118.YANG Xiaodong,WANG Chonglin,SHI Liping.Study of IGBT open-circuit fault diagnosis method for H-bridge inverter[J].Electric Machines and Control,2014,18(5):112-118.
[14]龚瑞昆,宁荣飞,陈磊,等.改进DOAS技术在混合气体中的定量分析[J].哈尔滨理工大学学报,2012,17(6):110GONG Ruikun,NING Rongfei,CHEN Lei,et al.Improvment of DOAS technology in the application of mixed gas quantitative analysis[J].Journal of Harbin University of Science and Technology,2012,17(6):110-113.
[15]YANG K H,ZHAO L L.A new optimizing parameter approach of LSSVM multiclass classification model[J].Neural Computing and Applications,2012,21(5):945-955.
[16]KALIN P,GUY L.Free search a comparative Analysis[J].Information Science,2005,172:173-193.
[17]张弦,王宏力.嵌入维数自适应最小二乘支持向量机状态时间序列预测方法[J].航空学报,2010,31(12):2309-2313.ZHANG Xian,WANG Hongli.Condition time series prediction using least squares support vector machine with adaptive embedding dimension[J].Acta Ateronautica Astronautica Sinica,2010,31(12):2309-2313.
[18]尚万峰,赵升吨,申亚京.遗传优化的最小二乘支持向量机在开关磁阻电机建模中的应用[J].中国电机工程学报,2009,29(12):65-69.SHANG Wanfeng,ZHAO Shengdun,SHEN Yajing.Application of LSSVM optimized by genetic algorithm to modeling of switched reluctance motor[J].Proceedings of the CSEE,2009,29(12):65-69.
[2]SUYKENS J A K,VANDEWALLE J.Least squares support vector machines classifiers[J].Neural Processing Letters,1999,9(3):293-300.
[3]田中大,高宪文,李琨.基于EMD与LS-SVM的网络控制系统时延预测方法[J].电子学报,2014,42(5):868-874.TIAN Zhongda,GAO Xianwen,LI Kun.Time-delay prediction method of networked control system based on EMD and LS-SVM[J].Acta Electronica Sinica,2014,42(5):868-874.
[4]田中大,高宪文,史美华,等.资源受限网络控制系统的模糊反馈调度[J].电机与控制学报,2013,17(1):94-101.TIAN Zhongda,GAO Xianwen,SHI Meihua,et al.Fuzzy feedback scheduling of resource constrained networked control system[J].Electric Machines and Control,2013,17(1):94-101.
[5]HU T J,HUANG X X,TAN Q.Time delay prediction for space teleoperation based on non-Gaussian auto-regressive model[C]//2012 Proceedings of International Conference on Modelling,Identification&Control(ICMIC),June 24-26,2012,Wuhan,China.2012:567-572.
[6]时为国,邵成,孙正阳.基于AR模型时延预测的改进GPC网络控制算法[J].控制与决策,2012,27(3):477-480.SHI Weiguo,SHAO Cheng,SUN Zhengyang.Improved GPC network-control algorithm based on AR model time-delay prediction[J].Control and Decision,2012,27(3):477-480.
[7]YANG M,RU J,LI X R,et al.Predicting Internet end-to-end delay:a multiple-model approach[C]//24th Annual Joint Conference of the IEEE Computer and Communications Societies,March 13-17,2005,Miami,USA.2005,4:2815-2819.
[8]TABIB S R S,JALALI A A.Modelling and prediction of internet time-delay by feed-forward multi-layer perceptron neural network[C]//Second UKSIM European Symposium on Computer Modeling and Simulation,Sept 8-10,2008,Liverpool,England.2008:611-616.
[9]RAHMANI B,MARKAZI A H D,MOZAYANI N.Real time prediction of time delays in a networked control system[C]//Proceedings of the 3rd International Symposium on Communications,Control and Signal Processing,March 12-14,2008,St Julians,Malta.2008:1242-1245.
[10]SADEGHZADEH N,AFSHAR A,MENHAJ M B.An MLP neural network for time delay prediction in networked control systems[C]//Control and Decision Conference,July 2-4,2008,Yantai,China.2008:5314-5318.
[11]田中大,高宪文,李琨.基于KPCA与LSSVM的网络控制系统时延预测方法[J].系统工程与电子技术,2013,35(6):1281-1285.TIAN Zhongda,GAO Xianwen,LI Kun.Networked control system timne-delay prediction method based on KPCA and LSSVM[J].Systems Engineering and Electronics,2013,35(6):1281-1285.
[12]李春茂,肖建,张玥.基于LS-SVM的网络化控制系统自适应预测控制[J].系统仿真学报,2007,19(155):3494-3498.LI Chunmao,XIAO Jian,ZHANG Yue.Approach of adaptive prediction control on networked control systems based on leastsquares support vector machines[J].Journal of System Simulation,2007,19(155):3494-3498.
[13]杨晓冬,王崇林,史丽萍.H桥逆变器IGBT开路故障诊断方法研究[J].电机与控制学报,2014,18(5):112-118.YANG Xiaodong,WANG Chonglin,SHI Liping.Study of IGBT open-circuit fault diagnosis method for H-bridge inverter[J].Electric Machines and Control,2014,18(5):112-118.
[14]龚瑞昆,宁荣飞,陈磊,等.改进DOAS技术在混合气体中的定量分析[J].哈尔滨理工大学学报,2012,17(6):110GONG Ruikun,NING Rongfei,CHEN Lei,et al.Improvment of DOAS technology in the application of mixed gas quantitative analysis[J].Journal of Harbin University of Science and Technology,2012,17(6):110-113.
[15]YANG K H,ZHAO L L.A new optimizing parameter approach of LSSVM multiclass classification model[J].Neural Computing and Applications,2012,21(5):945-955.
[16]KALIN P,GUY L.Free search a comparative Analysis[J].Information Science,2005,172:173-193.
[17]张弦,王宏力.嵌入维数自适应最小二乘支持向量机状态时间序列预测方法[J].航空学报,2010,31(12):2309-2313.ZHANG Xian,WANG Hongli.Condition time series prediction using least squares support vector machine with adaptive embedding dimension[J].Acta Ateronautica Astronautica Sinica,2010,31(12):2309-2313.
[18]尚万峰,赵升吨,申亚京.遗传优化的最小二乘支持向量机在开关磁阻电机建模中的应用[J].中国电机工程学报,2009,29(12):65-69.SHANG Wanfeng,ZHAO Shengdun,SHEN Yajing.Application of LSSVM optimized by genetic algorithm to modeling of switched reluctance motor[J].Proceedings of the CSEE,2009,29(12):65-69.