基于改进相空间重构原理的支持向量机月径流模拟
|
摘要
基于交叉验证支持向量机(CV-SVM)原理及方法,构建以相空间重构理论与支持向量机相结合的径流时间序列模拟模型。针对相空间重构中确定延迟时间τ和嵌入维数m的方法众多,且各方法确定结果不尽相同等缺点,本文采用试凑的方法,在延迟时间τ和嵌入维数m取值范围为2～10内依次构建81个基于相空间重构理论的CV-SVM月径流模拟模型,以南利河董湖站月径流模拟为例进行分析,确定最佳延迟时间τ和嵌入维数m,并与自相关函数法等相关延迟时间τ和嵌入维数m的确定方法确定结果进行比较,旨在探寻延迟时间τ和嵌入维数m对月径流模拟精度的影响及其规律,为基于时间序列的水文模拟及预测预报提供方法和参考。
Based on the CV-SVM principle and methods,the paper constructed the runoff time series model by combining phase space reconstruction theory with support vector machine.According to the many methods of phase space reconstruction to determine the delay time τ and embedding dimension m,and the shortcomings of determining different results by every method,this paper adopted the method of trial and error,the delay time τ and embedding dimension m value range from 2 to 10 in order to construct 81 monthly runoff simulation model based on the theory of phase space reconstruction CV-SVM.Taking monthly runoff simulation in Nanli River Donghu Lake station for example,it determined the optimal delay time τ and embedding dimension m and compared the results with that determined by autocorrelation function.The aim is to explore the influence of delay time τ and embedding dimension m of monthly runoff simulation on accuracy and its regularity,and provide method and reference for the hydrological simulation and forecast of time series.
Based on the CV-SVM principle and methods,the paper constructed the runoff time series model by combining phase space reconstruction theory with support vector machine.According to the many methods of phase space reconstruction to determine the delay time τ and embedding dimension m,and the shortcomings of determining different results by every method,this paper adopted the method of trial and error,the delay time τ and embedding dimension m value range from 2 to 10 in order to construct 81 monthly runoff simulation model based on the theory of phase space reconstruction CV-SVM.Taking monthly runoff simulation in Nanli River Donghu Lake station for example,it determined the optimal delay time τ and embedding dimension m and compared the results with that determined by autocorrelation function.The aim is to explore the influence of delay time τ and embedding dimension m of monthly runoff simulation on accuracy and its regularity,and provide method and reference for the hydrological simulation and forecast of time series.
引文
[1]谷晓平,王长耀,王汶,等.应用于水文预报的优化BP神经网络研究[J].生态环境,2004,13(4):524-527.
[2]庞博,郭生练,熊立华,等.改进的人工神经网络水文预报模型及应用[J].武汉大学学报(工学版),2007,40(1):33-36+41.
[3]蓝永超,康尔泗,徐中民,等.BP神经网络在径流长期预测中的应用[J].中国沙漠,2001,21(1):97-100.
[4]尹晔,梁川.改进的BP网络模型及其在日径流预测中的应用[J].云南水力发电,2005,21(3):14-17.
[5]邓霞,董晓华,薄会娟.基于BP网络的河道径流预报方法与应用[J].人民长江,2010,41(2):56-59.
[6]于国荣,夏自强.混沌时间序列支持向量机模型及其在径流预测应用[J].水科学进展,2008,19(1):116-122.
[7]陈南祥,黄强,曹连海,等.径流序列的相空间重构神经网络预测模型[J].河海大学学报(自然科学版),2005,33(5):490-493.
[8]战国隆,马孝义,刘继龙,等.泾河月径流量的混沌特征识别及预测[J].水利水电科技进展,2010,30(1):7-9(增刊).
[9]Kantz H,Schreiber T.Nonlinear time series analysis[M].Cambridge:Cambridge University Press,1997.
[10]Fraser A M,Swinney H L.Independent coordinates forstrange attractors from time series[J].Physical ReviewA:1986,33:1134-1140.
[11]Grassberger P,Procaccia I.Measuring the strangeness ofstrange attractors[J].Physica D:Nonliner Phenomena,1983,9:189-208.
[12]Kennel M B,Brown R,Abarbanel H D I.Determining em-bedding dimension for phase-space reconstruction usinga geometrical construction[J].Physical Review A.,1992,45:3403-3411.
[13]Kugiurmtzis D.State space reconstruction parameters inthe analysis of chaotic times series-the role of the timewindow length[J].Physica D:Nonlinear Phenomena,1996,95:13-28.
[14]Kim H S,Eykholt R,Salas J D.Nonlinear dynamics,delaytimes and embedding windows[J].Physica D:NonlinearPhenomena,1999,127:48-60.
[15]Vladimir N Vapnik.统计学习理论的本质[M].张学工译.北京:清华大学出版社,2000.
[16]田景文,高美娟.人工神经网络算法研究及应用[M].北京:北京理工大学出版社,2006.7.
[17]邵东国,刘丙军,阳书敏,等.水资源繁杂系统理论[M].北京:科学出版社,2012.6.
[18]吴学文.考虑生态的多目标水电站水库混沌优化调度研究[M].北京:中国水利水电出版社,2012.12.
[19]MATLAB中文论坛.MATLAB神经网络30个案例分析[M].北京:北京航空航天大学出版社,2010.4.
[20]史峰,王辉,等.MATLAB智能算法30个案例分析[M].北京:北京航空航天大学出版社,2011.7.
[21]中华人民共和国水利部.SL250-2000,水文情报预报规范[S].北京:中国水利水电出版社,2000.
[2]庞博,郭生练,熊立华,等.改进的人工神经网络水文预报模型及应用[J].武汉大学学报(工学版),2007,40(1):33-36+41.
[3]蓝永超,康尔泗,徐中民,等.BP神经网络在径流长期预测中的应用[J].中国沙漠,2001,21(1):97-100.
[4]尹晔,梁川.改进的BP网络模型及其在日径流预测中的应用[J].云南水力发电,2005,21(3):14-17.
[5]邓霞,董晓华,薄会娟.基于BP网络的河道径流预报方法与应用[J].人民长江,2010,41(2):56-59.
[6]于国荣,夏自强.混沌时间序列支持向量机模型及其在径流预测应用[J].水科学进展,2008,19(1):116-122.
[7]陈南祥,黄强,曹连海,等.径流序列的相空间重构神经网络预测模型[J].河海大学学报(自然科学版),2005,33(5):490-493.
[8]战国隆,马孝义,刘继龙,等.泾河月径流量的混沌特征识别及预测[J].水利水电科技进展,2010,30(1):7-9(增刊).
[9]Kantz H,Schreiber T.Nonlinear time series analysis[M].Cambridge:Cambridge University Press,1997.
[10]Fraser A M,Swinney H L.Independent coordinates forstrange attractors from time series[J].Physical ReviewA:1986,33:1134-1140.
[11]Grassberger P,Procaccia I.Measuring the strangeness ofstrange attractors[J].Physica D:Nonliner Phenomena,1983,9:189-208.
[12]Kennel M B,Brown R,Abarbanel H D I.Determining em-bedding dimension for phase-space reconstruction usinga geometrical construction[J].Physical Review A.,1992,45:3403-3411.
[13]Kugiurmtzis D.State space reconstruction parameters inthe analysis of chaotic times series-the role of the timewindow length[J].Physica D:Nonlinear Phenomena,1996,95:13-28.
[14]Kim H S,Eykholt R,Salas J D.Nonlinear dynamics,delaytimes and embedding windows[J].Physica D:NonlinearPhenomena,1999,127:48-60.
[15]Vladimir N Vapnik.统计学习理论的本质[M].张学工译.北京:清华大学出版社,2000.
[16]田景文,高美娟.人工神经网络算法研究及应用[M].北京:北京理工大学出版社,2006.7.
[17]邵东国,刘丙军,阳书敏,等.水资源繁杂系统理论[M].北京:科学出版社,2012.6.
[18]吴学文.考虑生态的多目标水电站水库混沌优化调度研究[M].北京:中国水利水电出版社,2012.12.
[19]MATLAB中文论坛.MATLAB神经网络30个案例分析[M].北京:北京航空航天大学出版社,2010.4.
[20]史峰,王辉,等.MATLAB智能算法30个案例分析[M].北京:北京航空航天大学出版社,2011.7.
[21]中华人民共和国水利部.SL250-2000,水文情报预报规范[S].北京:中国水利水电出版社,2000.