基于径流分类的流域降雨—径流过程动态神经网络建模

doi:10.13249/j.cnki.sgs.2012.01.74

地理科学 ›› 2012, Vol. 32 ›› Issue (1): 74-80.doi: 10.13249/j.cnki.sgs.2012.01.74

基于径流分类的流域降雨—径流过程动态神经网络建模

邵月红¹(), 林柄章¹, 刘永和²

1.南京信息工程大学应用水文气象研究院, 江苏南京 210044
2.河南理工大学资源环境学院, 河南焦作 454000

收稿日期:2011-01-17 修回日期:2011-06-02 出版日期:2012-01-20 发布日期:2012-01-20
作者简介:
作者简介: 邵月红（1977-）,女,山西侯马人,博士,主要从事水文、水资源、GIS与RS的研究。E-mail: syh@nuist.edu.cn
基金资助:
水利部公益性行业专项（201001047）、校科研基金项目（20100402）、国家青年科学基金项目（41105074）资助

Rainfall-runoff Simulation Based on Runoff Classification Using Dynamic Artificial Neural Networks

Yue-hong SHAO¹(), Bing-zhang LIN¹, Yong-he LIU²

1. Applied Hydrometeorological Research Institute, Nanjing University of Information Science and Technology, Nanjing, Jiangsu 210044, China
2. Institute of Resources and Environment, Henan Porytechnic university, Jiaozuo, Henan 454000, China

Received:2011-01-17 Revised:2011-06-02 Online:2012-01-20 Published:2012-01-20

摘要/Abstract

摘要：

利用聚类分析,将径流序列分为不同类型的子径流序列,对这些子序列建立神经网络模型,采用Elman动态神经网络对沂沭河流域上游临沂子流域日径流量进行预测分析,通过与不加分类的总体神经网络的模拟结果进行对比分析。确定性系数、相关系数、平均相对误差和平均相对均方根误差4个统计指数及流域径流过程线和次洪误差分析结果都表明：Elman动态神经网络能够对日径流量进行较好模拟,但基于径流分类的降雨—径流模型表现出更优良性能,能较大程度提高径流模拟精度。

关键词: 径流分类, 聚类分析, Elman神经网络, 降雨—径流, 局部神经网络, 总体神经网络

Abstract:

A runoff sequence is classified into several sub-runoff sequences with cluster analysis and practical application, and local dynamic neural network for each sub-runoff sequence is performed separately. An Elman recurrent neural network model (ENN) is constructed and applied for the daily runoff forecast in the Linyi sub-catchment of the upper Yishu river basin in this paper. In order to further evaluate the performance of local Elman neural network (LENN), global Elman neural network (GENN) is applied as a comparison at the same time in the study region. Based on analysis indexes such as Nash-Sutcliffe coefficient, correlation coefficient, mean relative error and root mean relative square error, the results of daily runoff and flooding processes are attained. It suggested that Elman recurrent neural network model has a high accuracy of forecast on the rainfall-runoff dynamic process and the maximum peak flow and peak occurrence time, whether it is configuration samples or evaluation samples. However, local Elman neural network based on runoff classification is more suitable and efficient to daily runoff forecasting than global Elman neural network, especially during the period of arid season and semi-humid season. The simulation precision of different periods decreases in the order of arid season, semi-humid season and humid season. Elman recurrent neural network model is feasible to effectively simulate the hydrological dynamic characteristic of the daily runoff and to reflect the complex nonlinear runoff regular pattern in the basin as a promising and efficient method. In order to further evaluate superiority of this method, longer series of data, more regions and hydrology model are need to study and analysis because of the highly nonlinear, spatial and temporal heterogeneity and dynamic uncertainty of rainfall-runoff process.

Key words: runoff classification, cluster analysis, Elman neural network, rainfall-runoff, local neural network, global neural network

中图分类号:

S273

邵月红, 林柄章, 刘永和. 基于径流分类的流域降雨—径流过程动态神经网络建模[J]. 地理科学, 2012, 32(1): 74-80.

Yue-hong SHAO, Bing-zhang LIN, Yong-he LIU. Rainfall-runoff Simulation Based on Runoff Classification Using Dynamic Artificial Neural Networks[J]. SCIENTIA GEOGRAPHICA SINICA, 2012, 32(1): 74-80.

图/表 7

图1

表1

图2

表2

图3

图4

表3

参考文献 16

[1]	Ju Q,Yu Z B,Hao Z C,et al.Hydrologic Simulations with Artificial Neural Networks[C]//Angela Burgess. Third International Conference on Natural Computation. Los Alamitos:IEEE Computer Society Publications,2007,2:22-27.
[2]	王芳,卓莉,黎夏,等.基于高光谱特征选择和RBFNN的城市植被胁迫程度监测[J].地理科学,2008,28(1):77~82.
[3]	Carcano E C,Bartolini P,Muselli M,et al.Jordan recurrent neural network versus IHACRES in modeling daily stream flows[J].Journal of Hydrology,2008,362(3-4):291-307.
[4]	Kin C L,Ball J E,Sharma A.An Application of Artificial Neural Networks for Rainfall Forecasting[J].Mathematical and Computer Modelling,2001,33(6):883-699.
[5]	苏安玉,李衡,濮励杰,等.基于RAGA-BP神经网络模型的三江平原地下水资源预测研究[J].地理科学,2009,29(2):283~287.
[6]	王钧,蒙吉军.黑河流域近60年来径流量变化及影响因素[J].地理科学,2008,28(1):83~88.
[7]	Tang W M.The study of the optimal structure of BP neural network[J].Systems Engineering Theory and Practice,2005,25(10):95-100.
[8]	Yu S W,Zhu K J,Diao F Q.A dynamic all parameters adaptive BP neural networks model and its application on oil reservoir prediction[J].Applied Mathematics and Computation,2008,195(1): 66-75.
[9]	张翀,李晶,任志远.基于Hopfield 神经网络的中国近40年气候要素时空变化分析[J].地理科学,2011,31(2):211~217.
[10]	Srinivasn B,Prasad U R,Rao N J.Back propagation through adjoins for the identification of nonlinear dynamic systems using recurrent neural models[J].IEEE Transactions on Neural Networks,1994,5(2):213-227.
[11]	Tommy W S,Chow Y F.A recurrent neural network based real time learning control strategy applying to nonlinear system with unknown dynamics[J].IEEE Transactions on Industrial Electronics,1998,45(1):151-161.
[12]	王玲,黄国如.基于径流分类的日径流量预测神经网络模型[J].灌溉排水,2002,21(4):45~48.
[13]	Elman J L.Finding structure in time[J].Cognitive Science,1990,14(2):179-211.
[14]	Shi X H,Liang Y C,Lee H P.Improved Elman networks and applications for controlling ultrasonic motors[J].Applied Artificial Intelligence,2004,18(7):603-629.
[15]	飞思科技产品研发中心.神经网络理论与MATLAB7实现[M]. 北京: 电子工业出版社, 2005.
[16]	Nash J E,Sutcliffe J V,River flow forecasting through conceptual model: a discussion of principles[J]. Journal of Hydrology,1970,10(3): 282-290.

组别	第1组	第2组	第3组
时间	11、12、1、2月	3、4、5、10月	6、7、8、9月
季节	枯水期	半湿润期	丰水期
代号	LENN1	LENN2	LENN3

样本	分组	LENN				GENN
样本	分组	DC	R	W_abs	R_MSE	DC	R	W_abs	R_MSE
训练样本	枯水期	0.962	0.981	0.104	0.208	0.447	0.670	0.498	0.885
	半湿润期	0.941	0.970	0.310	0.597	0.860	0.930	0.517	0.931
	丰水期	0.865	0.930	0.340	0.670	0.841	0.917	0.360	0.730
	全年	0.886	0.941	0.309	0.874	0.857	0.926	0.396	0.979
检验样本	枯水期	0.856	0.925	0.161	0.261	0.481	0.693	0.348	0.524
	半湿润期	0.719	0.845	0.261	0.441	0.175	0.418	0.555	0.985
	丰水期	0.781	0.884	0.363	0.549	0.757	0.870	0.373	0.583
	全年	0.820	0.905	0.332	0.972	0.792	0.890	0.393	1.050

样本	次洪编号	Q_max(m³/s)	LENN			GENN
样本	次洪编号	Q_max(m³/s)	模拟Q_max(m³/s)	误差 (%)	T_peak误差 (d)	模拟Q_max(m³/s)	误差 (%)	T_peak误差(d)
训练样本	2001-7-31	2030.0	1504.4	-25.89	+1	1369.8	-32.52	+1
	2002-7-25	138.0	134.4	-2.61	0	143.2	3.77	0
	2003-9-7	1150.0	1044.2	-9.20	0	934.9	-18.70	+1
	2004-8-5	1200.0	800.1	-33.33	+1	773.9	-35.51	+1
	2005-9-21	3760.0	3496.6	-7.01	0	3364.1	-10.53	0
文本检验	2006-8-29	827.0	755.7	-8.62	+1	1155.8	39.76	+1
	2007-8-19	1334.6	858.8	-35.65	0	1021.3	-23.48	0
	2008-7-24	1814.4	1776.6	-2.08	0	1547.0	-14.74	0

基于径流分类的流域降雨—径流过程动态神经网络建模

Rainfall-runoff Simulation Based on Runoff Classification Using Dynamic Artificial Neural Networks

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 7

参考文献 16

相关文章 14

Metrics

本文评价

推荐阅读 0

[1]	祁添垚, 张强, 王月, 肖名忠, 刘剑宇, 孙鹏. 1960~2005年中国蒸发皿蒸发量变化趋势及其影响因素分析[J]. 地理科学, 2015, 35(12): 1599-1606.
[2]	王富喜, 毛爱华, 李赫龙, 贾明璐. 基于熵值法的山东省城镇化质量测度及空间差异分析[J]. 地理科学, 2013, 33(11): 1323-1329.
[3]	周在明, 张光辉, 王金哲, 严明疆. 环渤海低平原微咸水区土壤盐渍化与盐分剖面特征[J]. 地理科学, 2011, 31(8): 929-934.
[4]	方修琦, 张学珍, 戴玉娟, 李蓓蓓, 侯光良. 1951~2005年中国大陆冬季温度变化过程的区域差异[J]. 地理科学, 2010, 30(4): 571-576.
[5]	欧素英. 珠江三角洲咸潮活动的空间差异性分析[J]. 地理科学, 2009, 29(1): 89-92.
[6]	李雪铭, 李明. 基于体现人自我实现需要的中国主要城市人居环境评价分析[J]. 地理科学, 2008, 28(6): 742-747.
[7]	罗传秀, 郑卓, 潘安定, 安放舟, BEAUDOUIN Celia, 黄康有. 新疆地区表土孢粉空间分布规律研究[J]. 地理科学, 2008, 28(2): 272-275.
[8]	张耀光, 张岩, 王宁. 长山群岛海洋农牧化海域环境分区研究[J]. 地理科学, 2007, 27(6): 768-773.
[9]	史春云, 张捷, 尤海梅, 单勇兵, 章锦河. 中国城市居民出游潜力的空间分异格局[J]. 地理科学, 2006, 26(5): 622-628.
[10]	苏勤. 旅游者类型及其体验质量研究——以周庄为例[J]. 地理科学, 2004, 24(4): 506-511.
[11]	王秀红. 多元统计分析在分区研究中的应用[J]. 地理科学, 2003, 23(1): 66-71.
[12]	宝音, 包玉海, 海山, 银山. 北方农牧林交错带土地利用区的分析与评价——以科尔沁右翼前旗为例[J]. 地理科学, 1996, 16(4): 377-384.
[13]	王俊, 姜建祥, 杨彬, 王宁. 吉林省河流水质分析与评价[J]. 地理科学, 1994, 14(2): 165-171,200.
[14]	刘幼慈, 应龙根. 环境质量区域分异模式的建立——以上海市川沙县为例[J]. 地理科学, 1991, 11(3): 253-260.