长江科学院院报 ›› 2019, Vol. 36 ›› Issue (3): 26-30.DOI: 10.11988/ckyyb.20170992

• 水资源与环境 • 上一篇    下一篇

基于BP神经网络的MIKE SHE模型参数率定

郭怡, 吴鑫淼, 郄志红, 冉彦立   

  1. 河北农业大学 城乡建设学院,河北 保定 071000
  • 收稿日期:2017-08-29 出版日期:2019-03-01 发布日期:2019-03-20
  • 通讯作者: 吴鑫淼(1972-),女,河北保定人,教授,博士,硕士生导师,主要从事节水灌溉和工程结构研究。E-mail:wuxinmiao001@163.com
  • 作者简介:郭 怡(1990-),女,河北保定人,硕士研究生,主要从事水资源规划管理与水环境研究。E-mail:15028201075@163.com

Parameter Calibration of MIKE SHE Model Based on BP Neural Network

GUO Yi, WU Xin-miao, QIE Zhi-hong, RAN Yan-li   

  1. College of Urban and Rural Construction, Agricultural University of Hebei, Baoding 071000, China
  • Received:2017-08-29 Online:2019-03-01 Published:2019-03-20

摘要: 为了更精细地对水文全过程进行描述和解析,更准确地构建分布式水文模型,以丹麦Karup流域为例,对MIKE SHE模型的饱和导水率、饱和带水平水力传导系数、河床透水系数进行了参数率定,模拟流域的日径流过程。结果表明:基于BP神经网络反分析的参数率定方法比MIKE SHE模型参数自动率定计算得到的均方根误差RMSE小,模型效率系数Ens更接近1;采用BP神经网络反演率定参数后,3组测试样本的日径流模拟过程的RMSE分别为0.04,0.03,0.08 m3/s,Ens均为0.99,且模拟结果能较好地反映径流的实际变化趋势。因此,这种基于BP神经网络反分析的参数率定方法对构建分布式水文模型具有一定的价值。

关键词: 径流模拟, 参数率定, MIKE SHE模型, BP神经网络, 反分析, 均匀设计, 丹麦Karup流域

Abstract: In order to describe and interpret hydrological processes in more detail, and at the same time to construct a more accurate distributed hydrological model, we took the Karup watershed in Denmark as an example and calibrated three parameters of MIKE SHE model, namely, saturated hydraulic conductivity, saturated horizontal hydraulic conductivity, and leakage coefficient of river bank, and simulated the daily runoff process in the watershed. Results demonstrate that the root mean square error (RMSE) obtained by the method of parameter calibration based on BP neural network is smaller than that by automatic parameter calibration in MIKE SHE model, with the model efficiency coefficient Ens closer to 1. Having been treated by parameter calibration by BP neural network, the values of RMSE of daily runoff of three test samples are 0.04 m3/s, 0.03 m3/s, and 0.08 m3/s, respectively, and the value of Ens is 0.99. As the simulated runoff displays a trend in agreement with the real runoff, the back analysis method of parameter calibration based on BP neural network is of certain value in runoff simulation.

Key words: runoff simulation, parameter calibration, MIKE SHE model, BP neural network, back analysis, uniform design, Karup watershed in Denmark

中图分类号: