[1]李小兰,曾献奎,王栋,等.水文模拟替代模型方法对比探究[J].江苏水利,2022,(08):57-61.
　LI Xiaolan,ZENG Xiankui,WANG Dong,et al.Comparative study of surrogate model methods for hydrological simulation[J].JIANGSU WATER RESOURCES,2022,(08):57-61.
点击复制

水文模拟替代模型方法对比探究()
分享到：

《江苏水利》[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2022年08期
页码:
57-61
栏目:
水文化
出版日期:
2022-09-01

文章信息/Info

Title:
Comparative study of surrogate model methods for hydrological simulation
文章编号:
1007-7839(2022)08-0057-0005
作者:
李小兰; 曾献奎; 王栋; 吴吉春
（南京大学地球科学与工程学院，江苏 南京 210023）
Author(s):
LI Xiaolan;  ZENG Xiankui;  WANG Dong;  WU Jichun
（School of Earth Science and Engineering, Nanjing University, Nanjing 210023, China）
关键词:
水文模型; 稀疏网格; 神经网络
Keywords:
hydrological model;  sparse grid;  neural network
分类号:
TV11
文献标志码:
B
摘要:
传统的水文模型参数识别需要多次调用水文模型，从而导致严重的计算负荷。替代模型具有和原始水文模型几乎相同的模拟精度且运行时间可以忽略不计，从而有效解决参数识别中的计算负荷问题。以长江流域上游水文模拟为案例分析，系统对比了当前几种常用的替代模型方法，如稀疏网格方法、Elman-NN方法、RBF-NN方法，为大尺度水文模拟的替代模型构建提供一定的参考依据。
Abstract:
Traditional parameter identification of hydrological models usually needs to call the hydrological model many times, which leads to the problem of computational load. The surrogate model has almost the same simulation accuracy as the original model, and the running time can be ignored. Therefore, the surrogate model can solve the problem of excessive calculation load. This study takes the hydrological simulation in the upper reaches of the Yangtze River Basin as a case study, and compares several commonly systematically used surrogate model methods, such as sparse grid method, Elman-NN method and RBF-NN method, which can provide a certain reference basis for the construction of surrogate models for large-scale hydrological simulation.

参考文献/References:

[1]PARK D，MARKUS M. Analysis of a changing hydrologicflood regime using the Variable Infiltration Capacity model[J]. Journal of Hydrology，2014（515）：267-280.
[2]BECKER A. Criteria for a hydrologically sound structuring of large scale land surface process models[M]//O[′]Kane JP. Advances in Theoretical Hydrology. Amsterdam：Elsevier，1992：97-111.
[3]COUSQUER Y，PRYET A，ATTEIA O，et al. Developing a particle tracking surrogate model to improve inversion of ground water Surface water models[J]. JOURNAL OF HYDROLOGY，2018（558）：356-365.
[4]HU Y，GARCIA-CABREJO O，CAI X，et al. Global sensitivity analysis for large-scale socio-hydrological models using Hadoop[J]. Environmental Modelling & Software，2015（73）：231-243.
[5]TSOUKALAS I，MAKROPOULOS C. Multi objective optimisation on a budget：Exploring surrogate modelling forrobust multi-reservoir rules generation under hydrologicaluncertainty[J]. Environmental Modelling & Software，2015(69):396-413.
[6]LIANG X，LETTENMAIER D P，WOOD E F，et al. A simple hydrologically based model of land surface water and energy fluxes for general circulation models[J]. Journal of Geophysical Research-Atmospheres，1994，99（D7）：14415-14428.
[7]SMOLYAK S. Quadrature and Interpolation Formulas forTensor Products of Certain Classes of Functions[J]. DokladyAkademiinauk SSSR，1963，4（5）：240-243.
[8]高鑫宇，曾献奎，吴吉春. 基于改进稀疏网格替代模拟的地下水DNAPLs运移不确定性分析[J]. 水文地质工程地质, 2020,47(1):1-10.
[9]ELMAN J L. Finding Structure in time[J]. CognitiveScience，1990，14（2）：179-211.

备注/Memo

备注/Memo:
收稿日期：2022-03-16
基金项目：江苏省水利科技项目（2020038）；国家重点研发计划（2016YFC0402802）
作者简介：李小兰（1998—），女，硕士研究生，主要从事水文学与水资源研究。E-mail:mg1929057@smail.nju.edu.cn
通信作者：曾献奎（1985—），男，副教授，博士，主要从事地下水数值模拟研究。E-mail: xiankuizeng@nju.edu.cn
更新日期/Last Update: 2022-08-15