Improved streamflow forecasting using self-organizing radial basis function artificial neural networks

Hamid Moradkhani, Kuo Lin Hsu, Hoshin Vijai Gupta, Soroosh Sorooshian

Research output: Contribution to journalArticle

141 Citations (Scopus)

Abstract

Streamflow forecasting has always been a challenging task for water resources engineers and managers and a major component of water resources system control. In this study, we explore the applicability of a Self Organizing Radial Basis (SORB) function to one-step ahead forecasting of daily streamflow. SORB uses a Gaussian Radial Basis Function architecture in conjunction with the Self-Organizing Feature Map (SOFM) used in data classification. SORB outperforms the two other ANN algorithms, the well known Multi-layer Feedforward Network (MFN) and Self-Organizing Linear Output map (SOLO) neural network for simulation of daily streamflow in the semi-arid Salt River basin. The applicability of the linear regression model was also investigated and concluded that the regression model is not reliable for this study. To generalize the model and derive a robust parameter set, cross-validation is applied and its outcome is compared with the split sample test. Cross-validation justifies the validity of the nonlinear relationship set up between input and output data.

Original languageEnglish (US)
Pages (from-to)246-262
Number of pages17
JournalJournal of Hydrology
Volume295
Issue number1-4
DOIs
StatePublished - Aug 10 2004

Fingerprint

stream flow
artificial neural network
neural networks
streamflow
water resources
water resource
engineers
control system
managers
river basin
salt
salts
simulation
testing
sampling

Keywords

  • Cross-validation
  • Neural network
  • Radial basis function
  • Self-organizing feature map
  • Streamflow forecasting

ASJC Scopus subject areas

  • Soil Science
  • Earth-Surface Processes

Cite this

Improved streamflow forecasting using self-organizing radial basis function artificial neural networks. / Moradkhani, Hamid; Hsu, Kuo Lin; Gupta, Hoshin Vijai; Sorooshian, Soroosh.

In: Journal of Hydrology, Vol. 295, No. 1-4, 10.08.2004, p. 246-262.

Research output: Contribution to journalArticle

Moradkhani, Hamid ; Hsu, Kuo Lin ; Gupta, Hoshin Vijai ; Sorooshian, Soroosh. / Improved streamflow forecasting using self-organizing radial basis function artificial neural networks. In: Journal of Hydrology. 2004 ; Vol. 295, No. 1-4. pp. 246-262.
@article{8c2ad9010c324df88ccb8a1b3ed26b22,
title = "Improved streamflow forecasting using self-organizing radial basis function artificial neural networks",
abstract = "Streamflow forecasting has always been a challenging task for water resources engineers and managers and a major component of water resources system control. In this study, we explore the applicability of a Self Organizing Radial Basis (SORB) function to one-step ahead forecasting of daily streamflow. SORB uses a Gaussian Radial Basis Function architecture in conjunction with the Self-Organizing Feature Map (SOFM) used in data classification. SORB outperforms the two other ANN algorithms, the well known Multi-layer Feedforward Network (MFN) and Self-Organizing Linear Output map (SOLO) neural network for simulation of daily streamflow in the semi-arid Salt River basin. The applicability of the linear regression model was also investigated and concluded that the regression model is not reliable for this study. To generalize the model and derive a robust parameter set, cross-validation is applied and its outcome is compared with the split sample test. Cross-validation justifies the validity of the nonlinear relationship set up between input and output data.",
keywords = "Cross-validation, Neural network, Radial basis function, Self-organizing feature map, Streamflow forecasting",
author = "Hamid Moradkhani and Hsu, {Kuo Lin} and Gupta, {Hoshin Vijai} and Soroosh Sorooshian",
year = "2004",
month = "8",
day = "10",
doi = "10.1016/j.jhydrol.2004.03.027",
language = "English (US)",
volume = "295",
pages = "246--262",
journal = "Journal of Hydrology",
issn = "0022-1694",
publisher = "Elsevier",
number = "1-4",

}

TY - JOUR

T1 - Improved streamflow forecasting using self-organizing radial basis function artificial neural networks

AU - Moradkhani, Hamid

AU - Hsu, Kuo Lin

AU - Gupta, Hoshin Vijai

AU - Sorooshian, Soroosh

PY - 2004/8/10

Y1 - 2004/8/10

N2 - Streamflow forecasting has always been a challenging task for water resources engineers and managers and a major component of water resources system control. In this study, we explore the applicability of a Self Organizing Radial Basis (SORB) function to one-step ahead forecasting of daily streamflow. SORB uses a Gaussian Radial Basis Function architecture in conjunction with the Self-Organizing Feature Map (SOFM) used in data classification. SORB outperforms the two other ANN algorithms, the well known Multi-layer Feedforward Network (MFN) and Self-Organizing Linear Output map (SOLO) neural network for simulation of daily streamflow in the semi-arid Salt River basin. The applicability of the linear regression model was also investigated and concluded that the regression model is not reliable for this study. To generalize the model and derive a robust parameter set, cross-validation is applied and its outcome is compared with the split sample test. Cross-validation justifies the validity of the nonlinear relationship set up between input and output data.

AB - Streamflow forecasting has always been a challenging task for water resources engineers and managers and a major component of water resources system control. In this study, we explore the applicability of a Self Organizing Radial Basis (SORB) function to one-step ahead forecasting of daily streamflow. SORB uses a Gaussian Radial Basis Function architecture in conjunction with the Self-Organizing Feature Map (SOFM) used in data classification. SORB outperforms the two other ANN algorithms, the well known Multi-layer Feedforward Network (MFN) and Self-Organizing Linear Output map (SOLO) neural network for simulation of daily streamflow in the semi-arid Salt River basin. The applicability of the linear regression model was also investigated and concluded that the regression model is not reliable for this study. To generalize the model and derive a robust parameter set, cross-validation is applied and its outcome is compared with the split sample test. Cross-validation justifies the validity of the nonlinear relationship set up between input and output data.

KW - Cross-validation

KW - Neural network

KW - Radial basis function

KW - Self-organizing feature map

KW - Streamflow forecasting

UR - http://www.scopus.com/inward/record.url?scp=3142538909&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=3142538909&partnerID=8YFLogxK

U2 - 10.1016/j.jhydrol.2004.03.027

DO - 10.1016/j.jhydrol.2004.03.027

M3 - Article

AN - SCOPUS:3142538909

VL - 295

SP - 246

EP - 262

JO - Journal of Hydrology

JF - Journal of Hydrology

SN - 0022-1694

IS - 1-4

ER -