A study on real-time forecasting of reservoir inflow based on Artificial Neural Network

Chang Sam Jeong, Won Jun Koh, Jun-Haeng Heo

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

For the most effective operation of multi-purpose reservoir at flood period, the forecasting of inflow must be preceded and a rainfall-runoff modeling is necessary for the forecasting of inflow. However, the rainfall-runoff process is nonlinear and complex so many errors can be occurred by uncertain parameter estimation in modeling procedure. In this study, a neural network theory was adopted for modeling rainfall-runoff process, and a real-time inflow forecast system was developed. The models developed in this study were based on the back-propagation algorithm and Cascade-Correlation algorithm for learning. We applied these models to Soyangang River basin, so we could get forecasted inflow values., 1 hour, 3 hour and 6 hour preceding inflows. In case of the back-propagation algorithm, many trials are required to find out the optimum structure, but Cascade-Correlation algorithm can make the optimum neural network structure automatically at a time. We applied this model to the August '95 flood event at Soyangang River basin by using Cascade-Correlation algorithm and back-propagation algorithm. In order to improve the accuracy of the flood forecasting, the filtering technique has been used at the neural network model. As a result, Cascade-Correlation filtering model shows better forecasting capability.

Original languageEnglish
Title of host publicationWatershed Management and Operations Management 2000
Volume105
DOIs
Publication statusPublished - 2004 Dec 1
EventWatershed Management and Operations Management 2000 - Fort Collins, CO, United States
Duration: 2000 Jun 202000 Jun 24

Other

OtherWatershed Management and Operations Management 2000
CountryUnited States
CityFort Collins, CO
Period00/6/2000/6/24

Fingerprint

artificial neural network
inflow
back propagation
rainfall-runoff modeling
river basin
flood forecasting
learning
runoff
rainfall
modeling

All Science Journal Classification (ASJC) codes

  • Water Science and Technology

Cite this

Jeong, C. S., Koh, W. J., & Heo, J-H. (2004). A study on real-time forecasting of reservoir inflow based on Artificial Neural Network. In Watershed Management and Operations Management 2000 (Vol. 105) https://doi.org/10.1061/40499(2000)82
Jeong, Chang Sam ; Koh, Won Jun ; Heo, Jun-Haeng. / A study on real-time forecasting of reservoir inflow based on Artificial Neural Network. Watershed Management and Operations Management 2000. Vol. 105 2004.
@inproceedings{037f8742183b4a8fbc7601c590742e3b,
title = "A study on real-time forecasting of reservoir inflow based on Artificial Neural Network",
abstract = "For the most effective operation of multi-purpose reservoir at flood period, the forecasting of inflow must be preceded and a rainfall-runoff modeling is necessary for the forecasting of inflow. However, the rainfall-runoff process is nonlinear and complex so many errors can be occurred by uncertain parameter estimation in modeling procedure. In this study, a neural network theory was adopted for modeling rainfall-runoff process, and a real-time inflow forecast system was developed. The models developed in this study were based on the back-propagation algorithm and Cascade-Correlation algorithm for learning. We applied these models to Soyangang River basin, so we could get forecasted inflow values., 1 hour, 3 hour and 6 hour preceding inflows. In case of the back-propagation algorithm, many trials are required to find out the optimum structure, but Cascade-Correlation algorithm can make the optimum neural network structure automatically at a time. We applied this model to the August '95 flood event at Soyangang River basin by using Cascade-Correlation algorithm and back-propagation algorithm. In order to improve the accuracy of the flood forecasting, the filtering technique has been used at the neural network model. As a result, Cascade-Correlation filtering model shows better forecasting capability.",
author = "Jeong, {Chang Sam} and Koh, {Won Jun} and Jun-Haeng Heo",
year = "2004",
month = "12",
day = "1",
doi = "10.1061/40499(2000)82",
language = "English",
isbn = "0784404992",
volume = "105",
booktitle = "Watershed Management and Operations Management 2000",

}

Jeong, CS, Koh, WJ & Heo, J-H 2004, A study on real-time forecasting of reservoir inflow based on Artificial Neural Network. in Watershed Management and Operations Management 2000. vol. 105, Watershed Management and Operations Management 2000, Fort Collins, CO, United States, 00/6/20. https://doi.org/10.1061/40499(2000)82

A study on real-time forecasting of reservoir inflow based on Artificial Neural Network. / Jeong, Chang Sam; Koh, Won Jun; Heo, Jun-Haeng.

Watershed Management and Operations Management 2000. Vol. 105 2004.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - A study on real-time forecasting of reservoir inflow based on Artificial Neural Network

AU - Jeong, Chang Sam

AU - Koh, Won Jun

AU - Heo, Jun-Haeng

PY - 2004/12/1

Y1 - 2004/12/1

N2 - For the most effective operation of multi-purpose reservoir at flood period, the forecasting of inflow must be preceded and a rainfall-runoff modeling is necessary for the forecasting of inflow. However, the rainfall-runoff process is nonlinear and complex so many errors can be occurred by uncertain parameter estimation in modeling procedure. In this study, a neural network theory was adopted for modeling rainfall-runoff process, and a real-time inflow forecast system was developed. The models developed in this study were based on the back-propagation algorithm and Cascade-Correlation algorithm for learning. We applied these models to Soyangang River basin, so we could get forecasted inflow values., 1 hour, 3 hour and 6 hour preceding inflows. In case of the back-propagation algorithm, many trials are required to find out the optimum structure, but Cascade-Correlation algorithm can make the optimum neural network structure automatically at a time. We applied this model to the August '95 flood event at Soyangang River basin by using Cascade-Correlation algorithm and back-propagation algorithm. In order to improve the accuracy of the flood forecasting, the filtering technique has been used at the neural network model. As a result, Cascade-Correlation filtering model shows better forecasting capability.

AB - For the most effective operation of multi-purpose reservoir at flood period, the forecasting of inflow must be preceded and a rainfall-runoff modeling is necessary for the forecasting of inflow. However, the rainfall-runoff process is nonlinear and complex so many errors can be occurred by uncertain parameter estimation in modeling procedure. In this study, a neural network theory was adopted for modeling rainfall-runoff process, and a real-time inflow forecast system was developed. The models developed in this study were based on the back-propagation algorithm and Cascade-Correlation algorithm for learning. We applied these models to Soyangang River basin, so we could get forecasted inflow values., 1 hour, 3 hour and 6 hour preceding inflows. In case of the back-propagation algorithm, many trials are required to find out the optimum structure, but Cascade-Correlation algorithm can make the optimum neural network structure automatically at a time. We applied this model to the August '95 flood event at Soyangang River basin by using Cascade-Correlation algorithm and back-propagation algorithm. In order to improve the accuracy of the flood forecasting, the filtering technique has been used at the neural network model. As a result, Cascade-Correlation filtering model shows better forecasting capability.

UR - http://www.scopus.com/inward/record.url?scp=74949113917&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=74949113917&partnerID=8YFLogxK

U2 - 10.1061/40499(2000)82

DO - 10.1061/40499(2000)82

M3 - Conference contribution

AN - SCOPUS:74949113917

SN - 0784404992

SN - 9780784404997

VL - 105

BT - Watershed Management and Operations Management 2000

ER -

Jeong CS, Koh WJ, Heo J-H. A study on real-time forecasting of reservoir inflow based on Artificial Neural Network. In Watershed Management and Operations Management 2000. Vol. 105. 2004 https://doi.org/10.1061/40499(2000)82