Hourly water level forecasting at tributary affecteby main river condition

Ji Youn Sung, Jeongwoo Lee, Il Moon Chung, Jun Haeng Heo

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

This study develops hourly water level forecasting models with lead-times of 1 to 3 h using an artificial neural network (ANN) for Anyangcheon stream, one of the major tributaries of the Han River, South Korea. To consider the backwater effect from this river, an enhanced tributary water level forecasting model is proposed by adding multiple water level data on the main river as input variables into the conventional ANN structure which often uses rainfall and upstream water level data. Four types of ANN models per each lead-time are built with increasing complexity of the input vector, and their performances are compared. The results indicate that the inclusion of multiple water level data on the main river to the network provides water level forecasts with greater accuracy at the Ogeumgyo gauging station of interest. The final best ANN models for water level forecasts with lead-times of 1 to 2 h show good performance with root mean square errors (RMSE) below 0.06 m and 0.12 m, respectively. However, the final best ANN model for forecasting 3 h ahead was unsatisfactory, showing underestimation at many rising parts of the hydrograph.

Original languageEnglish
Article number644
JournalWater (Switzerland)
Volume9
Issue number9
DOIs
Publication statusPublished - 2017 Aug 28

Fingerprint

Water levels
Rivers
surface water level
tributary
water level
river
neural network
artificial neural network
neural networks
water
rivers
Water
Neural Networks (Computer)
Neural networks
Republic of Korea
Gaging
backwater
South Korea
hydrograph
Mean square error

All Science Journal Classification (ASJC) codes

  • Biochemistry
  • Geography, Planning and Development
  • Aquatic Science
  • Water Science and Technology

Cite this

Sung, Ji Youn ; Lee, Jeongwoo ; Chung, Il Moon ; Heo, Jun Haeng. / Hourly water level forecasting at tributary affecteby main river condition. In: Water (Switzerland). 2017 ; Vol. 9, No. 9.
@article{44b7574f2def4818bc065b88b5b57b5b,
title = "Hourly water level forecasting at tributary affecteby main river condition",
abstract = "This study develops hourly water level forecasting models with lead-times of 1 to 3 h using an artificial neural network (ANN) for Anyangcheon stream, one of the major tributaries of the Han River, South Korea. To consider the backwater effect from this river, an enhanced tributary water level forecasting model is proposed by adding multiple water level data on the main river as input variables into the conventional ANN structure which often uses rainfall and upstream water level data. Four types of ANN models per each lead-time are built with increasing complexity of the input vector, and their performances are compared. The results indicate that the inclusion of multiple water level data on the main river to the network provides water level forecasts with greater accuracy at the Ogeumgyo gauging station of interest. The final best ANN models for water level forecasts with lead-times of 1 to 2 h show good performance with root mean square errors (RMSE) below 0.06 m and 0.12 m, respectively. However, the final best ANN model for forecasting 3 h ahead was unsatisfactory, showing underestimation at many rising parts of the hydrograph.",
author = "Sung, {Ji Youn} and Jeongwoo Lee and Chung, {Il Moon} and Heo, {Jun Haeng}",
year = "2017",
month = "8",
day = "28",
doi = "10.3390/w9090644",
language = "English",
volume = "9",
journal = "Water (Switzerland)",
issn = "2073-4441",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "9",

}

Hourly water level forecasting at tributary affecteby main river condition. / Sung, Ji Youn; Lee, Jeongwoo; Chung, Il Moon; Heo, Jun Haeng.

In: Water (Switzerland), Vol. 9, No. 9, 644, 28.08.2017.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Hourly water level forecasting at tributary affecteby main river condition

AU - Sung, Ji Youn

AU - Lee, Jeongwoo

AU - Chung, Il Moon

AU - Heo, Jun Haeng

PY - 2017/8/28

Y1 - 2017/8/28

N2 - This study develops hourly water level forecasting models with lead-times of 1 to 3 h using an artificial neural network (ANN) for Anyangcheon stream, one of the major tributaries of the Han River, South Korea. To consider the backwater effect from this river, an enhanced tributary water level forecasting model is proposed by adding multiple water level data on the main river as input variables into the conventional ANN structure which often uses rainfall and upstream water level data. Four types of ANN models per each lead-time are built with increasing complexity of the input vector, and their performances are compared. The results indicate that the inclusion of multiple water level data on the main river to the network provides water level forecasts with greater accuracy at the Ogeumgyo gauging station of interest. The final best ANN models for water level forecasts with lead-times of 1 to 2 h show good performance with root mean square errors (RMSE) below 0.06 m and 0.12 m, respectively. However, the final best ANN model for forecasting 3 h ahead was unsatisfactory, showing underestimation at many rising parts of the hydrograph.

AB - This study develops hourly water level forecasting models with lead-times of 1 to 3 h using an artificial neural network (ANN) for Anyangcheon stream, one of the major tributaries of the Han River, South Korea. To consider the backwater effect from this river, an enhanced tributary water level forecasting model is proposed by adding multiple water level data on the main river as input variables into the conventional ANN structure which often uses rainfall and upstream water level data. Four types of ANN models per each lead-time are built with increasing complexity of the input vector, and their performances are compared. The results indicate that the inclusion of multiple water level data on the main river to the network provides water level forecasts with greater accuracy at the Ogeumgyo gauging station of interest. The final best ANN models for water level forecasts with lead-times of 1 to 2 h show good performance with root mean square errors (RMSE) below 0.06 m and 0.12 m, respectively. However, the final best ANN model for forecasting 3 h ahead was unsatisfactory, showing underestimation at many rising parts of the hydrograph.

UR - http://www.scopus.com/inward/record.url?scp=85028340749&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85028340749&partnerID=8YFLogxK

U2 - 10.3390/w9090644

DO - 10.3390/w9090644

M3 - Article

AN - SCOPUS:85028340749

VL - 9

JO - Water (Switzerland)

JF - Water (Switzerland)

SN - 2073-4441

IS - 9

M1 - 644

ER -