Improved Water Level Forecasting Performance by Using Optimal Steepness Coefficients in an Artificial Neural Network

Muhammad Sulaiman, Ahmed El-Shafie, Othman A. Karim, Hassan Basri

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

Developing water level forecasting models is essential in water resources management and flood prediction. Accurate water level forecasting helps achieve efficient and optimum use of water resources and minimize flooding damages. The artificial neural network (ANN) is a computing model that has been successfully tested in many forecasting studies, including river flow. Improving the ANN computational approach could help produce accurate forecasting results. Most studies conducted to date have used a sigmoid function in a multi-layer perceptron neural network as the basis of the ANN; however, they have not considered the effect of sigmoid steepness on the forecasting results. In this study, the effectiveness of the steepness coefficient (SC) in the sigmoid function of an ANN model designed to test the accuracy of 1-day water level forecasts was investigated. The performance of data training and data validation were evaluated using the statistical index efficiency coefficient and root mean square error. The weight initialization was fixed at 0.5 in the ANN so that even comparisons could be made between models. Three hundred rounds of data training were conducted using five ANN architectures, six datasets and 10 steepness coefficients. The results showed that the optimal SC improved the forecasting accuracy of the ANN data training and data validation when compared with the standard SC. Importantly, the performance of ANN data training improved significantly with utilization of the optimal SC.

Original languageEnglish
Pages (from-to)2525-2541
Number of pages17
JournalWater Resources Management
Volume25
Issue number10
DOIs
Publication statusPublished - Aug 2011

Fingerprint

Water levels
artificial neural network
water level
Neural networks
Water resources
river flow
Multilayer neural networks
Network architecture
Mean square error
flooding
water resource
Rivers
damage
prediction

Keywords

  • Artificial neural networks
  • Sigmoid function
  • Steepness coefficient
  • Water level forecasting

ASJC Scopus subject areas

  • Water Science and Technology
  • Civil and Structural Engineering

Cite this

Improved Water Level Forecasting Performance by Using Optimal Steepness Coefficients in an Artificial Neural Network. / Sulaiman, Muhammad; El-Shafie, Ahmed; A. Karim, Othman; Basri, Hassan.

In: Water Resources Management, Vol. 25, No. 10, 08.2011, p. 2525-2541.

Research output: Contribution to journalArticle

@article{6f559c89a69d499fb0c2918f2c8c3ba2,
title = "Improved Water Level Forecasting Performance by Using Optimal Steepness Coefficients in an Artificial Neural Network",
abstract = "Developing water level forecasting models is essential in water resources management and flood prediction. Accurate water level forecasting helps achieve efficient and optimum use of water resources and minimize flooding damages. The artificial neural network (ANN) is a computing model that has been successfully tested in many forecasting studies, including river flow. Improving the ANN computational approach could help produce accurate forecasting results. Most studies conducted to date have used a sigmoid function in a multi-layer perceptron neural network as the basis of the ANN; however, they have not considered the effect of sigmoid steepness on the forecasting results. In this study, the effectiveness of the steepness coefficient (SC) in the sigmoid function of an ANN model designed to test the accuracy of 1-day water level forecasts was investigated. The performance of data training and data validation were evaluated using the statistical index efficiency coefficient and root mean square error. The weight initialization was fixed at 0.5 in the ANN so that even comparisons could be made between models. Three hundred rounds of data training were conducted using five ANN architectures, six datasets and 10 steepness coefficients. The results showed that the optimal SC improved the forecasting accuracy of the ANN data training and data validation when compared with the standard SC. Importantly, the performance of ANN data training improved significantly with utilization of the optimal SC.",
keywords = "Artificial neural networks, Sigmoid function, Steepness coefficient, Water level forecasting",
author = "Muhammad Sulaiman and Ahmed El-Shafie and {A. Karim}, Othman and Hassan Basri",
year = "2011",
month = "8",
doi = "10.1007/s11269-011-9824-z",
language = "English",
volume = "25",
pages = "2525--2541",
journal = "Water Resources Management",
issn = "0920-4741",
publisher = "Springer Netherlands",
number = "10",

}

TY - JOUR

T1 - Improved Water Level Forecasting Performance by Using Optimal Steepness Coefficients in an Artificial Neural Network

AU - Sulaiman, Muhammad

AU - El-Shafie, Ahmed

AU - A. Karim, Othman

AU - Basri, Hassan

PY - 2011/8

Y1 - 2011/8

N2 - Developing water level forecasting models is essential in water resources management and flood prediction. Accurate water level forecasting helps achieve efficient and optimum use of water resources and minimize flooding damages. The artificial neural network (ANN) is a computing model that has been successfully tested in many forecasting studies, including river flow. Improving the ANN computational approach could help produce accurate forecasting results. Most studies conducted to date have used a sigmoid function in a multi-layer perceptron neural network as the basis of the ANN; however, they have not considered the effect of sigmoid steepness on the forecasting results. In this study, the effectiveness of the steepness coefficient (SC) in the sigmoid function of an ANN model designed to test the accuracy of 1-day water level forecasts was investigated. The performance of data training and data validation were evaluated using the statistical index efficiency coefficient and root mean square error. The weight initialization was fixed at 0.5 in the ANN so that even comparisons could be made between models. Three hundred rounds of data training were conducted using five ANN architectures, six datasets and 10 steepness coefficients. The results showed that the optimal SC improved the forecasting accuracy of the ANN data training and data validation when compared with the standard SC. Importantly, the performance of ANN data training improved significantly with utilization of the optimal SC.

AB - Developing water level forecasting models is essential in water resources management and flood prediction. Accurate water level forecasting helps achieve efficient and optimum use of water resources and minimize flooding damages. The artificial neural network (ANN) is a computing model that has been successfully tested in many forecasting studies, including river flow. Improving the ANN computational approach could help produce accurate forecasting results. Most studies conducted to date have used a sigmoid function in a multi-layer perceptron neural network as the basis of the ANN; however, they have not considered the effect of sigmoid steepness on the forecasting results. In this study, the effectiveness of the steepness coefficient (SC) in the sigmoid function of an ANN model designed to test the accuracy of 1-day water level forecasts was investigated. The performance of data training and data validation were evaluated using the statistical index efficiency coefficient and root mean square error. The weight initialization was fixed at 0.5 in the ANN so that even comparisons could be made between models. Three hundred rounds of data training were conducted using five ANN architectures, six datasets and 10 steepness coefficients. The results showed that the optimal SC improved the forecasting accuracy of the ANN data training and data validation when compared with the standard SC. Importantly, the performance of ANN data training improved significantly with utilization of the optimal SC.

KW - Artificial neural networks

KW - Sigmoid function

KW - Steepness coefficient

KW - Water level forecasting

UR - http://www.scopus.com/inward/record.url?scp=79960702239&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79960702239&partnerID=8YFLogxK

U2 - 10.1007/s11269-011-9824-z

DO - 10.1007/s11269-011-9824-z

M3 - Article

AN - SCOPUS:79960702239

VL - 25

SP - 2525

EP - 2541

JO - Water Resources Management

JF - Water Resources Management

SN - 0920-4741

IS - 10

ER -