An augmented Wavelet De-noising Technique with Neuro-Fuzzy Inference System for water quality prediction

Ali Najah Ahmed, Ahmed El-Shafie, Othman A. Karim, Amr El-Shafie

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Johor River Basin is located in Johor state, Malaysia, which is significantly degrading due to human activities and development along the river. Accordingly, it is very important to implement and adopt a water quality prediction model that can provide a powerful tool to implement better water resource management. Several modeling methods have been applied to this research including: Multi Layer Perceptron Neural Networks (MLP-ANN), Radial Basis Function Neural Networks (RBF-ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS). Nevertheless, the data arising from monitoring stations and experiment may be polluted by noise signals owing to systematic errors and random errors. This noisy data often make the predict task relatively difficult. Therefore, this study suggests an augmented Wavelet De-noising Technique with Neuro-Fuzzy Inference System (WDT-ANFIS). In this study, the water quality parameters in the domain of interests are dissolved oxygen (DO), biochemical oxygen demand (BOD), chemical oxygen demand (COD). Two scenarios were introduced: Scenario 1 was to construct prediction model for water quality parameters at each station, while Scenario 2 was to construct prediction model based on the value of same parameter at previous station (upstream), and both were based on the value of the twelve input parameters. The WDT-ANFIS was verified based on field data from 2009-2010. The WDT-ANFIS model outperformed all the proposed models and improved predicting accuracy for all water quality parameters. Scenario 2 performed more adequately than Scenario 1 with significant improvement ranging from 0.5% to 3.1% for all water quality parameters at all stations. The verification of the proposed model showed that the model satisfactorily predicted all the parameters (R 2 values bigger than 0.9). ICIC International

Original languageEnglish
Pages (from-to)7055-7082
Number of pages28
JournalInternational Journal of Innovative Computing, Information and Control
Volume8
Issue number10 B
Publication statusPublished - Oct 2012

Fingerprint

Wavelet Denoising
Fuzzy Inference System
Neuro-fuzzy
Water Quality
Fuzzy inference
Water quality
Adaptive Neuro-fuzzy Inference System
Prediction
Scenarios
Prediction Model
Oxygen
Rivers
Neural networks
Water Management
Malaysia
Random errors
Radial Basis Function Neural Network
Water Resources
Systematic Error
Random Error

Keywords

  • MLP-ANN
  • RBF-ANN
  • WDT-ANFIS

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Information Systems
  • Software
  • Theoretical Computer Science

Cite this

An augmented Wavelet De-noising Technique with Neuro-Fuzzy Inference System for water quality prediction. / Najah Ahmed, Ali; El-Shafie, Ahmed; A. Karim, Othman; El-Shafie, Amr.

In: International Journal of Innovative Computing, Information and Control, Vol. 8, No. 10 B, 10.2012, p. 7055-7082.

Research output: Contribution to journalArticle

@article{77c77a6250e04e2ab0db7d9092f7c4ef,
title = "An augmented Wavelet De-noising Technique with Neuro-Fuzzy Inference System for water quality prediction",
abstract = "Johor River Basin is located in Johor state, Malaysia, which is significantly degrading due to human activities and development along the river. Accordingly, it is very important to implement and adopt a water quality prediction model that can provide a powerful tool to implement better water resource management. Several modeling methods have been applied to this research including: Multi Layer Perceptron Neural Networks (MLP-ANN), Radial Basis Function Neural Networks (RBF-ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS). Nevertheless, the data arising from monitoring stations and experiment may be polluted by noise signals owing to systematic errors and random errors. This noisy data often make the predict task relatively difficult. Therefore, this study suggests an augmented Wavelet De-noising Technique with Neuro-Fuzzy Inference System (WDT-ANFIS). In this study, the water quality parameters in the domain of interests are dissolved oxygen (DO), biochemical oxygen demand (BOD), chemical oxygen demand (COD). Two scenarios were introduced: Scenario 1 was to construct prediction model for water quality parameters at each station, while Scenario 2 was to construct prediction model based on the value of same parameter at previous station (upstream), and both were based on the value of the twelve input parameters. The WDT-ANFIS was verified based on field data from 2009-2010. The WDT-ANFIS model outperformed all the proposed models and improved predicting accuracy for all water quality parameters. Scenario 2 performed more adequately than Scenario 1 with significant improvement ranging from 0.5{\%} to 3.1{\%} for all water quality parameters at all stations. The verification of the proposed model showed that the model satisfactorily predicted all the parameters (R 2 values bigger than 0.9). ICIC International",
keywords = "MLP-ANN, RBF-ANN, WDT-ANFIS",
author = "{Najah Ahmed}, Ali and Ahmed El-Shafie and {A. Karim}, Othman and Amr El-Shafie",
year = "2012",
month = "10",
language = "English",
volume = "8",
pages = "7055--7082",
journal = "International Journal of Innovative Computing, Information and Control",
issn = "1349-4198",
publisher = "IJICIC Editorial Office",
number = "10 B",

}

TY - JOUR

T1 - An augmented Wavelet De-noising Technique with Neuro-Fuzzy Inference System for water quality prediction

AU - Najah Ahmed, Ali

AU - El-Shafie, Ahmed

AU - A. Karim, Othman

AU - El-Shafie, Amr

PY - 2012/10

Y1 - 2012/10

N2 - Johor River Basin is located in Johor state, Malaysia, which is significantly degrading due to human activities and development along the river. Accordingly, it is very important to implement and adopt a water quality prediction model that can provide a powerful tool to implement better water resource management. Several modeling methods have been applied to this research including: Multi Layer Perceptron Neural Networks (MLP-ANN), Radial Basis Function Neural Networks (RBF-ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS). Nevertheless, the data arising from monitoring stations and experiment may be polluted by noise signals owing to systematic errors and random errors. This noisy data often make the predict task relatively difficult. Therefore, this study suggests an augmented Wavelet De-noising Technique with Neuro-Fuzzy Inference System (WDT-ANFIS). In this study, the water quality parameters in the domain of interests are dissolved oxygen (DO), biochemical oxygen demand (BOD), chemical oxygen demand (COD). Two scenarios were introduced: Scenario 1 was to construct prediction model for water quality parameters at each station, while Scenario 2 was to construct prediction model based on the value of same parameter at previous station (upstream), and both were based on the value of the twelve input parameters. The WDT-ANFIS was verified based on field data from 2009-2010. The WDT-ANFIS model outperformed all the proposed models and improved predicting accuracy for all water quality parameters. Scenario 2 performed more adequately than Scenario 1 with significant improvement ranging from 0.5% to 3.1% for all water quality parameters at all stations. The verification of the proposed model showed that the model satisfactorily predicted all the parameters (R 2 values bigger than 0.9). ICIC International

AB - Johor River Basin is located in Johor state, Malaysia, which is significantly degrading due to human activities and development along the river. Accordingly, it is very important to implement and adopt a water quality prediction model that can provide a powerful tool to implement better water resource management. Several modeling methods have been applied to this research including: Multi Layer Perceptron Neural Networks (MLP-ANN), Radial Basis Function Neural Networks (RBF-ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS). Nevertheless, the data arising from monitoring stations and experiment may be polluted by noise signals owing to systematic errors and random errors. This noisy data often make the predict task relatively difficult. Therefore, this study suggests an augmented Wavelet De-noising Technique with Neuro-Fuzzy Inference System (WDT-ANFIS). In this study, the water quality parameters in the domain of interests are dissolved oxygen (DO), biochemical oxygen demand (BOD), chemical oxygen demand (COD). Two scenarios were introduced: Scenario 1 was to construct prediction model for water quality parameters at each station, while Scenario 2 was to construct prediction model based on the value of same parameter at previous station (upstream), and both were based on the value of the twelve input parameters. The WDT-ANFIS was verified based on field data from 2009-2010. The WDT-ANFIS model outperformed all the proposed models and improved predicting accuracy for all water quality parameters. Scenario 2 performed more adequately than Scenario 1 with significant improvement ranging from 0.5% to 3.1% for all water quality parameters at all stations. The verification of the proposed model showed that the model satisfactorily predicted all the parameters (R 2 values bigger than 0.9). ICIC International

KW - MLP-ANN

KW - RBF-ANN

KW - WDT-ANFIS

UR - http://www.scopus.com/inward/record.url?scp=84866049092&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866049092&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84866049092

VL - 8

SP - 7055

EP - 7082

JO - International Journal of Innovative Computing, Information and Control

JF - International Journal of Innovative Computing, Information and Control

SN - 1349-4198

IS - 10 B

ER -