Multiscale Gaussian process regression-based generalized likelihood ratio test for fault detection in water distribution networks

R. Fazai, Majdi Mansouri, K. Abodayeh, V. Puig, M. I.Noori Raouf, Hazem Nounou, Mohamed Nounou

Research output: Contribution to journalArticle

Abstract

This paper proposes a new leak/contaminant detection approach that aims to enhance the monitoring of water distribution network (WDN). The developed method relies on using machine learning (e.g Gaussian process regression (GPR)) as a modeling framework and generalized likelihood ratio (GLRT) for detection purposes. To improve the performances of the developed GPR model even further, multiscale representation of data will be used to develop multiscale extension of these method. Multiscale representation is a powerful data analysis technique that presents efficient separation of deterministic characteristics from random noise. Therefore, the multiscale GPR method, that combines the advantages of the GPR method with those of multiscale representation, will be developed to enhance the WDN modeling performance. We develop a new technique for detecting leak/contaminant in WDN using GLRT. For further enhance, the performance of GLRT, an exponentially weighted moving average (EWMA)-GLRT (EWMA-GLRT) chart is developed. The simulation results show that the MSGPR-based EWMA-GLRT method outperforms MSGPR-based GLRT and that both of them provide clear advantages over the neural networks (NN)- and support vector regression (SVR)- and GPR-based GLRT techniques.

Original languageEnglish
Pages (from-to)474-491
Number of pages18
JournalEngineering Applications of Artificial Intelligence
Volume85
DOIs
Publication statusPublished - 1 Oct 2019

Fingerprint

Fault detection
Electric power distribution
Impurities
Water
Learning systems
Neural networks
Monitoring

Keywords

  • Contaminant/leak detection
  • Gaussian process regression (GPR)
  • Generalized likelihood ratio test (GLRT)
  • Multiscale representation
  • Water distribution networks (WDNs)

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Artificial Intelligence
  • Electrical and Electronic Engineering

Cite this

@article{4c8780fc11b54419b1f2255889813773,
title = "Multiscale Gaussian process regression-based generalized likelihood ratio test for fault detection in water distribution networks",
abstract = "This paper proposes a new leak/contaminant detection approach that aims to enhance the monitoring of water distribution network (WDN). The developed method relies on using machine learning (e.g Gaussian process regression (GPR)) as a modeling framework and generalized likelihood ratio (GLRT) for detection purposes. To improve the performances of the developed GPR model even further, multiscale representation of data will be used to develop multiscale extension of these method. Multiscale representation is a powerful data analysis technique that presents efficient separation of deterministic characteristics from random noise. Therefore, the multiscale GPR method, that combines the advantages of the GPR method with those of multiscale representation, will be developed to enhance the WDN modeling performance. We develop a new technique for detecting leak/contaminant in WDN using GLRT. For further enhance, the performance of GLRT, an exponentially weighted moving average (EWMA)-GLRT (EWMA-GLRT) chart is developed. The simulation results show that the MSGPR-based EWMA-GLRT method outperforms MSGPR-based GLRT and that both of them provide clear advantages over the neural networks (NN)- and support vector regression (SVR)- and GPR-based GLRT techniques.",
keywords = "Contaminant/leak detection, Gaussian process regression (GPR), Generalized likelihood ratio test (GLRT), Multiscale representation, Water distribution networks (WDNs)",
author = "R. Fazai and Majdi Mansouri and K. Abodayeh and V. Puig and Raouf, {M. I.Noori} and Hazem Nounou and Mohamed Nounou",
year = "2019",
month = "10",
day = "1",
doi = "10.1016/j.engappai.2019.07.007",
language = "English",
volume = "85",
pages = "474--491",
journal = "Engineering Applications of Artificial Intelligence",
issn = "0952-1976",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Multiscale Gaussian process regression-based generalized likelihood ratio test for fault detection in water distribution networks

AU - Fazai, R.

AU - Mansouri, Majdi

AU - Abodayeh, K.

AU - Puig, V.

AU - Raouf, M. I.Noori

AU - Nounou, Hazem

AU - Nounou, Mohamed

PY - 2019/10/1

Y1 - 2019/10/1

N2 - This paper proposes a new leak/contaminant detection approach that aims to enhance the monitoring of water distribution network (WDN). The developed method relies on using machine learning (e.g Gaussian process regression (GPR)) as a modeling framework and generalized likelihood ratio (GLRT) for detection purposes. To improve the performances of the developed GPR model even further, multiscale representation of data will be used to develop multiscale extension of these method. Multiscale representation is a powerful data analysis technique that presents efficient separation of deterministic characteristics from random noise. Therefore, the multiscale GPR method, that combines the advantages of the GPR method with those of multiscale representation, will be developed to enhance the WDN modeling performance. We develop a new technique for detecting leak/contaminant in WDN using GLRT. For further enhance, the performance of GLRT, an exponentially weighted moving average (EWMA)-GLRT (EWMA-GLRT) chart is developed. The simulation results show that the MSGPR-based EWMA-GLRT method outperforms MSGPR-based GLRT and that both of them provide clear advantages over the neural networks (NN)- and support vector regression (SVR)- and GPR-based GLRT techniques.

AB - This paper proposes a new leak/contaminant detection approach that aims to enhance the monitoring of water distribution network (WDN). The developed method relies on using machine learning (e.g Gaussian process regression (GPR)) as a modeling framework and generalized likelihood ratio (GLRT) for detection purposes. To improve the performances of the developed GPR model even further, multiscale representation of data will be used to develop multiscale extension of these method. Multiscale representation is a powerful data analysis technique that presents efficient separation of deterministic characteristics from random noise. Therefore, the multiscale GPR method, that combines the advantages of the GPR method with those of multiscale representation, will be developed to enhance the WDN modeling performance. We develop a new technique for detecting leak/contaminant in WDN using GLRT. For further enhance, the performance of GLRT, an exponentially weighted moving average (EWMA)-GLRT (EWMA-GLRT) chart is developed. The simulation results show that the MSGPR-based EWMA-GLRT method outperforms MSGPR-based GLRT and that both of them provide clear advantages over the neural networks (NN)- and support vector regression (SVR)- and GPR-based GLRT techniques.

KW - Contaminant/leak detection

KW - Gaussian process regression (GPR)

KW - Generalized likelihood ratio test (GLRT)

KW - Multiscale representation

KW - Water distribution networks (WDNs)

UR - http://www.scopus.com/inward/record.url?scp=85068931261&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068931261&partnerID=8YFLogxK

U2 - 10.1016/j.engappai.2019.07.007

DO - 10.1016/j.engappai.2019.07.007

M3 - Article

VL - 85

SP - 474

EP - 491

JO - Engineering Applications of Artificial Intelligence

JF - Engineering Applications of Artificial Intelligence

SN - 0952-1976

ER -