Neural network identification of water pipe blockage from smart embedded passive acoustic measurements

Research output: Contribution to journalArticlepeer-review

Standard

Harvard

APA

Vancouver

Author

Bibtex

@article{4deee4e5d10e48f1a4430f4fd3920d57,
title = "Neural network identification of water pipe blockage from smart embedded passive acoustic measurements",
abstract = "This study presents a new neural network approach to identify the presence and type of obstruction in pipes from measurements of passive acoustic emissions. Inserts were used in a fluid re-circulation loop to simulate different types of blockage at various flow rates within the turbulent regime, generating patterns of acoustic emissions. The data were pre-processed using Fourier analysis, and two candidate sets of statistical descriptors were extracted for each measurement. The first set used average and spread of the Fourier transform amplitudes, the second used data binning to obtain a concise representation of the spectrum of amplitudes. Experimental evidence showed the second set of descriptors was the most suitable to train the neural network to recognize with accuracy the presence and type of blockage. The obtained results compare favourably with the literature, indicating that the approach provides a tool to enhance process monitoring in water supply systems, in particular early detection of upstream blockages.",
keywords = "acoustic emission, neural network, online monitoring, pressure drop, turbulent flow",
author = "Luca Baronti and Marco Castellani and Daniel Hefft and Federico Alberini",
note = "Publisher Copyright: {\textcopyright} 2021 The Authors. The Canadian Journal of Chemical Engineering published by Wiley Periodicals LLC on behalf of Canadian Society for Chemical Engineering.",
year = "2021",
month = may,
day = "24",
doi = "10.1002/cjce.24202",
language = "English",
journal = "Canadian Journal of Chemical Engineering",
issn = "0008-4034",
publisher = "Wiley",

}

RIS

TY - JOUR

T1 - Neural network identification of water pipe blockage from smart embedded passive acoustic measurements

AU - Baronti, Luca

AU - Castellani, Marco

AU - Hefft, Daniel

AU - Alberini, Federico

N1 - Publisher Copyright: © 2021 The Authors. The Canadian Journal of Chemical Engineering published by Wiley Periodicals LLC on behalf of Canadian Society for Chemical Engineering.

PY - 2021/5/24

Y1 - 2021/5/24

N2 - This study presents a new neural network approach to identify the presence and type of obstruction in pipes from measurements of passive acoustic emissions. Inserts were used in a fluid re-circulation loop to simulate different types of blockage at various flow rates within the turbulent regime, generating patterns of acoustic emissions. The data were pre-processed using Fourier analysis, and two candidate sets of statistical descriptors were extracted for each measurement. The first set used average and spread of the Fourier transform amplitudes, the second used data binning to obtain a concise representation of the spectrum of amplitudes. Experimental evidence showed the second set of descriptors was the most suitable to train the neural network to recognize with accuracy the presence and type of blockage. The obtained results compare favourably with the literature, indicating that the approach provides a tool to enhance process monitoring in water supply systems, in particular early detection of upstream blockages.

AB - This study presents a new neural network approach to identify the presence and type of obstruction in pipes from measurements of passive acoustic emissions. Inserts were used in a fluid re-circulation loop to simulate different types of blockage at various flow rates within the turbulent regime, generating patterns of acoustic emissions. The data were pre-processed using Fourier analysis, and two candidate sets of statistical descriptors were extracted for each measurement. The first set used average and spread of the Fourier transform amplitudes, the second used data binning to obtain a concise representation of the spectrum of amplitudes. Experimental evidence showed the second set of descriptors was the most suitable to train the neural network to recognize with accuracy the presence and type of blockage. The obtained results compare favourably with the literature, indicating that the approach provides a tool to enhance process monitoring in water supply systems, in particular early detection of upstream blockages.

KW - acoustic emission

KW - neural network

KW - online monitoring

KW - pressure drop

KW - turbulent flow

UR - http://www.scopus.com/inward/record.url?scp=85108723123&partnerID=8YFLogxK

U2 - 10.1002/cjce.24202

DO - 10.1002/cjce.24202

M3 - Article

JO - Canadian Journal of Chemical Engineering

JF - Canadian Journal of Chemical Engineering

SN - 0008-4034

ER -