Near–Real Time Burst Location and Sizing in Water Distribution Systems Using Artificial Neural Networks

Laura Monteiro,Dídia Covas,Miguel Capelo,Bruno Brentan

doi:10.3390/w13131841

Laura Monteiro, Dídia Covas + Show 2 more

Open Access

https://doi.org/10.3390/w13131841

Copy DOI

Journal: Water	Publication Date: Jul 1, 2021
Citations: 13	License type: CC BY 4.0

Affiliation: University of Lisbon, Universidade Federal de Minas Gerais

Abstract

The current paper proposes a novel methodology for near–real time burst location and sizing in water distribution systems (WDS) by means of Multi–Layer Perceptron (MLP), a class of artificial neural network (ANN). The proposed methodology can be systematized in four steps: (1) construction of the pipe–burst database, (2) problem formulation and ANN architecture definition, (3) ANN training, testing and sensitivity analyses, (4) application based on collected data. A large database needs to be constructed using 24 h pressure–head data collected or numerically generated at different sensor locations during the pipe burst occurrence. The ANN is trained and tested in a real–life network, in Portugal, using artificial data generated by hydraulic extended period simulations. The trained ANN has demonstrated to successfully locate 60–70% of the burst with an accuracy of 100 m and 98% of the burst with an accuracy of 500 m and to determine burst sizes with uncertainties lower than 2 L/s in 90% of tested cases and lower than 0.2 L/s in 70% of the cases. This approach can be used as a daily management tool of water distribution networks (WDN), as long as the ANN is trained with artificial data generated by an accurate and calibrated WDS hydraulic models and/or with reliable pressure–head data collected at different locations of the WDS during the pipe burst occurrence.

Highlights

Water distribution systems (WDS) are one of the most important public infrastructures that provide an essential service to populations: the provision of water in quantity and with adequate pressure and quality
The current paper aims at using a data mining and machine learning technique, Artificial Neural Networks (ANN), for near–real time burst location and sizing in water distribution networks
The solution to the lack of real pipe burst data is to artificially generate pipe burst scenarios using a reliable and calibrated network model. This model can be developed in EPANET, a public domain software applied to WDS modeling, and pipe burst scenarios systematically simulated with the support of any programming tool, such as the MATLAB

Summary

Introduction

Water distribution systems (WDS) are one of the most important public infrastructures that provide an essential service to populations: the provision of water in quantity and with adequate pressure and quality. Among the practical water losses control methods and leak detection techniques, acoustic signal analysis, minimum night flow monitoring and water balance calculation in network sectors are the most widely used and successfully implemented by many water utilities [4,5,6,7]. Though these methods undoubtedly allow efficient water losses control in WDS, they require a huge investment of water utilities in training human resources and in the installation of the necessary instrumentation for monitoring the

Methods

Results

Conclusion