Adaptive Multipopulation Evolutionary Algorithm for Contamination Source Identification in Water Distribution Systems

Abstract

Real-time monitoring of drinking water in a water distribution system (WDS) can eectively warn and reduce safety risks. One of the challenges is to identify the contamination source through these observed data due to the real-time, non-uniqueness, and large scale characteristics. To address the real-time and non-uniqueness challenges, we propose an adaptive multi-population evolutionary optimization algorithm to determine the real-time characteristics of contamination sources, where each population aims to locate and track a dierent global optimum. The algorithm adaptively adjusts the number of populations using a feedback learning mechanism. To eectively locate an optimal solution for a population, a co-evolutionary strategy is used to identify the location and the injection prole separately. Experimental results on three WDS networks show that the proposed algorithm is competitive in comparison with three other state-of-the-art evolutionary algorithms.

Partitioning Leakage Detection in Water Distribution Systems: A Specialized Deep Learning Framework Enhanced by Spatial–Temporal Graph Convolutional Networks

“…4 Undetected leaks result in wasted water, energy consumption for needless pumping, and contamination risks. 5,6 They also represent lost revenue for the utility. As pipes are buried underground, visual leak detection is complex, and many leaks remain unnoticed until they surface or cause obvious problems.…”

“…Leakages can occur due to corrosion, pressure transients, construction defects, and aging infrastructure. , Even small leaks can gradually intensify and lead to significant pipe ruptures . Undetected leaks result in wasted water, energy consumption for needless pumping, and contamination risks. , They also represent lost revenue for the utility. As pipes are buried underground, visual leak detection is complex, and many leaks remain unnoticed until they surface or cause obvious problems .…”

An Effective and Efficient Method for Identification of Contamination Sources in Water Distribution Systems Based on Manual Grab‐Sampling

“…These developments started by using traditional optimization techniques, such as linear programming (LP) scheme (Preis & Ostfeld, 2006). This was followed by the use of various evolutionary algorithms (EAs) as they possess superior search capabilities compared to the traditional LP and nonlinear programming (NLP) techniques (Hu et al., 2015; Li et al., 2021; Preis & Ostfeld, 2008). While these algorithms have reliable performance in locating contamination sources in hypothetical case studies, their practical application can be challenging.…”