Challenges, Methods, Data -- a Survey of Machine Learning in Water Distribution Networks

TL;DR

This survey reviews how machine learning is increasingly applied to water distribution networks, highlighting challenges, methods, and benchmarks, especially in leakage detection, amidst climate change impacts.

Contribution

It provides a comprehensive overview of machine learning applications in water networks, including a structured survey, challenges, and evaluation benchmarks for leakage detection.

Findings

Machine learning offers promising solutions for water network management.

Challenges include domain-specific data and system complexity.

Benchmarks facilitate standardized evaluation of ML methods.

Abstract

Research on methods for planning and controlling water distribution networks gains increasing relevance as the availability of drinking water will decrease as a consequence of climate change. So far, the majority of approaches is based on hydraulics and engineering expertise. However, with the increasing availability of sensors, machine learning techniques constitute a promising tool. This work presents the main tasks in water distribution networks, discusses how they relate to machine learning and analyses how the particularities of the domain pose challenges to and can be leveraged by machine learning approaches. Besides, it provides a technical toolkit by presenting evaluation benchmarks and a structured survey of the exemplary task of leakage detection and localization.