Multi-objective sensor placement for leakage detection and localization in water distribution networks

Abstract

A multi-objective approach for sensor placement in water distribution networks (WDNs) is presented in this paper. The performance of leakage detection and localization methods highly depends on the amount, type and placement of the sensors installed in the network. Sensor placement is a challenging task considering the demand uncertainties, measurement noise, and varying leakage size. While some sensor placement methods have been designed to maximize leakage detection, others have been developed for leakage localization. In practice, both tasks are relevant. Thus, the goal of obtaining a sensor configuration that establishes a compromise between leak detection and localization errors is considered here. The sensor placement task is then formulated as a multi-objective optimization problem. A real-life and large-scale network that represents a version of the distribution system of the Modena city is used as a case study with synthetically generated data. The feasibility of the proposal is demonstrated for scenarios with demand uncertainty, measurements noise and varying leakage size.