GA Based Multi-Objective Sensor Placement Approach for Contamination Detection in Water Distribution Networks

Abstract

Sensors are provided in water distribution networks to: (1) Assures quality of water delivered to consumers; (2) Early detection of contamination events so as to minimize its consequences. Most of the multi-objective techniques consider the second purpose and almost ignore the first purpose. In this study, a sensor placement problem is formulated to cover these two performance objectives through maximization of: (1) Demand coverage; and (2) Time-constrained detection likelihood. These two objectives are combined into a single objective by using weights. Several solutions are obtained by varying the weights of two objectives. Genetic Algorithm is used to obtain optimum sensor locations. The methodology is applied on a bench mark problem. The proposed formulation requires pre-specified value of time of detection as level of service that need to be judicially selected for the network.