Multi-objective model for optimal sensor placement in water distribution systems considering contamination probability variation-based contaminant impact

Contaminant intrusion in water distribution networks affects the safety of residential drinking water, and sensors are placed to detect a contaminant intrusion. As the number and location of sensors directly affect the detection efficiency, several studies focused on sensor placement optimization. Sensor placement optimization is an important mitigation measure, which is necessary to reduce the serious consequences of contaminant intrusion as different nodes impact differently on the water distribution networks after contaminant intrusion, that is, the risk levels of contaminant intrusion at nodes are different. Existing studies have considered changes in the probability of node contamination, but this probability is primarily based on the properties of the node itself. This study proposes a multi-objective sensor placement optimization method based on contamination risks, which involves the following steps. (1) Four different types of contamination probabilities are defined in terms of the impact of the contamination events. This is followed by solving a multi-objective optimization problem to obtain the Pareto fronts that are based on a variety of assumptions for the contamination probability distribution. (2) The Pareto fronts are ranked and clustered through a multi-criteria decision analysis, which leads to the optimal scheme in each cluster under different preferences. (3) The optimal schemes are compared in terms of the impact of the contaminant intrusion on the network. The proposed method is empirically evaluated by using the D-town network model. The results reveal that when node-to-node variation in the contamination probability is enabled, each sensor placement scheme exhibits a high true detection rate of contamination. The contamination probability based on the number of affected pipelines can yield optimal sensor placement for a fixed number of sensors to minimize the impact of contaminant intrusion on the pipeline network.