Pressure and Water Quality Integrated Sensor Placement Considering Leakage and Contamination Intrusion within Water Distribution Systems

Abstract

Integrated sensors are installed in water distribution systems for the real-time detection of pipe leakage and contamination. We design a novel strategy for sensor placement to monitor leakages and contaminant intrusion, ensuring stability and sanitation. First, the fuzzy C-means algorithm divides junctions into different classes. Second, various leakage events and contaminant intrusion events are generated. Then, one of the junctions in each cluster is selected randomly for the integrated sensors that experience these events to acquire the global identification metric values. The events are used as optimization objectives for deploying sensors by the evolutionary algorithm (EA) module. Different sensor failure scenarios are also generated and tested by the EA module. A sensor ranking method is used to select the optimal locations of sensors after considering the probability of each scenario occurring. Furthermore, two cases are employed to verify the reliability of this method, and two models─leakage zone identification and the water quality sensor placement strategy─are used for comparison. It is confirmed that the objective functions and computation time of the proposed model are superior to those of the other models, and the total number of detections and the detection stability are improved.