Identifying water quality monitoring stations in a water supply system

Abstract

Despite good quality assurance and conformance to the standards at the treatment plants, water quality could vary considerably within the distribution network. As water flows through the pipe network, its quality undergoes various transformations due to many factors such as the properties of the finished water, pipe materials, water temperature, water age and low level of disinfectant residuals. Sampling and monitoring of water quality is, therefore, important to ensure that clean and safe water is transported to the consumers. In this paper, a model based on genetic algorithms and fuzzy logic was developed to identify locations of water quality monitoring stations in a water distribution system. While identifying the monitoring locations, multiple sources of water supply, water age and constituent concentration were considered. The developed model was applied on a hypothetical network and results indicate that monitoring stations are proposed at locations with maximum coverage of water supply within the network and maximum violation for average water age and constituent concentrations.