Multi-objective Coverage-based ACO Model for Quality Monitoring in Large Water Networks

A numerical procedure is presented for the optimization of the position of water quality monitoring stations in a pressurized water distribution system (WDS). The procedure is based on the choice of the set of sampling stations which maximizes the monitored volume of water while keeping the number of stations at minimum. The optimization model is formulated in terms of integer programming, and the solution of the mathematical problem is efficiently approximated by means of a multi-objective multi-colony ant algorithm. A built-in routine is developed for calculation of the water fraction matrix and integrated into the general modeling structure to facilitate data entry and storage to minimize problems associated with water fraction matrix determination for varying scenarios and coverage criteria for any scenario. The proposed methodology is very robust in analyzing the effects of different scenarios and/or number of potential monitoring stations by eliminating the need of employing an off-line routine for coverage matrix identification. Robustness, ease of generalization, multi-objective nature, and computational efficiency are the main characteristics and novelty of the proposed approach. Monitoring stations are optimally located in a large-scale real-world network with 104 nodes and multiple demands using the proposed ACO models. The set of non-dominated solutions forming the Pareto front for a number of monitoring stations and the total coverage of the system are also presented.