Optimization in water systems: a PSO approach

In this paper, Particle Swarm Optimization (PSO), a very well established evolutionary optimization technique, is applied to different problems of industrial interest in the Water Industry. The design of both Water Distribution and Wastewater Systems, the calibration of a network and the identification and detection of leaks can be easily addressed by means of PSO. Originally designed to deal with continuous variables, the PSO derivative we consider here overcomes two typical features of this optimization technique. For one thing, it is adapted to consider mixed integer-continuous optimization since the problems we tackle here involve the use of both continuous and discrete variables. For the other, one of the main drawbacks associated with PSO comes from the fact that it is difficult to keep good levels of population diversity and to balance local and global searches. This formulation is able to find optimum or near-optimum solutions much more efficiently and with considerably less computational effort because of the richer population diversity it introduces. Needing a low number of generations is a major advantage in real water distribution or wastewater systems, where costs and time constraints prohibit repeated runs of the algorithm and hydraulic evaluations. The performance of the variant herein proposed is investigated by applying the model (i) to the design of two water distribution networks, which are already traditional benchmark problems in the literature, namely the Hanoi new water distribution network and the New York Tunnel water supply system; (ii) to the design of a wastewater network and (iii) to the calibration and identification of leaks in a water distribution network. The obtained results show considerable improvements regarding both convergence characteristics and the quality of the final solutions.