Author(s)

J. Sousa, M. da Conceição Cunha & A. Sá Marques

Abstract

Water supply systems usually include a number of components (treatment plants, pipelines, pumping stations and distribution reservoirs), each of which performs a specific function. The interconnection between components requires the integrated operation of the systems. The operation of these systems is quite expensive because the energy consumption of some components is rather high. Both reasons (interconnection between components and high energy consumption) justify the development of tools to help their integrated operation while minizing costs. This paper presents a mixed-integer optimization model aimed at minimizing systems operation costs, over a 24-hour period. The model includes constraints related to the physical behaviour of the systems, and takes the variable electricity tariff into account. The model is solved using two different methods: \“Branch-and-Bound” (an exact method) and \“Simulated Annealing” (a random search method), In order to compare the efficiency and effectiveness of the two methods, the model is applied to a hypothetical example. Even though the model has not yet been tested in practice, authors are convinced that the use of this type of tool could be of great help in the operation of water distribution systems, allowing significant savings in energy costs.

Keywords