Ant Colony Optimization for Optimal Control of Pumps in Water Distribution Networks

Abstract

Reducing energy consumption of water distribution networks has never had more significance than today. The greatest energy savings can be obtained by careful scheduling of operation of pumps. Schedules can be defined either implicitly, in terms of other elements of the network such as tank levels, or explicitly by specifying the time during which each pump is on/off. The traditional representation of explicit schedules is a string of binary values with each bit representing pump on/off status during a particular time interval. In this paper a new explicit representation is presented. It is based on time controlled triggers, where the maximum number of pump switches is specified beforehand. In this representation a pump schedule is divided into a series of integers with each integer representing the number of hours for which a pump is active/inactive. This reduces the number of potential schedules (search space) compared to the binary representation. Ant colony optimization (ACO) is a stochastic meta-heuristic for combinatorial optimization problems that is inspired by the foraging behavior of some species of ants. In this paper, an application of the ACO framework was developed for the optimal scheduling of pumps. The proposed representation was adapted to an ant colony Optimization framework and solved for the optimal pump schedules. Minimization of electrical cost was considered as the objective, while satisfying system constraints. Instead of using a penalty function approach for constraint violations, constraint violations were ordered according to their importance and solutions were ranked based on this order. The proposed approach was tested on a small test network and on a large real-world network. Results are compared with those obtained using a simple genetic algorithm based on binary representation and a hybrid genetic algorithm that uses level-based triggers.