Nonlinear Heuristic for Pump Operations

Abstract

A nonlinear heuristic is developed for use in obtaining least-cost pump-operation policies for multisource, multitank water-distribution systems. The proposed algorithm links a minimum-cost-constraint identification methodology with a network-simulation model in order to provide the resulting policy. The algorithm has the advantage of being computationally efficient while incorporating the nonlinear characteristics of the water-distribution network. In addition, the algorithm has the added advantage of providing several feasible solutions to the control problem, which then provides the system operator with increased flexibility in selecting a particular policy.