Problem of hydroelectric resources management: Numerical approach

Abstract

In this work we analyze an optimal control problem for a system of two hydroelectric power stations in cascade withreversible turbines. The objective is to optimize the proft of power production while respecting the system's restrictions. Some of theserestrictions translate into state constraints and the cost function is nonconvex. This increases the complexity of the optimal controlproblem. The problem is solved numerically and two different approaches are adopted. These approaches focus on global optimizationtechniques (Chen-Burer algorithm) and on a projection estimation refnement method (PERmethod). PERmethod is used as a techniqueto reduce the dimension of the problem. Results and execution time of the two procedures are compared. (c) 2016 NSP Natural Sciences Publishing Cor.