Abstract of Thesis presented at COPPE/UFRJ as a partial fulfillment of the requirements for the degree of Master of Science (M.Sc.)

Control of Parameters in Meta-Heuristics for Hydrothermal Coordination in the Horizon of Medium Period

Karla Crispina Augusta Roberto

March/2009

Advisor:

Alexandre Pinto Alves da Silva

Department:

Eletrical Engineering

      This work aims at solving the problem of hydrothermal coordination for the horizon of medium term. With that in mind evolutionary strategies and fuzzy adaptative simulated annealing meta-heuristics have been selected, motivated by their capacities of exploring the search space and adapting their control parameters. Therefore, such meta-heuristics are analyzed with the intention of verifying if they are efficient tools for the solution of the problem of interest considering the size of the examined test system. The test system is composed of seven hydroelectric power stations and six thermal units. Simulations have taken into account, deterministic river flows and multi-scenarios flow. The tests with evolutionary strategies have investigated eight configurations seeking to the autoadaptation of some parameters of search control. The fuzzy adaptative simulated annealing has been tested in two versions, depending on the river flows been assumed deterministic or not. The obtained results are analyzed, compared and exposed graphically. It has been concluded that fuzzy adaptative simulated annealing seems to be more appropriate in comparison with the other investigated alternatives. However, the evolutionary strategies, in spite of requiring more investigation, have the advantage of being easier to implement.