Despacho de potência ótimo em sistemas elétricos usando uma abordagem com muitos objetivos

Abstract

This work presents an optimization methodology to solve the power dispatch problem considering four objectives: i) minimization of power losses, ii) minimization of the cost of real and reactive power dispatch, iii) maximization of system voltage stability, and iv) maximization of power factor on transmission network step-down transformers. The optimization methodology integrates the Non-dominated Sorting Genetic Algorithm III (NSGA-III) and an efficient algorithm to calculate the power flow in the network. The methodology of simultaneous exploration of real and reactive power dispatch (for the optimization of the four objectives) is novel since it has not yet been addressed in the literature. Regarding the mathematical model, the proposal to calculate the opportunity cost associated with the reactive power dispatch stands out. This methodology is tested both on the 30-bus and 57-bus IEEE systems, as well as on a 10351-bus Brazilian system. Efficient solutions with good convergence and diversity were obtained for each system, which demonstrated improvements in each objective. Several power dispatches of practical interest in the tested systems are analyzed and discussed. The results showed that the simultaneous dispatch of real and reactive power is a successful approach to find solutions that improve the performance of power systems in several criteria. In addition, the simultaneous dispatch of real and reactive power supports analysis of reactive power impacts, such as dispatch cost analysis and voltage stability.