Máxima capacidade de hospedagem de geração distribuída em sistemas multifásicos: uma abordagem considerando um modelo de otimização usando o método de injeção de correntes trifásico

Abstract

This work presents a novel deterministic methodology to determine the maximum distributed generation capacity in power distribution systems. In the proposed approach, an optimization problem is formulated to calculate the maximum amount of power to be accommodated in a power system without violating operational limits. For the proposed formulation, the current injection method equations are considered as equality constraints to compute the optimal power flow solution. Maximum and minimum limits for the voltage magnitudes are considered as inequality constraints, as well as the powers at each busbar and the reverse power flow at the utility substation. Also as a contribution of this research, a new methodology is proposed to determine the MCH in unbalanced multiphase systems in which, through modifications in the optimal power flow, the aim is to determine the maximum amount of power to be hosted in three-phase four-conductor system. For this, inequality constraints incorporate limits associated with the voltage magnitudes on each phase, voltage unbalance factor and neutral voltage. Equations of the three-phase four-conductor current injection method are treated as equality constraints. The optimization problem is solved via interior-point method. Additionally, in order to account for uncertainties in the proposed methods, the system loads are modeled using Gaussian probability density functions considering a statistical dispersion with respect to their nominal values in order to evaluate the impact of uncertainties on the load values in the MCH values. Computational simulations are conducted using two systems: the single-phase equivalent distribution system of the Federal University of Juiz de Fora, composed of 23 buses, and the unbalanced three-phase IEEE system of 33 buses. Several analyses are carried out, considering the generation units allocated dispersedly in the grid to be evaluated. Still, in order to validate the proposed methodologies, the results are compared to related works in the literature, proving the versatility and applicability of the methodologies proposed in this dissertation.