Investigação de técnicas de controle modernas para mitigar instabilidade em sistemas de micro rede CC

Abstract

The DC microgrids, it has gained prominence due to their high efficiency and also due to the growth of renewable distribution systems such as solar and wind generation. The systems are composed of several switching devices, used for the regular level of voltage between sources and loads. These devices, when are strictly regulated, present themselves as constant power loads (CPL) and can lead to instability in the system due to their negative incremental impedance effect, which reduces system damping and may cause intability in the system. There are several analysis criteria to verify the stability of DC micro-grids, however, these criteria are not yet used to design a contorl methodology that guarantees stability and performance of the system. In this sense, this work presents a proposal for robust state feedback control based on LMI. The proposal is formulated as an optimization problem, where the cost function is to be minimized and obtained through the Middlebrook analysis criterion. In this way, four optimization problems are formulated, varying the restriction criteria of each one. These constraints are modified from the indices H_2, H_infinite and convex region is delimited for pole allocation applied in two types of mathematical modeling: the system model using power as a perturbation and a model of parametric uncertainty. The proposed controllers were evaluated through simulated and experimental tests and are compared with the classical Lyapunov control method. In the tests of input voltage variation and power variation of the CPL, it was observed that the proposed methods, when using modeling of the CPL power variation as a disturbance, is better than the methodology that use the CPL power variation as parametric uncertainties. All methods ensured stability and performance and outperforms the Lyapunov's method.