Controle de corrente em conversores conectados à rede elétrica por meio de filtro-LCL utilizando o controle adaptativo simples com dinâmicas suplementares discretas

Abstract

In applications involving the connection of converters in the electric network by LCL filter, the Proportional Resonant (PR) controller are among the most used for the rejection of disturbances from the electric grid and to achive asymptotic tracking of sinusoidal references. However, due to the inherent uncertainties of the physical system, the practical application of PR requires, in most cases, to make adjustments to its gains through trial and error in order to obtain the desired performance. In order to compensate the uncertainties of the plant are used controllers with adaptive topologies, since these techniques have the ability to modify their gains in order to obtain the desired performance when the plant have parametric variation. An adaptive control techinique is the Simple Adaptive Control (SAC), a control topology derived from the Command Generator Tracker (CGT), and must be used in plants with passivity criteria of SPR or ASPR; using the LCL filter as plant, the application of the SAC is not possible, since the LCL filter is classified as a non-ASPR plant. In this way, it is necessary to use the SAC with addition of the suplementary dynamics, additional structure in order to extend the application range of the adaptive algorithm. Combining PR with this control technique yelds an algorithm that supports parametric variations without loss of performance and uses a compensator, with a design procedure already consolidated in the literature, for stabilization of the plant. This work aims to apply, in the discrete time domain, the adaptive algorithm SAC with supplementary dynamics, using as structure the PR controller, to control the output current of the LCL filter. The proposed structure needs to be studied and proven in order to guarantee the stability of the algorithm, thus, in addition to the proposed objective, the development of the SAC with supplementary dynamics for the application in the discrete time is realized. The development of the algorithm resulted in a structure similar to that presented in continuous time, requiring only the modification of the stabilization term of the adaptive algorithm (v(k)) and the addition of a constant gain (0 < f ll1) in the structure of the supplementary dynamics. The validation of the control technique applied to the LCL filter consisted of computational simulations and experimental applications considering the parameters of the LCL filter with nominal conditions and with parametric variations. From the performance point of view, even with resonance problems of the LCL filter, the adaptive strategy using the PR as supplementary dynamic obtained a shorter transient time and reduced tracking error when compared to the same PR controller applied alone, ie without the addition of the adaptive topology.