Controle MPC robusto aplicado ao conversor Boost CCTE otimizado por inequações matriciais lineares

Abstract

This work proposes a MPC predictive control strategy applied to a three-state switching cell boost (3SSC) converter that gives more simplicity and systematization to the control design steps and analysis. The integral gain controller can adjusted, eliminating the re-design the weighting matrices. Besides, the stability elipsoid theory are used to simplify the stability analysis of converter. The proposed MPC begins using the linear time varying (LTV) model plant. The parameters variations are considered polytopic uncertainties using linear matrix inequalities approach (LMIs) with relaxations. Theoretical and experimental aspects are applied and analyzed to a 1 kW 3SSC boost converter with voltage input and load uncertainties. To 500 W experimental test, it is considered load and voltage input steps variations. To 1000 W experimental test, it is only considered load steps variation. Also, the proposed MPC-LMI is compared with a classical LQI known in the literature. Theoretical-experimental aspects are applied and analyzed in a 1 kW converter with uncertainties in the input voltage and load. For the experiments, the voltage and load step tests for the 500 W project and a load step test for the 1000 W one were considered. In addition, to establish a performance reference, the proposed MPC is compared with the classic LQR controller known in the literature. The proposed control strategy presents advantages considering the variations of the model due to load tests in DC-DC 3SSC boost applications, being a feasible proposal in solving problems for high-performance converters.