Contribuições ao controle adaptativo robusto por modelo de referência

Abstract

This work proposes to combine RMRAC (Robust Model Reference Adaptive Control) control techniques to improve controller performance and ensure system robustness. Furthermore proposed the development of stability proofs by Lyapunov Theory. Initially, a direct RMRAC is applied, in the continuous time domain, based on a Gradient type parameter identification algorithm, which is designed by input-output approach. A Sliding Mode Control action type is also included, named RMRAC-SM. After this, a RMRAC algorithm is presented in discrete time, with stability analysis following the steps of (IOANNOU; TSAKALIS, 1986b). Adding up with previous algorithm, a Slide Mode Control action is inserted to the RMRAC, originating the RMRAC- SM, in discrete time. Finally, a RMRAC-SM controller combined with a Model Predictive Control (MPC) is developed. Stability proofs of the proposed controllers are presented, in continuous and discrete time that considered unmodeled plant dynamics. To corroborate the performance of the controller and discuss its robustness, simulation results are presented.