Implementação de controle preditivo para supressão ativa de flutter em um aerofólio na presença de restrições

Abstract

This work presents an AFS methodology for flutter suppression, given the fact that this phenomenon’s unstable nature can lead to catastrophic structural failure. Particularly, it is adopted a model-based predictive control method with two modes in the presence of control and control increment constraints. The system is modeled based on a typical section airfoil with a flap control surface in the trailing edge. The aeroelastic forces are modeled and the unsteady aerodynamic loads are approximated via rational functions sum. Furthermore, with the addition of delay terms in the equations of motion, a state space linear model is obtained. Simulations of the system without AFS (i.e. in open loop) and at flutter speed show that the output with a non-zero initial condition is a sustained oscillation, given the existence of marginally stable poles. In closed loop simulations, the conventional states feedback is capable of suppressing flutter. Nonetheless, in the presence of constraints, the LQR is insufficient, considering it is not possible to stabilize the system with input saturation. Even so, the results showed that the dual mode MPC was successful, since it is possible to eliminate flutter with control and control increment constraints. On top of that, the system’s aeroelastic effect is amplified, with the controller being capable of actuating in above flutter speeds. It is verified that, with the predictive control based AFS usage, the oscillation speed can be raised by up to 23 % in relation to the flutter speed in open loop.