Adaptive and neural control of a wing section using leading- and trailing-edge surfaces

摘要

This paper treats the question of control of nonlinear aeroelastic responses of a prototypical wing section with structural nonlinearity using leading- and trailing-edge control surfaces. It is assumed that all the aerodynamic, structural and inertia parameters are unknown to the designer. As such the limitation of a recent control design reported in the literature, which requires complete knowledge of aerodynamic derivatives and inertia parameters, is removed. An adaptive controller and a neural control system are designed for the trajectory control of the plunge displacement and pitch angle. For the derivation of the adaptive control law, a linearly parameterized model is used but the neural controller is designed by treating the stiffening-type structural nonlinearity as an unstructured function (not parameterizable). It is shown that the adaptive and neural controllers accomplish trajectory control in the closed-loop system. Simulation results are presented which show that these controllers are effective in regulating the nonlinear responses to the origin in the state space in spite of large model uncertainties. Moreover unlike the model with a single trailing-edge surface, two control surfaces provide flexibility in shaping both the plunge and pitch responses.