Dynamic recurrent neural networks for stable adaptive control of w ing rock motion

摘要

Wing Rock or limit cycle oscillation (LCO) in an aircraft is often caused by aerodynamic nonlinearities or mechanical hystersis. A dynamic recurrent radial basis function neural networks (DR-RBF) are proposed for modelling the nonlinear hysteresis. The concept based on Preisach hystersis model is used in the design of neural networks. The structure and memory mehanism in the hysteresis mode, is analysogous to parallel connectivity and memory formation in neural networks. Adaptive control law based on minimisation of the energy function ot ensure stability in the Lyapunov sense is presented.