The Study Of Self-Repairing Flight Control Technology

While aircraft flying, if actuator assembly as elevator and aileron and rudder are locked or control surfaces are damaged accidentally, it will lead to serious reduction of flight quality, even to damage the stability of flight control system and cause prang of the aircraft. This thesis takes a certain model of aircraft as the plant. The purpose is to study the self-repairing flight control technology. When the actuator assembly fault or fighting damnification of aircraft appear, the technology will automate the stabilization and resuming process so as to avoid occur of accident.The main contributions of this thesis:First, the nonlinear six degree-of-freedom model is established. A great deal of aerodynamics and moment parameters are processed by cubic spline interpolation under Matlab and a flight simulation model of normal aircraft and fault aircraft are established.Second, neural network dynamic inverse systems of pitch channel and tumbling channel and yaw channel of airplane are established based on combination of the neural network ability of approximating unknown nonlinear systems and the inverse theory, which overcomes the abuse of neglecting nonlinear relationship that can't be neglected or replacing it with linear relationship in using traditional control method to design the inverse model.Third, self-repairing control of some kind of fault states of plane is studied by using the plan of model reference adaptive tracking self-repairing control for nonlinear systems based on neural network dynamic inversion, which has solved the problem of weak robust when uses linear control theory to control the complicated nonlinear systems like airplane. Self-organizing fuzzy cerebellar model articulation controller neural network is used to adaptively regulate the error in the plant inversion which may be caused by modeling uncertainties and disturbances in order to achieve prospective control effect and carry self-repair point.Finally, the real time expectation is proved by embedding control law program of the normal aircraft and fault aircraft self-repairing to the simulation flatform.