| The design of wavelet neural-networks flight control system for helicopter is studied in this dissertation, which involving several aspects about building flight dynamics nonlinear model for helicopter, inverse dynamic design for helicopter flight control system, compensating inverse error by wavelet neural-networks and maneuver flight simulation for helicopter.As different components of helicopter has the own motion characteristic, a group of nonlinear motion equations are established firstly, and then analyzed unloop motion characteristic for free helicopter and the infection of the dynamic couple and the control couple. This work supplied the physics concept for designing control system of helicopter.In the process of designing flight control system, the control loop of attitude angle and the control loop of collective angle are designed separately in this dissertation. The control loop of attitude angle is designed with dynamic inversion method, and is divided into fast loop and slow loop at two time scales. In the end, the result of simulation showed that the three channels of attitude angle are discoupled by dynamic inversion method.Wavelet neural-networks is used to compensate inverse error of dynamic inversion controller brought by error of model accuracy in this dissertation. The dissertation described the theory of wavelet in detail and presented the method of building wavelet function, presented the learning method and adjusting factor law of BP wavelet neural-networks, designed neural-networks controller. The result of simulation showed the controller could compensate dynamic inverse error.At the end of the dissertation, the fuzzy PD track tail controller is designed. The PD controller and controller of attitude angle are made up of flight control system of helicopter. Two swift and violent maneuver are simulated, the result of simulation showed that maneuver performance could be improved by using flight control system designed in the dissertation.
|
PDF Full Text Request