Research On Modeling And Tracking Control Technology For Hypersonic Vehicle During Cruise Phase

Hypersonic vehicle has significant military strategic value and wide civil prospects,and it has become the aerospace research focus in the 21 st century. There is a strong coupling among the aerodynamics, propulsion and structure due to the application of lightweight materials and airframe/engine integration technology, and the dynamics model of hypersonic vehicle is highly nonlinear. At the same time, uncertainties and unknown disturbances exist in the model because of the harsh flight conditions and large flight envelope. All the above factors have brought great challenge to the disign of the control system for a hypersonic vehicle. In this dissertation, robust tracking control problems are studied for commands tracking during cruise flight phase, based on the generic hypersonic concept model.As the preparation for the following control system development, the six degrees of freedom motion equations of hypersonic vehicle are deduced using Newton-Euler method,and the longitudinal model is acquired under certain simplifying assumptions. Then, the control models for speed and altitude tracking, longitudinal overload and speed tracking and the angle of attack tracking are obtained.In the speed and altitude tracking problem, feedback linearization method is used to obtain the state feedback exact linearized model for the hypersonic vehicle longitudinal nonlinear model. And then dynamic inverse control is used as inner loop controller to obtain nonlinear decoupling between inputs/outputs. When there are matching uncertainties,outer loop controller is designed adopting sliding mode control technology. First, a classic sliding mode control law is designed; then an index time-varying integral sliding mode control law is proposed to eliminate the reach stage of sliding mode and achieve global robust. The simulation results show that the speed and altitude commands are accurately tracked adopting the designed control laws, with good decoupling capability between inputs/outputs and excellent robustness against matching uncertainties. The simulation results show that the dynamic inverse controller achieves good decoupling capability between inputs/outputs and the speed and altitude commands are accurately tracked adopting the designed control laws, with excellent suppression of disturbance due to matching uncertainties. When there are non-matching uncertainties, an adaptive backstepping variable structure control law is designed by the combination of backstepping,sliding mode and adaptive law. Then, in order to improve “differential explosions” problem in the classical backstepping method, first order low-pass filters are adopted and an adaptive dynamic surface control law is designed. The simulation results show accurate commands tracking ability and good suppression of disturbance due to non-matching uncertainties.In the longitudinal overload tracking problem, the longitudinal overload contol model is modified to reach full vector relative degree when state feedback linearization is taking.There are two modifies for the model, the first is to ignore the height in the states and ignore the partial derivatives of V?, ?? and ?? to h, the other is to take the integral of longitudinal overload as a virtual output during the feedback linearization. Then, based on the linearized model, a quadratic optimal controller and a sliding mode controller are designed to track both the longitudinal overload command and speed command. The simulation results verify the feasibility and robustness of the designed control system.In the angle of attack tracking problem, Jacobi method is used to obtain Jacobi linear model, and based on the Jacobi linear model, a model reference adaptive sliding mode control law is designed to track the command of angle of attack. Then, by ignoring the height in the states, ignoring the change of the acceleration of gravity with height and considering the integral of the angle of attack as a virtual output, the angle of attack tracking model reaches full vector relative degree when state feedback linearization is taking. Based on the feedback linearized model, an adaptive sliding mode control law is proposed to track the commands of both the angle of attack and flight speed. Considering parameter uncertainties, simulation is taken to verify the tracking capability for command signals and disturbance suppression ability.