Researh On Longitudinal Flight Control Of Rotated Wing Aircraft During Transition

The tilt-wing aircraft studied in this paper is similar to the tilt-rotor aircraft.It is a new type of aircraft which combines the advantages of conventional fixed-wing aircraft and helicopter.The difference is that the structure design of tilt-wing aircraft uses the rotor and part of the fixed wing to tilt at the same time.The rotor is a non-variable-moment propeller with variable speed.It has simple structure and is more flexible.The advantage of this new type of aircraft is that the down-wash flow of the rotor can be avoided to hit the fixed wing in helicopter mode,which reduces the lift loss and enlarges the range in disguise.However,it also has some problems,such as the model is more complex and the actual control is more difficult.Especially when it is in transitional mode,it has more remarkable distinguishing features of time varying,strong coupling and non-linearity.This makes the whole flight control of tilt-wing aircraft more complex.Therefore,in order to make the tilt-wing aircraft fly safely and steadily in the transitional portion,it is the key and difficult point to face at present.This paper mainly studies the longitudinal flight control of tilt-wing aircraft in transition portion.Sliding mode control is essentially a control method to deal with non-linear problems.Its non-linearity is mainly manifested in the discontinuity of control.This method has great advantages in dealing with the problems of time-varying,coupling and under-actuated nonlinear systems.Therefore,on the basis of the sliding mode controller designed in this paper,RBF neural network is added to further improve the controller by utilizing its strong non-linear mapping ability and fault-tolerant ability.By comparing the above two simulation experiments,it can be concluded that the new controller makes up for the shortcomings of sliding mode variable structure control and further improves the control effect of the system because it combines the characteristics of the two control methods.The main work is as follows:1.Firstly,the motion characteristics of tilt-wing aircraft are analyzed,and the corresponding non-linear dynamics and kinematics equations are established.After the aerodynamic models of every section of the tilt-wing aircraft are got by the split method,the corresponding non-linear simulation model of the aircraft is constructed inSimulink.2.Then,under certain ideal assumptions,the appropriate control scheme of the transition section of the tilt-wing aircraft is adopted,and the corresponding sliding mode controller is designed to simulate the longitudinal flight control of the transition section of the tilt-wing aircraft model.The results show that the sliding mode control can deal with such time-varying,strong coupling and under-actuated non-linear problems,but there are still some problems such as chattering and over-dependence on model,which lead to low control accuracy.3.On this basis,RBF neural network is added to improve the designed controller,and the desired control results are obtained through simulation experiments.Finally,under the environment of MATLAB,the gust interference which may occur in the process of aircraft longitudinal control is reasonably analyzed.Experiments show that tilt-wing aircraft can still fly stably and safely with gust disturbance,which shows that the designed controller has good robustness.The results of this research lay a foundation for follow-up research on tilt-wing aircraft,and also have important significance for dealing with such time-varying,strong coupling and under-actuated nonlinear problems in the future.