Research On The Aircraft Control Method For The Improvement Of The Comprehensive Control Performance

The attitude tracking control of modern high-speed aircraft is a hot topic in the field of flight control.With the rapid development of related fields,the flight environment of high-speed aircraft is becoming more and more complicated,and the flight process is filled with stronger nonlinearity and uncertainty,which makes it difficult for the conventional control methods to achieve satisfactory control performance.At the same time,more and more new flight missions have put forward higher requirements for the comprehensive control performance of the flight control system: in the complex flight environment,the system is supposed to track the flight attitude commands faster and more accurately while maintaining its stability all the time.In order to deal with these new challenges,aiming at the attitude tracking control problem of the modern high-speed aircraft in the reentry flight segment,this paper studies the methods to improve the comprehensive control performance of the aircraft in the uncertain flight environment.The work done mainly includes:Aiming at a generic high-speed aircraft,the aerodynamic of the aircraft is analyzed,and the nonlinear six-degree-of-freedom motion equations of the aircraft are derived,as well as the corresponding small-deviation linear model and affine nonlinear control model,and a complete flight numerical simulation platform is built.From the perspective of control performance assessment,a gain-adaptive control scheme based on the online performance assessment is designed and validated.Firstly,multi-level linear correction controllers with different control performances are designed,and the network training data set of the flight state and the corresponding control performance level is established through offline simulation,and then the trained neural network is used to monitor the actual flight state online and evaluate the system in real time.Finally,based on the performance evaluation results,the controller gain is adaptively adjusted to improve the comprehensive control performance of the aircraft.From the perspective of system model identificat ion,an adaptive control scheme based on the instantaneous linearization of the neural network is designed and validated.Firstly,the dynamic characteristic of the aircraft is identified offline and online by a neural network,then the identification network is instantaneously linearized to obtain the equivalent linear model of the system at that moment,and finally based on the linear model and the generalized minimum variance control index,the controller parameters are adaptively adjusted to improve the comprehensive control performance of the aircraft.From the perspective of nonlinear robust control,a sliding mode control scheme based on the improved prescribed performance control is designed and validated.Firstly,the performance constraint and error transformation schemes are improved on the basis of prescribed performance control theory,and a control strategy for restarting performance constraints is proposed.Then,a double-loop sliding mode controller is designed based on the virtual control errors,so that the system can always operate in accordance with the preseted control performance.