Research On Model Predictive Fault-tolerant Control Technology For Hypersonic Vehicles Under Actuator Failures

Hypersonic vehicle is a kind of high-speed,high-precision aircraft with broad development prospects.As the important actuator of the hypersonic vehicle,the surfaces have great influence on the stability of the system.In this paper,the model predictive control-based fault tolerant controllers are proposed for the longitudinal dynamics of hypersonic vehicle with input constraints and control surface faults.The controllers ensure the stability of the control system while tracking the desired commands.Firstly,the background and significance of the project are expounded.Then,the development of hypersonic vehicle at home and abroad is introduced briefly.Furthermore,an overview of the the current status of the hypersonic vehicle longitudinal control and the model predictive control are described respectively.Besides,the research status of the fault tolerant control for the hypersonic vehicle is presented.Secondly,controllers based on generalized model predictive control are designed for both altitude and velocity subsystems with control input constraints.An adaptive generalized model predictive fault tolerant controller is designed for the altitude subsystem,which realizes the altitude tracking control with control surface fault.Through a large number of simulation studies,the ranges of controller parameters under typical flight conditions are concluded to ensure that the controller output satisfies the constraints of the system control input.Thirdly,based on the previous research,an adaptive generalized model predictive fault tolerant controller is proposed for the longitudinal multi-input multi-output(MIMO)nonlinear model of hypersonic vehicle in the presence of input constraints,the uncertainties of system parameters and unknown control surface faults.Based on the scheme proposed in Chapter 3 to solve the problem of control input constraints,aiming at hypersonic multiple input multiple output nonlinear system,the paper discusses how to choose controller parameters in the presence of uncertain system parameters.Through a large number of simulation experiments,it concludes the ranges of the controller parameter under different parametric uncertainties.Besides,it also explores the fault tolerant ability of the proposed controller.Finally,a fault-tolerant controller based on multi-model predictive control is designed for the the longitudinal dynamic model of the hypersonic vehicle with control surface fault.First,a discrete linear model set of longitudinal dynamics of hypersonic vehicle with control surface fault is established based on the theory of gap metric.Then,constrained model predictive controllers are established for each sub-model to achieve the equilibrium point of the linear model established by the system individually.The sub-controllers are swiched based on the dynamic pressure to realize the control target of tracking the desired commands gradually.