Research On Adaptive Tracking And Safety Control Of Switched LPV Systems

With the advancement of science and technology,the complexity of the existing system is gradually increasing,and accordingly,people’s demand for the control accuracy of complex systems is also increasing.The traditional linear system model can no longer meet people’s control precision requirements for sophisticated and complex systems.Linear parameter-varying(LPV)systems were proposed because they own real-time measurable scheduling parameters and can model complex systems more accurately.However,when the parameters change in a large range,the control effect of the LPV systems is often unsatisfactory.For this reason,the switched LPV systems were proposed,and the parameter space is divided into several small subspaces to limit the variation range of the parameters of each subsystem.Further,the control effect is improved,and the degree of design freedom is improved also.Because of the characteristics of accurate modeling and easy control of switched LPV systems,it has been widely used in the control field.By constructing a time-varying parameter-dependent multiple-Lyapunov function method,this thesis studies the model reference adaptive tracking control problem of switched LPV systems with parameter uncertainties and the security control problem of switched LPV systems,in which the actuators and sensors encounter the deception attacks simultaneously.The main research contents are as follows:(1)This thesis firstly studies the stability of switched LPV systems with parameter uncertainties.By constructing time-varying positive definite matrices,a switching LPV controller with parameter projection adaptive laws and a dwell time switching law are designed so that the state of the switched LPV systems can achieve bounded tracking of the state of the reference model.(2)Considering both sensors and actuators suffer deception attacks at the same time during network transmission,the security control problem of switched LPV systems is studied.Each subsystem is allowed to be unstable.By designing a parameter-dependent switching adaptive controller and a state and parameter-triggered switching strategy with dwell time constraints,the security of the switched LPV systems under the deception attacks is guaranteed.(3)Taking the GE-90 aviation turbofan engine as an example,the theoretical methods proposed above are verified and analyzed by simulating and adjusting the controllable factors such as blade speed,core speed,and fuel flow.