Analysis And Design Of Event-triggered Controller Based On Interval Type-2 T-S Fuzzy Network Systems

The systems appeared in real world are almost nonlinear one with complex structures,which is difficult to be modeled accurately.The interval type-2 fuzzy model has the advantages in dealing with the parameter uncertainties of membership function involved in type-1 fuzzy model and reducing computational complexity of type-2 fuzzy system,which makes it one of the powerful tools to handle nonlinear systems.Network control system is widely used in science and engineering for its high efficiency and resource sharing.Based on the existing work,aiming at networked control system,event trigger controller under fading channel as well as network attack are investigated for interval type-2 fuzzy systems.The main contributions of this dissertation are summarized as follows:Part ?: The event-based state-feedback control problem is addressed for interval type-2fuzzy systems subject to the fading channel.For saving communication resources,a dynamic event-triggered mechanism is utilized to determine the data transmission from sensors to controller.A time-varying random process is employed to characterize the fading phenomenon in the unpredictable communication network.A nonparallel distribution compensation interval type-2 fuzzy controller is synthesized and its number of rules and membership functions can be freely selected to the effects of channel fading.By taking the global membership boundary information into stability analysis,the membership-function-dependent analysis method is employed to handle these imperfectly matched membership functions and to obtain relaxed criteria.Besides,sufficient criteria are obtained so that the resulting closed-loop interval type-2 fuzzy system can achieve stochastic stability despite fading measurements.Finally,the effectiveness and superiority of the proposed method are illustrated by a numerical example.Part ?: The resilient event-triggered control problem is studied for interval type-2 networked systems subject to randomly occurring deception attacks.First,a new memory eventtriggered scheme membership functions is proposed to reduce the utilization of communication resources while maintain the desired system performance.Different from some existing eventtriggered schemes,some recently released packets are firstly utilized in the proposed memory event-triggered scheme,which provides a better flexibility to improve the system dynamics.Then a randomly occurring deception attack model is employed by considering the security problem of the networked control systems.Considering the effects of memory event trigger and deception attack,a new type of networked control system model is constructed and the corresponding memory state-feedback controller is designed.Thus the sufficient conditions for asymptotic stability of the systems are derived by using a Lyapunov functional method.Finally,the obtained results are verified through a numerical example,which demonstrates the effectiveness of the proposed methods.Part ?: The security control problem is investigated for the networked control system subjected to denial of service(DoS)attacks.Different from previous work,the influence of packet dropouts due to DoS attacks are treated as a uncertainty of triggering condition and a novel resilient triggering strategy is proposed.Under the resilient triggering strategy,the event-based security controller is designed while the DoS-based security performance is preserved.At last,the simulation results show that the proposed resilient triggering strategy is resilient to DoS attacks and guarantees the security performance.