Observer Based Fault-tolerant Control For Markovian Jump Systems

With the development of modern industry,science and technology in recent years,the scale of various control systems has become more complex,intelligent and large-scale,which has required a higher standard to keep the system reliability,maintainability and safety.Otherwise,if the system suffers external interference or malfunction,it will affect the stability of the system and reduce the performance of the system,which lead to catastrophic accidents.In that case,the control system needs to detect and isolate system failures in a timely and effective manner,and ensure the stability of the closed-loop system,which can effectively improve the reliability and safety of the system.The observer-based fault-tolerant control technology studied in this paper can reconstruct the system state,diagnose whether the system has faults by residuals,and the stability of the closed-loop system is guaranteed by a fault-tolerant controller,which has an important theoretical significance and application value.Markovian jump systems,known as a special type of hybrid dynamic systems,which are systems consisting of several subsystems.Due to it can be used to describe systems that subjected to random abrupt and environmental changes,it has a very wide range of applications in the fields of economics,power and communications.In practice,Markovian jump systems may be affected by unknown inputs,such as actuator fault,sensor fault,or external interference,which will reduce the stability of the system.Therefore,the research of observer-based fault-tolerant control technology has great significance to ensure the safety and reliability of Markovian jump systems.The detail content of this thesis is as follows:1.The adaptive sliding mode observer-based fault-tolerant control is designed for a class of nonlinear Markovian jump systems with generally uncertain transition rates.The nonlinear function considered in the system satisfies the one-sided Lipschitz condition.An adaptive sliding mode observer is first constructed to reconstruct the system state and obtain the estimated value of the original system state.The observer will suppress the influence of unknown input on the system through adaptive sliding mode technology,without knowing any information of the unknown input.Based on the estimated value of the system state,a fault-tolerant control strategy is therefore proposed to ensure the random stability of the closed-loop system.In the end,a practical example is given to illustrate the effectiveness of the designed observer and controller.2.The Luenberger observer-based fault-tolerant control is designed for a class of stochastic Markovian jump systems with time-varying generally uncertain transition rates.These systems are subjected to mixed mode-dependent time delays and unknown input.To deal with the time-varying generally uncertain transition rates,a novel quantization mechanism is developed,thus lowering the conservatism of the proposed method.First,a Luenberger observer is designed to ensure the error dynamics being finite-time bounded,so as to achieve estimations of the original system states against the unknown input.Based on these estimated states,a fault-tolerant controller is then proposed to allow the closed-loop system to be finite-time bounded.Finally,a numerical example is brought to verify the of the feasibility proposed method.