Event-Triggered Muiti-Objective Tiny Fault Robust Observer Design

With the gradual development of automatic control technology,modem control systems are gradually becoming more complex and refined,as well as networked.The safety and stablization of system have become an increasingly important indicator.This paper is mainly aimed to solve the problem of discretetime sensor networked system and discrete-time linear timevariant systems fault detection considering the networked transmission constraints.The methods of zonotope based and event-triggered based multi-objective fault detection observer design are given,which achieve the disturbance robust fault detection.The whole design scheme of tiny fault robust observer is unfolded as:First of all,considering that the effects of noise signals and tiny faults on the residual signal are difficult to distinguish directly,disturbance robust and fault sensitive design is required to make the fault signal more clearly reflected on the residual signal;And then,the range of residual value is estimate by the designed zonotope based algorithm.The fault detection logic is given to improve detection rate.In the mean time,in order to solve the constraint of communication bandwidth and improve communication efficiency,an adapted event-triggered mechanism is introduced to improve the effectiveness of transmission.Finally,the observer design of each research object is verified through simulation experiments.The main innovations of this study are:1.The multi-objective fault detection gain are designed base on the proposed disturbance robust and fault sensitive index,which makes the residual system robust to disturbance and sensitive to fault.2.The satic event-triggered mechanism and adaptively dynamic eventtriggered mechanism are designed for discrete-time sensor networked systems and linear parameter varing systems to reduce the total amount of data transmission.3.It is assumed that the peak value of the disturbance signal is bounded in this study.The zonotopic algorithm is designed to calculates the upper and lower bounds of residual signal.In turn,the number of false detections and missed faults can be reduced.