Fault Diagnosis And Fault-tolerant Control Of Complex Stochastic System In Networked Environment

The networked control system is a closed-loop feedback control system connected by system components via a shared communication network.With the rapid development of Internet technology,the cross-integration between network technology and modern in-dustrial production process is deepening.The vigorous development of scientific theory makes the networked control system actually become increasingly complex and usually possesses high non-linearity.The complex random system is an important component in the field of control systems,which can more accurately describe the existence of sudden structural change in network communication systems,spacecraft systems,and motion control systems,which has become a long-standing research hotspot and has attracted extensive attention from scholars.During the long-term and high-load operation of the system,various components will inevitably fail.The most common fault is the actuator failure,which often brings about system safety and reliability problems that may not be avoided,causing serious Economic loss.Therefore,fault diagnosis and fault-tolerant con-trol technology are used to compensate system failures and improve system performance.In recent years,it has become a key research topic at home and abroad,effective appli-cation and integration between research results and industrial production processes have been reached.The thesis takes the complex random system in the network communication envi-ronment as the research object,considers the communication constraints such as signal quantization,event triggering,data packet losses,actuator failure and saturation and actu-ator physical constraints,develops fault diagnosis filter design and fault tolerance control.The main research contents of this paper are summarized as follows:Firstly,the H_? filtering problem is investigated for a class of discrete-time Markov jumping nonlinear systems with partly unknown transition probabilities and subject to sensor saturation over unreliable communication.The description of researched plan-t includes global Lipschitz nonlinearities,state-dependent random noise and external-disturbance.A decomposition approach is used to deal with the characteristics of sensor saturation.Since the communication links between the plant and filter lack enough relia-bility,the effects of output quantization and data packet losses should both be considered.Compared with the control results obtained by using the traditional logarithmic quantizer,the proposed quantizer's parameter is on-line updating and the corresponding practical adjusting rule can ensure the dynamic performance of the controlled system.Among different operation modes,the cross coupling between system matrices and Lyapunov matrices is disposed by introducing proper slack matrix variables.The purpose of this work is to design a full-order filter based on incomplete output measurements in order to guarantee the stochastic stability of the estimation error.Precise expression of the filters and related analysis are depicted in this paper.Finally,a numerical simulation is provided to show the effectiveness of the designing filtering method.Secondly,this paper investigates the adaptive fault-tolerant control problem for a class of continuous-time Markoving jump systems with digital communication constraints,parameter uncertainties,disturbances and actuator faults.In this study,the exact informa-tion for actuator faults,disturbances and the time-varying stuck faults are totally unknown.The dynamical uniform quantizer is utilized to perform the design work and the mis-matched initializations at the coder and decoder sides are also considered.In this paper,a novel quantized adaptive fault-tolerant control design method is proposed to eliminate the effects of actuator faults,parameter uncertainties and disturbances.Moreover,it can be proved that the solutions of the overall closed-loop system are uniformly bounded,which is asymptotically stable almost surely.Finally,numerical examples are provided to verify the effectiveness of the new methodology.In addition,this paper investigates the adaptive fault-tolerant control problem of a class of Markov jumping systems with high frequency sampling.In this design,matched nonlinearity,unknown actuator degradation factor and unmeasurable states are consid-ered in a unified framework.The backstepping control method and sliding mode control method were used to stabilize the system.An adaptive sliding mode observer approach is developed to obtain the estimates of system state vector.In the next step,based on the state estimation,an integral-type sliding surface for the overall closed-loop Delta operator system is presented,a sliding mode control law is proposed to guarantee the stochastic stability of the fault closed-loop system.Finally,a simulation example is provided to verify the effectiveness of the designed fault-tolerant control method.In the end,this paper addresses an event-triggered adaptive fuzzy sliding mode con-trol for Ito stochastic systems with actuator failures.In this study,a novel event-triggered criterion and a new adaptive sliding mode control scheme are both applied to stabilize the resulting fault stochastic system,where the unknown nonlinearities is approximated by using the fuzzy mechanism.After appropriately estimating the loss of effectiveness of actuator faults,the proposed robust sliding mode controller is able to ensure that the trajectory finally moves in the sliding region and guarantee that the state variable of the closed-loop system is bounded stable and can be arbitrarily small.Finally,an inter-event time lower bound between two sampling points is derived and a practical example is il-lustrated to prove validity of the mentioned method.