Research On Fault-Tolerant Control Method For Time-Delay Systems Based On Adaptive Technique

In recent years,many researchers devote themselves to the study of time-delay systems due to the great economic value of time-delay systems for industry,military areas and robotics.In this thesis,the main purpose is to design a class of controllers for time-delay systems against actuator failures.The contributions of this thesis are summarized as follows:Firstly,construct a model on the condition that the actuator fault and mismatch external disturbance are taken into account.In order to greatly handle time-varying actuator failures,the two-step method is employed to deal with the issue.In the first part,an adaptive observer is given to estimate the actuator fault of the system.Here,the designed mechanism can not only increase the speed of fault estimation but also can eliminate the high frequency oscillation effect on the performance of the considered system.In the second part,the estimation error is treated as external disturbance,and then the corresponding controller is given to compensate for the estimated error so that the closed-loop system is stable with a prescribed performance index.Secondly,a new scheme is proposed in order to overcome the disadvantage of the two-step method.In order to cope with the time varying actuator failures,a novelty Lyapunov functional with a compensation term is employed,and the time-varying actuator failures of the considered system are handled by this compensation term.Besides,the theoretical results can degenerate into different situations by turning some weighting parameters of the compensation term.In addition,the proposed actuator fault estimation principle can not only increase the speed of fault estimation,but also eliminate the high frequency oscillation effect on the performance of the system.Compared with the two-step method,this processing method has the following advantages:1)Eliminate the restrictive rank condition.2)The controller is designed simply,which can be implemented in practice easily.3)When the system does not exist the external disturbance,the states of the system exponentially converges to zero.4)The theoretical results can degenerate into different situations by turning some weighting parameters of the compensation term which have wider application.Thirdly,the stochastic property of the system is also taken into account besides the considered actuator failures and the mismatch external disturbance.The designed controller can achieve stable even if the actuator failures and the external disturbance coexist.By combine the Lyapunov-Krasovskii method with some novel inequalities,some controller design conditions are obtained by solving the optimization problems.Finally,an example is given to show the effectiveness of the theory in this chapter.Concluding remarks are presented at the end of this thesis.Some open problems are also pointed out,which deserve further study.