Observer-based H_∞ Fault Estimation And Accommodation

With the increasing demands of safety and reliability for equipments and control systems,active fault tolerant control for complex systems has been receiving more and more attention and has become one of the active research topics.Reviewing the developments over the past times,most of achievements are about separated design of fault diagnosis subsystems and fault tolerant controllers,which is the so-called open-loop design of active fault tolerant control system.On the other hand, un-modeled dynamics,parameters' perturbations,and nonlinear are unavoidable in real systems and,in these cases,it is always impossible to design fault diagnosis subsystems and fault tolerant controllers complete separately.From the view point of integrated design,fault estimation and accommodation provides an important scheme to active fault tolerant control and has attracted some research interests in recent years. To the best of author's knowledge,however,quite a few achievements have been obtained.Especially,no result of observer-based H_∞fault estimation and accommodation has been published in open literature,which motivates this study.This dissertation is supported in part by National Natural Science of China.The main purpose is to investigate observer-based H_∞fault estimator and fault tolerant control for a class of linear time invariant systems,model uncertain linear systems, and a class of nonlinear systems,respectively,by applying generalized internal model control architecture.The main works and contributions in the dissertation are given as follows:1.The problem of observer-based H_∞fault estimation and active fault tolerant control is studied for linear time invariant systems with L₂-norm bounded disturbances. By incorporating some prior knowledge of fault and the internal model principle into the design procedure,a reference fault model is first proposed to describe the dynamic behavior of the possible fault.Then the problem of fault estimation can be transformed as H_∞filtering problem and,by using the online fault estimation,a new H_∞controller in the generalized internal model control architecture is presented to realize fault accommodation.Both a sufficient solvable condition and a solution to the observer and controller gain matrices are derived in terms of linear matrix inequality. Through the simulation results of abrupt fault,ramp fault and sine wave fault,a numerical example is given to show the effectiveness of the proposed method.2.Based on the established reference fault model and the generalized internal model control architecture,the problem of active fault tolerant H_∞fault estimation and accommodation is formulated as a two-objective H_∞optimization and sufficient condition is derived in terms of certain matrix inequalities.An iterative linear matrix inequality algorithm is developed to implement the integrated design of an active fault tolerant H_∞controller and fault estimator.Through the simulation results of abrupt fault,ramp fault and sine wave fault,a numerical example is given to show the effectiveness of the proposed algorithm.3.The observer-based H_∞fault estimation and accommodation scheme incorporating reference fault model and generalized internal model control architecture is further extended to a class of non-linear systems.By applying Lyapunov approach,a sufficient condition on the existence of such an active fault tolerant control and a solution to the integrated design of observer and controller gain matrices are derived in terms of linear matrix inequality.A numerical example is also considered to illustrate the developed results.