Research On The Methods Of Fault Diagnosis In Finite Frequency Domains For Linear Uncertain Systems

With the increasing demand for reliability and safety for modern control systems,fault diagnosis technology provides a new way to improve the safety and reliability of complex system?The problem of fault detection is studied for linear uncertain system model.An adaptive mechanism is introduced to construct a new fault detection(FD)filter with varying gains for the fault detection in the finite frequency domains.An event triggering mechanism is introduced to study the fault detection problem in finite frequency domains.The main innovations of the thesis are listed as follows:A fault detection filter design approach is developed in finite frequency domains for multi-delay systems subject to affine uncertainties.An adaptive mechanism is introduced to construct a new FD filter with varying gains,and the H? tracking method is used to generate the residual signal.By defining appropriate Lyapunov functional,the performance analysis is firstly conducted in finite frequency domains via Parseval's theorem.Subsequently,the FD filter design conditions are derived based on the result of performance analysis.Moreover,it is shown that,for the considered multi-delay uncertain systems,the FD filter with varying gains can achieve better FD performances than the existing ones with fixed gain.An event-based FD filter design approach is developed in finite frequency domains for affine uncertain systems.An event-trigger is introduced to release sampled measurement outputs only when the event condition is violated.Under the event triggered scheme and the network transmission delay,the event-based fault detection model is constructed.A new bounded real lemma(BRL)is established by using the Lyapunov-Krasovskii functional theory and Parseval's theorem to describe the finite frequency disturbance attenuation performance.Similar to the event-based H?performance analysis,the H? FD scheme is applied to maximize the sensitivity of faults.Then,convex FD filter design conditions are obtained via an invertible linear transformation of the filtering error system.Moreover,the proposed approach can be reduced to the existing method in full frequency domains and achieve better FD performances.Finally,a simulation example on a ground vehicle is given to illustrate the advantages of the proposed theoretical result.