Study On The Algorithms Of Adaptive Fault Tolerant Control With Partial Actuator Faults

Modern control systems trend to be more scaled and more complicated withdevelopment of science and technology and requirements of practical application.It will result in seriously damage to operators or machines in presence of faults insuch systems. Strong reliability and security mean that the safety of operators andmore profits can be guaranteed. Fault tolerant control (FTC) schemes compensateinfluence of faults automatically in presence of faults such that stability can beguaranteed and dynamic performance can be recovered as possible in the post-faultsystems, and the systems run stable. Hence, FTC schemes provide new ways toimprove the reliability and security of complex systems.Originally, FTC is used in aerospace field, such as reliable flights control, etc.And gradually, its application is enlarged into other fields, for example, industryprocess control, which is significant. In early days, the faulty actuators are thrownoff ignoring their remaining actuation and only healthy actuators are used for faulttolerance, which is not always brain way. Because it will reduce design freedomsand system driving power ignoring the remaining actuation in faulty actuators. Andif possible, the remaining actuations should be used for fault tolerance via actuationreconfiguration.It is necessary that redundancy exist in systems for fault tolerance. Namely,There exists the actuation redundancy for actuator fault tolerance, existsmeasurement redundancy for sensor fault tolerant. Actuators are subject to be inpresence of faults for consisting operation. Fault tolerance for actuators should beconsidered more carefully.The task of FTC systems design is how to such redundancy. Generallyspeaking, depending on how redundancy is being utilized, fault tolerant controlsystems can be categorized into passive and active types. In active fault tolerantcontrol, active fault tolerant control systems rely on an adaptive algorithm totracking faults or a Fault Detection and Isolation (FDI) scheme to detect theoccurrence of faults in the system to obtain information on the faults, such as thesource and seriousness of the faults. Based on the FDI information, a faultaccommodation scheme can then be designed to maintain a certain degree ofcontrol performance of the post-fault system.Adaptive control as an effective method can solve the problem of the systemswith uncertainty and robustness, so this control method has a promising foregroundin fault tolerant control fields, especially for some systems focusing on their safetysuch as airplane system. When something has been wrong with the actuator orsensor of the system, where nothing about the failure is known previously, adaptivecontrol scheme is applied to the system and always can get satisfied performance,even under the condition that failure pattern is unknown.Model Reference Adaptive Controller (MRAC) based on the Lyapunovmethod as actuator failure compensator is designed for a class of continuous-time,multi-input and single-output linear time-invariant complex plants. Global stabilityis proved in theory for the design of the adaptive control law. The control task is todesign a controller and accommodate the rest actuators in order to attain the desiredperformance under the condition that failure parameters are unknown. Theparameters of the system and failure pattern need to be analyzed and something ofthe desired controller should be known in order to design the adaptive controlscheme. The plant model match parameters are discussed before and after thefailure happens, so the designed controller can adaptively regulate arbitraryactuator failure. According to the states and output of the reference modelrespectively, adaptive control schemes are presented. Adaptively parameters of thestate feedback controller are regulated continuously, so ultimately the states andoutput both could follow after the given reference signals. Simulation results for alinearized airplane model with actuator failure are shown to verify the efficiency ofthe designed adaptive methods.In this thesis, the adaptive state-feedback based fault tolerant control schemes,incorporated reference models with single input and multiple inputs, are deeplyinvestigated. The main researches are arranged as follows.1.The background and significance of the research are introduced. And thecurrent developing situation of fault tolerant control is synthesized. A case ofclassification and implementation is studied in detail.2.Typical fault models of control systems are analyzed, and the fault modelfor our special purpose is introduced, in which the partial actuation faults areconsidered. A general design procedure of reference model based adaptive controlalgorithm, taking into account Lyapunov stability theory, is mentioned here.3.Considering a reference model with single input, a model for partialactuation faults with effectiveness factors is built, and the associating referencemodel based adaptive control law is designed. The preconditions for control lawexistence are given firstly. Subsequently, the faulty actuator effectiveness factorsare estimated on-line. Furthermore, the actuator reconfiguration is performed in theclosed-loop post-fault system. The numerical simulation results verified theeffectiveness of the proposed designed. Namely, the tracing to reference states issuccessful.4.Considering a reference model with multiple inputs and a single output, amodel for partial actuation faults with effectiveness factors is built, and theassociating reference model based adaptive control law is designed. Thepreconditions for control law existence are given firstly. Subsequently, the faultyactuator effectiveness factors are estimated on-line. Furthermore, the actuatorreconfiguration is performed in the closed-loop post-fault system. The proposedcontroller adjusts the performance of the post-fault control systems to the desireobject. The numerical simulation results verified the effectiveness of the proposeddesigned. Namely, the tracing to reference states is successful even though withoutany prior information of actuator faults.5.The conclusion of the whole thesis is provided. And research points in thefuture is pointed out.