Font Size: a A A

Fault Diagnosis And Fault Tolerant Control For Sampled-data Flight Control Systems

Posted on:2014-12-29Degree:MasterType:Thesis
Country:ChinaCandidate:M ZhouFull Text:PDF
GTID:2252330422950647Subject:Control Science and Engineering
Abstract/Summary:PDF Full Text Request
With the development of science technology, the structures and functions ofaircraft are becoming more and more complex, which greatly increases theprobability of failure. And then with the growing requirements of security andreliability, the design of fault detection and fault tolerant control for aircraftcontrol systems has become more and more important. In recent years, computertechnology has been extended quickly and used in many control systems, almostall systems are using discrete time controllers, thus the study of sampled-datasystems receives more attention of many researchers. In this thesis, we will studythe fault detection and fault tolerant control of aircraft for sampled-data systems.First, fault detection and fault tolerant control problem for aircraft arebriefly introduced, include the classification and the development and theapplication of the technology. The thesis mainly introduces the robust faultreconstruction technology. Next introduce some basic knowledge.The aircraft continuous-time model is discretized by the theory ofdiscretization. And by considering the fault as an auxiliary state vector, anaugmented system is constructed. According to the theory of robust control andregional pole assignment we design a discrete-time robust proportional integralobserver (PIO) and compare it with the common PIO, the results demonstrate therobust PIO improve the robustness for abrupt fault and the poles can be located inthe desired regional. As the RPIO cannot eliminate the influence of disturba nce, arobust unknown input observer (UIO) are proposed, which eliminate thedisturbance of unknown input and improve the precision of the faultreconstruction. This paper gives a sufficient and necessary condition of the twoobservers and solves them based on the linear matrix inequality (LMI). Besides,the two observers designed for the augmented systems can not only estimate theactuator fault but also the sensor fault, they need not build the residual andevaluation threshold, do not affected by the unknown input and robust for abruptfault.As in many practical applications not all the states are needed, in order tomake the process of design simplicity and reduce the amount of calculation, thisthesis presents a discrete-time reduced-order observer for actuator faultestimation. This method overcomes the using conditions and restrictions on thephysical implementation of the common reduced-order observer.Finally this thesis researches the fault tolerant controller based on the faultreconstruction, first we introduce the design of passive fault tolerant control, andthen according to the control law reconfiguration design method, an active fault tolerant controller is proposed based on the state feedback, the state is based on theobserver. We compare the active fault tolerant controller with the passive faulttolerant controller. Through the simulation results we can see the advantages of thedesign method.
Keywords/Search Tags:flight control, sampled-data system, regional pole assignment, robustproportional integral observer (RPIO), robust unknown input observer (RUIO), reduced-order observer, fault tolerant control
PDF Full Text Request
Related items