Dissertation
Dissertation > Mathematical sciences and chemical > Mathematics > Control theory, information theory ( mathematical theory ) > Cybernetics ( control theory, mathematical theory )

Robust Fault Tolerant Control for Networked Control Systems

Author HeTao
Tutor ZhangQingLing
School Northeastern University
Course Operational Research and Cybernetics
Keywords networked control systems robust fault-tolerant control linear matrix inequality time-delay uncertainty Markovian characteristics
CLC O231
Type Master's thesis
Year 2010
Downloads 36
Quotes 0
Download Dissertation

Firstly, some basic issues, the current research and development trend on networked control systems (NCSs) are introduced in this thesis, this thesis explain the use of fault-tolerant control theory to study the method based on networked control systems.Secondly, considering the existence of time delay and packet dropout, networked control systems is modeled as asynchronous dynamical systems. Based on the unilateral and bilateral network, the problem of robust fault-tolerant control of NCSs with time delay and packet dropout is studied, and the sufficient conditions for the asymptotic stability of the system with the sensor or actuator failure is presented, The approach of controller design is given through the linear matrix inequality (LMI).Thirdly, taking the impact of uncertain parameter time delay and uncertain disturbance into account, deducting a sufficient condition for closed-loop networked control systems with uncertainty against actuar or sensor failures was subjected to the robust integrity. The design approach of controller were also given, and show the correctness and validity of fault-tolerant control with simulation examples.Finally, the model of NCSs with discrete-time plant is proposed. Then, the definition of "the quantity of packet-disordering" is given; the random network-induced time delay and the quantity of continuous packet-disordering are described as two independent Markov chains. Then the augmented vector is introduced, and the obtained closed-loop systems are jump linear systems with two modes characterized by two Markov chains. Furthermore, the stability of the closed-loop system is investigated; and an LMI method is given for the State feedback controller design.

Related Dissertations
More Dissertations