Research On Fault Diagnosis And Fault-tolerant Control Of Time-delay Systems

There exists time delay in many practical industrial processes and distributed networks. It is necessary to investigate fault diagnosis and fault-tolerant control methods for time delay systems in order to guarantee their safety and reliability. Both minimum error entropy filter and H∞ filter are designed for networked control systems with induced delay, packet dropout and noise. Filter based fault diagnosis method is then studied for networked control systems. Some fault-tolerant control methods are investigated for some nonlinear control systems with several kinds of typical time delays, sensor/actuator failures. Simulation results testify the effectiveness of the proposed methods.The main research work of this thesis is listed as follows:1.A minimum error entropy based filter is presented for networked control systems with stochastic noises, induced delay and packet dropout. The objective of filtering is to estimate the states of networked control systems so that the entropy of output error is made monotonically decrease. Analysis is carried out to explicitly determine the convergent condition of the filtering algorithm. Fault detection is carried out by evaluating the residuals produced by the proposed filter. Simulation results demonstrate effectiveness of the fault detection method.2. The reliable filter is designed for network-based linear continuous-time systems with sensor failure. The model of the disabled sensor is established. The reliable H∞ filter is designed. The parameter matrices of the filter can be obtained by using Lyapunov theorem of stability and linear matrix inequality (LMI) method. Simulation results illustrate effectiveness of the proposed method.3. A class of fault-tolerant controller is designed for interval time-varying delay systems with probabilistic actuator failure. A new type reliable control system model is proposed by introducing stochastic variables. Based on the new models, sufficient conditions for the exponential mean square stability of the system are derived by using the Lyapunov functional method and linear matrix inequality (LMI) technique. The derived criteria, which are expressed in terms of a set of LMIs, are fault-occurrence rate and delay-dependent. The derived controller ensures that the system with or without frequent actuator fault is stable. Simulation results demonstrate the feasibility and effectiveness of the proposed design synthesis.4. The reliable fault-tolerant controller is designed for nonlinear discrete/ continuous system with time-varying delay and actuator intermittent failures. The fault model is established by assuming the intermittent fault occurrence obeying bernoulli distribution. An approach to fault-tolerant control is then presented for discrete/continuous networked control systems with time-varying delay and actuators intermittent failures. Two illustrative examples are given to demonstrate the validity of the proposed design procedures.