Research Of Iterative Learning Fault-tolerant Control Based On 2D System Theory

With the increasingly complicated field operations in modern industrial processes, the possibility of system failures occurring is enhanced due to various working conditions and operating modes, and there is an urgent requirement of ensuring safe and reliable operation of the industrial control process to improve the control quality of the system, therefore,fault-tolerant control attaches great importance. Iterative learning control is an important control technology in intelligent system, which is suitable for the industrial control process with repetitive operations on the finite interval, and the learning law is established through the previous batch information of system to improve the control performance. Iterative learning fault-tolerant controller can still maintain the ideal characteristics, and guarantee the stable operation of the closed-loop system in the case of the repetitive control system suffering from faults, so it has good practical value.Based on the stability theory of two-dimensional(2D) system, this paper investigates the analysis and synthesis problems of iterative learning control fault- tolerant control system for various fault repetitive systems. And sufficient conditions for the monotone convergence of the dynamic process and the controller parameters of the controller are derived in terms of linear matrix inequality(LMI). The major research work of this thesis is summarized as follows:(1) A robust iterative learning fault-tolerant control scheme is proposed for linear repetitive processes with actuator failures and external disturbances based on the 2D system theory. And the sufficient conditions ensuring monotonic convergence along both time and cycle directions as well as robust H?performance is derived in terms of LMI, which also balance and optimize the robust fault-tolerant performance of the system. In addition, this kind of performance is tested by the simulation experiments of the rotational control system.(2) The problem of a robust dissipative iterative learning fault-tolerant control for a class of multi-rate sampling linear batch process with disturbances and actuator faults is researched.Firstly, the multi-rate sampling process is equivalently transformed into state-space model in low sampling rate by using the lifting technology. Then the 2D Roesser fault model under iterative learning control is constructed, combining the 2D system theory, and the stability and robust dissipative performance of the closed-loop fault system are analyzed by constructing Lyapunov functional. The analysis results are further validated by the simulations of injection molding process.(3) The problem of iterative learning fault-tolerant control for linear time-delay differential process is presented from the perspective of a finite frequency range. Firstly, the2 D linear time-delay Roesser model is constructed in time domain, then the sufficient conditions for the asymptotically stability of the learning process and the designed parameters are derived in terms of the generalized Kalman-Yakubovich-Popov(KYP) lemma. In addition,the method is also extended to the time-delay differential system with structure uncertainties,which accord with the engineering application effect. Finally, the simulations on the control of a two-stage chemical reactor with delayed recycle streams are presented to verify the results.