Research On Control Algorithms For Systems With Unmodeled Dynamic Uncertainty

Many controlled systems in practical problems are complex and difficult to be modeled into simple linear models,especially for the typical system of stochastic system.On the other hand,adaptive dynamic surface control has been one of the important problems in the research because it effectively solves the problem of "complexity explosion" in the process of reverse step design.Therefore,it is of practical significance to control and analyze stochastic systems with unmodeled dynamics.This paper mainly studies the regulation control problem of uncertain systems with unmodeled dynamics.The specific research content is summarized as follows:(1)This chapter considers the global adaptive regulation by dynamic partial state feedback method for a class of uncertain stochastic time-delay cascade system with asymmetric structure and nonlinear parameterization.The global asymptotic stability can be achieved under input-to-state stability like the condition based on the zero-dynamics with time delay.The partial state adaptive control scheme with a delay-independent has been developed by employing the adaptive feedback domination design.By constructing suitable Lyapunov-Krasovskii functionals,it is proven that all the states of the closed-loop system signals can be regulated to the origin.Meanwhile,the boundedness of the closed-loop system signals is kept.Simulation results further demonstrate the effectiveness of the proposed method.(2)For single-phase full-bridge inverter system,the bounded stability tracking control problem is studied.For the system with unmodeled dynamics and actuator faults,firstly,based on Lyapunov stability theory,a fault estimation algorithm is proposed,which can make the observation error converge to the neighborhood of the origin.Secondly,the self-triggering control is designed through the adaptive dynamic surface tracking control strategy.The theoretical derivation proves that any signal of the closed-loop system is uniformly bounded.Simulation examples prove the feasibility of the scheme.Figure [10] Reference [97]...