Decentralized Control For Interconnected Nonlinear Time-delay Systems With Triangular Structure

With the rapid development of economy and science, large-scale systems have been integrated into the development of all walks of life. The wide application of large-scale system model has been gradually received the attention of scholars in the control field. It is challenging to control the large-scale system well because of the complexity of the system. However, the control problem of large-scale system is handled effectively with the emergence of the decentralized control method. In addition, there are time delays, disturbances and nonlinear input and other factors in the control systems which have a serious impact on the operation of the system. Therefore, the research on the decentralized control of the large-scale systems with multiple factors is of great significance. Inspired by the above reasons, an effective decentralized control strategy is proposed for the interconnected nonlinear systems with complex factors.Firstly, the interconnected systems and the Lyapunov stability theory are briefly introduced. And then some techniques and methods, such as the method of handling input saturation, the principle of neural network and the related knowledge of prescribed performance, are introduced and analyzed.Secondly,the decentralized control problem for a class of triangular configuration interconnected time-delay nonlinear systems with input saturation and unmeasured states is studied. A reduced-order observer is firstly designed to estimate the unmeasured states. Then an auxiliary design system is employed to address the input constraint. Based on the backstepping approach, an adaptive neural network output feedback prescribed performance control strategy is proposed to guarantee the stability of the closed-loop system and the transient and steady performances of the output in each subsystem.Finally,the problem of decentralized control for a class of triangular configuration interconnected time-delay nonlinear systems with unmodeled dynamics and uncertain parameters is investigated. After a linear transformation, a decentralized filter is designed for the transformed system. Then, the Lyapunov Krasovskii and changing supply function methods are adopted to deal with the problems of time-varying delay and unmodeled dynamics, respectively. Combining neural network with adaptive control method, an output feedback prescribed performance controller is designed to guarantee that the interconnected systems are stable and the output of each subsystem meets the prescribed performance requirements.