Researches On Networked Control Systems With Long Delays And Packet Loss

Network Control System (Networked Control Systems, NCS) refers to the formation of aclosed-loop feedback control system by adding real-time communications network controlsystem. It generally serves as a Distributed Control System (Distributed Control System, DCS)of a realization form, directly connected to the member (including sensors, controllers on thenetwork) are regarded as a node on the network. Compared with the traditional point-to-pointcontrol, network control systems have a good prospect which with resource sharing, remoteoperation, low cost, easy maintenance. However, because of the introduction of thecommunication network, the control system has lots of new issues that need to be resolvedwhich includes network-induced delay, packet loss. Then because of the communication islimited, the analysis and design of the NCS are complicated.Network latency and packet loss caused many problems in the NCS system thatperformance degradation and even instability of the important reasons. Random varying delayand uncertain in the actual control system, packet loss are more universal. Long time delayand uncertain loss makes the system a random, making it difficult to use conventional lineartime-invariant systems theory and methods of analysis and design.In this paper, the stochastic stability of networked control systems with long delay andpacket loss is given. Markov stochastic control theory and linear matrix inequalities Shurcomplement rationale as well as the cone complementarity linearization theory, non-theidealized network control system modeling, stochastic stability analysis and controller designproblem.The main contributions of this thesis are listed as follows:First of all, for a linear time invariant plant, the network induced delay assumption forthe long delay with sampling period time-varying uncertain parameters. Discrete model of thesystem is established based on the stochastic stability sufficient condition for stabilizingcontroller design of linear matrix inequality and Lyapunov function of closed loop networkcontrol system is given, finally the effectiveness by simulation examples verify method.Secondly, considering the state feedback control law, the closed-loop network controlsystem is modeled. The closed-loop network control system feedback and forward of the twochannels in the network delay of this article to describe the Under the conditions of existingtheorem proposed a sufficient condition for stability of the system random conecomplementarily linearization algorithm, while the use of the controller design. Then, anexample is given to demonstrate the effectiveness of the proposed method.Finally, the modeling and stabilizing control of networked control systems withstochastic packet loss and time-varying delays are addressed. In specific, a novel Markovianjump linear system with time-varying delays is used to model the networked control system (NCS). Without using the augmented state method,sufficient conditions for the stochasticstabilization of the NCS with packet loss and time-varying delays are obtained via themode-dependent Lyapunov function method. The mode-dependent controller for theclosed-loop NCS is presented in the linear matrix inequalities (LMIs) formulation via the Shurcomplement theory. A numerical example is given to illustrate the effectiveness of theproposed method.