Study On A Class Of Networked Control Systems With Packet Disordering And Quantized Feedback Control

With the constant penetration and development of modern science and technology in the military,industrial and commercial areas,the traditional point-to-point control system has not kept up with the growing demand of people.Therefore requires a new control system to make up for the shortage of the traditional control system,and then promote the development of the networked control systems.Networked control systems are the closed loop feedback control systems where the spatially distribution of system components such as sensors,controllers,actuators and so on are connected through a real-time network.This kind of network-based control system provides more advantages such as low cost,less wiring,simple installation and maintenance as well as remote operation and control.However,since the communication networks have been inserted into the control loop,which brought many new problems and challenges to systems,such as network-induced delay,packet loss and disordering,quantization,etc.These issues have an undesirable effect on the control performance of the systems to a varying degree and even destroy the stability of systems.Currently,the research on the network control systems has become a research hotspot of the study of the control field internationally.This paper mainly analyses and studies the problem of packet disordering and quantization in networked control systems from three aspects of system modeling,stability analysis and controller design respectively.The main contributions are summarized as follows:Firstly,the problem of packet disordering in a class of networked control systems is conducted a targeted study.Put forward a compensation control scheme to eliminate the adverse effect of packet disordering on the system.First of all,utilizing the time-stamp method to label the time sequence of the data packets sent from the source node,and set a comparator in the controller and actuator side separately,the combination of above is determined whether the data packet is disordering.If disordering,the packet is processed accordingly,which keep the control signals actually applied to the plant are always up to date.As a result,the close-loop system is modeled as a Markov jump system.The Lyapunov stability theory is utilized to obtain the necessary and sufficient conditions for the stochastic stability of system.This condition is described by a set of linear matrix inequalities(LMIs)with inverse constraints.Finally,the problem of controller design is converted into linear matrix inequality convex optimization problem.Then the cone complementary linearization(CCL)algorithm is applied to obtain the state feedback controller gain.Secondly,this part is concerned with the stabilization of networked control systems subject to both network-induced delays and data quantization.Based on logarithmic quantized feedback controller and the working mechanism of zero-order hold(ZOH),utilizing sector bound approach,the controlled system is modeled as an uncertain time-delay system with input delay.Among them,both feedback channel and forward channel of NCSs was separately jointed the logarithmic quantizer in order to quantify system state signal and control input respectively.In term of the given model,sufficient conditions for the stability and stabilization of NCSs are derived by using a delay-dependent Lyapunov-Krasoskii function.The condition is described by two sets of linear matrix inequalities.Thus,the corresponding quantized feedback controller was designed to make system asymptotically stable and pointed out that the density of the quantizer directly affects the control performance of the system.Finally,a numerical example is given to illustrate the effectiveness of the developed method.