Design Of Networked Control Systems With Input Saturation And Time-Varying Sampling Periods

Networked control systems not only realize the industrial field visualization,achieve a better human-computer interaction,but also can implement remote control and diagnosis using information obtained from the process of production,which makes the system more flexible,the production process safer and the quality of the product better.It has been successfully used in aerospace,remote robot control,wireless sensor network,process control and others.Nevertheless,the communication delay,packet losses and other networked-induced issues caused by the communication constraint would degrade the system performance or even destabilize the control system.On the other hand,the actual systems always have control input saturation constraint,which would also affect the stability and control performance of the system.Hence,it is necessary to consider the problems of input saturation constraint and other networked issues in the application of the actual system.This thesis is mainly concerned with the networked level control system with control input saturation constraint,time-varying sampling periods and short time-varying delay.The main work of this thesis is summarized as follows:1)The thesis is concerned with the guaranteed-cost controller design problem of the NCS with input control constraint and time-varying sampling periods.The system is modeled as a discrete-time switching system according to the variations of different sampling periods.Based on the Lyapunov stability theory and the average dwell-time method,optimally guaranteed-cost feedback controller with control input saturation constraint is designed.2)The problem of stability analysis and guaranteed-cost controller design of the NCS with control input saturation constraint and short time-varying delay(STVD)are investigated in chapter 3.The NCS with STVD is transformed into a time-invariant system with norm bounded uncertainty.Estimation of the mean value of the time-varying delay and the upper norm bound of the uncertainty is performed when the lower and upper bounds of the time delay are given.Based on the robust control method,optimally guaranteed-cost feedback controller design with control input saturation constraint is presented.3)The networked level control experiment platform is developed in the fourth chapter to verify the effectiveness of the algorithms proposed in this thesis.A mathematical model of the level system is established using the mechanism modeling method.Then,optimally guaranteed-cost feedback controllers with control input saturation constraint are designed.After that,under the.NET 4.0 framework and Visual Studio 2010 development environment,C# language is used to design and develop the time-varying sampling periods level control system.Design and development of the client/server architecture based remote level control system are discussed,in which the server communicates with the client via TCP/IP protocol.Finally,experiments on the real level control system demonstrate the effectiveness of the designed guaranteed-cost controllers with control input saturation constraint.