网络控制系统的Smith预估时滞补偿方法

Smith Predictive Method for Time Delay of Networked Control Systems

针对网络控制系统中存在网络延时、非线性系统本身滞后以及延时对系统控制的影响,建立了时滞被控对象的网络控制系统模型,并分别从被控对象已知和控制器已知两个方面为出发点,提出了改进的Smith预估时滞补偿方法,无需对网络时延进行在线测量、辨识,并将系统前向通道中的时延因子移到了系统的闭环回路之外.最后,通过仿真实验对比了无时延和有时延两种情况下同一系统的稳定性情况,通过仿真结果验证文中提出的Smith预估补偿方法是可行的,具有实际应用的指导作用.

Against to time-delay in networked control systems,delay of nonlinear system itself and the effect on system＇s control,in this paper,it set up a model of networked control system with time-delay controlled objects and proposed improved Smith predictive methods based on known controlled objects and known controller.It needn＇t measure delay at real time and removed forward path＇s delay of system from system＇s closed loop.Finally,it compared the same system＇s stability with delay and no-delay through simulation and proved that the Smith predictive methods in this paper were feasible and had practical function.