基于改进Smith预估补偿的网络控制系统研究

为了有效减少网络时延对网络控制系统的影响,提出了一种新的改进Smith动态预估补偿方法,实现了将前向通路的时延从闭环回路中移到闭环回路外,将反馈通路(传感器到控制器)的时延从系统中彻底消除,无需对反馈通路实施网络调度。由于补偿模型不包含网络时延,因此无需在线测量、辨识或估计网络时延的大小,适用于网络时延是随机、时变和不确定,大于一个、甚至数十个采样周期的应用场合。基于CSMA/CD(Ethernet)和CSMA/AMP(CAN bus)网络协议,仿真研究验证了该方法的有效性。

Study of networked control systems based on modified Smith prediction compensation

In order to effectively restrain the impact of network delay on the NCS,a new approach is proposed that modifies Smith dynamic prediction compensation.The delays of forward path are removed from the closed loop,and time-variant uncertain return path(sensor to controller) delay is totally eliminated from the system.Therefore,the traffic on the return path does not need to be scheduled.Because this compensation model does not include network delays,thus network delays do not need to be measured,identified or estimated on-line,and it is applicable to some occasions that network delays are random,time-variant and uncertain,possibly large compared to one,even tens sampling periods.Based on network protocols of CSMA/CD(Ethernet) and CSMA/AMP(CAN bus),the simulation results show that approach is effective.