Research The Variable Sampling Period Scheduling Strategy Of Networked Control System

Networked Control System(NCS)is a new type of distributed,intelligent,and networked control system.Specifically,it is a real time feedback control system by scattered in different locations of sensors,actuators,controllers and controlled object connected through the network,which based on communication network as the transmission medium.This system is widely used with the advantages such as strong anti-interference ability,low connection and adaptability,easy to expand etc,but at the same time,it also produces many problems,such as network-induced delay,packet loss etc,which influences the stability performance of the system.Therefore,how to solve the effects of these problems is of great significant nowadays.According to the control performance and stability of network control system,scholars mainly depend on control algorithms and scheduling algorithms for related research.Considering the limited bandwidth resources,this paper takes a method that makes full use of network bandwidth and reasonable scheduling transmit data in this system to meet the real-time requirements of data,which reduces negative influence on the system performance caused by the network intervention.Therefore,this paper carries out the study of variable sampling period scheduling strategy,the research ideas as following:Considering networked control system of the limited bandwidth,a kind of variable sampling period scheduling algorithm based on grey-neural-network prediction network bandwidth is proposed.Firstly,according to the history and current network bandwidth values of monitoring period,original time series of different lengths are built,then grey prediction method is used to get different grey prediction data;Secondly,grey prediction data as the input of the BP neural network is used to realize secondary prediction for network bandwidth,it ensures that prediction data can reflect the actual situation of the network;Finally,absolute error integral parameter is used for the dynamic allocation of bandwidth resources,then this can adjust the sampling period of the control system in real-time.The simulation results show that the scheduling algorithm proposed can ensure the control performance and the stability of the system.In view of the random time-varying of time delay,a method based on parameter correction grey model is proposed for time delay estimation and prediction in NCS.Firstly,let the time delay data as a non-stationary random time sequence,then by using this traditional grey prediction method can calculates the parameters a and u;Secondly,adjusting the parameters based on the correction method on-line,then this can establishes a new grey prediction model used by the modified parameters to improve the prediction precision of time delay.Finally,let the time delay estimation data as an important parameter of network bandwidth,and combining the variable sampling period scheduling algorithm based on grey-neural-network prediction bandwidth can realize variable sampling period scheduling,which can adjust the sampling period of the control system in real-time.Compared with the above scheduling algorithm,the simulation results show that this scheduling algorithm has a better control performance and stability of system.