多新息优化MFAC控制律在网络控制系统中的应用

MFAC Control Law Optimization Using Multi-Innovation and its Application in Network Control System

摘要:

网络控制系统建模难度大或代价高,并且网络环境受时延、丢包、其他节点干扰的影响时刻都在变化,使用传统的控制方法已经难以满足控制要求,在网络控制系统中引入仅利用当前在线数据的无模型自适应控制,从而避开了系统建模问题,同时为了更好地适应网络环境变化,将多新息理论应用于无模型方法的控制律计算,加快其收敛速度,达到准确快速控制的目的；利用TrueTime仿真工具箱,搭建了一个以太网网络控制系统的仿真平台；仿真结果证实了无模型自适应控制算法在非线性系统下的控制效果优于PID,并且证实了使用多新息理论改进无模型控制律算法的有效性。

Abstract:

Networked control system modeling is difficult with large cost, and the network enviroment is changeable because of delays, package loss and interference from other nodes. The traditional control method can’t meet the network control system demand. Model-free adaptive control algorithm is introduced to network control system which only use system’s online data in calculation, to avoid modeling problem.And in order to adapt the network enviroment faster and more accurately, using multi-innovation theory to improve MFAC control law, to accelerate the convergence rate. We use truetime toolbox to bulid an ethernet network control system simulation platfom. And simulation results confirm the model-free adaptive control algorithm results in a nonlinear system is superior to PID, and confirmed the validity of the MFAC control law improved multi-innovation theory.