具有长时延网络系统内模控制策略研究  

Research on Internal Model Control Strategy of Networked System with Long Delay

作　　者：雷何芬 时维国[1] LEI He-fen;SHI Wei-guo(School of Electrical and Information Engineering,Dalian Jiaotong University,Dalian 116028 China)

机构地区：[1]大连交通大学,电气信息工程学院,辽宁大连116028

出　　处：《自动化技术与应用》2022年第1期75-78,101,共5页Techniques of Automation and Applications

基　　金：辽宁省教育厅科学研究项目(JDL2019011)。

摘　　要：NCS(网络控制系统)中的随机时延会导致系统的不稳定。本文从控制策略的角度,在NCS中加入内模控制。在内模通道中增加AR-BP神经网络时延预测方法;针对被控对象数学模型与被控对象不一致而引起的控制性能降低问题,提出利用模糊控制规则实时调整滤波器参数的方法。仿真结果表明,在内模通道中增加AR-BP神经网络时延预测环节,能够很好的补偿内模NCS中的随机长时延,提高系统的稳定性;当数学模型不匹配时,采用模糊控制规则实时调整滤波器参数,优化了系统的动、静态性能,提高了系统的控制精度。The random delay in NCS even leads to the instability of the system. In this paper, internal model control is added to NCS from the perspective of control strategy. The method of adding AR-BP neural network delay prediction in internal model channel is adopted;Aiming at the problem of control performance degradation caused by the inconsistency between the mathematical model of the controlled object and the controlled object, a method of real-time adjusting filter parameters by using fuzzy control rules is proposed. The simulation results show that adding AR-BP neural network delay prediction link in the internal model channel can well compensate the random long delay in the internal model NCS and improve the stability of the system;When the mathematical models do not match, the fuzzy control rules are used to adjust the filter parameters in real time, which optimizes the dynamic and static performance of the system and improves the control accuracy of the system.

关 键 词：网络控制系统 长时延 内模控制 时延预测 模糊控制 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]