%0 Journal Article
%T 基于PSO-ELM网络控制系统容错控制研究<br>Research on Fault Tolerant Control of Network Control System Based on PSO-ELM
%A 李吉萌
%A 孙海义
%J Dynamical Systems and Control
%P 1-12
%@ 2325-6761
%D 2021
%I Hans Publishing
%R 10.12677/DSC.2021.101001
%X <div style=\"text-align:justify;\">
	针对动态系统执行器任意单通道故障问题，提出一种以闭环极点作为故障信息设计故障诊断系统并实现容错控制。在执行器任意单通道发生故障或不同通道的故障增益发生变化时，通过极点观测器得到极点在给定区域内的变化判断故障。将闭环系统极点作为极限学习机ELM (Extreme Learning Machine)故障诊断模型的输入，同时利用粒子群算法PSO (Particle Swarm Optimization)优化极限学习机，实现对系统执行器不同通道的容错控制。通过网络控制系统NCSS (Networked Control Systems)模拟，验证容错控制系统的准确性，极点观测器的精确性和可靠控制器的有效性。<br />
Aiming at the problem of any single channel failure of dynamic system actuators, a closed-loop pole as fault information is proposed to design a fault diagnosis system and realize fault-tolerant control. When any single channel of the actuator fails or the failure gain of different channels changes, the pole observer obtains the change of the pole in a given area to judge the fault. The closed-loop system poles are used as the input of the fault diagnosis model of the extreme learning machine, and the particle swarm algorithm is applied to optimize the extreme learning machine to realize the fault-tolerant control of the diverse passage of the system actuator. Using network control system simulation, verify the accuracy of the fault-tolerant control system, the exactness of the vertex observer and the validity of reliable controllers.
</div>
%K 故障诊断，容错控制，极限学习机算法，粒子群算法，网络控制系统<br>Malfunction Diagnosis
%K Fault-Tolerant Control
%K Extreme Learning Machine Algorithm
%K Particle Swarm Optimization Algorithm
%K Networked Control Systems
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=39520