吉林大学学报(理学版)

• 计算机科学 • 上一篇    下一篇

基于最小二乘支持向量机的网络控制系统建模

孙丹1, 秦贵和1, 董劲男1,2, 陈虹2   

  1. 1. 吉林大学 计算机科学与技术学院, 长春 130012; 2. 吉林大学 通信工程学院, 长春 130012
  • 收稿日期:2013-12-13 出版日期:2014-11-26 发布日期:2014-12-11
  • 通讯作者: 董劲男 E-mail:dongjinnan@jlu.edu.cn

A Universal Modeling Method for Networked Control SystemBased on Least Squares Support Vector Machines

SUN Dan1, QIN Guihe1, DONG Jinnan1,2, CHEN Hong2   

  1. 1. College of Computer Science and Technology, Jilin University, Changchun 130012, China;2. College of Communication Engineering, Jilin University, Changchun 130012, China
  • Received:2013-12-13 Online:2014-11-26 Published:2014-12-11
  • Contact: DONG Jinnan E-mail:dongjinnan@jlu.edu.cn

摘要:

针对网络控制系统实时信号的传输问题, 提出一种采用最小二乘支持向量机理论预测和补偿非理想条件下网络传输导致的不良影响, 建立一个较通用网络控制系统模型的方法, 并对所提出模型的正确性和可应用性进行验证. 仿真结果表明,  基于最小二乘支持向量机预测模型能在一定程度上提高网络控制系统的动态性和稳定性,
弥补了周期性传输采样信号占用大量网络带宽和不必要网络通信的缺陷.

关键词: 网络控制系统, 最小二乘支持向量机, 预测补偿, 建模

Abstract:

In connection with the realtime signal transmission of networked control system, least squares support vector machine theory was used to predict and compensate bad effects caused by transmission uncertainties so as to establish a model of networked control system. Experiments were carried out to validate the correctness and applicability of this method. The simulation results show that least squares support vector machine modeling can improve the dynamic performance and stability of networked control system, which makes up the periodic transmission of sampled signals taking up a lot of network bandwidths and unnecessary network traffic defects, and improve the system reliability.

Key words: networked control system (NCS), least squares support vector machine (LS-SVM), predictive compensation, modeling

中图分类号: 

  • TP13