NSSC OpenIR  > 空间技术部
晶体生长炉神经元网络PID解耦控制系统
Alternative TitleNeural Network PID Decoupling Controller of Coupled Multi-zone Crystal Growth Furnace
于强; 石柱; 吕旭涛; 北京8701信箱
Department空间科学技术研究室
Source Publication航天控制
2008
Volume26Issue:2Pages:59-62
ISSN1006-3242
Language中文
Keyword多变量系统 解耦控制 神经元网络 Pid控制
Abstract多温区晶体生长炉是一种多输入多输出、强耦合的被控对象。根据多温区晶体生长炉内温场分布的要求,提出一种神经元网络PID多变量强耦合系统控制器,通过在线学习自动调节控制器参数,同时实现解耦和控制作用。仿真结果验证了本方法对多变量强耦合时变对象具有良好的解耦性能和自学习控制特性,达到了系统所要求的控制指标。
Other AbstractA new k ind of neu ral network P ID for m u lti2variable and strong2coupled control system is proposed in this paper. Con trol param eters are autom a tically ad justed through learning on line and it can be used to decouple the strong coupling and control the m ultivariable system sim ultaneously. B y apply ing this controller to con trol the tem peratu re of a m u lti2zone crystal grow th fu rnace, the sim ulation results show perfect decoupling and self2learning con trol perform ance for strong2coupled tim e2va rying system w ithout knowing the system dynam ic m odel and training off2line, and the tem peratu re accuracy is better than that of designing.
Indexed ByCSCD
Funding Project中国科学院空间科学与应用研究中心
Citation statistics
Document Type期刊论文
Identifierhttp://ir.nssc.ac.cn/handle/122/617
Collection空间技术部
Corresponding Author北京8701信箱
Recommended Citation
GB/T 7714
于强,石柱,吕旭涛,等. 晶体生长炉神经元网络PID解耦控制系统[J]. 航天控制,2008,26(2):59-62.
APA 于强,石柱,吕旭涛,&北京8701信箱.(2008).晶体生长炉神经元网络PID解耦控制系统.航天控制,26(2),59-62.
MLA 于强,et al."晶体生长炉神经元网络PID解耦控制系统".航天控制 26.2(2008):59-62.
Files in This Item: Download All
File Name/Size DocType Version Access License
200826259.pdf(369KB) 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[于强]'s Articles
[石柱]'s Articles
[吕旭涛]'s Articles
Baidu academic
Similar articles in Baidu academic
[于强]'s Articles
[石柱]'s Articles
[吕旭涛]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[于强]'s Articles
[石柱]'s Articles
[吕旭涛]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 200826259.pdf
Format: Adobe PDF
This file does not support browsing at this time
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.