期刊文献+

基于高斯混合模型的多模态过程监测

Online Monitoring for Multiple Mode Processes Based on Gaussian Mixture Model
下载PDF
导出
摘要 本文针对多模态复杂过程的多变量、多工序、变量时变性以及模态转换时间不确定等多种特性,提出基于高斯混合模型的多模态过程监测算法;针对离线数据没有模态标签的问题,提出离线数据分类算法;针对在线数据无法对应模态类型的问题,提出在线数据模态识别算法.并在以上方法的基础上建立多模态过程监测模型,以连续退火机组为背景,利用实际生产过程数据验证了算法的有效性. Considering the process high dimensionality, multi -operation, time -variant characteristics, and unknown mode duration, the article proposes the multiple mode monitoring algorithm based on the gaussian mixture model. It also proposes the offline data classification algorithm aiming at the problem that offline data has no modal tag. For online data to corresponding modal type of problem, online data mo- dal identification algorithm is put forward. And on the basis of the above methods establishing the model of multimodal process monitoring in continuous annealing line as the background, the effectiveness of the algorithm was validated by actual production data.
作者 张艳芬 谭帅 李彬彬 ZHANG Yah- fen, TAN Shuai, LI Bin -bin (1. Yingkou Vocational & Technical College, Yingkou Liaoning 115000, China; 2. School of Information Science & Engineering, Northeast University, Shenyang Liaoning 110004, China)
出处 《长春师范学院学报(自然科学版)》 2014年第1期21-26,共6页 Journal of Changchun Teachers College
基金 中央高校基本科研专项资金(N120304004) 中国博士后科学基金(2013M530937).
关键词 多模态过程 过程监测 模态识别 连续退火机组 multiple mode processes process monitoring mode identification continuous annealing line
  • 相关文献

参考文献2

二级参考文献15

  • 1何宁,谢磊,郭明,王树青.基于独立成分的动态多变量过程的故障检测与诊断方法[J].化工学报,2005,56(4):646-652. 被引量:8
  • 2牛征,刘吉臻,牛玉广.动态多主元模型故障检测方法在变工况过程中的应用[J].动力工程,2005,25(4):554-558. 被引量:17
  • 3陆宁云,王福利,高福荣,王姝.间歇过程的统计建模与在线监测[J].自动化学报,2006,32(3):400-410. 被引量:61
  • 4Ge Z. Q. and Song Z. H. Online monitoring of nonlinear multiple mode processes based on adaptive local model approach [ J ]. Con- trol Engineering Practice ,2008,16 ( 2 ) : 1427-1437.
  • 5Yu J. and Qin S. J. Multimode process monitoring with Bayesian in- ference-based finite Gaussian mixture models[ J]. AIChE J. ,2008, 54(7) :1811-1829.
  • 6Aapo Hyvarinen, Juha Karhunen, Erkki Oja. Independent compo- nent analysis[ C]. New York: John Wiley & Sons,Inc. ,2001.
  • 7Kano M. , Hasebe S. , Hashimoto I. , Ohno H. Evolution of multiva- riate statistical process control: Independent component analysis and external analysis [ J ]. Computers & Chemical Engineering, 2004,28 ( 2 ) : 1157-1166.
  • 8Lee J. M. , Yoo C. K. and Lee I. B. Statistical process monitoring with independent component analysis [ J ]. Journal of process con- trol ,2004,14(3 ) :467-485.
  • 9Lee J. M. , Qin S. J. and Lee I. B. Fault detection and diagnosis based on modified independent component analysis [ J ]. AIChE J. ,2005,52(10) :3501-3514.
  • 10Ge Z Q,Song Z. H. Process monitoring based on independent com- ponent analysis - principal component analysis ( ICA-PCA ) and similarity factors [ J ]. Industrial and Engineering chemistry Re- search ,2007,46 (7) :2054-2063.

共引文献7

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部