全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

一种基于聚类分析的3MAD-MMMD过失误差侦破方法

DOI: 10.3969/j.issn.1005-3026.2015.08.006, PP. 1089-1092

Keywords: 软测量建模,过失误差,聚类分析,3MAD,MMMD

Full-Text   Cite this paper   Add to My Lib

Abstract:

软测量建模时所使用的数据集中若含有过失误差,将在很大程度上影响所建模型的精确度.因此,在建模之前,针对建模所使用的数据集,提出了基于聚类分析的集成3MAD-MMMD过失误差的侦破方法.在采集无缝钢管穿孔过程中不同变量不同时刻的数据后,将其排列成数据矩阵.首先运用3MAD算法剔除其中的单变量大误差得到新的数据矩阵,之后采用欧氏距离公式求得新矩阵中同一变量的数据到其最近点的距离,最后以所有变量最近距离的中位值d○med为检测标准,对新的数据矩阵进行过失误差侦破处理.实验和仿真图表明,3MAD-MMMD侦破方法有效地剔除了采集数据中的过失误差.

References

[1]  Tham M A,Montague G A.Soft-sensors for process estimation and inferential control[J].Automotive Engineer,1991,1(1):3-14.
[2]  Wei J T,Chen F T,Jiang M L.Application of neural network in fault diagnosis of radar device[J].Modern Electronics Technique,2012,12(19):131-134.
[3]  Serth R,Heenan W.Gross error detection and data reconciliation in steam metering systems[J].American Institute of Chemical Engineering Journal,1986,32(12):733-742.
[4]  王建勇,郝勇生,彭兴.基于质量平衡的煤气数据校正在能源管控系统中的应用[J].冶金自动化,2013,37(2):138-142.(Wang Jian-yong,Hao Yong-sheng,Peng Xing.The application of the data correction of gas in the energy system of pipeline control based on mass balance[J].Metallurgical Industry Automation,2013,37(2):138-142.)
[5]  Zhang Z J,Shao Z J,Chen X.Quasi-weighted least squares estimator for data reconciliation[J].Computers and Chemical Engineering,2010,34(2):154-162.
[6]  Yang X S,He X S.Intelligence and smart optimization algorithms[J].Basic Sciences Journal of Textile Universities,2013,26(3):287- 296.
[7]  Rajab J M,Matjafri M Z,Lim H S.Combining multiple regression and principal component analysis for accurate predictions for column ozone in peninsular Malaysia[J].Atmospheric Environment,2013,71(1):36-43.
[8]  Malumfashi A.Education expenditure and economic growth in Nigeria:co-intergration and correction technique[J].International Journal of Research in Commerce,Economics and Management,2012,2(8):34-37.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133