|Table of Contents|

The Research and Application of Neural Network Predicting Model of Molecule Weight of Casting Slice Belt(PDF)

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

Issue:
2000年02期
Page:
139-142
Research Field:
Publishing date:

Info

Title:
The Research and Application of Neural Network Predicting Model of Molecule Weight of Casting Slice Belt
Author(s):
LuYunxu LuBaochun ZhangShiqi
School of Manufacturing Engineering,NUST,Nanjing 210094
Keywords:
neural netw ork multiple reg ression structural optimizat ion genet ic algorithm backpropagation algorithm on-line predict ing model
PACS:
TP183
DOI:
-
Abstract:
Polymerization is a complicated and important chemical process in the polyamide fibre product ion. It. s signif icant to build the predict ing model of molecule w eight of casting slice belt to guide the product ion. The improved hybrid g enetic algorithm and backpropagat ion algorithm are combined to t rain neural netw ork, and the node numbers of input layer are opt imized based on mult iple stepw ise regression. An on-line predicting neural netw ork model of molecule w eight of cast ing slice belt in polymerizat ion reaction is presented. Its applicat ion in certain chemical factory shows that this model has faster convergence, higher predict ion accuracy and more network generalization than those of the least square.

References:

1 罗红宝. 扬州有机化工厂CIMS 示范工程锦纶- 6 帘子布生产优化子系统的研究及开发:[ 硕士学位论文] .南京: 南京理工大学, 1998. 41~ 43
2 陈国良, 王煦法, 庄镇泉. 遗传算法及其应用. 北京: 人民邮电出版社, 1996
3 Davidon W C. New least square algorithm. Journal of Optimization Theory and Applications,1976, 18( 2) : 187~ 197

Memo

Memo:
-
Last Update: 2013-03-25