Development and Application of Self-learning Model for HSM

DOI

10.2991/iccia-17.2017.15How to use a DOI?

Keywords

Hot strip rolling, Profile and flatness control, Self-learning.

Abstract

In recent years, based on maturity of AGC technology, the accuracy of Thickness Control has been improved greatly. The poor profile control is the major factor which restricts the dimensional accuracy. The mathematical model of plate shape-setup is the basis for realizing the process control of strip mill. Because the deviation of the mathematical model, the device and the parameter is inevitable during actual rolling, self-learning is an important mean to reduce the error of the set-up model. On a production line, the self-learning algorithm is not in line with actual production, which results in control chaos. To solve this problem, combined with the actual production process, a new self-learning algorithm has been developed, which considers operator revision, unstable error interference and actual parameters optimization. New algorithms has been applied, it has solved the phenomenon of self-learning disorder, improved the hit rate of profile greatly, and improved the quality of the production.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Ce Wang
AU  - Ziying Liu
AU  - Lijie Dong
AU  - Changli Zhang
AU  - Fengqin Wang
PY  - 2016/07
DA  - 2016/07
TI  - Development and Application of Self-learning Model for HSM
BT  - Proceedings of the 2nd International Conference on Computer Engineering, Information Science & Application Technology (ICCIA 2017)
PB  - Atlantis Press
SP  - 99
EP  - 103
SN  - 2352-538X
UR  - https://doi.org/10.2991/iccia-17.2017.15
DO  - 10.2991/iccia-17.2017.15
ID  - Wang2016/07
ER  -