2008
DOI: 10.1016/j.jmatprotec.2007.05.014
|View full text |Cite
|
Sign up to set email alerts
|

Hybrid neural–GA model to predict and minimise flatness value of hot rolled strips

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3

Citation Types

0
17
0

Year Published

2011
2011
2023
2023

Publication Types

Select...
7

Relationship

0
7

Authors

Journals

citations
Cited by 43 publications
(19 citation statements)
references
References 3 publications
(6 reference statements)
0
17
0
Order By: Relevance
“…Liu et al adopted the error propagation neural network optimized by GA to establish the transfer matrix between the flatness error character parameters and the flatness adjustment parameters, and the transfer matrix is successfully applied into the flatness setting control for a 900 mm HC 6‐high mill. John et al directly applied ANN to establish the model between input parameters and strip flatness, and optimized the control parameters with GA. A mixed GA‐ANN flatness model was obtained to predict and minimize flatness value of hot rolled strips. ANN was used in shape pattern recognition in cold strip rolling by Peng et al, and good results had been achieved on an 8000 kN HC mill.…”
Section: Introductionmentioning
confidence: 99%
“…Liu et al adopted the error propagation neural network optimized by GA to establish the transfer matrix between the flatness error character parameters and the flatness adjustment parameters, and the transfer matrix is successfully applied into the flatness setting control for a 900 mm HC 6‐high mill. John et al directly applied ANN to establish the model between input parameters and strip flatness, and optimized the control parameters with GA. A mixed GA‐ANN flatness model was obtained to predict and minimize flatness value of hot rolled strips. ANN was used in shape pattern recognition in cold strip rolling by Peng et al, and good results had been achieved on an 8000 kN HC mill.…”
Section: Introductionmentioning
confidence: 99%
“…PC (Pair Cross) mills can increase the range of crown control by crossing the conventional rolls. However, they do not have the ability for controlling the work roll wear proposed by John et al The K‐WRS (Kawasaki‐Work Roll Shifting) mill is able to reduce the U‐type wear to some extend by using a symmetrical roll contour and periodic shifting. But the K‐WRS technology can not fully eliminate the U‐type wear and meet the requirement of strip crown control.…”
Section: Introductionmentioning
confidence: 99%
“…One is the difficulty of wear control in the schedule-free rolling (SFR) process, and the other is the increasingly stringent requirement on the strip quality including profile and flatness in the non-oriented electrical steel production. [1,2] In order to solve the problem, the new-generation hightech rolling mills, such as Continuously Variable Crown (CVC), K-WRS (Kawasaki-Work Roll Shifting) and PC (Pair Cross), with effective profile and flatness technologies, are established and have been applied widely. The K-WRS mill could make work roll wear dispersive and uniform by the symmetrical roll contour and periodic shifting, but the work roll "U" groove-type wear still exists and the K-WRS technology is incapable of the crown control as suggested by Cao et al [3] The CVC and PC mill can provide a wide crown control range by shifting the CVC work roll or crossing the conventional rolls, respectively, but it is incapable of the wear control as suggested by John et al [2,3] Combined with the developed lubricating rolling technology, high speed steel (HSS) roll, on-line roll grinder (ORG) and other devices, these mills can improve the strip profile and flatness with the complicated system.…”
Section: Introductionmentioning
confidence: 99%