Data mining and state monitoring in hot rolling

Author

Cser, L. ; Korhonen, A.S. ; Gulyás, J. ; Mantyla, P. ; Simula, O. ; Reiss, Gy ; Ruha, P.

Author_Institution

Bay Zoltan Inst. of Logistics & Production Syst., Miskolc-Tapolca, Hungary

Volume

1

fYear

1999

fDate

36342

Firstpage

529

Abstract

An overview of state monitoring in hot rolling is reviewed, and a new concept of state monitoring is shown. Based on a detailed analysis of all factors a state monitor is proposed. A system state corresponds to the proper product quality. If the system is leaving the area of required quality in the state space, a signal is given with the evaluation of situation. Self-organising maps (SOM) are especially suitable in analysing the very complex process of hot rolling. Application of SOM helps to discover hidden dependencies influencing the quality parameters, such as flatness, profile, thickness and width deviation as well as wedge and surface quality. Results from the analysis of more than 70 parameters in 16,000 strips gave the state space used in state monitoring based on online data sampling. The coloured visualisation map shows the state space enabling prediction of product quality

Keywords

computerised monitoring; data mining; hot rolling; process control; quality control; real-time systems; self-organising feature maps; state-space methods; steel industry; data mining; flatness; hot rolling; quality control; real time systems; self-organising maps; state monitoring; state space; thickness; Artificial neural networks; Automatic control; Data mining; Finishing; Milling machines; Monitoring; Production systems; State-space methods; Steel; Strips;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Processing and Manufacturing of Materials, 1999. IPMM '99. Proceedings of the Second International Conference on

Conference_Location

Honolulu, HI

Print_ISBN

0-7803-5489-3

Type

conf

DOI

10.1109/IPMM.1999.792534

Filename

792534