Title

IMPROVEMENT OF CASTER PERFORMANCE BY REALTIME DEFECT PREVENTION UTILIZING AI AND ML

IMPROVEMENT OF CASTER PERFORMANCE BY REALTIME DEFECT PREVENTION UTILIZING AI AND ML

Authorship

KEMPKEN, JENS

KEMPKEN, JENS

I am this author

DOI

10.5151/2594-5300-40239

Downloads

Abstract

Multiple defects, such as slivers, laminations or surface inclusions, can occur during steel production causing losses from reduced quality and rejected production. The Cracs Preventer analyses data from the entire process chain with the goal of reducing losses due to those defects. It does so by combining metallurgical expertise with artificial intelligence. The defects are predicted before they occur and the necessary countermeasures to avoid the defects are suggested. The Cracs preventer can currently be applied to flat products and long products production.

Multiple defects, such as slivers, laminations or surface inclusions, can occur during steel production causing losses from reduced quality and rejected production. The Cracs Preventer analyses data from the entire process chain with the goal of reducing losses due to those defects. It does so by combining metallurgical expertise with artificial intelligence. The defects are predicted before they occur and the necessary countermeasures to avoid the defects are suggested. The Cracs preventer can currently be applied to flat products and long products production.

Keywords

continuous casting; quality and performance improvement; surface defects; machine learning

continuous casting; quality and performance improvement; surface defects; machine learning