BEGIN:VCALENDAR
VERSION:2.0
METHOD:PUBLISH
BEGIN:VEVENT
ORGANIZER;CN=ESTAD 2023:mailto:info@metec-estad.com
LOCATION:Room 1
SUMMARY:Utilization of deep-learning models to predict defects during the continuous casting process
DESCRIPTION:During continuous casting longitudinal facial cracks (LFC) can occur due to various root causes. LFCs are difficult to capture during the production process and are visible only in the final stages of a process. Some of the LFCs can open and lead to a breakout. This creates safety and environmental issues due to hot, liquid metal flowing outside the casting machine. It also leads to a loss in production time and potential damage to the equipment. The goal of this paper is the prediction of those breakouts during operations using deep learning models. The deep learning algorithm uses different signals at the mold to calculate a probability of a breakout. When the probability exceeds a threshold value, a breakout alarm will be triggered. Upon detection of an upcoming breakout, suitable countermeasures are suggested to prevent the breakout from happening. Early results suggest a reduction of breakout events of 50%.
CLASS:PUBLIC
DTSTART:20230615T135000
DTEND:20230615T141000
END:VEVENT
END:VCALENDAR