BEGIN:VCALENDAR
VERSION:2.0
METHOD:PUBLISH
BEGIN:VEVENT
ORGANIZER;CN=ESTAD 2023:mailto:info@metec-estad.com
LOCATION:Room 1
SUMMARY:Risk estimation for surface defects of long products
DESCRIPTION:A concept for estimating the risk of steel long products to develop surface defects during the production is presented, based on an advanced product tracking system and machine learning algorithms. The aim of the risk estimators is to identify products with a high risk for defects as early as possible in the production chain and to support operator decisions on corrective actions or the immediate recycling of a product in order to save resources for the further processing. In our use case, the secondary metallurgy, continuous casting and hot rolling processes are considered of relevance for the surface quality, and besides the raw process data also the results of a set of soft sensors are evaluated, which integrate physical process knowledge into the otherwise data-driven risk estimators to increase the reliability. Some challenges of the approach are discussed, such as difficult product tracking conditions, the need to retrieve aligned data from different sources and the availability of sufficiently broad datasets for training. A digital twin-based software platform for the integration of different data sources, soft sensors and risk sensors is presented, along with a data model based on international standards but adapted to our use case. 
CLASS:PUBLIC
DTSTART:20230615T145000
DTEND:20230615T151000
END:VEVENT
END:VCALENDAR