BEGIN:VCALENDAR
VERSION:2.0
METHOD:PUBLISH
BEGIN:VEVENT
ORGANIZER;CN=ESTAD 2023:mailto:info@metec-estad.com
LOCATION:Room 15b
SUMMARY:Application of data science methods on deep thermocouple readings for monitoring of the blast furnace hearth state
DESCRIPTION:The wear of the blast furnace hearth lining defines its campaign life. Due to the excessive costs of the relining, this is one of the most important economic factors of blast furnace ironmaking. As current state of art, hearth wear cannot reliably be measured directly. It is deduced from the temperature measurements in the refractory wall. Increases in the maximum observed temperatures are interpreted as additional wear. Nevertheless, the analysis of hearth wall temperatures suggests that other operational factors such as thermal hearth state and liquid flow also affect those temperature readings. Due to the harsh environment inside the hearth, it is impossible to directly measure factors describing such hearth processes.
Dillinger BF5 is equipped with deep thermocouples reaching more than 1.6m inside the refractory. Dillinger and BFI have developed data streaming techniques enabling analysis of data from such thermocouples with comparable high time resolution of a few seconds. In parallel, Dillinger has performed continuous temperature measurements of the tapped hot metal. 
Applying methods from data science like analysis of data clusters and search for correlations, repeating temperature fluctuation patterns have been found. These patterns were related to either the tapping regime or to hot blast stove changes. The comparison of the process data distributions with and without such patterns provided valuable information on the inner hearth processes and the health of the hearth lining. This enables better and more reliable monitoring of the blast furnace hearth state.
CLASS:PUBLIC
DTSTART:20230614T121000
DTEND:20230614T123000
END:VEVENT
END:VCALENDAR