Neural networks for process condition prediction in basic oxygen steel-making

Project Information

Grant agreement ID: 7210-CB-818

Project closed

Start date 1 July 1996

End date 31 December 1999

Funded under

• Medium-term guidelines for the ECSC steel RTD programme of research and pilot/demonstration projects, 1996-2002

Total cost € 888 500,00

EU contribution € 533 100,00

Coordinated by

British Steel plc
United Kingdom

Objective

The project objective is to improve end-point analysis and temperature control in the BOS process through the application of neural networks to current computer models for primary charge balancing, sub-lance end-blow control and end-point analysis prediction. Conventional statistical process models based on thermal or physical logic are limited in their ability to account for interrelated variables whose influence is not well understood. Neural networks with their self-learning capabilities are expected to be more accurate, faster and more precise.

Programme(s)

• ECSC-STEEL C - Medium-term guidelines for the ECSC steel RTD programme of research and pilot/demonstration projects, 1996-2002

Topic(s)

• C1 - Primary steelmaking

Call for proposal

Data not available

Funding Scheme

CSC - Cost-sharing contracts

Coordinator

British Steel plc

• EU contribution: No data
• Address: Moorgate ; S60 3AR Rotherham ; United Kingdom
• Total cost: No data

Participants (2)

INSTITUT FÜR BERGWERKS- UND HÜTTENMASCHINENKUNDE DER RWTH AACHEN

Germany

• EU contribution: No data
• Address: Wüllnerstrasse, 2 ; 52056 AACHEN
• Total cost: No data

KONINKLIJKE HOOGOVENS

Netherlands

• EU contribution: No data
• Address: 1970 CA IJMUIDEN
• Total cost: No data