Next-Generation Quality Prediction and Defect Detection Systems for Continuous CASTing Driven by Fundamental and Data-Driven Modeling

NextGenCAST

During continuous casting the formation of defects not only undermines the product quality and flow of production but also increases the production costs and deteriorates the energy efficiency and environmental aspects of steelmaking. The presence of cracks and other defects seriously affects the material properties such as strength, hardness, ductility and toughness. The causes of cracks in various steel grades are also different, which necessitates the development of novel solutions for detection and prediction purposes.

The NextGenCAST project brings together a diverse consortium including leading universities, SMEs, and industrial partners, combining multidisciplinary expertise in solidification, mathematical modeling, sensors, machine learning, artificial intelligence, and steelmaking domain knowledge. The primary aim is to develop novel quality prediction and detection systems to mitigate and eventually prevent defect formation in continuous casting of steel.

Project information

Project duration

-

Project funder

EU Research Fund for Coal and Steel (RFCS)

Funding amount

3 875 778 EUR

Project coordinator

University of Oulu

Unit and faculty

Process Metallurgy
Faculty of Technology

Contact information

Contact person

Project description

One of the greatest challenges in continuous casting of steel is to avoid defects in cast products. The formation of different defects is generally caused by a combination of the numerous fundamental phenomena which are present in continuous casting. The primary objective of NextGenCAST project is to develop novel, phenomena-based quality prediction and detection systems towards minimized defect formation and sustainable continuous casting and so to efficient steel production. Scientific tools, sensor data, laser cameras, laser scanners, and machine learning will be utilized. The core of these systems is quality indices computed with scientific tools. The effect of steel chemistry is a subject of special interest. The reason is that even if the defects eventually are caused by a combination of many factors, steel chemistry is generally the most significant factor contributing to the sensitivity of defect formation, if irregular casting conditions are not considered. With these systems, improvements in quality, energy efficiency and yield are achieved in the industry leading to efficient steelmaking and for defect-free casting. The developed systems also promote digitalization in the steel industry.

During this project, the following main systems and concepts will be developed and evaluated:

  1. Scientific, model-based (phenomenological) quality prediction systems for continuous casting.
  2. Feedback information to operators regarding the root causes of the defects and preventative actions.
  3. Offline re-engineering of steel compositions within the tolerance ranges of each current and future steel grade to minimize the risks of defect formation.
  4. Coupling of laser cameras and scanners with phenomenological quality prediction systems for improved detection and prediction of defects.

Research groups

Continuous casting and non-metallic inclusions

Partners

University of Oulu; Casim Consulting Oy; Sapotech Oy; Sidenor Investigacion Y Desarrollosa; M.I.K. S.Coop; Acciaierie Di Calvisano Spa; Arlenico; Feralpi Siderurgica Spa; Visiorobotics S.R.L; Ergolines Lab Srl; University of Brescia; SSAB Europe Oy.