In the process of flat steel production, a significant amount of energy is required for the treatment of the slabs produced in the continuous casting process. For the industrial partner voestalpine Stahl GmbH, the goal is a ?zero defect production?.
The liquid steel flow in the mould, especially the flow near the mould level, significantly influences the quality of the produced steel. Therefore the proper setting of process parameters (e.g. casting speed, argon injection rate but also the geometry of the submerged entry nozzle) is important. Due to the covering slag layer not even the flow near the mould level is visible in the casting process. In order to analyse the mould flow, computational flow simulations and water model experiments are state of the art. But neither of these two approaches is able to provide the relevant information in real-time. This is the reason why empirical and semi empirical models are in use to determine the risk of slag entrainment, the risk of liquid steel break out or disadvantageous mould level fluctuations . (Semi)empirical models face two major limitations:
? They can only cover the set of parameters investigated by computational fluid dynamics (CFD), plant observations or physical models.
? The plant data available from measurements is limited ? spatially, temporally and by quantity. So far only mould level sensors and thermocouples in the mould are state of the art, providing partial information about the mould level and the temperature distribution in the mould.
The goal of this project is to address both limitations. For the (semi)empirical models the goal is to extend the range of validity using CFD, physical modelling and plant measurements. The modelling approaches will also investigate the feasibility of coarse but fast 2D CFD, to represent the mould flow. In case of the plant data, the goal is to get the maximum of information and to evaluate the possibilities for additional measurements.