SURF supports Tata Steel casting technology calculations

Tata Steel in IJmuiden serves customers who demand high-quality steel. This requires optimum production processes, which is in turn reliant on research. Eelco van Vliet is one of the 1,100 Tata Steel researchers worldwide, and performs his computations on SURF’s Lisa compute cluster.

Stalen balken bij Tata Steel

Quality products

“Tata Steel produces high-quality steel. Processed steel for use in batteries, for example, must meet high quality standards: a battery mustn't develop holes due to causes such as small particles in the steel. Construction steel, on the other hand, has very different quality requirements,” says Van Vliet. His research deals with processes related to fluid mechanics: “Steel casting is an important research topic right now. It is a vital process, because it determines the ultimate quality of the steel.”

The casting process

How does the casting process work? Liquid steel is poured into a mould from above. Around the outside, heat escapes from the steel creating a layer of solidified steel which is extracted from the bottom of the mould. Magnetic fields ensure the correct flow of steel through the mould. This casting process is called ‘continuous casting’.

“We can now run a whole heap of production settings within a short time, allowing us to understand the effects of the magnet on the pouring process much more quickly.”

Magnetic field research

Little is known about the effects of magnetic fields. For example: what is the ideal strength of the magnet? And how should the poles be oriented? Van Vliet performs these computations on SURF's Lisa compute cluster. “This is our first research project with SURF. Previously we calculated this kind of computational fluid dynamics using commercial software on our own computers. But that became too expensive, which was part of the reason why we switched to the open-source software package ‘OpenFoam’. In summer 2010 I contacted SURF about running OpenFoam on their machines.”

Tenfold calculating speed

At SURFsara, Van Vliet performs computations 10 times faster than before. “We can now run a whole heap of production settings within a short time, allowing us to understand the effects of the magnet on the pouring process much more quickly.” Another benefit is that Van Vliet now performs computations that he was unable to do before due to long processing times: “I no longer need to say no to my customers at production departments who have complex issues.”

“SURFsara modified the OpenFoam code for me so that the software would run more effectively. They also compile ParaView, which I use to produce customer simulations.”
Eelco van Vliet, researcher at Tata Steel

Fruitful collaboration

Van Vliet is satisfied with the partnership: “SURF modified the OpenFoam code for me so that the software would run more effectively. They also compile ParaView, which I use to produce customer simulations. That requires expertise that I don’t have.” But there is more: “Two other companies and I have submitted a project proposal to the European Research Fund for Coal and Steel. Part of the proposal involves us producing an even more detailed description of the casting process in OpenFoam. Using this model, we will be performing large-scale computations with SURF as part of a four-year project. That means a lot of computing work coming up, and I’m very much looking forward to working with them!”