Abstract
The steel industry faces increasing pressure to reduce CO2 emissions and transition away from blast furnaces, while maintaining efficient and profitable production of high-quality products. To support this transition, comprehensive computational models of new processes are needed. This presentation looks into how thermochemical equilibrium and physical property calculations are combined and accelerated to support large scale computational models of Direct Reduced Iron (DRI) smelting furnaces (DSFs), a key technology for greener steelmaking.

The RapidThermo accelerator concept [1,2] was developed to act as an intermediary between computational models and thermochemical calculators like ChemApp. It stores equilibrium calculation results as geometric objects in an n-dimensional phase diagram. As the phase diagram is populated to cover more composition and temperature ranges, the algorithm interpolates rather than performing more time-consuming direct calculations. This allows large-scale multiphysics models developed with tools like OpenFOAM to benefit from thermochemical calculations to describe high-temperature processes more realistically, without suffering from prohibitive computation times that can run into years or even hundreds of years for large models. Other applications that will benefit from acceleration in the future include dynamic process simulators and advanced process control systems.

In this presentation we share the progress made with RapidThermo implementation since the concept was first published in 2022. The current version provides more acceleration than the original Python implementation, and is now able to provide slag physical properties such as density, electrical conductivity, and viscosity, in addition to equilibrium results that include phase fractions, phase compositions, heat capacity, and enthalpy.

Click here to watch the presentation now.

References
1. Roos W.A., Zietsman J.H. 2022. Geometric acceleration of complex chemical equilibrium calculations – Algorithm and application to two- and three-component systems. CALPHAD, vol. 77, DOI: 10.1016/j.calphad.2022.102420.
2. Roos W.A., Zietsman J.H. 2022. Geometric acceleration of complex chemical equilibrium calculations – Performance in two- and five-component systems. CALPHAD, vol. 82, DOI: 10.1016/j.calphad.2023.102584.

Biography
Dr Johan Zietsman has more than 25 years of extractive metallurgy experience, both in industry and academia, as well as software engineering related to extractive metallurgy. He is the founder and CEO of Ex Mente Technologies, a pyrometallurgy consulting and systems development company. His interests in extractive metallurgy include pyrometallurgical and hydrometallurgical processes, thermochemistry, computational modelling, slag freeze linings, and new process technology development. In software engineering, he is involved in design and development of information systems, and software for process modelling, financial modelling, operational enhancement, and advanced process control. He has a Ph.D. in metallurgical engineering from the University of Pretoria, and has published several papers at peer-reviewed conferences and refereed journals.