Modeling of a thermo-mechanically controlled virtual finishing mill and coil cooling processes

Emissions from iron and steel production account for up to nine percent of total greenhouse gas emissions globally. Consequently, steel production is facing significant changes in the coming decades as it transitions to lower-emission production methods. Modern hot strip rolling lines for steel are based on electric arc furnace technology, which requires large melting sizes and excludes the rolling of individual test strips. To continue the development of new steel grades, experimental test strips increasingly need to be replaced with multi-physical, phenomenon-based mathematical models that describe the processes and phenomena in steel production.
The rolling of steel has been researched and modeled using various methods for over a century. The oldest foundational theories still in use today also originate from that time. Over the years, models of the phenomena in steel rolling have been integrated and more accurate models have been developed. However, these models often overlook the rolling process itself, such as changes in the boundary conditions of the steel strip brought about by rolling automation. The significance of these changes is considerable even during a rolling process lasting only minutes.
The aim of this thesis was to develop virtual phenomenon-based process models of finishing rolling, coiling, and coil cooling. Special attention was given to the boundary conditions caused by the process itself in the production of the steel strip. Therefore, finishing rolling was modeled using the finite element method by implementing virtual rolling automation to control the rolling simulation. This enabled the creation of process conditions and boundary conditions for the simulated steel strip that were as closely comparable to the actual rolling process as possible. The rolling simulation was also developed to be interactive. Sensors placed in the finite element method model measured the progress of the rolling simulation, transmitting data to the virtual rolling automation which controlled and adjusted the operation of the rolling mill as necessary.
Coil cooling is a crucial process for multiphase steels. The final microstructure of these steel grades forms during coil cooling and is thus dependent on the cooling rates of different areas of the coil. In the finite element method model developed for simulating coil cooling, a 36-stage coil conveyance path from the coiler to the coil field was considered. Only by accounting for the most significant heat transfer mechanisms and boundary conditions affecting the heat transfer of the coil could cooling rates corresponding to the actual coil cooling be produced for the calculation of phase transformations. Using an accurate coupled temperature and a phase transformation model, the thickness variations in the cold rolling process for multiphase steels could be explained. The differences in cooling rates between the top and bottom parts of the coil caused by coil conveyance led to differences in phase fractions and therefore differences in strength, creating issues in thickness control during the rapid cold rolling process.