*5.3. Further Research*

The Simulator mode allows direct user input and feedback during the simulation through a simple graphical interface. Although the current demonstrator is limited, all operation chart values can easily be added and manipulated during the simulation and additional settings such as compositions of charged materials for each basket could be added. All intermediate results of the model, such as temperatures, heat flows, and masses, are available during the simulation.

A target value is selected (for example 100 MW of electrical power) for both the automatic control and the simulator modes, and that exact value was used during the simulation, whereas in reality many operation chart values, especially the electrical power, current, and voltage, will fluctuate due to random influences such as the behavior of the electric arcs. In further work, a stochastic element could be added to introduce such fluctuations during simulations and produce more realistic behavior. Furthermore, it may be necessary to adjust the density of the scrap for each basket separately to better reproduce the behavior seen in real-life operation of the furnace.

In addition, further improvement of the underlying process model itself is of course possible. For example, a better representation of the melting behavior of scrap and adjustments for the use of other input materials, such as direct reduced iron, hot briquetted iron, or hot metal; improved

description of the foaming of the slag; as well as the automatic adjustment of model parameters for different furnaces are possible areas of further research.

#### **6. Conclusions**

An automatic control and a simulator were developed on the basis of a comprehensive dynamic EAF process model that has previously been validated exhaustively using data from industrial furnaces. The automatic control is capable of reproducing real-live operation charts and adjusting them for different operating conditions. It can be used to evaluate various scenarios such as new control strategies, different materials for injection and charging, or the installation of new equipment. The rules currently used are partially based on parameters that cannot be measured or observed directly in the real-live process, and adjusted rules that are based more closely on the actual parameters used in process control such as total electric energy input and off-gas composition may be more useful for future work. The rules and parameters can be adjusted easily to allow different furnaces and operating strategies to be replicated and evaluated.

The simulator was shown to be stable and fast enough to run simulations based on user input in real-time as well as with higher speeds selected by the user with no noticeable delay apart from a few seconds during initialization after charging. Additional inputs and outputs can be added to make the simulator more versatile. For use in teaching and training, a more refined user interface and documentation would also be useful.

Due to the modular structure of the EAF process model simulations based on measured data, automatic control and simulator input can be run using the same basic model and structure with changes only necessary to the input module and the evaluation and output of the results, with the flexible solver algorithm allowing for stable and fast simulations for all cases. The model can therefore be used to evaluate the current state of the process and identify potential for improvement, to create and assess different scenarios and operating strategies, and for training and teaching using the simulator.

**Author Contributions:** Conceptualization, methodology, software, and writing—original draft preparation by T.H.; supervision, project administration, and writing—review and editing by T.E.; writing—review and editing by V.-V.V.

**Funding:** This research received no external funding

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
