Authors:
Maria Gabriela Garcia Campos | Tata Steel Nederland | Netherlands
Dr. Bruno Luchini | Tata Steel Nederland | Netherlands
Dr. Paul van Beurden | Tata Steel Nederland | Netherlands
Tom van der Velde | Tata Steel Nederland | Netherlands
Hilbrand Kuiken | Quantillion Technologies | Netherlands
Dr. Sido Sinnema | Tata Steel Nederland | Netherlands
The steel industry is highly complex. Steel manufacturers have to monitor all kinds of processes, maintain the operation and reduce the emission of CO2. Advanced digitalization tools are vital to evolve toward a more sustainable, optimized, integrated, and agile operating model. Representing processes and systems as a digital twin with artificial intelligence-based optimization and scheduling models greatly enhances the decision-making on logistics, refractory maintenance, and energy efficiency. In this context, the thermal management of the torpedo ladle cars (used to transfer pig iron from the blast furnace to the steelmaking plant) has an important role. In this paper, the development of an artificial intelligence system to support decision-making on selecting the most energetically favorable torpedo car is presented. The application uses a simulation-based digital twin consisting of reduced order models of validated FEM models to forecast the torpedoes' refractory lining and pig iron temperatures. In addition, the system includes the life, location, and contents of the torpedoes to indicate the most suitable to be selected for pig iron transportation. This increases the life of the equipment and avoids unnecessary heat losses, thereby reducing production costs and energy consumption. Thus, these new technologies lead us to solve existing logistics management challenges, improve process control, and prepare steel companies for the future.