© Adobe Stock / MPStudio
Use Case 1 (UC1) is focused on the heavy melting scrap (HMS) verification. Currently, the HMS characterization is based on visual inspection by experienced employees and random sample spectroscopic analysis with hand-held X-Ray fluorescence. An improved and reliable scrap characterization is needed to allow operator-friendly sorting and better separation to reduce impurities in the targeted steel heat.
Use Case 2 (UC2) focuses on optimizing the production of crude steel by electric arc furnace (EAF). Currently, the process control of the scrap-based EAF is highly empirical, relying mainly on the individual skills of the operators and on fixed operating patterns, but not on real-time measured process data, making the current process inefficient in terms of material and energy. An innovative solution is needed to monitor and control sustainability of the running process conditions, to set up countermeasures to stay into the optimal process window.
Use Case 3 (UC3) is focused on the quality assurance of semi-finished products particularly steel sheets, and the optimization of the process parameters. The current process control is based on long-term system behavior, lacks regular updates and does not consider changes in the production system or inline sensor data, and the final quality of the product is determined after its leveling, blanking and annealing. Around 4 % of the steel sheets need to be reworked (leveling and annealing), that leads to up to 20 % of additional plant utilization because a rework of plate material is significantly more time consuming than the processing of the coil material. A new approach is needed to optimize the leveling process using data from previous process steps, as well as predicting final product quality before annealing, to reduce the re-work by half (down to 2 %).
Work package 1 will deal with analyzing the current state of the use case environments, both with regards to physical and digital aspects. Data sources, interfaces and data sets pertaining to the use cases will be reviewed and gaps and needs for implementing Digital Twins and process models will be identified. Solutions will be evaluated, and data structures, soft sensors and new hardware specified to fill the gaps previously found. Lastly, the WP will close with the procurement of the specified hardware.
Work package 2 will be focused on the mechanical and digital adaption of the sensors, and data sources defined in WP1 (e.g. process and environmental parameters) so that they meet the defined requirements and can be adapted and integrated in existing process lines and infrastructure in the respective use case.
Work package 3 will implement suitable communication interfaces, structure data sets, and network local data archives for offline and online data sharing among industrial use-case parties and modeling experts. This will create a central data source for the entire project's analysis and Digital Twin information and will serve as the central data target for storing results such as prognoses, clustering results, and control commands. These steps consider relevant security aspects to minimize risks associated with cross-company data exchange.
Work package 4 will develop, adapt, and extend use case specific process models using physical/analytical as well as data-based/ML approaches with inputs from process data and novel sensor information. The models will then be transferred into Digital Twins for integration into the architecture set up in WP3.
Work package 5 will focus on the testing and evaluation of the technologies integrated in WP1–WP4 under the real conditions of UC1. Individual adjustments will be made to the subsystems to meet the requirements defined in WP1. Finally, the validation of the collected relevant characteristics of scrap within a container without direct extraction of individual components will be carried out.
The Digital Twin (WP4) will provide process parameters for the process behavior, with a focus on improving the quality of foamy slag. In a second step, optimized set points for control parameters will be calculated to increase the reliability of slag foaming and slag height that will increases the energy input into the melt by reducing radiation losses. A user interface will then be implemented to assist operators in adjusting control parameters based on real-time data and to facilitate further process improvements. Additionally, the feasibility of integrating a closed-loop control system for carbon and oxygen injection will be tested.
The Digital Twin model(s) developed in WP4 will be further developed to predict the final product quality and therefore optimize the process parameters to increase the process yield of the final product as well as predicting and optimizing the energy system behavior. Finally, the developed models will be implemented in a real operational environment at Voestalpine and the preliminary work of WP2–WP4 will be evaluated.
The results of the three use cases that cover different stages of the steel production value chain, from material selection to processing will be virtually integrated, validated and evaluated. The work package will explore the benefits of information exchange and cross-process optimization for improving sustainability, quality, and productivity. The work package will also develop concepts for transferring the results to other domains.
These WPs will provide a clear organizational framework, guidance and all support mechanisms to enable a smooth project workflow and to ensure that objectives and milestones will be met in time. As a pre-requisite for all technical WPs, they will ensure that all contractual commitments will be met in time. They will also coordinate communication, dissemination, will provide optimal visibility and a wide outreach to relevant stakeholders, particularly relevant co-programmed partnerships and uptake, use and exploitation of results for three reporting periods.
Demonstration of Digital Twins for a Green Steel Value Chain