Imagine trying to bake a cake using leftover ingredients, but you aren't sure exactly what's in each scrap. This project creates a 'digital ID card' and smart cameras to identify exactly what's in recycled steel. By knowing the ingredients perfectly, factories can use more recycled metal and less coal, cutting down on pollution.
Steel producers struggle to increase recycled scrap usage because they cannot accurately identify impurities or track scrap quality, leading to high CO2 emissions and inefficient use of expensive alloying elements.
The project developed Deep Learning computer vision for scrap identification, a Digital Scrap Information Card for tracking, and high-speed sampling systems for liquid steel analysis.
If you are an integrated steel mill dealing with high CO2 emissions from coal-fired blast furnaces — this project developed AI-based scrap identification and charge optimization that allows you to increase scrap uptake and reduce pig iron usage.
If you are a scrap supplier dealing with low-value mixed metal piles — this project developed a Digital Scrap Information Card that tracks and classifies scrap quality, allowing you to valorize materials like Mn steels.
If you are an equipment provider dealing with long production delays during metal analysis — this project developed high-speed sampling and analysis tools that eliminate waiting times for liquid steel results.
Based on available project data, specific pricing or implementation costs are not provided.
Yes, the project involves 8 industrial partners and targets three real-world European steelmaking routes to ensure the solutions are relevant to all steelmakers.
Based on available project data, the specific licensing terms for the Deep Learning and Computer Vision tools are not disclosed.
Yes, it directly supports the EU climate goals for 2050 and the European Clean Steel Partnership by reducing the 1.8 tons of CO2 typically emitted per ton of steel.
Based on available project data, the integration timeline is not specified, though the project runs from 2022 to 2025.
The consortium is heavily industry-driven, with 8 out of 9 partners being industrial entities (89% ratio). This high concentration of steelmakers, scrap suppliers, and technology providers across 5 countries (AT, DE, EL, IT, SE) suggests the outputs are designed for immediate commercial application rather than theoretical research.
Contact Kungliga Tekniska Hoegskolan (KTH) in Sweden
Contact us to connect with the HIYIELD industrial partners for technology transfer.
