AI Powered Characterisation and Modelling for Green Steel Technology

The scope of the tools developed through the AID4GREENEST project includes the steel design (chemistry and microstructure), process design (processing parameters), product design (processing and heat treatments), and product performance stages. The proposed tools are complemented by a roadmap designed to enable model-based innovation processes, from materials design to product development while considering industry needs: enhanced material quality, a reduction of carbon emissions, limiting waste generation, and reducing risks in the supply of critical raw materials. A consortium of 10 partners composed of leading European universities, research centers, steel companies, and a small enterprise steer the project.

Hot Rolling of Steels

AID4GREENEST allows the steel industry to accelerate the decarburisation of the manufacturing of hot rolled steels with a martensitic matrix. With a total market size of about 5 million tonnes per year globally, these AHSSs make up only a small portion of the total flat carbon market (1800 million tonnes). However, redesigning existing grades (and hence existing tonnages) towards low CO2 versions represents several billions of euros worth of tonnage in absolute values. In addition, thanks to the excellent mechanical properties of martensitic steels, they are of particular interest to the flourishing market segments of yellow and green goods and heavy transport applications.

Forging and quenching

Steel will remain important for sector large-size forgings, mainly as elements of turbines and generators. Therefore, the optimisation of production (including energy-intensive heat treatment) will allow for a significant reduction of costs and environmental impact. Achieving higher efficiency, long durability using sustainable steels, and lower costs in the manufacturing processes of components such as wind turbine shafts, are all essential factors in ensuring improved competitiveness. In this regard, increasing the efficiency and effectiveness of the materials and of the product and the manufacturing process for these fundamental components will have a significant positive effect in the sector: more reliable components at a reduced cost and reduced time to market.

Research objectives:

Develop AI-Based Chemistry-Process-Structure Modelling Tools: The project seeks to create three AI-driven tools that facilitate the design and processing of robust materials. These tools will also optimize production routes, leading to reduced CO2 impact.

Predict Creep Performance with Machine Learning: A machine learning-based tool will predict creep performance, and a rapid data generation methodology based on accelerated creep testing will enhance R&D efficiency. This focus extends to heat-resistant steels and their manufacturing processes.

Model Microstructure Evolution for Meter-Scale Products: A sequential model will predict microstructure changes during forging and quenching of meter-scale products. By improving existing manufacturing methods, it aims to eliminate re-processing of defective parts and reduce energy/material waste associated with trial-and-error R&D tests.

Validate AI-Based Tools and Characterization Methods: The project will validate the AI-based tools and characterization techniques using both existing and newly generated data within the AID4GREENEST framework.

Create an AI-based Online Platform for Knowledge Transfer: An online platform will facilitate knowledge sharing, standardization, and guidance related to characterization methods and modeling tools. It will provide open access publications and annotated data from microstructural and mechanical analyses.

Perform Life Cycle Assessment and Foster Model-Based Innovation: The project will conduct life cycle assessments and promote model-based innovation processes across all stages, from materials design to product development