We are seeking a motivated and talented PhD student to work alongside our Digital Twins team within the Advanced Steel Research Centre at WMG, University of Warwick. Our Digital Twin team are developing a suite of models to deliver a through-process microstructural and mechanical property prediction framework for steel. By integrating these models into a cohesive digital twin architecture, the work will enable steel producers to rapidly predict and respond to live production conditions, supporting fast, intelligent decision-making to optimise process parameters, product performance, and operational efficiency. This PhD will focus on developing a specific Digital Twin for steel section production working with our industry partners.
Rising energy prices, volatile raw material costs, and increasingly demanding mechanical property specifications have placed the steel industry under significant pressure to modernise its production methods. The UK steel industry is transitioning to using more, and potentially greater variability, scrap steel as feedstock, therefore digital twins are required to offer support in understanding the potential effects on processing and properties, and adopting appropriate control strategies. Current steel processing typically has very tightly controlled processes with little variability and control approaches optimised to minor changes. The combined challenges of cost competitiveness, quality assurance, and sustainability targets along with greater variations being introduced into the process mean that new approaches are required.
To transition into an era of smart steel production, greater predictability is required, which can be gained by utilising the complementary strengths of empirical modelling, finite element analysis, and artificial intelligence. By combining these tools within a real-time digital twin framework, steel producers can access rapid, data-driven insights that support optimised process control, reduced reject and downgrade rates, and meaningful improvements in energy efficiency and sustainability across the production chain.
The Advanced Steel Research Group at WMG, University of Warwick, has developed a suite of models covering various stages of the steelmaking process. The aim of this project is to develop new insight in the application of a Digital Twin for steel section production. This will involve combining several of existing models into a cohesive through-process framework, working closely with industry partners and real-world production data to create a model process route capable of optimising steel properties within the practical constraints of mill and production operations.
The scientific challenge will be to use the model and machine learning alongside live mill data (temperature, rolling loads etc) to reverse engineer the current microstructural state of the material and provide feedback on optimised next stage processing.
Given the computationally intensive nature of finite element modelling, the outputs of these models will be used to train a machine learning tool. This will enable rapid integration, feedback and process optimisation without the need to directly run the core finite element models at every step, making the framework suitable for deployment in fast-paced industrial environments.
The successful candidate will be responsible for defining the architecture, robustness and limitations of this combined modelling framework, as well as its implementation within an industrial setting. As such, the project will include a placement with the industrial partner, the timing and duration of which can be agreed in discussion with all parties.
Funding information
DigitalMetals CDT
