Can Industrial Maths help with issues in the steel industry?
Posted on 23/04/2019
Driving translation of mathematical and statistical research advances into high value applications in industry is vital to unlocking key societal and economic challenges in clean growth and sustainability.
To aid this process, a range of problems will be tackled by researchers at a three-day Clean and Sustainable Growthstudy group which will take place in Nottingham from 29 April – 1 May.
The organisers, KTN, alongside the University of Nottingham, are looking for researchers to work on the following conundrums, both presented by Tata Steel, one of the world’s most geographically diversified steel producers, with operations in 26 countries and commercial offices in over 35 countries:
Evolution of the thermal state of the steel transfer ladle and its liquid contents
The real-time modelling and predicting the evolution of the thermal state of the steel transfer ladle, its refractory lining and liquid steel content is fundamental to being efficient and even optimal in managing energy (and speed of work) in integrated steelmaking.
The existing approach to representing the ladle and its refractory lining has been to use finite difference approach and to reduce the computational load by assuming and applying symmetry. Alternative approaches that may be more insightful or computationally feasible but with less simplification through symmetry can be of value.
Greater insight from new approaches might yield better ability to understand and control energy flows and to look with greater granularity at the heat fluxes and forces that the expensive refractory lining materials are exposed to.
Optimal scheduling of steel-making and casting operations
Integrated steelmaking delivers a semi-finished slab product by connection of a series of batch unit operations to the semi-continuous casting process. Steel transfer ladles and transport operations via transfer cars and cranes connect and bind these discrete process steps together.
Nominal scheduling to optimise throughput or robust delivery of a specific product mix is straight forwards. Successful identification of optimal schedules proves much more difficult as plant reliability and variable, human decision making are brought into scope.
Previous attempts, with Goldratt Research labs, to apply discrete event and agent-based simulation to reveal constraints and countermeasures has proved unsuccessful.
Key to the difficulty was the premise that inconsistent human decision-making could be represented repeatability by sufficiently complex logic but that the complexity in turn proved to be a barrier preventing validation of the veracity of the logic and that continued efforts to refine such logic led to divergence from rather than convergence with benchmarked historical performance.
Again, new approaches that might reveal optimum strategies, concepts or rules for scheduling in such complexity can provide a valuable planning and forecasting tool.
The latter two are problems related to the Industrial Strategy Clean Growth Grand Challenge, Transforming Foundation Industries: this challenge will transform the UK’s foundation industries (glass, metals, cement, ceramics, chemicals) to make them internationally competitive, securing more jobs and greater sector growth by 2025.
If you are a researcher working in a UK university who would like to work on this problem at the study group in Nottingham from 29 April – 1 May, please register here. Early stage career academics, Ph.D students, and postdocs are particularly welcome.
This Study Group is fully-funded by the University of Nottingham Leverhulme Doctoral Scholarships programme ‘Modelling and Analytics for a Sustainable Society’ with support from KTN.