NURC
Projects
Project 1 – Superslab, modelling of inclusion formation and floating, tundish flow optimisation
Persons involved
- Jean-Christophe Gebelin
- Xiaoan Yang
- Nanfu Zong
- Andrew Dunsmore
- Chinnapat Panwisawas
Summary
About 95% of steel production worldwide is made using the continuous casting process. In this process, the steel that has been prepared to meet chemical specification is poured from a ladle into a tundish, which will distribute the steel to the different strands of the continuous caster. As well as distributing the steel, the tundish improves the steel quality by floating any non-metallic inclusions that could have get into the melt in the previous step of the process. A good tundish design will minimise the differences at the different strand for steel temperature and residence time in the tundish, while enabling a steady flow that enable the floatation of light non-metallic inclusions.
The project so far has investigated nine different geometries, analysing temperature differences at the outlets, flow patterns, downward velocity index, plug and dead flow volume fraction and standard deviations of plug flow and dead flow between the two outlets. This enables us to check that the steel getting into the different strands of the continuous caster have small differences in terms of temperature and residence time, and that the flow does not entrain any light inclusion into the strands.
We now concentrate on a systematic study of the effect of seven design variables on the performance of the tundish for steel cleanness. In April 2022, we will start a physical modelling, using water model, for four selected geometries in collaboration with University of Science and Technology Beijing. This will enable us to validate the modelling done so far and get more confidence in our models’ predictions, before starting an industrial trial later this year.
Project 2 – Data driven modelling BOF process and hardness prediction
Persons involved
- Bogdan Nenchev
- Zihui Dong
- Xiaoan Yang
- Jun Fu
- Chinnapat Panwisawas
- Qing Tao
Summary
The mission our NURC’s data science team is to promote and implement machine learning for steelmaking in order to improve the industry competitiveness. Steel, an alloy of iron and carbon containing less than 2% carbon, is our most important construction and engineering material, with over 1.9 billion tonnes of crude steel produced in 20201. About 70% of this steel is produced by Basic Oxygen Steelmaking (BOS). In a highly competitive industry, steel producers have to ensure their furnaces are operating at maximum efficiency, maximizing product quality and process safety while minimizing energy consumption and environmental impact. A key requirement is to monitor, control and optimize the final carbon content while maximising the yield.
The project facilitates knowledge transfer and collaborations between academic researchers, engineers, IT departments, R&D specialists, and businesses. As part of our long-term collaborative programme, NURC performs data extraction and processing enabling the application of machine learning algorithms. In the work package of NURC, we have developed bespoke neural network algorithms handling multiple data types and solving challenges within big industrial datasets. We are aiming to provides accurate, real time data analysis to furnace control systems and dynamic control models, resulting in significant process benefits. A 1% increase in throughput is worth around $20,000 a day for a plant producing 10,000 tons of steel a day, so payback is extremely fast.
Project 3 – Process simulation welding and digital twining of continues casting
Persons involved
- Jean-Christophe Gebelin
- Jun Fu
- Nanfu Zong
- Andrew Dunsmore
Summary
95% of steels are produced using the continuous casting route. Maintaining the quality of the products while maximising productivity can be helped using Digital Manufacturing. Increasing the withdrawal rate on the strands of a continuous caster will have an impact on the solidification of the steel, changing segregation pattern and microstructure produced, on the inclusion being drawn with the melt, on the requirement for magnetic steering, on the secondary cooling requirements, etc.
In this project we will build up on our previous modelling work and look in details at the behaviour of the steel and heat transfer mechanisms in the mould to better understand the impact of changing casting speed on the dynamic of the steel in the mould. We will also build model for the secondary cooling to investigate the effect on microstructure formed and segregation. Effect of the casting speed change on the performance of the tundish will also be considered using models developed in project 1.
These models can then be used to optimise casting parameters and understand impact on quality.
They also have the potential to help with casting difficult steel grade such as high manganese steels.
Schematic diagram of the mould region showing the different phenomena to be considered (from Modelling of the continuous casting of Steel – Past, present and future by Brian G. Thomas, Met.Mat.Trans.B, 33, 2002, pp795-812)
Project 4 – Digital-enabled Decarbonisation for Steel Industry
Persons involved
- Chinnapat Panwisawas
- Qing Tao
- Andrew Dunsmore
- Hongbiao Dong
Summary
Reduction in carbon emission to meet the net-zero target in 2050 in the UK and 2060 in China can be achieved using digital technology to accelerate the circular economy (reuse, recycling or remanufacturing) and maximise materials and process efficiency. The global steel industry accounts for more than 10% of CO2 emission, which is needed to be revolutionised by strategic decarbonisation pathway to be environmentally friendly.
In this project, hydrogen technology and digital-enabled decarbonisation are to be studied and researched in strategic collaboration with NISCO, University of Cambridge, WMG as well as other global industry and academic partners. This is to accelerate the development of net-zero carbon steelmaking and hydrogen technology for green steel metallurgical industry such as scrap utilisation and injection of hot reducing gases into blast furnace, and to provide an international research consortium for scientists, engineers, technologists and industry leaders to achieve the sustainable, resilient, and intelligent manufacturing.