|(NSF) High Fidelity Numerical Investigations of Tailored Magnetic Fields for Defect Reduction in Continuous Casting of Steel
S.P. Vanka, B.G. Thomas, S.M. Cho, K. Jin, R. Lui, R. Singh
National Science Foundation CMMI 1130882
National Science Foundation Support
9/1/11 to 8/31/15
NSF Program Director: Mary Toney, (703) 292-7008
Industry Support and Partnerships
Continuous Casting Consortium
The Continuous Casting Consortium in the Department of Mechanical Science and Engineering at the University of Illinois at Urbana-Champaign is directed by Brian G. Thomas. Its objectives are to develop comprehensive mathematical models of continuous casting processes and to apply those models to problems of practical industrial interest. Steel-related companies are members of the Consortium and contribute funding, measured plant data obtained at the steel plant, and advice, to the Consortium, which is helpful to the success of the project. Companies most relevant to this project are Severstal, ABB, Baosteel, POSCO, Nucor and Arcelor-Mittal. Severstal Recent collaborations with Severstal provided Ramnk Singh and Rui Liu an opportunity to conduct nail board measurements at one of the slab caster at Dearborn, MI, USA. CUFLOW was then applied to study mold flow in this caster geometry and the calculated results were compared with nail board measurements. Furthermore, CUFLOW was applied to predict the effect of FC mold EMBr configuration on this caster. ABB ABB has been using magnetic fields to control mold flow in industrial casters. They are the leaders in this technology and have been producing commercial products known as the FC Mold. They will be providing our research with data of the initial fields applied in the FC Mold to enable accurate modeling of the fluid flow at industrial conditions. Baosteel, Postech, Nucor and Arcelor-Mittal Experiments on steel casting using EMBr are carried out at several steel companies by technical personnel from the company with the help of students associated with the CCC. Specifically Seong-Mook Cho and Kai Jin have been analyzing the results of experiments to study the surface flow in continuous slab casting mold under the influence of EMBr using nail boards to measure the surface profile and surface velocity.
Provided plant nailboard measurement; provided captured bubble distribution report.
MHD Department, Forschungszentrum Dresden-Rossendorf
We have collaborated with researchers from the MHD department at FZD, Dresden and compared with their experimental data for solver validation. The researchers directly involved are Klaus Timmel, PhD. Student, and Sven Eckert, Department Head of MHD Department.
Provided plant measurements of fluid flow with nailboard measurements.
CUFLOW is currently limited to use of a single GPU when performing calculations. Utilizing a cluster of GPUs will allow for faster calculations, as well as a greater resolution. With multiple GPUs, we will be able explore many more parameters, such as such as magnetic field strength, position, type and combinations of mold geometry, nozzle shape and casting speed.
1. Implement multi-GPU data structure into CUFLOW.
2. Test multi-GPU CUFLOW performance on different platforms: desktop with
multiple GPUS connected to the same motherboard and supercomputers with
CPU as co-processor.
3. Use the developed multi-GPU code to study the effect of double ruler EMBr in
flow pattern, surface velocity and fluctuations in a real commercial caster.
A multi-GPU version of CUFLOW has been developed and tested. The code is written with CUDA Fortran and works on multi core desktop which has multiple GPUs as well as supercomputers with NVIDIA Compute Unified Device Architecture (CUDA) compatible GPUs as co-processors. Message Passing Interface (MPI) is used for data transfer between different computer nodes. The multi-GPU code was validated by simulating the lid-driven cavity problem and comparing with many existing published results and good agreement has been achieved. After validation, the code was compiled and tested on different platforms including a desktop computer which has 4 GPUs connected to a single motherboard, and Blue Waters supercomputer which has CPU-GPU as a pair in each node. The performance of the code was evaluated by running with different number of GPUs and on different platforms.
To study the effect of EMBr on real commercial caster, several high fidelity LES simulations were performed by using this multi-GPU CUFLOW. Total 8 LES simulations were conducted with different settings of casting speed and different magnetic field which comes from plant measurement. The flow fields were studied, and comparisons of surface velocities, surface flow patterns, surface level profiles, surface level fluctuations, mold flow patterns and Reynolds stresses were made.
In coordination with the CCC, synergistic research have been conducted to compare the accuracy and efficiency of different computational methodologies. In addition to the effects of EMBr on mold flows, other phenomena such as multiphase flow have been studied. Measurements from nail board experiments have been analyzed to determine temporal variations of top surface profile and velocity to understand free surface behavior in real casters under the effect of various EMBr configurations, and to compare with numerical results. Two papers discussing the results from this study have been presented.
CUFLOW has been used in the past year to correctly predict the significant effect of conducting side walls on the flow pattern in the continuous casting mold, study scaling criteria in presence of applied magnetic fields and understand the mold flow in real commercial casters. It has also been used to predict the effect of a FC-Mold EMBr on a commercial caster and in future could be used to optimize the applied field to tailor specific characteristics of the flow pattern such as surface velocity.
Training and Professional Development Provided
Support from NSF on this project has trained several graduate students.
1. Ramnik Singh, MS degree (Mechanical Engineering), 2014.
2. Rui Liu, PhD (Mechanical Engineering), 2014 (partially supported)
3. Kai Jin, Ph.D. (Mechanical Engineering), expected in 2016
4. Seong-Mook Cho, Visiting Researcher, (Postech), (partially supported)
Additional support was provided through the Continuous Casting Consortium, which also benefited all the students in this project via mutual interactions, and from visiting the industrial plants to obtain measurements and discuss with industry personnel.
Dissemination of Results to Communities of Interest
1. The results of this study have been presented at CCC annual meetings, which are attended by representatives from the CCC member companies, and consist of a majority of the largest steel manufacturing companies in the world.
2. In addition, results from this work have been incorporated into the short course, “Brimacombe Continuous Casting Course” (Vancouver, Canada every May), presented annually to industry attendees by a team of 5 instructors (which includes Co-PI B.G. Thomas). Attendance exceeded 140 people in 2012, and ranged from operators who implement technology on the shop floor to company researchers who work on improving that technology. The attendees have been mainly from steel companies in the U.S.
3. The GPU computational methodology was presented (with code demonstration) as a tutorial in the ASME 2014 Fluids Engineering Summer Meeting at Chicago, IL, USA in May. Many conference participants attended the short course including researchers from companies and universities who are interested in GPU programing and multi-GPU implementation. Sample codes were handed out to the audience.
4. The detailed research methodologies (including both the computational methods, and plant measurement methods, such as nail-boards), results, and practical implications were presented to and discussed with several industry members of the consortium including two companies whose casters were simulated and measured: Severstal (now AK Steel) and Baosteel. Recommendations to improve the plant operations were made, and some suggestions have been implemented in plant. Follow-up research is being conducted at these plants.
5. The methodology used in this work and the important findings were published in journal papers and conference proceedings. Several presentations were given to engineers and researchers at several worldwide conferences.
CUFLOW – a comprehensive CFD tool for efficient LES modeling.
24 Publications (11 journal papers, 10 conference papers, and 3 theses)
B. G. Thomas, Q. Yuan, S. Mahmood, R. Liu, and R. Chaudhary, “Transport and Entrapment of Particles in Steel Continuous Casting.” Metallurgical and Materials Transactions B-Process Metallurgy and Materials Processing Science, 2014. 45(1): p. 22-35. Click here for a PDF version (2 MB)
R. Singh, B. G. Thomas, and S. P. Vanka, “Effects of a Magnetic Field on Turbulent Flow in the Mold Region of a Steel Caster.” Metallurgical and Materials Transactions B-Process Metallurgy and Materials Processing Science, 2013. 44(5): p. 1201-1221. Click here for a PDF version (2.7 MB)
S.-M. Cho, S. H. Kim, and B. G. Thomas, “Transient Fluid Flow during Steady Continuous Casting of Steel Slabs: Part II. Effect of Double-Ruler Electro-Magnetic Braking.” ISIJ International, 2014. 54(4): p. 855-864. Click here for a PDF version (2.2 MB)
R. Singh, B. G. Thomas, and S. P. Vanka, “Large Eddy Simulations of Double-Ruler Electromagnetic Field Effect on Transient Flow During Continuous Casting.” Metallurgical and Materials Transactions B-Process Metallurgy and Materials Processing Science, 2014. 45(3): p. 1098-1115. Click here for a PDF version (2 MB)
K. Jin, S. P. Vanka, and B. G. Thomas, “Three-Dimensional Flow in a Driven Cavity Subjected to an External Magnetic Field.” Journal of Fluids Engineering-Transactions of the ASME, 2015. 137(7).
S. P. Vanka, 2012 Freeman Scholar Lecture: “Computational Fluid Dynamics on Graphics Processing Units.” Journal of Fluids Engineering-Transactions of the ASME, 2013. 135(6).
S. P. Vanka, Liu, R., and B. G. Thomas, “Particle Transport and Deposition in a Turbulent Square Duct Flow with an Imposed Magnetic Field.” Journal of Fluids Engineering-Transactions of the ASME, 2014. 136(12). Click here for a PDF version (4.6 MB)
R. Liu, B. G. Thomas, L. Kalra, T. Bhattacharya, and A. Dasgupta, “Slidegate Dithering Effects on Transient Flow and Mold Level Fluctuations.” Iron and Steel Technology, 2014. 11(7): p. 87-98. Click here for a PDF version (4.2 MB)
S.-M. Cho, S.-H. Kim, and B. G. Thomas, “Transient Fluid Flow during Steady Continuous Casting of Steel Slabs: Part I. Measurements and Modeling of Two-phase Flow.” ISIJ International, 2014. 54(4): p. 845-854. Click here for a PDF version (2.2 MB)
B. G. Thomas, R. Singh, S. P. Vanka, K. Timmel, S. Eckert, and G. Gerbeth, “Effect of Single-Ruler Electromagnetic Braking (EMBr) Location on Transient Flow in Continuous Casting.” J. Manufacturing Science and Production, 2015. 15(1): p. 93-104.
K. Jin, B. G. Thomas, X.-M. Ruan, “Modeling and Measurements of Multiphase Flow and Bubble Entrapment in Steel Continuous Casting.” accepted by Metallurgical and Materials Transactions B.
K. Jin, B. G. Thomas, R. Liu, S. P. Vanka and X.-M. Ruan, “Simulation and Validation of Fluid Flow and Particle Transport in Continuous Slab.” IOP Conf: Materials Sci. & Eng. Modeling of Casting, Welding, and Advanced Solidification Processes (MCWASP XIV), 2015. 84(1): p. 012095:1-8.
S.-M. Cho, S.-H. Kim, R. Chaudhary, B. G. Thomas, H.-J. Shin, W.-Y. Choi and Sung-Kwang Kim, “Effect of Nozzle Clogging on Surface Flow and Vortex Formation in the Continuous Casting Mold.” Iron and Steel Technology, 2012. 9(7): p. 85-95. Click here for a PDF version (986 KB)
R. Liu, S. P. Vanka, and B. G. Thomas, “Particle Transport in a Turbulent Square Duct Flow with an Imposed Magnetic Field.” 2013. Lake Tahoe, Nevada, July 7-11, 2013: Proceedings of the ASME 2013 Fluids Engineering Division Summer Meeting, FEDSM2013. Click here for a PDF version (6.4 MB)
R. Liu, S. P. Vanka, and B. G. Thomas, “Eulerian-Lagrangian Simulations of Bubbly Flows in a Vertical Square Duct.” 66th Annual Meeting of the APS Division of Fluid Dynamics. 2013. Pittsburgh, Pennsylvania, November 24–26, 2013.
S. P. Vanka, B. G. Thomas, A. F. Shinn, R. C. Chaudhary, R. Singh, R. Liu, K. Jin and P. Kumar, “CUFLOW: A Collaborative Computational Tool for Transport Phenomena in Materials Processing.” MHD 2015: Nagy El-Kaddah Memorial Symposium on Magnetohydrodynamics (MHD) in Materials Processing. 2015. Orlando, FL, Mar. 12-15
B. G. Thomas, R. Singh, R. Chaudhry and S. P. Vanka, “Flow Control with Ruler Electromagnetic Braking (EMBr) in Continuous Casting of Steel Slabs.” in Fifth Baosteel Biennial Academic Conference (BAC2013). 2013. Shanghai, PRC, June 4-6, 2013. Click here for a PDF version (765 KB)
R. Singh, B. G. Thomas, and S. P. Vanka, “Effect of Electromagnetic Braking (EMBr) on Turbulent Flow in Continuous Casting.” AISTech 2013 Proceedings, Pittsburgh, PA, May 6-8, 2013. 2013. Assoc. Iron Steel Technology, Warrendale, PA. Click here for a PDF version (1 MB)
R. Singh, S. P. Vanka, and B. G. Thomas, “Effect of a Magnetic Field on Turbulent Flow in Continuous Casting Mold.” in Bulletin of the American Physical Society 57. 2012.
K. Jin, B. G. Thomas, and X.-M. Ruan, “Effect of EMBr on Multiphase Flow and Bubble Entrapment in Steel Continuous Casting.” Sixth Baosteel Biennial Academic Conference (BAC2015). 2015. Shanghai, PRC, October 20-23, 2015.
K. Jin, S. P. Vanka, and B. G. Thomas, “Large Eddy Simulations of the Effects of Double-Ruler Electromagnetic Braking and Nozzle Submergence Depth on Molten Steel Flow in a Commercial Continuous Casting Mold.” TMS Annual Meeting. 2016. Nashville, TN, Feb. 14-18, 2016.
Thesis and Dissertations
R. Singh, M.S. thesis, High fidelity numerical investigations of tailored magnetic fields for defect reduction in continuous casting of steel, University of Illinois at Urbana-Champaign, 2013
R. Liu, Ph.D. dissertation, Modeling transient multiphase flow and mold top surface behavior in steel continuous casting, University of Illinois at Urbana-Champaign, 2014
K. Jin, Ph.D. dissertation, Large Eddy Simulations of Inclusion Transport and Capture in Continuous Casting of Steel Subject to an External Magnetic Field, University of Illinois at Urbana-Champaign, 2016 (in preparation)