B.G.

Thomas: 1 Baosteel Annual Academic Conference, May 27-28, Shanghai, China

st

1

COMPUTATIONAL MODELING OF FLOW, HEAT TRANSFER, AND DEFORMATION IN THE CONTINUOUS CASTING OF STEEL
Brian G. Thomas Mechanical Engineering, University of Illinois at Urbana-Champaign, IL, USA Abstract: Further technology improvements to complex mature processes, such as the continuous casting of steel, require a combination of careful laboratory experiments, plant trials on the commercial process, and advanced computational models. As computer power increases, computational models are able to contribute more to the understanding, design, and control of these complex processes. Fluid flow models can now include phenomena such as transient behavior during steady casting, including particle transport, capture and removal. Heat flow models can include interfacial slag layer heat, mass and momentum balances, and nonequilibrium crystallization behavior, and can predict slag structure and friction within the mold. Stress models can now make quantitative predictions such as ideal mold taper and maximum casting speed to avoid problems such as off-corner longitudinal cracks. This work presents recent examples of these models, and their comparison with experimental measurements. Key words: Computational modeling, continuous casting, turbulent flow, heat transfer, interfacial friction, stress analysis, crack formation 1. Introduction Understanding and controlling the continuous casting process is important because it may introduce defects such as inclusions and cracks, which persist into the final product, even after many later processing steps. In addition to plant experiments and water models, advanced computational models are increasingly able to generate this understanding. Helped by recent improvements in computer power and software, computational models can simulate a wide range of phenomena ranging including fluid flow, heat transfer, and stress generation. This paper shows recent examples of these from the U. of Illinois Continuous Casting Consortium. 2. Simulations of Fluid Flow in Nozzle and Mold Three-dimensional turbulent flow and particle motion is being predicted in both the nozzle and mold using the Large Eddy Simulation code, UIFLOW [1]. This approach is more accurate than the standard K- turbulence model for transient flows [2], and not as computationally intensive as DNS (direct numerical simulation). An example for a trifurcated nozzle submerged 127mm deep into an 984mm wide x 132 mm thick stainless-steel caster is shown in Figure 1 (model domain) and Figure 2 (typical result) [3]. The velocities compare well with quantitative velocity measurements in a water model of this caster [4]. As shown in Figure 3, the model can simulate the top surface contour in both water model (measured from video frames) and in the actual steel caster (measured by inserting a steel sheet into the mold [5]). The shape follows expectations for a typical double-roll flow pattern. The jets impinging the narrow face wall turn upwards and downwards along the narrow face. The momentum of the upward jet lifts the meniscus by a height that can be approximated with a simple energy balance based on the pressure difference at the meniscus [4]. The surface

Shanghai.4 SEN SEN -2 -10 -12 0 -4 Instant t = 20. Typical instantaneous velocity vector plot at the center plane between wide faces Steel surface: liquid level profile (mm) 10 8 6 4 2 0 -2 -4 -6 -8 Predicted Level.1 132mm -0.4 (m) 0.3 0. Note that the particles move with an asymmetrical distribution. China st 2 contour is similar to that of the water model. The approach was first validated by a successful comparison of transport in a water model where over 50% of the 15. Comparison of predicted and measured top surface liquid levels Distance from center.4 0.5 9 8 4 mm Turbulent jets 2400mm Liquid pool Narrow face Wide face x y z 0.6 0. x (mm) 0. in spite of the symmetrical boundary conditions.2 (m) Figure 1.3 0. In the actual thin-slab caster. Portion of tundish Stopper rod SEN 0 0. After a 9s sudden burst of particles entering the steel caster. Figure 4 shows a snapshot of the distribution of moving particles in the liquid pool.B. which can either be entrapped into the solidifying shell to form slivers and cracks.1 1. except that the slag layer buoyancy increases the magnitude of the variations.000 particles floated out in 100s. rather than by inlet variations at the nozzle port. Particle transport were simulated using a Lagrangian trajectorytracking approach. Computational domain of tri-furcated nozzle and thinslab casting mold Water surface: liquid level profile (mm) 4 Measured Level (Instant 1) 2 Measured Level (Instant 2) Measured Level (Instant 3) Predicted Instantaneous Level 0 Figure 2.4 Distance from center. ~ 4 minutes were needed for all of them to be captured or removed.8 0. Right Predicted Level.2 0.000 small particles were removed to the top surface [4]. Left Measurement 0.3 0. May 27-28.2 0. .4 -0.1 Water model Steel caster Figure 3. Thomas: 1 Baosteel Annual Academic Conference.2 0 1. based on the time-dependent flow fields just discussed [6].7 0. or can be safely removed into the top surface slag layer. only about 8% of the 40. This is caused by transients in fluid turbulence in the lower recirculation region.0m/s: 0.9 1 9 3 4 mm Domain Bottom 80mm 1.2 0. Level fluctuations can also be predicted.2 0. x (m) The model can also predict the motion of inclusion particles.1 0 0.3 in simulation 0.G.

Distribution of 7.1 .5 5 0 Figure 6.2 0. Thomas: 1 Baosteel Annual Academic Conference.5 0 -5 0. Figure 6 shows how steel that initially contains 10ppm at the nozzle ports will ultimately range from only 6ppm in the interior to over 50ppm in spots near the narrow face. Predicted total oxygen concentration averaged along the slab length 0.5 moving particles.3 -0.05 0 The results of this work confirm the important role of flow transients in the transport and -0.4 -0.B. Shanghai.1 0.G.5 0 -5 0.4mm/s -4 Distance down solid slab (m) -2 Distance down solid slab (m) -3 -3 -2 -1 Strand downward withdrawl during 9s particle injection -1 0 Oxygen concentration distribution 0 Nozzle port Oxygen concentration: 1ppm Total Oxygen (ppm) 100 50 25 15 10 Figure 4. China st 3 The computational results were further processed to predict the ultimate distribution of impurity particles in the solid thin slab after a short burst of inclusions entered the mold [4]. Figure 5 shows that most of the particles end up between 1m below and 1m above the location of the meniscus when they entered the mold.05 0 -0. May 27-28.1ppm average for -0. especially near the narrow faces. 6.5 1 1.0s 1 Meniscus location at start of injection 1 (m) 2 2 (m) Figure 5.05 capture of particles during continuous casting.2 -0. Further details are given elsewhere [3. and can serve as a benchmark (m) future -0. The greatest inclusion concentration is found near the surface.1ppm average for uncomputed interior region -0.5 (m) Thickness (m) -0. The results were reprocessed to reveal the distribution of total oxygen content for a steady inclusion supply from the nozzle.05 (m) 6.5 (m) -4 Casting speed: 25. 6] Width (m) -1 -0.3 for -0.3 0. Note that the asymmetric particle distribution during the flow only leads to slight asymmetries in the distribution in the final product. Final entrapment locations for a 9s-burst of particles: view from wide face View from narrow face Total Oxygen (ppm) 0.1 0 0.4 simplified models.4 (m) 100 50 25 15 10 7.2 -0. 48s after 5 injection 0 48.

finding the metallurgical length. It features a detailed treatment of the interface between the solidifying strand and the water-cooled mold which governs heat transfer in the process. the thermal history and interfacial friction can be predicted in the gap between the shell and mold wall.8mm 1. mold taper. May 27-28. d =2mm l CON1D FDM Model -50 Velocity (mm/s) t=3/2 top surface meniscus element for force balance mold ferrostatic pressure 0 t=/2 steel shell 50 molten steel t=2 shear stresses solid slag Vc liquid slag 100 0 0. Modeling the Mold / Strand Interface Heat transfer and solidification models are useful for many purposes. marks Distance from mold hot face 0. and a vertical-direction force balance between friction against the mold wall and shell. To study these phenomena. To investigate mold flux behavior in the gap.9% Al O 2 3 Figure 10 -Effect of casting speed on friction Figure 9 . The accuracy of this part of the model is shown by its matching of results from a finite difference model of flow in the gap. and axial stresses in each slice through the solid flux layer (Figure 7). the design of spray-cooling. including the design of cooling water slots.5 1 x (mm) 1. as shown in Figure 9. a model (CON1D) has been developed that includes a 1-D transient solidification model in the shell coupled with a 2-D steady model of the model [7]. During each oscillation cycle. . x z Vm slag rim slag powder liquid slag pool h0 -100 t= (d) n=1.B. 1400 Slag in Osc.2mm 0.4mm 0.G. and trouble-shooting the location of internal defects (hot tears that form at the solidification front).6.6 and film thickness = 2mm) From the velocity profile and flux consumption.5 2 Figure 7 -Schematic of interfacial gap phenomena in continuous casting mold Figure 8 -Velocity profiles in liquid flux layer (for viscosity exponent = 1.Slag layer cooling history with force: measurement and prediction attached slag and measured TTT curves Time (s) 1000 .0mm Temperature ( C) 1200 1000 800 600 400 200 1 10 100 o 7. shell thickness. China st 4 3. Thomas: 1 Baosteel Annual Academic Conference. Shanghai. which is predicted based on the measured TTT diagram [9] . with a curved profile according the temperature-dependent viscosity of the liquid flux [8].6mm 0. The mold flux properties change greatly if the flux crystallizes. the model shows (in Figure 8) that the velocity profile in the liquid flux layer varies from the casting speed on the shell side (right) to the mold speed on the solid side attached to the mold wall (left). Further details are given elsewhere [8]. the model includes a momentum balance on the solid and liquid flux layers in the gap.

non-uniform superheat dissipation due to turbulent fluid flow. which creates high friction forces. China st 5 By integrating the predicted shear stress distribution down the mold over the entire oscillation cycle. and austenite phases. the flux against the steel shell is molten. 100 200 300 400 500 600 700 800 Distance Below Top of Mold (mm) Figure 13: Temperature contours at 285mm meniscus compared with Figure 12 Computed & measured mold below corresponding sulfur print from plant trial. Super Heated Liquid Steel L Shape Domain (CON2D) Ferrostatic Pressure Solidified Shell Liquid Steel Casting Direction Gap Tapered Distroted Mold Billet Figure 11. May 27-28. Schematic of billet casting 160 140 120 100 80 60 40 20 0 Predicted Measured Hot face Cold face Heat flux across the inter-face depends on the size of the shrinkage gap. and the strand roughness. CON2D. Thomas: 1 Baosteel Annual Academic Conference. Total heat removal was calibrated to match the measured heatup of the cooling water. such as insufficient taper. The model was validated with analytical solutions. A contact algorithm is used to prevent penetration of the shell into the mold wall due to the internal liquid pressure. It includes the effects of heat conduction. s 4. solidification. Shanghai. taper and thermal distortion of the mold. friction increases due to factors not currently included in the model. temperatures . High in the mold. as with the CON1D model. both in and below the mold [10]. and shape development during the continuous casting of steel. The model simulates a transverse section of the strand in generalized plane strain as it moves down at the casting speed (Figure 11). deltaferrite. semi-solid. has been developed to simulate temperature. the model can predict the total friction force in mold. It is further validated by simulating con-casting of a 120mm square billet [10]. Stress Modeling of the Solidifying Steel Shell A coupled finite-element model. This powerful modeling tool can help the interpretation of plant measurements such as friction. as shown in Figure 10. the flux layer may become completely solid. Figure 12 shows that the predictions match with thermocouple measurements of mold wall temperature [11]. At high speed. It can also be measured in the plant using accelerometers. Low in the mold.G. so friction is low. mutual dependence of the heat transfer and shrinkage on the size of the interfacial gap. Increasing speed increases the liquid region.B. The stress model features an elastic-viscoplastic creep constitutive equation that accounts for the different responses of the liquid. thereby lowering friction. the conductivity of the gap vapor. stress. including thinning of the corner [11]. An efficient two-step algorithm is used to integrate these highly non-linear equations. which agrees with measurement. Figure 13 shows that the model is able to predict shell thickness.

13%C. owing to their deeper oscillation marks.5 0. excessive bulging. The heat flux profiles for different steel grades and mold fluxes are shown in Figure 14. 60 50 0. Flux solidification T: 1120 C 0. this causes them to have less shrinkage and less taper. especially of the narrow face [12]. One of these is the avoidance of longitudinal cracks.8 0.1 1300 0.7 0.1 1 0.27%C. Flux solidification T: 1120 oC 0. As shown in Figure 15.1 40 0. and mold distortion. Shanghai.6 0. Flux solidification T: 1120 oC 300 400 500 600 700 800 X | X | X | X | 1 0 0 100 200 300 400 500 600 700 0 800 100 200 DISTANCE BELOW MENISCUS (mm) DISTANCE BELOW MENISCUS (mm) Figure 14: Heat flux profiles for different Figure 15: Ideal taper for different steel steel grades (1.4 1.6 0. Flux solidfication T: 1215 oC 0.4 Y(mm) 30 20 -0.2 60 0 X(mm) 30 50 Figure16: Damage Strain Contours at 100mm below Mold Exit (4. such as the one pictured in Figure 17. gradedependent creep strain.2 Maximum casting speed to avoid hot-tear cracks Another application of the stress model is predicting the maximum casting speed to achieve quality steel at maximum productivity [13]. | 5X 5 | X SHRINKAGE .5 m/min 0.B.79 0.13%C.2 1. Flux solidfication T: 1120 oC 0. reheating in the spray zones and many other problems. insufficient metallurgical length. China st 6 4.4 0.3 0.1 0.5m/min) grades (1. Thomas: 1 Baosteel Annual Academic Conference.02 00 12 -0.1 Optimization of mold taper design One application of the thermal stress model is to optimize mold taper design.5 1. which depends on steel grade (temperature range across the mushy zone) and strain rate.27%C.07%C.9 0.5m/min) 4. Flux solidfication T: 1120 C 0. Flux solidfication T: 1120 oC o 4 3 | X X | X | | X 3 2 X | X | X | X | X | 2 X X X X X X X X | X | 1 X | X | X | Casting speed: 1.1 | 0X 0 | |X | X | X X | X | | X X | | | | | | | | | | | X X X X X | | X | | X 4 | X | X | X HEAT FLUX (MW/m2) Casting speed: 1. in spite of the phase transformation contraction from delta-ferrite to austenite experienced by these steels. Flux solidification T: 1215 oC o 0.8 0.3 1. Casting speed can be limited by factors ranging from excessive level fluctuations.4 10 0 1.47%C.11411.5 m/min 0. Peritectic steels have lower heat flux. Ideal mold taper is affected by thermal shrinkage of the shell.47%C.2 1.G.72 1477.4 40 1500.2 0.4 10 20 -0.07%C. May 27-28. mold friction. and high-solidification mold fluxes usually used for these grades. Hot-tear cracks are predicted to form when the “damage strain” (tensile strain accumulated at the solidification front within the temperature range between 90 – 99% solid) exceeds a critical value.4 m/min) Figure 17: Con-cast billet after breakout showing offcorner sub-surface crack .NARROW FACE DISTORTION (%) 1.

P. Dallas. Y. Warrendale. 2001. 34B (5). B. ISS. Thomas and C.G. Thomas & S. Thomas. 2003).-k. B.G. B. 12.. TX. Thomas and S. Samarasekera. & Mat. "Interfacial Friction-Related Phenomena in Continuous Casting with Mold Slags. 86. Thomas. The quantitative ability of the model is demonstrated by comparison with plant measurements. 41 (10). Vanka. C. 2004. Q. (Indianapolis. 9. B. & Material Trans. 2003. . L. TN. Middletown. 29 (5). Yuan. Steelmaking. 2002. Warrendale.G. Vol." Ironmak." in Steelmaking Conf. Thomas. 2003.. Southern Methodist Univ. Particle Transport. "Study of Transient Flow and Particle Transport during Continuous Casting of Steel Slabs.W. Service.G. Proc. Mat." in ISSTech 2003 Manfred Wolf Memorial Symposium Proceedings.. & Materials Trans. B. An implication of the model results is that higher casting speed is possible with carefully aligned foot rolls. 359-375.B. 2002). Meng and B. and Chunsheng Li. Proc. B. 589-606. Y. Q. Vol. so can be cast at higher speed. Li and B. Ya Meng. R. Longer molds allow a thicker shell to form. 913-927. Thomas. and by avoiding mold wear. and to the National Center for Supercomputing Applications (NCSA) at UIUC for computer facilities. 86. 7. (Indianapolis. Fluid Flow. 6. "Ideal Taper Prediction for Slab Casting.G. Apr." in ISSTech 2003 Process Technology Proceedings. The model can be used for many other practical applications. J. Y." Metal. ISS. submitted August. S. 185-191. Thomas. Special thanks are due to graduate students who worked on these projects. personal communication. O'Malley. II. Meng and B. IN. 2003. Vol.. Proceedings. Vol. March 10-13. Vanka. S.P. "Analysis of Thermo-Mechanical Behavior in Billet Casting with Different Mold Corner Radii. 86. I. PA. 81." Metall. and the bulging from internal ferrostatic pressure causes corner rotation and tensile strain at the solidification front (Figure 16). Vanka.J. B. 2003. ISS. Cicutti and A. IN.[13] assuming ideal mold taper and uniform temperature below mold (ideal spray cooling)." Metall. 11. Trans." in Steelmaking Conf. 295-308. B. Vol. thus increasing the allowable maximum casting speed. Warrendale.P.P. Ojeda. 1998. 6. "Heat Transfer and Solidification Model of Continuous Slab Casting: CON1D. 2003). submitted Aug." in ISSTech 2003 Steelmaking Conf. Thomas & S. PA. Proc. May 27-28. Sivaramakrishnan. Zhang. 10." Metal. 85.G. 109-130. and Manufacturing Grantees and Research Conf. Yuan. 2003. "Flow Dynamics and Inclusion Transport 2. 3. Kashiwaya. the shell thickness at mold exit decreases. C. Casting speed was increased until the maximum damage strain reached the critical value for hot-tear cracks.G. Warrendale. T. Yuan. Apr. 27-30. IN. 27-30." in 2004 NSF Design. TX). "Turbulent Flow and Particle Motion in Continuous Slab-Casting Molds. China st 7 With increasing casting speed. "Comparison of Four Methods to Evaluate Fluid Velocities in CC. Q. "Crystallization Behavior of Mold Slags. B.P. 8. 2003. Vanka. 2003). C. "Study of Transient Flow and Particle Transport during Con. 1266-1276. (Dallas. "Maximum Casting Speed for Continuous Cast Steel Billets Based on Sub-Mold Bulging Computation. Shanghai. Shi. Thomas: 1 Baosteel Annual Academic Conference. PA. Yuan. Acknowledgements The author thanks the National Science Foundation (Grant DMI-01-15486) and the member companies of the Continuous Casting Consortium at the University of Illinois at Urbana-Champaign (UIUC) for research support. Q. "Thermo-Mechanical Finite-Element Model of Shell Behavior in Continuous Casting of Steel. 685-705.G. Vol. B. Trans.B. Results for the critical maximum casting speed were calculated as a function of mold length and section size. Vol. Thomas and I. Similarly. References 1. Apr 27-30. The highest damage strains arise at the subsurface off-corner location where the longitudinal cracks are ultimately observed (Figure 17). Cramb. (Nashville. PA. M.G. Zhao.. S. Casting of Steel Slabs. 41p. 5. 2003. Vanka.V. Yuan. Thomas. 7. Assar. (Indianapolis.G.G. accepted Oct. PA. OH. Park. ISS. B. Li and B. Q. Warrendale. 2003. Iron and Steel Soc. 13. in Continuous Casting of Steel.G." ISIJ Internat. including Quan Yuan. 2002. 4. smaller section sizes experience much less bulging.

Sign up to vote on this title
UsefulNot useful