M.BOUREBIA, H.MAOUCHE, S.BOUHOUCHE, M.CHAOUR,
S.BOULKROUNE.
Welding and NDT Research, Centre (CSC)B.P 64 Cheraga/ Algeria
mounirabourbia@gmail.com, chaourmed@yahoo.fr, sofiane25000dz@yahoo.fr
s.bouhouche@csc.dz
Abstact - Continuous casting of steel is a complex industrial process, involving many factors and mechanisms. For steel makers it is very difficult to control the process parameters optimally to obtain a product free defect i.e. a quality with the best possible return. Numerical simulation is a tool that can be used to optimise the process parameters. This work has for objective to model the temperature field at the level of solidified skin at the outlet of the mold "Ts". There are , three important key parameters affecting the cooling process in the mold: The "immersion distance of the nozzle Di", the "casting temperature Tc" and the "casting speed Vc". Thes parameters are used as input parameters during a parametric study by the experiments: "box behenken" Plans. The "outlet temperature Ts" is measured on the consolidated layer of steel on leaving the mold. The simulation of continuous casting phenomen is carried out by the Fluent code 6.0. A relation between the above cited key parameters and the "outlet temperature Ts," was established by using the required algorithm.
Thus, the optimal parameters were obtained by predicting a minimum threshold temperature (TSmin) at the outlet of the mold.
.
Keywords: continuous casting, output tempetature, casting temperature, casting speed, distance immersion, model "box behenken."
I.INTRODUCTION
The financial stakes of the steel market are colossal, the main objective is to increase industrial productivity of the casting machine. During the solidification process in continuous [1], many factors should be taken in to account such as the casting speed, the casting temperature [2], the immersion nozzle [3], the geometry of the mold [4]
as well as other factors that influencing primary cooling [5].
The liquid steel is casted at a temperature "Tc"
through a "submerged nozzle at a distance Di" in the bottom of the cooled mold. A crust begins to form (see in Figure 1) during descent of the slab in the mold at a "casting speed Vc" [6]. So the temperature parameter is an important factor affecting the formation of a thick skin. The objective of this work is to model the temperature fields "Ts" on output the mold. Therefore, three parameters were considered: the "casting temperature Tc" "casting speed Vc" and the
"immersion of the nozzle Di" distance. A mathematical model was established between these parameters (inputs parameters) and the exit temperature "Ts", experimental plan-based Box Behenken with 3-factor and 3 levels. The phenomenon of continuous casting was simulated with the Fluent code 6.0. The outlet temperature was recorded at the mold wall and used as an output parameter in the used algorithm.
Matlab software was used to optimize the inputs parameters to obtain the minimum effective temperature "TSmin".
Proceedings of the 2015 International Conference on Industrial Engineering and Operations Management Dubai, UAE, March 3 – 5, 2015
Modeling of temperature field at outlet of continuous casting mold
L.LAOUAR
Badji Mokhtar University BP12 -2300, Industrial Mechanics Laboratory
Annaba, Algeria lakla_55@yahoo.fr
Fig1: Formation of the solid skin in the primary cooling solidified skin
mold entry nozzle
liquid steel lubricant
II.SIMULATION BY FLUENT CODE A- Material
The used material is a stainless steel 434, its main physical properties are indicated on the Table 1 Table 1. Physical properties of steel [4]
B.Mold geometry
It has been considered for this simulation three molds of the same dimension, with a variation of the immersion nozzle distance [7].
Dimension of molds :(0.700x0.300) m2 Distance of immersion:
a) submerged nozzle 0.070m b) submerged nozzle 0.160m c) submerged nozzle :0.250m
C.Mathematical model
The selected mathematical model is the turbulent model realizable k-ε [1] defined by the fundamental equation of heat conduction. The temperature distribution during the casting process is described by the equation of heat conduction in its general form expressed by the expression (1) [8] is given by the folowing equqtion.
Where:
k (T): the thermal conductivity (W / m.K) Cp (T): the specific heat (J / kg.K) ρ (T): the density (kg / m3) T: temperature (K)
• q: the source term (W / m3) t: time (s)
x, y, and z rectangular coordinates (m)
The latent heat (temperature between the liquidus and solidus) • q is expressed by [8].
Where:
L: latent heat of fusion (J / kg) fs: faction solidify
ρ (T) s: the (m3 kg /) density of the solid Boundary conditions
- At the mold wall is imposed the no-slip condition for the tangential velocity components and flow:
with
Фs: heat flux (W / m2)
- The flow conditions and stresses are zero for the free surfaces
The turbulent kinetic energy and initial dissipation rate of input are estimated using equations 4 [2].
Thermal conditions: inlet temperature is imposed, Tinlet k = 1832 or 1559 ° C.
The heat flux removed by the faces of the mold, decreases linearly when moving the meniscus level at the bottom of the mold, it is expressed by the relationship 5 [9]:
Where
- H is the position of the steel wafer according to the considered level of the meniscus.
NB: -The Unit of speed Vc used by the code Fluent is expressed in m / s
-Ts: temperature of the skin at the exit of the mold III MODELING PLANS BOX BEHENKEN -The Tests were conducted with the planning model multifactorial experiments "Box Behnken plans" for 15 trials with 3 factors (X1, X2, X3) at 3 levels [10]. The input parameters are: "distance of immersion Di", "casting speed Vc" and "casting temperature Tc".
Property Value
Thermal conductivity W/m.K 26
Density (ρ)( Kg /m3) 7020
Specific heat (J/Kg.K ) 680
Liquidus temperature ( K) 1775
Dynamic viscosity (m2/s) 0.0056
Casting temperature (K) 1832
(1)
(2)
(3)
(4)
(1,0-0,7.h)106 (5)
Figure2: Geometry moulds
Simulation tests are performed according to the matrix (Table 1).
Table 1 Testing Matrix
Simulation by the Fluent code was conducted according to the combination indicated on the matrix of experiments (Table 2). The values of the outlet temperature Ts were recorded on the skin solidified level of the wall - mold and noted on the table 2.
Table 2 Matrix of experiences
A.Simulation and prediction
The prediction model allowing the highlighting of a relationship between the parameters of primary cooling (Di, Vc, Tc) and the temperature Ts [11]
given by follows equation:
Tsprédite=1740+(9.75)*x1+(61)*x2+(18.75)*x3+
(0.5)*x1x2+(0)*x1*x3+(2.5)*x2*x3+(16.5)*(x1)² +(-41)*(x2)²+(-11.5)*(x3)²+ e
e: Predicted Error
Tspredicted: predicted temperature at the solidified skin at the output of the mold.
The optimum parameters obtained from the model are:
Dioptimal = 134.77mm, Vcoptimal = 0.1m / s, Tcoptimal = 1795 k.
These parameters predict a temperature equal to output:
Tspredicted = 1608.9 ± 15. 26 k with R2 = 0.99 and the Fischer score is Fcal = 4.3
Code parameters
levels
-1 0 +1
X1 Di (m) 0.07 0.160 0.250
X2 Vc (m/s) 0.1 0.237 0.375
X3 Tc (°K) 1795 1815 1835
No test
input parameters Output Parameter Di(m) Vc (m/s) Tc (°K) Ts (°K)
1 0.070 0.1 1815 1662
2 0.250 0.1 1815 1670
3 0.070 0.375 1815 1760
4 0.250 0.375 1815 1770
5 0.070 0.237 1795 1710
6 0.070 0.237 1835 1750
7 0.250 0.237 1795 1740
8 0.250 0.237 1835 1780
9 0.160 0.1 1795 1600
10 0.160 0.375 1795 1740
11 0.160 0.1 1835 1630
12 0.160 0.375 1835 1780
13 0.160 0.237 1815 1750
14 0.160 0.237 1815 1740
15 0.160 0.237 1815 1730
Figure3. Variation of the outlet temperature, Ts=F(Di, Vc, Tc) ; a)Di=0.07m, b)Di=0.16m, c) Di=0.25m
Di=0.070m b) Di=0.160m c) Di=0.250m
a)
B. Results and interpretations
Figure 3(a,b,c) illustrates the variation of the outlet temperature Ts based on the parameters of the continuous casting process (Di, Vc, Tc).
a-According to the figure3a, when the nozzle is submerged at a distance Di = 0.07m and the
"casting temperature Tc", the increase in "casting speed Vc" causes an elevation of the"outlet temperature Ts" about 1740 °k. A reduction by the
"casting speed Vc" associated with Rating decrease
"casting temperature Tc" causes a decrease in the
"Ts" temperature about at 1600 ° k, this assumes a thick to be suficient to overcome the ferro-static pressure.
b- By immersing the nozzle at a distance
Di = 0.160m, (figure3b), the effect of the "casting temperature Tc" is not significant. Nevertless increasing the "casting speed Vc" significantly affects the" outlet temperature Ts". Where Ts>
1760 k, the skin is affected: When the "casting speed Vc" tends to its minimum value associated with a "casting temperature Tc": The outlet temperature of the mold "Ts" decreases to a temperature lower than 1620 ° k.
c-In the case where the input nozzle is introduced to a distance Di = 0.250m (figure3c), increasing the
"casting temperature Tc" and the "casting speed Vc", are not recomended to the cooling process. A considerable increase in outlet temperature Ts exceeding 1800 ° k and a decrease in the "casting speed Vc" helps to reduce the "outlet temperature Ts". Aalso when the temperature of "casting Tc
"reaches its minimum value, the "output temperature Ts "decreases until 1640 ° k.
Continuous casting of steel is governed by many interacted parameters. By studying the effect of the considered parameters (Di, Vc, Tc) on the outlet temperature Ts for the considered material, it was found that for immersed input nozzles at the distances, Di =0.070mm, and Di = 0.160mm, the high values of the casting speed Vc generate an increase of Ts whatever the casting temperature Tc.This is explained by the fact that increasing of Vc induces a reduction of the cooling timei.e. the increase of the temperature Ts.
Low speeds are recomended for the cooling process, which favors the decrease of temperature Ts by allowing the formation of a thick skin.
Furthermore a submerged nozzle for a considerable distance Di = 0.250m, the high values of Tc and Vc are harmful for the process and even for them low casting speeds. Temperature Ts is always greater than that predicted, because the distance traveled by the steel casting is not suficient to allow the formation of a thick crust for retaining liquid steel.
CONCLUSION
In this contribution, the simulation and modeling of temperature fields are used to study the influence of process parameters (Di, Vc, Tc) on the variation of the temperature Ts of the solidified skin of the slab at the outlet of mold. Moreover, the mathematical model obtained by the experimental plans, offers the advantage of predicting the more interesting parameters to attain the minimum value of the outlet temperature Ts.The results of the simulation for the considered steel grade shown that the reduction of temperature and speed (minimum good speed) during casting of immersing input nozzle at an average distance, tend to minimize the temperature Ts. This involve the formation of a skin solidified.The predicted parameters, allowing optimum temperature Ts = 1608.95 ± 15.26 ° k are:
Di = 0.134m .; Vc = 0.1 m / s; Tc = 1795 k.
Simulation and modeling design of experiments can be used as a tool for steelmakers to allow optimal operating conditions such as the temperature Ts at the outlet of the mold, the crust solidified.
Practical experiments will be carried out in the future to confirm the obtained simulation results, modeling the thickness of the solidified skin is also envisaged.
REFERENCES
[1] Marc HENRI November 2009 "3D finite element modeling of the primary cooling during the continuous casting of steel", the Higher National School of Mines of Paris.
[2] Ahmed.Bellaouar, Omar.Kholai, Fatima. Daoud "Modeling of Thermal Transfer in the Secondary of a continuous casting machine Radial Zone." 13th International Conference on Thermal Author manuscript, published in Jith 2007, Albi France (2007).
[3] M.Bourebia, M.Chaour,, S.Boulkroune, S.Bouhouche, L.Laouar "Influence of the immersion of the nozzle distance in continuous casting of steel during the primary cooling" VIIIth Study Days Techniques - JET'2014 Marrekech / Maroc (2014).
[4] Akni.Ahcéne, Bellaouar.Ahmed, Lachi.Mohammed
"Numerical Modeling of 2D Temperature Field in the Area of Primary Cooling of Continuous Casting Machine" Days of National Studies Mechanics, JENM'2011 Ouargla, Algeria 07-08 March, 2011.
[5] Mounira Bourebia, L.Laouar, M.Chaour, H.Maouche
"0ptimisation the healing rate in the mold during continuous casting of steel" International 2ièmeConférence the Maintenance and Industrial Skikda / Security Algeria (2013) .
[6] Fréderic.Costes "Modeling Three-Dimensional Finite Element Thermomechanics of Continuous Casting of Steel" Ecole des Mines de Paris -ENSMP PhD 2004 [7] Technical Engineering
[8] N. Cheung, CA Santos, JA Spim b, A. Garcia "Application of a heuristic search for the technical improvement of cooling spray zones in terms Continuously cast steel billets" .2005
[9] A.Bellaouar, O.Kholai, P.Valentin, "Numerical modeling of a cooling process continuous stainless steel slab Coulen"
2005
[10] Jaques Goupy, Design of Experiments for Response Surface Collection Dunod 1999
[11] Laouar, H.Hamadache, S. Saad, A. Bouchelaghem S.
Mekhilef, “Mechanical Surface treatment of steel- Optimization parameters of regime, Physics Procedia 2 (2009) 1213-1.
BIOGRAPHY
. Mounira Bourebia is researcher in the Welding and NDT Research, Centre (CSC) it work on the project for the simulation of phenomena of breakthrough in continuous casting of steel. PhD student in the laboratory of an affiliated industrial mechanics, university Badji Mokhtar Annaba, Algeria, she obtained her magistere integrated manufacturing mechanics in 2010 Badji Mokhtar al university Annaba. She a member of the research team on the surface states. She works under the directives of Professor Lakhder .Laouar, their research focuses on the influence of surface roughness condition on a mechanical contact, the processes of mechanical surface treatment, the numerical simulation, optimization and manufacturing.
Dr.Bouhouche Salah is a senior researcher (Directeur de Recherche) at URASM-CSC Annaba, a department of the main organisation of welding and NDT Research centre -CSC. He obtained a Dipl.-Ing from the National Institute of Hydrocarbon and Chemistry INHC Boumerdes in the year 1985 and Magister thesis from the National Polytechnic School Algiers in 1995. He received a PhD (CUM LAUDE) from Mining and technical university Freiberg (TU BAF) Germany and a Doctor of Sciences (DSC) from National Polytechnic School Algiers in the year 202 and 2005 respectively.
Maouche Hichem: is a searcher attached in the Welding and NDT Research, Centre (CSC) and member in the laboratory of foundry metallurgy institute University of Annaba, Algeria. I earned Engineer in metallurgy Option: foundry at theBADJI MOKHTARAnnaba University, Algeria, Magister in metallurgy Option: Process modeling and thermal in foundry in the Department of metallurgy at the BADJI MOKHTAR Annaba University, Algeria. And preparation Doctor Metallurgy Option:
foundry from University of Annaba. I have participated conference papers. Ihave done research projects with CSC Research Centre. My research interests include in metallurgy, foundry, alloy development, heat treatment, development of foundry equipment, modeling,currently I'm working on the influence of the chemical element on the characteristic of service for cast Iron and manganese steel.
Lakhar.Laouar is Professor, Director of Research in the Department of Mechanical Engineering, Laboratory of Industrial Mechanical University Badji Mokhtar - Annaba , Areas of interest: Mechanical surfaces, Mechanical & Industrial Engineering, Industrial Maintenance and Diagnostics, Mechatronics.
Chaour Mohamed is a searcher attached in the Welding and NDT Research, Centre (CSC) and Temporary teaching in the department of science technology at the Constantine 1 University, Constantine, Algeria. He earned Engineer in Mechanical Engineering Option: Energy from Mentouri Constantine University, Algeria, Magister in Mechanical Engineering Option: Energy applied engineering in the Department of Mechanical Engineering at the Constantine 1 University, Constantine, Algeria. And preparation Doctor in Energy from University of Constantine 1. He has participated conference papers. Mohamed has done research projects with CSC Research Centre. His research interests include energy, simulation, combustion, optimization, Heat transfer, Fluid Mechanics, and Thermodynamics. He is a Mastered technical of Programming in FORTRAN and Simulation software in ANSYS and FLUENT.
Sofiane Boulkroune currently researcher in the research center and welding control. I have a Magister degree in mechanical engineering, energy option (2010), inscribed in 4th year PhD in science, University of Constantine 1. Previously temporary lecturer at the University of Constantine 1.
