Modeling SEN Preheatingccc.illinois.edu/s/2014_Presentations/16_Li-Y... · 2014. 8. 18. · Modeling SEN Preheating ... a chemical equilibrium program which can predict adiabatic

CCC Annual ReportUIUC, August 20, 2014

Yonghui Li

Department of Mechanical Science & EngineeringUniversity of Illinois at Urbana-Champaign

Modeling SEN Preheating

University of Illinois at Urbana-Champaign • Metals Processing Simulation Lab • Yonghui Li • 2

Objectives

• Develop an accurate preheating model to optimize preheating process:– Fuel composition;– Preheating time;– Torch configuration;– Insulation;– Refractory conductivity.

• Obtain air entrainment, flow and temperature distributions from combustion model.

• Evaluate Flame Temperature Model (in spread-sheet).


Burner tip

TC 1 SEN

TC 5TC 3

TC 4 TC 6

Gas temperature

Wall temperature

Premixed natural gas(mainly CH4)/ O2

Two-port SEN

Preheating experiment setup[1]

Stand-off distance

97

197

341

Unit: mm

324oC

249oC

174oC

23oC

91oC

Infra-red photo of SEN outside wall


Measurements[2] (for model validation)2. Wall temperature (transient are not listed here. )

1. Gas temperature 197mm below SEN top

3. The shape of flame

4. SEN outside wall temperature

324oC

249oC

174oC

23oC

91oC

Thermocouple

TC3 TC4 TC5 TC6

X* (mm) 394 538 394 538

Y* (mm) 48 48 69 69

Temp. (oC) 584 554 453 397

X: Distance from top air inlet;Y: Distance from SEN centerline.


Thermal – Flow Model

• FLUENT simulation is 2D axisymmetric. – The two-port SEN is simplified as a ring shaped port with the

same exit area.• The burner tip is assumed as annular shape with 3× bigger area.

– To avoid supersonic and mesh refinement at burner tip, accounting for gas expansion.

24*1.6mm diameter

1*0.8mmdiameter

Simplified as

Mixture inlet

17.4mm diameter of outer ring of holes

Rosebud tip surface


Model geometry and mesh97mm Validation Case 147mm Case

1-mm coating layer with 4 cells through thickness

Stand-off distance

88843 quadrilateral cells total

14797

Insulation Case


Material Properties

• Flow • SEN [1,4,5]

0

5

10

0 200 400 600 800 1000 1200

Ther

mal

Con

duct

ivity

(W

/m-K

)

Temperature (°C)

50010001500200025003000

0 200 400 600 800 1000 1200Spe

cific

Hea

t (J/

kg-K

)

Temperature (°C)

16% Porosity Doloma-Graphite

Glaze

Glaze

16% Porosity DG

Gas average [3]thermal conductivity2.7006 W/mK viscosity 9.32×10-5 kg/m

Species thermodynamic properties: thermo.db


Key features--combustion

• Non-premixed species model. • Fuel inlet: perfectly mixed CH4 and O2 in

1:2 mole ratio in a total mass flow rate of 3.8 g/s[2].

• Ambient air entrainment.• Non-adiabatic energy treatment.• GRI-Mech 3.0[6] natural gas combustion

mechanism, contains 325 reactions and 53 species.

8


Model validation

9


Temperature across SEN: 97mm Validation Case


Temperature across SEN: High-k Refractory Case

Outer glaze

SEN outer wall

SEN refractory wall

Nozzle inner bore with gasInner glaze

SEN inner wall

Measured TC3

TC 3-5line at 5min

TC5TC6TC4

TC 3-5line at 10min

TC 3-5& 4-6 line at SS


Temperature across SEN: Insulation Case


Transient temperature validation

TC 3 error= 25oCTC 4 error= 45oC

TC 5 error= 32oC

TC 6 error= 83oC

Error causes:

• Excessive thermal conductivity/diffusivity due to uncertain refractory properties

• Neglect of Zirconia sleeve at lower part of SEN

• Neglect of contact resistance at TC tip0 10 20 30 40 50 60 70 80 90 100 110

0

100

200

300

400

500

600

700

S

EN

wal

l tem

pera

ture

(°C

)

Time (min)

TC3 MEASURE TC4 MEASURE TC5 MEASURE TC6 MEASURE TC3 FLUENT TC4 FLUENT TC5 FLUENT TC6 FLUENT


Flame shape & Outer wall temperature contour validation

Flame shape comparisons of predicted temperature contours and close-up

photographSEN outer wall temperature comparison


Model Parametric study

15


Inputs of 4 cases

Model Inputs97mm

Validation Case

147mm Case

Insulated Case

High-kCase

Thermal conductivity DG, Glaze DG, Glaze

DG, Glaze, Insulation DG, Glaze

Specific heat DG, Glaze DG, Glaze DG, Glaze, Insulation DG, Glaze

Stand-off distance 97mm 147mm 97mm 97mm

Insulation layer No No Yes No


Flow distribution

Velocity (m/s)

(a) Direction arrows in the whole domain (b) Velocity vector inside SEN

(c) Zoom-in vector near SEN top

60 m/s

5 m/s

147mm Case

97mm Case 147mm Case

Air entrainment154% 135%

Stand-off distance

Flame spreads

Air entrainment

Flame temperature97mm Validation Case


Temperature distribution

(d)

(a) 97mm Validation Case(b) 147mm Case

(c) Insulation Case(d) High-k Refractory Case

(c)(b)(a)

Temperature (oC)


Transient temperature comparisons among 4 cases

0 10 20 30 40 50 60 70 80 90 100 1100

200

400

600

800

1000

1200

SE

N w

all t

empe

ratu

re (°

C)

Time (min)

Measure TC3 Measure TC5 97mm Case TC3 97mm Case TC5 147mm Case TC3 147mm Case TC5 High-k Refractory Case TC3 High-k Refractory Case TC5 Insulation Case TC3Insulation Case TC5


Flame Temperature Modelin Excel VBA

20


Flame Temperature Model(VBA)

InputsFuel typeOxygen sourceOxygen source fractionAir entrainmentReactants temperatureReactants pressure

OutputsFlame/Products temperatureProducts pressureSpecies componentProducts propertiesForce convection coefficient Free convection coefficient

Gaseq[7]

• Gaseq[7]: a chemical equilibrium program which can predict adiabatic temperature and composition at constant pressure.

• Oxygen Source Fraction = mole of oxygen inputmole of oxygen required for stoichiometric reaction

• Air Entrainment = mole of entrained airmole of air needed for stoichimetric reaction


Flame Temperature Model Results

Simple spread-sheet model can predict flame temperature approximately without sophisticated chemical reactions and thermal hydraulic models.

Air entrainment

Oxygen sourcefraction

Reactants Temp.

Flame Temp.

Flametemp.

97mm Validation Case 154% 100% 19 oC 1328 oC 1343oC

147mm Case 135% 100% 19 oC 1451 oC 1587oC

MeasurementCombustion

Model ResultComb.Model

Reactants, products pressure is 1atm.


Conclusions• A 2D axisymmetric model of nozzle preheating is developed, including

325 chemical reactions with 53 species of methane combustion.• Steady-state fluid flow, heat transfer and gas combustion, and transient

heat conduction in the SEN walls are simulated. • The model predictions were validated with a preheating experiment,

including the gas temperature across the flame, SEN wall temperature histories, flame shape, and SEN outer wall temperature distribution.

• Moving the burner further away from the SEN top leads to higher SEN temperature, due to flame expansion causing less air entrainment.

• Adding an insulation layer causes higher SEN wall temperatures and milder temperature gradients.

• Increasing refractory conductivity causes milder temperature gradient at SEN.

• To optimize preheating, a proper stand-off distance, stoichiometric fuel composition, proper refractory thermal properties , and insulation layers are recommended.

• A simple spread-sheet model of the adiabatic flame temperature predicts gas temperature approximately, based on knowing the air entrainment.


Acknowledgements

• The authors are grateful to R. Nunningtonand other personnel at Magnesita Refractories for providing the measurement data.

• The authors appreciate the support from the Continuous Casting Consortium at the University of Illinois.


References[1] R. Nunnington, Magnesita Refractories, private communication, Jun, Aug 2012 [2] Magnesita Refractories, PB10 SEN Temperature Data for CCC Heat Flow Model, Report, York, March 31st, 2010[3] FLUENT 13.0[3] Charles E. Heat Transfer in Industrial combustion, p469[4] T. Shimizu, Thermal conductivity of high porosity alumina refractory bricks made by a slurry gelation and foaming method, Journal of the European Ceramic Society, 2013[5] Hayashi K, Fujino Y, Nishikawa T. Thermal conductivity of Aluminiumand Zirconia fiber insulators at high temperature. Yogyo Kyokai Shi 1983;91:450–6.[6] Gregory P. Smith, David M. Golden, Michael Frenklach, Nigel W. Moriarty, Boris Eiteneer, Mikhail Goldenberg, C. Thomas Bowman, Ronald K. Hanson, Soonho Song, William C. Gardiner, Jr., Vitali V. Lissianski, and Zhiwei Qin http://www.me.berkeley.edu/gri_mech/[7] Gaseq, Chemical equilibrium program, available at http://www.gaseq.co.uk/[8] Y. Li, MS Thesis, Transient Model of Preheating a Submerged Entry Nozzle, 2014

Documents

Modeling SEN Preheatingccc.illinois.edu/s/2014_Presentations/16_Li-Y... · 2014. 8. 18. · Modeling SEN Preheating ... a chemical equilibrium program which can predict adiabatic