Simulation of high temperature raw materials processing using CFD - education and research Benelux PHOENICS User Meeting – Eindhoven, May 24th 2005 Yongxiang.

Simulation of high temperature raw materials processing using CFD- education and research

Benelux PHOENICS User Meeting – Eindhoven, May 24th 2005

Yongxiang YangResource EngineeringFaculty of Civil Engineering and Geosciences

Overview

• Introduction

• Examples of using CFD for raw materials processing• Simulation of hazardous waste combustion• Simulation of transient heating of dredging impellers

• Summary

Introduction

• History of using PHOENICS• Simulation of off-gas cooling in waste-heat boilers for

copper flash smelting process. (1992 – 1997 HUT)

• Simulation of gas flow and slab heating of steel reheat furnace. (1997 – 2000, HUT and TU Delft)

• Flow and combustion modeling of hazardous waste incineration in rotary kilns. (1999 – 2004, TU Delft)

• Predicting the fuel combustion and transient heating process of metal components. (2003 – 2004, TU Delft)

Introduction: history of using PHOENICS

Off-gas cooling & dust flow in WHBs

Sym m etric centerlineG as coo ling from 1370°C to586°C

2D view o f partic le tracks

50 m partic les , 20 injection po in ts

Y

Z

TEM1

181

281

380

480

580

679

779

879

978

1078

1178

1278

1377

1477

1577

Temperature distribution between the centerline and the side-wall (°C)

1477 1577

157712781377

1377

147712771178

127811781078

1078 1278181

Z

Y

Transient heat of steel slabs in re-heat furnace

0

200

400

600

800

1,000

1,200

1,400

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28

Furnace position in z- direction, m

Center temperature

Top surface temperature

Gas temperature near slab surface

Gas temperature 0.6 m above slab surface

Tem

per

atu

re (

°C)

Introduction

• Use of CFD in Raw Materials Processing at TU Delft• PHOENICS:

• Education (1998 -): “transport phenomena in raw materials processing” , Case studies in Computer Practicals.

• Steel flow in tundish, • Off-gas cooling in boilers, • Heat loss from furnaces and reactors, • Transient cooling of steel slabs

• MSc. thesis research: 5 students• Project research: modelling of hazardous waste incineration

• PhD research projects• Hot metal flow in an ironmaking blast furnace hearth (CFX)• Aluminium scrap melting in a rotary furnace (CFX)• Submerged arc furnace for phosphorus production (Fluent)

Example 1: Simulation of hazardous waste combustion in a rotary kiln• Key Issues and

challenges• Fuel stream modelling• Waste combustion modelling

(CFD)• Construction of cylindrical kiln

and rectangular secondary combustion chamber (SCC)

R otary k iln

S econdarycom bustioncham ber(S C C )

A shsum p

Waste combustion in the rotary kiln• Rotary kiln + secondary combustion chamber

(SCC)

• Waste supply • ~30 MW, 7.2 t/h• Main burner ~11.5 MW• Sludge burner ~4 MW• Solid burner ~11.5 • SCC burner ~3 MW• Fuel lances: non-regular

• Air supply• Load chute• Cooling ring• Air lances (SCC)

Rotary Kiln

Secondary Com bustionCham ber(S CC)

To RadiationChamberof W HB

A sh Sum p

Burner flame

Solid-waste flam e

Load chute

Main burner

Sludgeburner

SCC burner

Air lances(x 7)

A ir lances(x 7)

Off-gas

Slag

H O/air2

Kiln front-viewShreddedsolid-wasteburner

Approach Waste characterization

Virtual fuel definition

Inlet stream definition

(wastes/water, air)

CFD turbulent combustion

model (7-gas)

Temperature

validation

Species

validation

Post-processing

Statistical analysis

Parametric studies

Advice on operation

CFD combustion model

• Assumptions and physical models• Gas phase only (negligence of gasification and vapourisation)• Turbulence flow: k-model• Walls: adiabatic (small wall heat loss)• Radiation was neglected (adiabatic walls, no effect)• Kiln rotation: neglected

• Waste-combustion: • Global combustion model (SCRS, 3-gas model)

• Extended global combustion model (ESCRS, 7-gas model)

• Real gas/fuel compounds have to be used

7-Gas model: CmHn, H2, O2, CO, CO2, H2O, N2

1 Fuel Oxidant (1 ) Product Heatkg s kg s kg

Virtual fuel definition

Modelled virtual fuel

C3H4 87.37% 46.3 MJ/kg

CO 12.38% 10.1MJ/kg

H2 0.25% 120 MJ/kg

% C xH y % C O % H 2

A ssu m e d F u e l: C H1 .7

H ea t o f co m b u s tio n :42,000 kJ/kg

O ff g a s ca l cu la tio n s

C 1 02 H 1 31 O 11 0

H e a t o f c o m b u s tio nd e p e n d s o n c o m p .

H e a t o f c o m b u s tio n :1 0 ,1 0 0 k J /k g

H e a t o f c o m b u s tio n :1 2 0 ,0 0 0 k J /k g

% C O 2 , O 2

G ene ra lize d fu e l:C xH yO z

B ette r defined Chem ically

Therm ally

Original waste

Less know n, p oorly de fined C h e m ic a lly

T h e rm a lly

From the incineration plant

Combustion model

Wat

er

Re-defined waste

Propyne Allene

Average off-gas composition: 7.5%O2, 69.5%N2, 9%H2O, 14%CO2

7-gas combustion model

• 1 primary reaction• 2 secondary reactions

• Turbulent combustion rate:• Eddy break-up model

• Dependent on turbulent mixing

• Fuel concentration + turbulence

• Regulated by CEBU constant (0.1 – 4.0 tested, 1.0 used)

3 4 2 21.5 3 2C H O CO H

2 22 2CO O CO

2 2 22 2H O H O

min , /mfu fu oxS CEBU m m sk

Stream definition

• Input streams• Waste streams: multi-fuel

streams

• Water content: mixed with wastes

• Combustion air: multi-feeding ports

• 2 stream constraint (code)• Fuel rich stream• Fuel lean stream

• Individual stream

Ash

sum

p

SCC

Air lances(7x2)

SCC burner

G asoutlet

Load chute

Sludge burner

Main burner

S-burner

FUEL RICH STREAM

Compound C3H4 O2 CO H2 CO2 H2O N2 Total

Percentage 21,64% 15,64% 3,09% 0,00% 0,75% 4,41% 54,47% 100,0%

FUEL LEAN STREAM

Compound C3H4 O2 CO H2 CO2 H2O N2 Total

Percentage 0,28% 20,97% 0,06% 0,00% 0,00% 5,64% 73,06% 100,0%

x% Fuel-rich + y% fuel-lean

Computational grid

• Cartesian grid• 230,000 – 360,000 cells• Using solid blockages to

form the kiln and SCC,• Difficult to set B.C. for

non-rectangular shapes.

• Burner definition• 2 stream approximation

Fuel-rich Fuel-lean

Example: main burner

Load chute

Cooling ring

Main burner

Sludge burner

Solid burner

Typical CFD predictions

Across the SCC burner

Across the m ain burner

Typical CFD predictions

SCC burner

So lid w asteburner

S ludgeburnerM ain

burner

SC C back w allS CC side w all

A cross SCC

burner plane

1098

1247

1396

1396

1247

1098

1247

1396

1247

1247

1098

1396

58

504

653

1998

1247

2139

2139

2139

2139

2139

1544

1544

1396

1693

1098

2139

1544

58

1098

1098

A shsum p

1098

Across the load-chute

Between the solid-waste burnerand the sludge burner

Across the main burner

Across the SCC-burner Near the outlet to WHB

Centra colder zone due to air injection lances

1247

1098

1247139 6

109 8

1247

1 396

154 41693

1247

1098

801

124715441 693

184 11 990

213 9

801653504

1841

1544

801653

1396

1247

1098

1247

1098

1396

154 41 693

1841

1693

1 544

2 139

653

801

18411 6931 990

2139

8 011247653

10981 544

1 098

950801

1247

12471396

1 3961 5441693

19002 139

21391 693

1693

5043552075 81 098

1 841

15441 247

801653

1 0986 53

124 71 396

1544

1 3961 544

109 8

130 1

128 3

1301

1319

13541372

1390

13371283

128 3

1 265

131 9

11401176

1194

1 229

12471265 126 5 1301 13191337

1372139 0

131 9

122 9

1354130 1

9 50

950

Temperature distribution

Typical CFD predictions

CO0.000 - 0 .054

0.0040.0080.01 2

0.0040.054 0.027

0.0160.00 4

0.016

~0.00

~0.00

0.016

Main burner axis (X-), m

Mas

s fra

ctio

n

CO2

O2

C3H4

H2O

CO

H2

Outside the SCC

Species distribution

Typical CFD predictions

C3H 40 .0 0 - 0 .16 8

C O0 .000 - 0 .0 54

CO 20 .0 0 0 .2 1

H2O0 .0 00 - 0 .1 05

Species distribution

N ear the top Z85: 0 - 1 .8x10 -6

B elow the topZ80: 0 - 3 .5x10

-6Above the a ir lances

Z 60: 0 - 1 .3x10-4

N ear the a ir lancesZ52: 0 - 4 .3x10 -4

A cross the S C C -bu rne rZ 40: 0 - 5 .2x10 -2

B elow the S C C burnerZ36: 0 - 3 .3x10 -4

N ear the O utle t X 570 - 2 .7x10

-6

CO overall range: 0 .00 - 0.071

Z85

Z80

Z60

Z52

3 6

Z40

Z20

Z10

Z1

X1

X15

X30

X57

SC C b u rne r

H igh

Low

H igh

Low

Low

H igh

H igh

H igh

H igh

H igh

H igh

H igh

Low

Low

H igh

H igh

H igh

H igh

H igh

H igh

H igh

H igh

H ighH igh

H igh

H igh

H igh

Low

Low

Low

Low

In term ed iate

H igh

H igh

Low

Low

Low

H igh

H igh

H ighH igh

H igh

Low

Low

Low

Model validation

• Temperature measurements

Ash

sum

p

S C C

A ir lan ce s(7 x2 )

G a so utle t

L o ad ch u te

S lu d g e b u rn er

M a in b u r n e r

S o lid b u r n e r

Ash

sum

p

S C C

A ir lan ce s(7 x2 )

G a so utle t

L o ad ch u te

S lu d g e b u rn er

M a in b u r n e r

S o lid b u r n e r5-m longWater-cooledThermocouple4-view ports

0

200

400

600

800

1000

1200

1400

-300 -200 -100 0 100 200 300

Distance from SCC centreplane (mm)

Te

mp

era

ture

(°C

)

Measured3-gas model (no heat loss)

7-gas model (10% heat loss)7-gas model (no heat loss)

Waste energy input: 37 MWWaste mass input: 6.67 t/hAir supply assumed: 43.5 t/h atstoichiometric air/waste ratio 6.52.Excess air in the model: 75%(including leakage air)

SCC-burner side

Temperature profiles: statistical• Temperature averaging

• Area averaging (directly available in VR-Viewer)• Mass flow averaging: highly required for

heterogeneous flow

Average temperature weighted by mass, Kiln

100

300

500

700

900

1100

1300

1500

1700

1900

2100

2300

0 1 2 3 4 5 6 7 8 9 10 11

Flow direction, X (m)

T

( , , )T F x y zTmax

Tarea

Tmass

Tmin

Average temperature weighted by mass, SCC

0

200

400

600

800

1000

1200

1400

1600

1800

2000

0 2 4 6 8 10 12 14 16 18

Flow direction, z (m)

T

T_massT_areaTmaxTmin

Example 2: Simulation of transient heating of dredging impellers in a mobile furnace

• Fuel combustion• Oil (red diesel) simplified as gas,

3-gas model (CEBU=25)

• Radiation• IMMERSOL model (Ka=0.1, Kb=0.001)

• Metal surface emissivity: 0.05-1.0

• Wall emissivity: 0.8

• Conjugate heat transfer• CAD geometry for the impeller

Simulation of a heat treatment furnace

• Construction of furnace geometry and grid• Cylindrical furnace, curved

roof (spherical cap)• BFC or Cartesian?• Handling of furnace walls

and boundary conditions• Handling of furnace roof

• Using high conducting solids combined with normal refractories to form the cylindrical and curved walls

S and

Fan

Chimneyes

B urners (x4)

Simulation of a heat treatment furnace• Comparison of predicted and pre-set surface temperature profile on the dredging impeller• 3 steps of fuel supply• Varying excess air rate

Period(hour)

ExcessAir (%)

MixtureFraction

Fuel(kg)

Energy per burner (kW)

0-8 226% 0.02 214 80

8-10.5 160% 0.025 84 100

10.5-20.5 85% 0.035 356 85

Total - - 654 -

Temperature profile final model

0

100

200

300

400

500

600

700

800

900

1000

1100

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Time (hour)

Tem

per

atu

re (

ºC)

980 oC

CFD modelling of a heat treatment furnace

• Transient heating process of a dredging impeller in a mobile furnace

End of the transient m odel s te a d y -s ta te m od e l

12600 seconds7200 seconds1800 seconds

240 seconds30 seconds

73800 seconds45000 seconds36000 seconds

18000 seconds

4 seconds 20 seconds

Temperature range: 25 -1175°C

SandSand

Simulation of a heat treatment furnace• Surface temperature evolution of the impeller (0-

20 hrs)

Simulation of a heat treatment furnace

• Energy balance• Heating: 50 – 10%• Wall loss: 0 – 15%• Off-gas: 50 – 75%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0 2 4 6 8 10 12 14 16 18 20 22

Time (hrs)

En

erg

y d

istr

ibu

tio

n

solids

out offgas

Wall heat loss

Energy efficiency: LOW!

Simulation of a heat treatment furnace• Alternative heating source:

Using radiation plate• Substantially reducing heat loss

by off-gas• Easy to regulate and control• Energy saving:

• Oil: 43 MJ/kgx654 kg– 28 GJ• Radiation plates:

2233 kWh – 8 GJ (70% saving)

€ 178 vs € 520 (65% saving)

R a d ia tio n p la te

Im p e lle r

Time period Hours Energy setting Total energy

0-2 2 200 kW constant 400 kWh

2-8 6 75 – 105 kW linear 540 kWh

8-15 7 105 – 150 kW linear 893 kWh

15-20 5 80 kW constant 400 kWh

Total (0-20) 20 2233 kWh0

100

200

300

400

500

600

700

800

900

1000

1100

0 2 4 6 8 10 12 14 16 18 20 22

Time (hour)

Te

mp

era

ture

(ºC

)

Temperature impellermeasuredTemperature radiation 4inputs

Summary

• CFD is a useful and handy tool in education and research at universities, as well as in industrial design and process R&D.

• Various aspects of high temperature materials processing operations could be reasonably simulated.

• More insight understanding of the transport phenomena and process performance can be obtained.

• Still a lot of challenges: complex geometry, reactive and multi-phase flows in industrial applications.

• We need more functionality and power of the Code!