DEER 2004

A New CFD Model for Understanding and Managing Diesel Particulate Filter

Regeneration

Z. Jason Hou and Ted Angelo

Donaldson Company, Inc.

DEER 2004

DPF Technology Development

• Understand Regeneration– When?

– How?

• Characterize Soot Loading and Pressure Drop vs. Time

• System Control– Model-Based Feed-Forward Adaptive Control

• Component Development

• Flow Distribution and Thermal Management

DEER 2004

DPF Technology Development

• Understand Regeneration– When?

– How?

• Characterize Soot Loading and Pressure Drop vs. Time

• System Control– Model-Based Feed-Forward Adaptive Control

• Component Development

• Flow Distribution and Thermal Management

DEER 2004

Why Model Regeneration?• Goal: Achieve Quick, Complete and Safe

Regeneration with Minimal Fuel Penalty

• Regeneration is Very Complex

• Experiments are Difficult and Costly

• Modeling Allows Us to Optimize System and Regeneration Strategy

DEER 2004

Regeneration Is Complex• 3-D, Transient Flow

• Three Modes of Heat Transfer

• Porous Layers: Substrate and Soot

• Soot Combustion / Catalyst Effect

• Variable Properties: Porosity, Permeability, etc.

• Sensitive to Operating Parameters:

– Exhaust flow rate, gas temperature, soot load, oxygen, filter design, etc.

DEER 2004

Filter Failure: Want to Avoid

• Ring-off Cracking of Cordierite

DEER 2004

Regeneration Model: I/OInput:• Filter configuration: dimensions, cell density, wall thickness• Soot loading / distribution• Regeneration condition: inlet gas temp, exhaust flow rate, O2

• Substrate property: porosity, permeability, thermal, etc.• Soot property: packing density, permeability, etc.• Soot reaction kinetics and catalyst effect

Output:• Spatial and temporal profiles of key variables

– Temperature, velocity, oxygen, reaction rate, etc.

• Soot distribution (regen efficiency) vs. time

• A tool for parametric and what-if studies

DEER 2004

Regeneration Modeling: State of the Art• Ongoing, lots of good efforts to date

• Mostly 2-D, some 3-D (multiple 2-D channels)

• Lack details at channel level

• Simplified flow equations / Many assumptions

• Single temperature field for all phases: gas, substrate and soot

DEER 2004

Features in The New Model• 3-D for one inlet/outlet channel unit

• Porous medium model for substrate and soot layer

• Generic conservation equations for whole domain

• Capable of separate temperature fields for gas and solid via a heat transfer coefficient

• Arrhenius soot reaction with catalytic effect

• Custom CFD code

DEER 2004

Computational Domain

3-D view

Plug

Inlet

Inlet

Outlet

Outlet

Cross-section View

Inlet

Outlet

Side View

Plug

DEER 2004

Conservation Equations

Continuity equation:

Momentum equations:

Energy equation (single temperature):

Species continuity equation:

( ) ( ) muxt i

i

′′′=∂∂

+∂∂

&βρβ

βρβ

11

( ) ( ) kki

k

iki

ik S

xp

xu

xuu

xu

t+

∂∂

−⎟⎟⎠

⎞⎜⎜⎝

⎛∂∂

∂∂

=∂∂

+∂∂ βµ

ββρ

ββρ

β111

Hi

effii

ifPP S

xTk

xxTu

ctTc

+⎟⎟⎠

⎞⎜⎜⎝

⎛∂∂

∂∂

=∂∂

+∂∂

βββρ

βρ 1)( ,

( ) ( ) SxYD

xxYu

tY

iii

i +⎟⎟⎠

⎞⎜⎜⎝

⎛∂∂

∂∂

=∂

∂+

∂∂ βρ

ββρ

ββρ

β111

DEER 2004

Required Model Input:Soot Reaction Kinetics by TGA

Overlay of Kinetic Study of soot oxidationSecond Method - No Pre Treatment

0

10

20

30

40

50

60

70

80

90

100

0 200 400 600 800 1000 1200

Temperature (C)

Wei

ght (

%) 1 C/min

4 C/min7 C/min10 C/min

Temperature ( C)

Wei

ght (

%)

Arrhenius Plot of Heating RateSecond Method - No pretreatment

y = -9.8638x + 11.843R2 = 0.9995

y = -9.7341x + 11.827R2 = 0.9994

y = -9.65x + 11.905R2 = 0.9993

y = -9.4977x + 11.977R2 = 0.9977

0

0.2

0.4

0.6

0.8

1

1.2

1.08 1.10 1.12 1.14 1.16 1.18 1.20 1.22 1.24 1.26 1.28

1000/T (K)

Log

Hea

ting

Rat

e (C

/min

)

70% Oxidation60% Oxidation50% Oxidation40% Oxidation

Log

heat

ing

rate

(C/m

in)

1000/T (K-1)

DEER 2004

Required Model Input:Soot Cake Shape and Porosity by SEM

DEER 2004

Model Validation: Step 1. Heat-up w/o Regeneration

Heat Up at 160 liter/min - Cordierite

0

100

200

300

400

500

600

700

0 50 100 150 200 250 300

Time (sec)

Tem

pera

ture

(°C

)

DPF inletMeasured - DPF Mid-lengthModeled - DPF Mid-length

DEER 2004

Model Validation:Step 2. Regeneration

Temperature Measurements at 3 Axial Locations

200

300

400

500

600

700

800

0 60 120 180 240 300

Time (sec)

Tem

pera

ture

(C)

Measured at 3" down

Measured at 6" down

Measured at 9" down

DEER 2004

Model Validation:Step 2. Regeneration

Comparing Measurement and Model at 3 Axial Locations

200

300

400

500

600

700

800

0 60 120 180 240 300

Time (sec)

Tem

pera

ture

(C)

Measured at 3" downModeled at 3" downMeasured at 6" downModeled at 6" downMeasured at 9" downModeled at 9" down

DEER 2004

Model Results:Velocity and Temperature Fields

3-D View

Colored by Temperature

Cross-sectional View

DEER 2004

Model Results:Velocity and Pressure Field

Colored by Pressure

Side view – Porous Zone

Inlet

Outlet

Inlet

Outlet

Side view – Gas Zone

DEER 2004

Model Results:Temperature and Oxygen Distribution

DEER 2004

Model Results:Reaction Rate and Soot Distribution

Soot distribution

DEER 2004

Parametric Study:Effect of Flow Rate

200

300

400

500

600

700

800

900

1000

0 60 120 180 240 300

Time (sec)

Tem

pera

ture

(C)

Measured at 600 kg/h

Modeled at 600 kg/h

Modeled at 900 kg/h

Modeled at 300 kg/h

DEER 2004

Parametric Study:Effect of Soot Loading

200

300

400

500

600

700

800

900

1000

0 60 120 180 240 300

Time (sec)

Tem

pera

ture

(C)

1 gram/liter Soot

3 gram/liter Soot

5 gram/liter Soot

DEER 2004

200

300

400

500

600

700

800

0 60 120 180 240 300 360 420 480

Time (sec)

Tem

pera

ture

(C)

650C DPF inlet625C DPF inlet600C DPF inlet

Parametric Study:Effect of DPF Inlet Temperature

DEER 2004

200

300

400

500

600

700

800

900

1000

0 60 120 180 240 300

Time (sec)

Tem

pera

ture

(C)

10% Oxygen

5% Oxygen

Parametric Study:Effect of Oxygen

DEER 2004

Engine goes to idle

200

300

400

500

600

700

800

900

1000

0 60 120 180 240 300

Time (sec)

Tem

pera

ture

(C)

3" downstream

6" downstream

9" downstream

Mass flow rate

What-If Study:Runaway Regeneration?

DEER 2004

Summary • A new regeneration model was developed and validated• Features include: (1) 3-D; (2) A porous model; and (3)

Generic equations solved • The model provides detailed prediction of spatial and

temporal distributions of key parameters such as temperature, as well as regeneration efficiency

• The model was shown to be effective in parametric and what-if studies

• A good model is useful in the design and operation of DPF systems, hence shortening development cycle