Xin Q. Diesel Engine System Design

Подождите немного. Документ загружается.

508 Diesel engine system design

 A group of researchers at Michigan Technological University

(Huynh et al., 2003; Kladopoulou et al., 2003; Singh et al., 2005;

Mohammed et al., 2006a, 2006b; Premchand et al., 2007).

 A group of researchers at General Motors and the University of

Wisconsin–Madison (Kapparos et al., 2005; Strzelec et al., 2006;

England et al., 2006; He, 2007; Rutland et al., 2007; Gurupatham

and He, 2008; Singh et al., 2009).

 A group of researchers at Gamma Technologies for the software GT-

POWER (Tang et al., 2007, 2008; Wahiduzzaman et al., 2007).

 Other sources (Rumminger et al., 2001; Millet et al., 2002; Kandylas

and Koltsakis, 2002; Liu and Miller, 2002; Guo and Zhang, 2005a,

2005b; York et al., 2005, 2009; Reader et al., 2006; Yi, 2006;

Bouteiller et al., 2007; Cunningham and Meckl, 2007; Chiatti et

al., 2008; Subramaniam et al., 2009).

8.1.6 Exhaust thermal management

The pressure drop of the engine exhaust system is important for engine

performance. The thermal loss of the exhaust system also has a large

impact on aftertreatment performance. Modeling the exhaust system with

one-dimensional methods is discussed by Massey et al. (2002). The thermal

losses of the diesel engine exhaust system are investigated by Kapparos et

al. (2004) and Fortunato et al. (2007).

Figure 8.1 shows the exhaust gas temperature drop from the turbine

outlet to the DOC inlet for a heavy-duty diesel engine. It is observed that

Exhaust temperature drop at 77°F ambient,

45 mph vehicle speed

1000 1500 2000 2500

Engine speed (rpm)

Engine brake torque (ft.lb)

700

600

500

400

300

200

100

140

120

100

8.1 Simulation of exhaust temperature drop from turbine outlet to

DOC inlet.

Diesel-Xin-08.indd 508 5/5/11 11:52:07 AM



509Diesel aftertreatment integration and matching

the temperature drop is primarily a function of the engine speed, which is

related to the time scale in the exhaust pipe heat transfer. The temperature

drop becomes greater as the vehicle speed decreases. The temperature drop

is affected by the heat transfer loss of the exhaust pipe, which is in turn

affected by engine exhaust ow rate, ambient temperature, vehicle speed and

the orientation of the aftertreatment conguration (horizontal or vertical).

The exhaust gas temperature decreases when the engine operates in cold

climate (Fig. 8.2).

Flexibility of controlling the exhaust gas temperature and the ow rate is

another important topic in air system design for engine thermal management

and optimum aftertreatment performance. For example, the functions of

adsorber during the adsorption phase and reduction phase require

different exhaust temperatures and oxygen concentrations for the optimum

performance. In order for the converter to reach the light-off temperature

more easily or to carry out the aftertreatment regeneration more quickly, the

following measures can be used to raise the exhaust temperature:

∑ minimizing the heat losses by either insulating the exhaust pipe or

installing the aftertreatment device close to the exhaust manifold or

runner

∑ raising the intake manifold gas temperature by exible cooling or

bypassing the coolers on the gas side

∑ raising the exhaust temperature by bypassing the turbine

∑ retarding the fuel injection timing

∑ using the in-cylinder post fuel injection (with engine torque balance) to

reduce the air–fuel ratio and provide the reductant needed for the NO

adsorber regeneration or the DPF regeneration

Turbine outlet temperature ratio of –20 °F

cold ambient to 77 °F normal ambient

1000 1500 2000 2500

Engine speed (rpm)

Engine brake torque (ft.lb)

700

600

500

400

300

200

100

0.95

0.9

0.85

0.8

8.2 Simulation of exhaust temperature ratio between cold ambient

and normal ambient at turbine outlet.

Diesel-Xin-08.indd 509 5/5/11 11:52:07 AM



510 Diesel engine system design

∑ using external fuel dosing in the exhaust stream

∑ reducing the engine air ow rate by intake throttling, exhaust throttling,

or variable valve actuation to raise the exhaust temperature.

In some regeneration strategies, the EGR is turned off or reduced during the

regeneration in order to make the exhaust gas hotter. Increasing the NO

ratio by raising the concentration of feed-gas NO

can assist soot burning.

Hotter exhaust gas at the DOC outlet due to its exothermic reaction and

faster light-off also helps to initiate the DPF regeneration.

8.1.7 Aftertreatment calibration

The DoE method can be used in aftertreatment calibration for the DPF active

regeneration in order to optimize the effect of aftertreatment calibration

parameters on engine performance and emissions. The main fuel injection

timing and the air system control parameters such as intake throttle opening,

VGT vane opening, and EGR valve opening are adjusted in order to increase

the turbine outlet exhaust gas temperature to obtain a high light-off temperature

of the DOC. Once the DOC is heated up, it can produce high-efciency

oxidization of the post-injected fuel for the DPF regeneration. Based on

the DOC light-off temperature achieved, the in-cylinder post fuel injection

along with further modulation of the air system parameters can be used to

reach a sufciently high inlet gas temperature for the DPF regeneration.

The objective of the aftertreatment calibration optimization is to minimize

the BSFC under the constraints of acceptable NO

, soot, and HC emissions

at the DOC inlet, a sufciently high turbine outlet temperature, and an

undetectably small variation of the engine torque for driving smoothness.

Figure 8.3 shows an example of the advanced DoE optimization data at part

load conditions for the DPF regeneration of a heavy-duty diesel engine. The

control factors used in the DoE test include intake throttle opening, VGT

vane opening, EGR valve opening, and main fuel injection timing. The

‘T-DOC-In’ shown on the vertical axis of the gures represents the turbine

outlet exhaust gas temperature before the DOC. Engine BSFC is minimized

in the entire domain of ‘NO

vs. T-DOC-In’ by using the emulators built

from the DoE test data. These maps illustrate the parametric dependency

of the emission parameters and the exhaust temperature upon the air–fuel

ratio and the EGR rate.

8.1.8 Cold-start emissions control

The engine-out and aftertreatment-out emissions during transient cold-start

have received much greater attention since the 1990s after the type approval

procedures in emission regulations became more stringent, especially for

Diesel-Xin-08.indd 510 5/5/11 11:52:07 AM



0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

BSNO

(g/hp.hr)

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

BSNOx (g/(hp.hr))

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

BSNO

(g/(hp.hr))

0 1 2 3 4 5 6 7 8 9

BSNO

(g/(hp.hr)

0 1 2 3 4 5 6 7 8 9

BSNO

(g/(hp.hr))

0 1 2 3 4 5 6 7 8 9

BSNOx (g/(hp.hr))

0 1 2 3 4 5 6 7 8 9

BSNO

(g/(hp.hr))

0 1 2 3 4 5 6 7 8 9

BSNO

(g/(hp.hr))

0 1 2 3 4 5 6 7 8 9

BSNOx (g/(hp.hr))

HC concentration (ppm)

Brake speciﬁc HC (g/(hp.hr))

Smoke (avl FSN)

1800 rpm 15% load

1800 rpm 15% load 1800 rpm 15% load 1800 rpm 15% load

1800 rpm 15% load 1800 rpm 15% load

air–fuel ratio EGr rate (%)

BSFC (lb/(hp.hr))

air–fuel ratio EGr rate (%)

BSFC (lb/(hp.hr))

1000 rpm

20% load

1000 rpm

20% load

1000 rpm

20% load

T-DOC-ln (°F]

T-DOC-ln (°F]T-DOC-ln (°F]

T-DOC-ln (°F]

650

600

550

500

450

400

350

300

250

900

800

700

600

500

400

300

200

100

900

800

700

600

500

400

300

200

100

900

800

700

600

500

400

300

200

100

900

800

700

600

500

400

300

200

100

900

800

700

600

500

400

300

200

100

900

800

700

600

500

400

300

200

100

650

600

550

500

450

400

350

300

250

650

600

550

500

450

400

350

300

250

0.55

0.5

0.45

0.4

0.7

0.6

0.5

0.4

150

100

0.7

0.6

0.5

0.4

0.3

0.2

0.5

8.3 advanced

aftertreatment calibration

optimization for DPF

regeneration.

Diesel-Xin-08.indd 511 5/5/11 11:52:07 AM



512 Diesel engine system design

gasoline engines. The diesel engine is less sensitive than the gasoline engine

in cold-start emissions, although the diesel engine may experience problems of

‘white smoke’ formed from evaporated fuel droplets. Close-coupled catalysts

for cold-start emissions control may affect the HC:NO

ratio in the exhaust

 ow and hence the aftertreatment recipe. Overall, the emissions control at

cold-start in a cold climate is more of a local design issue related to catalyst

warm-up or cold-start aid rather than a system design focus.

8.2 Diesel particulate ﬁ lter (DPF) regeneration

requirements for engine system design

The DPF removes PM including the nanoparticles if a high  ltration ef ciency

is achieved. Engine–DPF matching requires careful integration. The DPF is

more complex than the DOC because the DOC is entirely passive, reacting

continuously, and does not require regeneration. The DPF is usually the

opposite.

A soot load factor is de ned as the ratio of the DPF pressure drop at the

fully loaded condition to the pressure drop at the clean condition (i.e., no

soot accumulated). The regeneration ef ciency is de ned as the ratio of the

pressure drop reduction after the regeneration to the pressure drop reduction

after an ideal complete removal of the soot. The soot load factor and the

DPF size are determined by the following: (1) the exhaust restriction level

allowed for acceptable engine performance; (2) the regeneration frequency;

(3) the peak burning temperature; (4) the temperature gradient during the

exothermic regeneration process related to DPF thermo-mechanical durability;

and (5) the regeneration ef ciency.

The regeneration ef ciency is de ned by

egen

lean

loade

ppDDpp

loade

–

ppDDpp

loade

–

8.1

where D p

loaded

is the pressure drop of a soot-loaded DPF, D p

regen

is the

pressure drop after the regeneration, and Dp

clean

is the pressure drop of a

clean DPF associated with certain incombustible ash loading. The  lters that

are only partially loaded are more dif cult to regenerate and to maintain

self-sustaining soot combustion. When the soot load accumulated in the DPF

exceeds a certain level, it is desirable to have a regeneration, either passive

or active, in order to burn off the soot to reduce the exhaust restriction.

Uncontrolled regeneration occurs with too much accumulation of soot and

deposited hydrocarbons. It generates excessively high burning temperatures

or temperature gradients to melt or crack the  lter. The related factors in

catalytic soot regeneration include mainly the following: exhaust temperature;

the concentrations of O

and NO

(for C+O

and C+NO

reactions of the

Diesel-Xin-08.indd 512 5/5/11 11:52:08 AM



513Diesel aftertreatment integration and matching

carbon burning); the NO

:NO (1) ratio; and the NO

:C mass ratio. The

following means can assist soot burning: higher temperature and higher O

concentration in the exhaust ow; (2) more deposited hydrocarbons (SOF)

and its exothermic burning reaction to provide energy to ignite the soot; (3)

lower space velocity or volumetric ow rate; and (4) lower soot loading.

DPF durability is related to the lter material and the design, the regeneration

burning temperature, the temperature spatial and temporal gradients, and

the regeneration frequency (related to the fatigue life). The regeneration

frequency is affected by the installation position and the soot loading. It is

also related to the lter size, the soot load-up rate and the driving cycle.

Small lters located closer to the engine under high-load driving need to

regenerate more frequently. DPF durability is discussed in more detail in

Chapter 2.

In the non-catalyzed DPF regeneration, the soot burning lasts a few minutes

and requires approximately 550–650°C exhaust temperature to start, which

cannot be reached in most part-load operating conditions. It was reported that

a complete burn of soot requires an exhaust temperature of above 600°C and

an oxygen concentration of above 7% (Basshuysen and Schafer, 2004). In

the catalyzed DPF (CDPF) regeneration, additives in the fuel or the catalytic

coating inside the lter may be used in order to reduce the thermal energy

required to initiate the regeneration. The soot ignition temperature can be

signicantly reduced to the range of 275–450°C. The catalyst, either for the

SOF or for the carbon soot, is usually more effective at a higher oxygen

concentration in the exhaust ow. The ignition and burning performance of

the catalytic regeneration also depends on the amount of NO

, the fuel sulfur

level, and the compositions of SOF and PM.

Passive regeneration, also known as self regeneration, is usually assisted

by catalytic means to lower the soot oxidation temperature. It offers certain

advantages such as not using sensing/control systems and hence eliminating

their durability troubles. In the passive catalytic regeneration, there is no

control over the soot load, the pressure drop of the DPF, and the regeneration

efciency before and after the regeneration because the driving condition

can initiate or halt the regeneration. Once started, the combustion of soot

may become self-sustaining, and thermal runaway could happen. Therefore,

low soot loading is preferred in the passive regeneration in order to avoid

catastrophic thermal failures.

In the active DPF regeneration, when the Dp

loaded

exceeds a certain

preset threshold, regeneration is triggered to occur. In addition to the above-

mentioned measures of raising the exhaust temperature, other methods may

also be adopted, some with additional costs, in order to precisely control

the course of the regeneration and the regeneration efciency. The methods

include: (1) using a fuel-fed burner; (2) electrical heating; (3) microwave;

(4) compressed air; (5) using controlled pre-DOC or post-DOC dosing of

Diesel-Xin-08.indd 513 5/5/11 11:52:08 AM



514 Diesel engine system design

hydrocarbons as a catalyst to generate sufcient heat energy to regenerate

the DPF; (6) using a transmission to force the engine to operate at a high

load; and (7) producing oxidants such as NO

turnover from the NO in the

DOC or the CDPF in order to oxidize the carbon soot (with reactions: NO

+ 0.5O

= NO

, 2NO = NO

+ 0.5N

, C + NO

= NO + CO). It is noted

that NO

is much more toxic than NO; but after leaving the tailpipe, NO

will be converted to NO

in the atmosphere anyway. For low engine-out

emissions, there is an impact on the NO

oxidant available for the DPF

regeneration.

A dynamic and precise control of the DPF regeneration process is needed

for several reasons. A dangerous phenomenon called ‘hill-cresting’ refers to

the thermal runaway (meltdown) with uncontrolled, excessively high soot-

burning temperatures inside the DPF when the regeneration occurs immediately

before the end of a high-load condition. In other words, thermal runaway

occurs when the engine speed/load decreases during the course of the DPF

regeneration. After the engine load or the EGR rate decreases or the speed

changes, the air–fuel ratio may increase. The exhaust ow rate may become

lower, resulting in a reduction in the convective cooling rate inside the DPF.

Consequently, the temperature may rise to unacceptably high levels to cause

durability issues. Moreover, it is important to slow down the regeneration

burn-off rate and to control the temperature spatial gradient because they

affect the thermal stress and the fatigue life of the DPF. The burn-off rate

can be reduced or optimized by changing the exhaust ow rate. It can also be

decreased by reducing the oxygen concentration via a reduction in air–fuel

ratio or an increase in EGR rate.

Secondary emissions may occur during the regeneration, such as the

following species: (1) the CO from incomplete soot burning; (2) the

evaporation of the adsorbed HC and SOF during a slow exhaust heating

phase without being oxidized; (3) the unburned hydrocarbons of post-injection

during the DPF heating; (4) the NO

due to shutting off EGR or due to the

soot oxidation reaction (C + NO

= CO + CO

+ NO); (5) the sulfuric acid

aerosol nanoparticles created by the catalyst; and (6) the organic compounds

nucleated downstream of the DPF. Using an ultra-low sulfur fuel, a low-

sulfur lube oil (McGeehan et al., 2002, 2006, 2007), and a sulfur trap may

eliminate the sulfur-containing aerosol nanoparticles. The effectiveness of

DPF regeneration also depends on the sulfur content in the diesel fuel.

An important goal of engine air system design is to robustly adapt to the

varying level of exhaust restriction caused by DPF soot loading changes,

and to achieve the capability required for optimum DPF regeneration at all

ambient conditions and operating modes. Vehicle driving cycle simulation

plays a critical role in analyzing the exhaust temperature windows as

functions of time and occurrence frequency for optimized DPF matching

and regeneration.

Diesel-Xin-08.indd 514 5/5/11 11:52:08 AM



515Diesel aftertreatment integration and matching

8.3 Analytical approach of engine–aftertreatment

integration

The analysis tasks of engine–aftertreatment integration or matching need

to start with analyzing the requirements of exhaust temperature, ow rate,

and constituent concentrations (e.g., O

, HC, CO, NO) at the inlet of each

aftertreatment device (e.g., DOC, SCR or LNT, DPF) for their normal

operation and regeneration in the entire engine speed–load domain. The

common requirements need to be consolidated, and the differences need to

be identied. Finally, the optimum air handling system needs to be designed

to minimize the BSFC and maintain good durability for the whole engine–

aftertreatment system. A similar approach of engine–vehicle matching and

engine–turbocharger matching may be used in engine–aftertreatment matching,

i.e., plotting the engine operating data on aftertreatment characteristic maps

for the non-regeneration and regeneration phases.

The process of engine–aftertreatment performance matching is summarized

in the following ve steps (Fig. 8.4):

1. Understand the aftertreatment design and the operating characteristics for

both normal operation and regeneration, and create the characteristic maps

for each device (e.g., DOC, DPF, SCR, or LNT) where the characteristics

need to be presented as functions of engine performance-related parameters

(such as ow rate, temperature and gas mass, and certain fundamental

combinations of these parameters) and chemical parameters (such as

the NO

:NO

ratio) in order to facilitate the matching in the next steps.

The characteristic responses may be presented in contours, for example

the chemical reactions involved, the light-off temperature, the DPF

‘balance point temperature’, temperature windows, species conversion

efciency, other efciencies, component size, storage capacity, ow

restriction, loading performance (e.g., DPF or LNT load-up rate), heat

losses, regeneration energy and temperature, thermal durability, etc.

Identifying the fundamental inuential parameters and constructing such

characteristic maps are the important tasks for aftertreatment suppliers

when they cooperate with the engine manufacturer.

2. Analyze the engine–aftertreatment calibration DoE test data to understand

the aftertreatment performance on the engine. The engine manufacturer

is responsible for this step.

3. Match the engine and each aftertreatment device by plotting the engine

performance characteristic data (e.g., exhaust gas temperature, ow rate

and engine-out emissions in the engine speed–load domain, at various

ambient and altitude conditions) on the aftertreatment characteristic maps;

or vice versa, whichever is more convenient. The engine manufacturer

holds a major responsibility for this step.

4. Discover any mismatch in both normal operation and regeneration modes

Diesel-Xin-08.indd 515 5/5/11 11:52:08 AM



vehicle driving

duty cycle

air system

hardware design

and calibration

Engine operation when

aftertreatment operates in

non-regeneration phase

Engine operation when

aftertreatment operates in

regeneration phase

DPF

DOC

SCr

lNT

aftertreatment

Engine related parameter a

Engine related parameter B

Mismatch

aftertreatment

characteristics

Driving cycle

Engine

characteristics

Good match requires

after treatment and

engine characteristics of

driving cycle stay closer

8.4 Engine–aftertreatment matching.

Diesel-Xin-08.indd 516 5/5/11 11:52:08 AM



517Diesel aftertreatment integration and matching

from the maps. Revise the engine air system controls and reconstruct

the matching maps. Alternatively, reselect the aftertreatment. Finally,

optimize all the links between different aftertreatment devices and the

engine–aftertreatment interface. The suppliers and the engine manufacturer

need to work together to accomplish this step. During the matching,

engine air system performance simulation and vehicle transient driving

cycle simulation need to be conducted. In addition to the packaging and

cost attributes of the aftertreatment design, consideration should be given

to the following important performance and durability topics during the

matching:

∑ Select the optimum location and sequence of each aftertreatment

device based on the steady-state and transient performance criteria of

the engine and the aftertreament (e.g., during turbocharger lag).

∑ Minimize the mismatch in the temperature windows and the space

velocity windows.

∑ Arrange exhaust thermal management to minimize the thermal energy

required (e.g., bypass the gas ow or conducting in-line regeneration).

Note that turbocharger efciency, air–fuel ratio and EGR rate all

have a direct impact on the turbine outlet exhaust temperature and

aftertreatment performance.

∑ Select the optimum amount of the precious metal used in catalytic

reaction based on the balance of product cost between the base

engine and the aftertreatment devices.

∑ Control back-pressure variations and the pressure drop of the

aftertreatment devices for optimum engine performance.

∑ Optimize the DPF regeneration methods (either active or passive)

and control the regeneration process.

∑ Minimize the fuel economy penalty due to regeneration by balancing

the engine–aftertreatment as a whole.

∑ Minimize the risk of temperature-induced durability problems for

the aftertreatment devices.

∑ Minimize the engine power surge during aftertreatment regeneration

for good vehicle drivability.

5. Finalize the engine-out emissions target and the air system requirements

from an aftertreatment perspective for the designs of the engine air

system and electronic controls.

Integrated system-level modeling to combine the engine and the aftertreatment

as a whole is the direction for future advanced simulations. Analytical tools

play a vital role in the engine–aftertreatment matching and integration.

Simulation models of system-level aftertreatment integration are presented

in Stamatelos et al. (1999), Peters et al. (2004), Kapparos et al. (2005),

England et al. (2006), Strzelec et al. (2006), Wahiduzzaman et al. (2007),

Diesel-Xin-08.indd 517 5/5/11 11:52:08 AM

