U.S. patent application number 11/408428 was filed with the patent office on 2007-04-12 for narrow speed range diesel-powered engine system w/ aftertreatment device.
This patent application is currently assigned to Eaton Corporation. Invention is credited to Haoran Hu, James Edward JR. McCarthy, Thomas Stover.
Application Number | 20070082783 11/408428 |
Document ID | / |
Family ID | 36586338 |
Filed Date | 2007-04-12 |
United States Patent
Application |
20070082783 |
Kind Code |
A1 |
Hu; Haoran ; et al. |
April 12, 2007 |
Narrow speed range diesel-powered engine system w/ aftertreatment
device
Abstract
One or more parameters for denitration or desulfations of a LNT
is varied, whereby the saturation of NOx and/or SOx in the LNT is
reduced to a lower level when an operating state makes the LNT
otherwise less effective or places a greater demand for conversion
efficiency on the LNT. The operating state can relate to whether
the LNT is at the limit of its effective operating temperature
range, a degree of poisoning, or an engine operating state.
Selectively tolerating high degrees of sulfur poisoning or NOx
saturation during periods of low exhaust flow allows the efficiency
of denitrations and/or desulfations to be increased over a large
portion of a vehicle's operating cycle, particularly when the
engine uses a CVT. Another concept relates to a power generation
system comprising an engine tuned to efficiently operate in a
narrow speed range for all levels of power output.
Inventors: |
Hu; Haoran; (Novi, MI)
; Stover; Thomas; (Milford, MI) ; McCarthy; James
Edward JR.; (Canton, MI) |
Correspondence
Address: |
PAUL V. KELLER, LLC
4585 LIBERTY RD.
SOUTH EUCLID
OH
44121
US
|
Assignee: |
Eaton Corporation
Cleveland
OH
|
Family ID: |
36586338 |
Appl. No.: |
11/408428 |
Filed: |
April 21, 2006 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11246974 |
Oct 7, 2005 |
7063642 |
|
|
11408428 |
Apr 21, 2006 |
|
|
|
Current U.S.
Class: |
477/100 |
Current CPC
Class: |
F01N 3/0885 20130101;
F01N 3/2066 20130101; F16H 2059/743 20130101; F02D 31/001 20130101;
F01N 3/0842 20130101; F02D 41/00 20130101; Y02T 10/12 20130101;
F16H 61/66259 20130101; F16H 2061/0018 20130101; Y10T 477/663
20150115; F01N 2610/03 20130101; F01N 3/0871 20130101; F01N 2250/12
20130101; Y02A 50/20 20180101; F01N 2250/02 20130101; F02D 41/0235
20130101; F01N 2250/14 20130101; F01N 3/2073 20130101; F01N 2240/30
20130101; F01N 13/0097 20140603 |
Class at
Publication: |
477/100 |
International
Class: |
F16H 59/74 20060101
F16H059/74 |
Claims
1. A method of operating an exhaust aftertreatment system for a
vehicle, comprising: removing NOx from the engine exhaust using an
exhaust aftertreatment system comprising a NOx adsorber-catalyst;
from time-to-time, denitrating the NOx adsorber-catalyst;
determining whether the exhaust aftertreatment system or a part
thereof has a satisfactory level of activity in relation to the
vehicle's operating mode; desulfating the NOx adsorber-catalyst if
its activity level cannot be made satisfactory by denitrating;
wherein a satisfactory level of activity is determined in relation
to the vehicle's operating mode whereby a greater degree of
reversible sulfur poisoning is tolerated in vehicle operating modes
that require lower NOx conversion efficiency.
2. A method of operating an exhaust aftertreatment system,
comprising: removing NOx from the engine exhaust using an exhaust
aftertreatment system, wherein the aftertreatment system has a
subsystem having an activity level relating to the efficiency with
which it removes NOx; maintaining an estimate for the current
activity level; and initiating a regeneration of the exhaust
aftertreatment based at least on both the current activity level
and an operating state; wherein regeneration is delayed until lower
activity levels occur if the current operating state places less
than a peak demand on the exhaust aftertreatment system and
regeneration is begun earlier and while higher activity levels
remain if the current operating state places a peak demand on the
exhaust aftertreatment system; and the operating state relates to
one of a vehicle operating mode, a degree of irreversible poisoning
or loss of activity in the NOx adsorber-catalyst, or a temperature
of the NOx adsorber-catalyst.
3. The method of claim 2, wherein the aftertreatment system
comprises a NOx adsorber catalyst and the regeneration is a
desulfation.
4. The method of claim 2, wherein the aftertreatment system
comprises a NOx adsorber catalyst and the regeneration is a
denitration.
5. The method of claim 2, wherein the aftertreatment system
comprises a NOx adsorber catalyst and a SCR catalyst.
6. The method of claim 2, wherein the operating state is a vehicle
operating state that relates to demand for engine power.
7. A method of operating an exhaust aftertreatment system,
comprising: removing NOx from the engine exhaust using an exhaust
aftertreatment system; from time-to-time, denitrating the NOx
adsorber-catalyst; from time-to-time, desulfating the NOx
adsorber-catalyst; wherein an endpoint for at least one type of
regeneration is selected in a manner that causes the degree of
saturation of the substance being removed at the endpoint of the
regeneration to depend on an operating state; wherein the endpoint
results in higher saturations if the current operating state places
less than a peak demand on the exhaust aftertreatment system and
the endpoint results in lower saturations if the current operating
state places a peak demand on the exhaust aftertreatment system;
and the operating state relates to one of a vehicle operating mode,
a degree of irreversible poisoning or loss of activity in the NOx
adsorber-catalyst, a temperature of the NOx adsorber-catalyst, or
an exhaust flow rate.
8. The method of claim 7, wherein the endpoint for denitration is
selected in a manner that causes the degree of NOx saturation at
the endpoint of denitration to depends an operating state.
9. The method of claim 7, wherein the endpoint for desulfation is
selected in a manner that causes the degree of SOx saturation at
the endpoint of desulfation to depends an operating state.
10. The method of claim 7, wherein the endpoint is selected to
reduce the saturation of the substance being removed to a target
value.
11. The method of claim 7, wherein the aftertreatment system
comprises a NOx adsorber catalyst and the regeneration is a
desulfation.
12. The method of claim 7, wherein'the aftertreatment system
comprises a NOx adsorber catalyst and a SCR catalyst.
13. The method of claim 7, wherein the operating state is a vehicle
operating state that relates to demand for engine power.
Description
CROSS-REFERENCE TO RELATED APPLICATION
[0001] The present application is a divisional of U.S. patent
application Ser. No. 11/246,974, filed Oct. 7, 2005.
FIELD OF THE INVENTION
[0002] The present invention relates to pollution control systems
and methods for diesel engines and lean burn gasoline engines.
BACKGROUND
[0003] NO.sub.x, emissions from diesel engines are an environmental
problem. Several countries, including the United States, have long
had regulations pending that will limit NO.sub.x, emissions from
trucks and other diesel-powered vehicles. Manufacturers and
researchers have put considerable effort toward meeting those
regulations.
[0004] In gasoline powered vehicles that use stoichiometric
fuel-air mixtures, three-way catalysts have been shown to control
NO.sub.x emissions. In diesel-powered vehicles, which use
compression ignition, the exhaust is generally too oxygen-rich for
three-way catalysts to be effective.
[0005] Several solutions have been proposed for controlling NOx
emissions from diesel-powered vehicles. One set of approaches
focuses on the engine. Techniques such as exhaust gas recirculation
and partially homogenizing fuel-air mixtures are helpful, but these
techniques alone will not eliminate NOx emissions. Another set of
approaches remove NOx from the vehicle exhaust. These include the
use of lean-burn NO.sub.x catalysts, selective catalytic reduction
(SCR), and lean NO.sub.x traps (LNTs).
[0006] Lean-burn NOx catalysts promote the reduction of NOx under
oxygen-rich conditions. Reduction of NOx in an oxidizing atmosphere
is difficult. It has proved challenging to find a lean-burn
NO.sub.x catalyst that has the required activity, durability, and
operating temperature range. Lean-burn NO.sub.x catalysts also tend
to be hydrothermally unstable. A noticeable loss of activity occurs
after relatively little use. Lean-burn NOx catalysts typically
employ a zeolite wash coat, which is thought to provide a reducing
microenvironment. The introduction of a reductant, such as diesel
fuel, into the exhaust is generally required and introduces a fuel
economy penalty of 3% or more. Currently, peak NOx conversion
efficiencies for lean-burn catalysts are unacceptably low.
[0007] SCR generally refers to selective catalytic reduction of NOx
by ammonia. The reaction takes place even in an oxidizing
environment. The NOx can be temporarily stored in an adsorbant or
ammonia can be fed continuously into the exhaust. SCR can achieve
high levels of NOx reduction, but there is a disadvantage in the
lack of infrastructure for distributing ammonia or a suitable
precursor. Another concern relates to the possible release of
ammonia into the environment.
[0008] LNTs are NOx adsorbants with catalysts that reduce NOx
during regeneration. The adsorbant is typically an alkaline earth
oxide adsorbant, such as BaCO.sub.3 and the catalyst is typically a
precious metal, such as Pt or Ru. In lean exhaust, the catalyst
speeds oxidizing reactions that lead to NOx adsorption. Accumulated
NOx is removed and the LNT is regenerated by creating a reducing
environment within the LNT. In a rich environment, the catalyst
activates reactions by which adsorbed NOx is reduced and
desorbed.
[0009] Regeneration to remove accumulated NOx may be referred to as
denitration in order to distinguish desulfation, described below.
The reducing environment for denitration can be created in several
ways. One approach uses the engine to create a rich fuel-air
mixture. For example, the engine can inject extra fuel into the
exhaust within one or more cylinders prior to expelling the
exhaust. A reducing environment can also be created by injecting a
reductant into the exhaust downstream of the engine. In either
case, a portion of the reductant is generally expended to consume
excess oxygen in the exhaust. To lessen the amount of excess oxygen
and reduce the amount of reductant expended consuming excess
oxygen, the engine may be throttled, although such throttling may
have an adverse effect on the performance of some engines.
[0010] Reductant can consume excess oxygen by either combustion or
reforming reactions. Typically, the reactions take place upstream
of the LNT over an oxidation catalyst or in a reformer. The
reductant can also be oxidized directly in the LNT, but this tends
to result in faster thermal aging. As an example, U.S. Pat. Pub.
No. 2003/0101713 describes an exhaust system with a fuel reformer
placed inline with the exhaust and upstream of a LNT. The reformer
includes both oxidation and reforming catalysts. The reformer both
removes excess oxygen and converts the diesel fuel reductant into
more reactive reformate.
[0011] In addition to accumulating NOx, LNTs accumulate SOx. SOx is
the combustion product of sulfur present in ordinarily diesel fuel.
Even with reduced sulfur fuels, the amount of SOx produced by
diesel combustion is significant. SOx adsorbs more strongly than
NOx and necessitates a more stringent, though less frequent,
regeneration. Desulfation requires elevated temperatures as well as
a reducing atmosphere. The elevated temperatures required for
desulfation can be produced by oxidizing reductant.
[0012] It is known that a NOx adsorber-catalyst can produce ammonia
during denitration and from this knowledge it has been proposed to
combine a NOx adsorber-catalyst and an ammonia SCR catalyst into
one system. Ammonia produced by the NOx adsorber-catalyst during
regeneration is captured by the SCR catalyst for subsequent use in
reducing NOx, thereby improving conversion efficiency over a
stand-alone NOx adsorber-catalyst with no increase in fuel penalty
or precious metal usage. U.S. Pat. No. 6,732,507 describes such a
system. U.S. Pat. Pub. No. 2004/0076565 describes such systems
wherein both components are contained within a single shell or
disbursed over one substrate. WO 2004/090296 describes such a
system wherein there is an inline reformer upstream of the NOx
adsorber-catalyst and the SCR catalyst.
[0013] It is known that LNTs function optimally only within limited
temperature ranges. U.S. Pat. Pub. No. 2003/0074888 states that NOx
reduction by a LNT is particularly efficient in the temperature
range from 300 to 350 .degree. C. The disclosure suggests heat
exchange within the exhaust gas treatment system to maintain
temperatures within a desired range. U.S. Pat No. 5,404,719
suggests another method of maintaining the temperature of a LNT
within a range where adsorption is efficient. When the temperature
needs to be increased, fuel is injected into the exhaust. When the
temperature needs to be decreased, air is injected into the
exhaust.
[0014] U.S. Pat. No. 6,866,610 suggests using a continuously
variable transmission (CVT) to prevent a catalytic converter having
a NOx storage reduction catalyst from cooling below an activation
temperature. In general, the CVT system is controlled to provide
torque multipliers at which the engine produces a required power
with optimal fuel economy. If, however, the optimal fuel economy
operating point would place the exhaust temperature in a low range,
a different torque ratio and engine operating point is selected to
increase the exhaust temperature.
[0015] Some other uses of a CVT in connection with exhaust
aftertreatment have been proposed. U.S. Pat. No. 6,135,917
describes using CVT to select operating points to speed the
light-off of a catalytic converter. U.S. Pat. No. 6,157,885
describes using a CVT system to avoid high exhaust temperatures
that would damage an exhaust gas purification system. U.S. Pat. No.
6,188,944 suggests using CVT to mitigate torque variations when a
lean-burn gasoline engine is run rich in order to regenerate a
LNT.
[0016] In spite of advances, there continues to be a long felt need
for an affordable and reliable exhaust treatment system that is
durable, has a manageable operating cost (including fuel penalty),
and is practical for reducing NOx emissions from diesel engines to
a satisfactory extent in the sense of meeting U.S. Environmental
Protection Agency (EPA) regulations effective in 2010 and other
such regulations.
SUMMARY
[0017] The present disclosure includes several concepts in which a
transmission, preferably a continuously variable transmission
(CVT), is used to facilitate the operation of an exhaust
aftertreatment system. Additional concepts relate to methods of
controlling exhaust aftertreatment. These methods are, in general,
particularly useful when used in connection with power generation
systems having CVTs.
[0018] One concept relates to a power generation system having a
fuel reformer positioned inline with an engine exhaust stream. A
transmission controller is configured to select operating points
for the engine in order to facilitate start-up or operation of the
fuel reformer. In one embodiment, the controller is configured to
select operating points to heat the exhaust and thus the reformer
prior to starting the reformer. This approach can reduce the
reformer start-up time and allows the reformer to be started in
situations where the reformer would otherwise be too cool to start
with fuel injection alone.
[0019] Another concept relates to a method of starting a fuel
reformer configured inline with an exhaust system. The transmission
is used to shift the operating point of the engine to produce a
hotter exhaust without affecting the engine's power output. The
hotter exhaust is allowed to heat the reformer for a period. If
necessary, fuel can be injected and combusted in the reformer to
provide further heating. The fuel injection rate is then set,
optionally in conjunction with reducing the oxygen content of the
exhaust, to provide lambda less than 1.0, whereby the reformer
begins to produce substantial quantities of reformate.
[0020] Another concept relates to a power generation system
comprising a LNT for exhaust aftertreatment. A transmission
controller is configured to select operating points for the engine
to reduce or limit the oxygen concentration in the exhaust during
denitration or desulfation of the LNT. The operating point
selection generally reduces the fuel penalty associated with
consuming excess oxygen in the exhaust during regeneration. This
approach is particularly useful when the aftertreatment system has
an inline reformer. Reducing the oxygen concentration may also
prevent the reformer and/or the LNT from overheating.
[0021] Another concept relates to a method of regenerating a LNT in
which a reductant is injected into the exhaust to provide a
reducing environment for the LNT. A transmission is used to reduce
or limit the exhaust oxygen concentration during the regeneration.
In one embodiment, the transmission is directed to shift an engine
operating point prior to beginning reductant injection in order to
allow time for a desired operating point to be reached prior to
injecting the reductant.
[0022] Another concept also relates to a method of regenerating a
LNT. The method involves injecting a reductant into the exhaust to
consume excess oxygen and reduce NOx stored in the LNT. A
transmission is used to reduce a fuel penalty for the regeneration.
The fuel penalty includes at least a contribution associated with
consuming excess oxygen in the exhaust. In one embodiment, the fuel
penalty also includes a contribution associated with operating an
engine at points apart from its optimal fuel economy operating
points. This method can take into account complex effects of both
exhaust oxygen concentration and exhaust flow rate on a fuel
penalty associated with regeneration.
[0023] Another concept relates to a power generation system
comprising a LNT for exhaust aftertreatment. The LNT has an
effective operating temperature range. When the LNT is near a limit
of its effective operating temperature range, the transmission is
used to select operating points that increase the LNT's
effectiveness. Generally, these operating points reduce the exhaust
flow rate, although other factors such as the exhaust temperature
may also be taken into account in selecting the operating points.
Preferably, the LNT's effective operating temperature range
includes exhaust temperatures produced by the engine at its point
of peak power output, whereby the LNT does not approach the limits
of its effective operating temperature range except when the engine
is at less than peak power. At lower power levels, it is generally
possible to select operating points that provide lower exhaust flow
rates than the flow rate occurring at the peak power level.
Reducing the exhaust flow rate can be more effective than adjusting
the exhaust temperature in maintaining the LNT's effectiveness.
[0024] Another concept relates to a method of operating an exhaust
aftertreatment system comprising removing NOx from a vehicle's
engine exhaust; from time-to-time, regenerating to remove NOx from
a NOx adsorber-catalyst of the aftertreatment system; and from
time-to-time, regenerating to remove SOx from the NOx
adsorber-catalyst. According to this concept, one or more
parameters for one or both types of regeneration varies, whereby
the saturation of NOx and/or SOx in the NOx adsorber-catalyst is
reduced to a lower level when an operating state makes the NOx
adsorber-catalyst otherwise less effective or places a greater
demand for conversion efficiency on the NOx adsorber-catalyst.
Where the operating state reduces the LNT efficiency or creates a
high demand for LNT efficiency, more extensive regenerations can be
used to compensate. On the other hand, where the operating state
allows for lower LNT efficiency, less frequent or less extensive
regeneration can be used. Less frequent or less extensive
regenerations can reduce the fuel penalty associated with
regeneration. Less frequent and/or shorter desulfations may also
increase the life of the LNT.
[0025] The operating state can relate to whether the LNT is at the
limit of its effective operating temperature range, a degree of
poisoning, or an engine operating state. The engine operating state
generally relates to power demands. Particularly where a CVT is
used, the exhaust flow rate is generally relatively low for all but
the highest levels of power demand. Lower exhaust flow rates place
lower demands on the aftertreatment system. Selectively tolerating
high degrees of sulfur poisoning or NOx saturation during periods
of low exhaust flow allows the efficiency of denitrations and/or
desulfations to be increased over a large portion of a vehicle's
operating cycle. In addition, the number and/or duration of
desulfations can be significantly reduced.
[0026] Another concept relates to a power generation system
comprising an exhaust aftertreatment system, a transmission and an
engine tuned whereby the engine can be efficiently maintained
within a narrow speed range, e.g. within a 300 RPM range, for all
levels of power output. The narrow speed range is generally a low
speed range, whereby the peak volumetric flow rate of the exhaust
is low in comparison to a conventional power generation system and
the demands on the aftertreatment system are correspondingly less.
In one embodiment, the lower demands on the aftertreatment system
are used to reduce catalyst loading in an aftertreatment device. In
another embodiment, the lower demands on the aftertreatment system
are used to operate a LNT with more efficient, less frequent,
and/or shorter regenerations.
[0027] The primary purpose of this summary has been to present
certain of the inventors' concepts in a simplified form to
facilitate understanding of the more detailed description that
follows. This summary is not a comprehensive description of every
one of the inventors' concepts or every combination of the
inventors' concepts that can be considered "invention". Other
concepts of the inventors will be conveyed to one of ordinary skill
in the art by the following detailed description and annexed
drawings. The concepts disclosed herein may be generalized,
narrowed, and combined in various ways with the ultimate statement
of what the inventors claim as their invention being reserved for
the claims that follow.
BRIEF DESCRIPTION OF THE DRAWINGS
[0028] FIG. 1 is a schematic illustration of an exemplary power
generation system.
[0029] FIG. 2 is a flow chart of an exemplary method of operating a
power generation system.
[0030] FIG. 3 is a plot showing optimal fuel economy operating
points for conventional and narrow-speed range diesel engines.
[0031] FIG. 4 is an exemplary plot of LNT effectiveness as a
function of temperature with examples of fresh and aged (poisoned)
catalysts.
[0032] FIG. 5 is a flow chart of an exemplary method for selecting
regeneration parameters.
[0033] FIG. 6 is an exemplary plot of NOx saturation for an LNT
through a series of lean/rich cycles.
[0034] FIG. 7 is a flow chart of an exemplary method for
determining whether to initiate a denitration.
[0035] FIG. 8 is a flow chart of an exemplary method for heating a
reformer.
[0036] FIG. 9 is a plot of showing an exemplary variation of
exhaust temperature with engine operating point.
[0037] FIG. 10 is a plot of showing an exemplary variation of
engine air-fuel ratio with engine operating point.
[0038] FIG. 11 is a flow chart of an exemplary method for
determining whether to initiate a desulfation.
[0039] FIG. 12 is a flow chart of an exemplary method of
desulfating an LNT.
DETAILED DESCRIPTION
[0040] FIG. 1 is a schematic illustration of an exemplary power
generation system 5, representing one of many systems in which
various concepts of the inventors can be implemented. The system 5
comprises an engine 9, a transmission 8, and an exhaust
aftertreatment system 7. The exhaust aftertreatment system 7
includes a controller 10, a fuel injector 11, a lean NOx catalyst
15, a reformer 12, a diesel particulate filter (DPF) 13, a lean
NOx-trap (LNT) 14, an ammonia-SCR catalyst 16, and a clean-up
catalyst 17. The controller 10 receives data from several sources,
include temperature sensors 20 and 21 and NOx sensors 22 and 23.
The controller 10 may be an engine control unit (ECU) that also
controls the transmission 8 and the exhaust aftertreatment system 7
or may include several control units that collectively perform
these functions.
[0041] The transmission 8 is generally of a type that allows
selection from among a large number of widely ranging torque
multipliers and makes available a range of operating points at
which the engine 9 can meet a given power demand. Typically, the
transmission 8 is a continuously variable transmission (CVT).
[0042] The lean-NOx catalyst 15 removes a portion of the NOx from
the engine exhaust using reductants, typically hydrocarbons that
form part of the exhaust or have been stored by the lean-NOx
catalyst 15. The DPF 13 removes particulates from the exhaust.
During lean operation (a lean phase), the LNT 14 adsorbs a second
portion of the NOx. The ammonia-SCR catalyst 16 may have ammonia
stored from a previous regeneration of the LNT 14 (a rich phase).
If the ammonia-SCR catalyst 16 contains stored ammonia, it removes
a third portion of the NOx from the lean exhaust. The clean-up
catalyst 17 may serve to oxidize CO and unburned hydrocarbons
remaining in the exhaust.
[0043] From time-to-time, the LNT 14 must be regenerated to remove
accumulated NOx (denitrated). Denitration may involve heating the
reformer 12 to an operational temperature and then injecting fuel
using the fuel injector 11. The reformer 12 uses the injected fuel
to consume excess oxygen in the exhaust while producing reformate.
The reformate thus produced reduces NOx adsorbed in the LNT 14.
Some of this NOx is reduced to NH.sub.3, most of which is captured
by the ammonia-SCR catalyst 16 and used to reduce NOx during a
subsequent lean phase. The clean-up catalyst 17 oxidizes unused
reductants and unadsorbed NH.sub.3 using stored oxygen. During
regeneration, the lean-NOx catalyst 15 may store reductant for
later use. The DPF 13 may serve to protect the LNT 14 from
excessive temperatures by providing a buffer between the reformer
12 and the LNT 14. Reducing the number and/or magnitude of
temperature excursions in the LNT 14 may extend the life of the LNT
14.
[0044] From time-to-time, the LNT 14 must also be regenerated to
remove accumulated SOx (desulfated). Desulfation may involve
heating the reformer 12 to an operational temperature, heating the
LNT 14 to a desulfating temperature, and providing the heated LNT
14 with a reducing atmosphere. A typical desulfation temperature is
in the range from about 500 to about 800.degree. C., more typically
in the range from about 650 to about 750.degree. C. Below a minimum
temperature, desulfation is very slow. Above a maximum temperature,
the LNT 14 may be damaged. A desulfation temperature is generally
obtained by combustion of injected fuel in the reformer 12. The
reformer 12 can generally be operated continuously unless it is
necessary to pulse the fuel supply rate to prevent either the
reformer 12, the DPF 13, the LNT 14, or the ammonia-SCR catalyst 16
from overheating. Pulsing allows devices to cool between fuel
pulses.
[0045] FIG. 2 is a flow chart of a process 100 embodying several of
the inventors' concepts for operating the aftertreatment system 7
in conjunction with the engine 9 and the transmission 8. The
controller 10 can be configured to implement the process 100. The
process 100 begins in step 101, wherein a default choice is made
for the engine operating point. In the present disclosure, an
operating point selection and in particular an engine operating
point selection made through a transmission, should be understood
as a selection of a torque multiplier. The selection of a torque
multiplier determines one from a plurality of engine speed-torque
combinations that can produce a given power level, the power level
generally being determined by vehicle operation. Accordingly,
operating points can be characterized in terms of engine speed and
power level or engine speed and torque multiplier. Typically, the
default operating point selection implemented by step 101 provide
optimal fuel economy operating points. The default operating points
can be influenced by other factors, such as mitigating NOx
emissions. Fuel economy can be defined in any suitable fashion. In
one embodiment, the fuel economy is measured strictly in terms of
the engine's fuel consumption. In another embodiment, the fuel
economy includes a fuel penalty for exhaust aftertreatment, which
is a function of the engine's NOx production rate.
[0046] FIG. 3 includes a rough plot of optimal fuel economy
operating points as a function of power level for a typical diesel
engine having a CVT. The engine speed for optimal fuel economy is
relatively low for most power demands, as are exhaust flow rates.
At peak power demands, which typically occur only over a small
fraction of a vehicle's operating cycle, engine speeds are much
higher. Exhaust flow rates generally vary in a similar manner to
the engine speed, although their variation is generally wider due
to concomitant variations in factors such as, EGR, turbo-charging,
and exhaust temperature. An aftertreatment system must be deigned
to handle the extremes of exhaust temperature and flow rate that
occur over the course of vehicle operation.
[0047] Exhaust aftertreatment devices have efficiencies that depend
on temperature. FIG. 4 is an exemplary plot of NOx removal
efficiency for a LNT as a function of temperature. If 50%
conversion is considered the limit of effectiveness, the effective
temperature range is from about 220.degree. C. to about 460.degree.
C. for a fresh catalyst and from about 220.degree. C. to about
390.degree. C. for an aged catalyst. Near the range limits, the LNT
effectiveness varies rapidly with temperature. Effectiveness can be
defined in any suitable fashion. Different LNT compositions give
different effective temperature ranges. A LNT can be designed to
operate efficiently at the temperatures occurring during peak power
demand, but such a design may not perform as well at temperatures
that occur at lower power demands.
[0048] Exhaust aftertreatment device efficiencies also depend on
volumetric flow rate of the exhaust. Flow rate also has a
significant effect on LNT performance. The conversion of a LNT will
generally depend on flow rate according to a formula similar to: f
NOx = 1 - e - kV F ( 1 ) ##EQU1## where f.sub.NOx is the fractional
conversion of NOx, k is a reaction rate constant, V is the LNT
volume, and F is the volumetric flow rate through the LNT. When F
is decreased, the effect can be substantial. For example, if
conversion is at 50%, Equation (1) indicates that halving the flow
rate will increase the conversion to 75%.
[0049] Another of the inventors' concepts is to select operating
points to enhance the efficiency of an aftertreatment system over
at least part of a vehicle's operating cycle, for example at points
in the cycle where the aftertreatment system is near a limit of its
effective operating range and would not have a satisfactory
effectiveness at an optimal fuel economy operating point. In this
regard, the method generally involves selecting operating points
that depart from an optimal brake-specific fuel economy operating
point choice. A distinguishing feature of this concept is that it
takes into consideration the effect of exhaust volumetric flow rate
on the aftertreatment system efficiency. Thus, in departing from an
optimal fuel economy operating point, the exhaust temperature can
actually move in a direction of decreasing exhaust aftertreatment
efficiency, provided the effect is more than offset by a decrease
in the exhaust flow rate. This concept can be used to reduce the
design requirements for an aftertreatment system or the frequency
or extent to which an aftertreatment system is regenerated.
[0050] In an exemplary implementation of this concept, step 102
performs a check to determine whether the LNT 14 is in a
satisfactory operating range. A satisfactory operating range can be
defined in any suitable fashion. In one embodiment, it is defined
by upper and lower temperature limits. In another embodiment, it is
defined by an area on a temperature-exhaust flow rate map. In a
further embodiment, it is defined in terms of the LNT 14's ability
to effectively reduce NOx, as determined in any appropriate manner.
The temperature of the LNT 14 can be measured directly through
temperature sensor 21. Optionally, the temperature is determined
from the exhaust temperature on the basis that the temperature of
LNT 14 rises and falls with exhaust temperature. Exhaust
temperature can be measured directly, or determined based on the
engine 9's operating point. If the LNT 14 is not in a satisfactory
operating range, the process proceeds to step 103.
[0051] Step 103 implements the concept of selecting operating
points to improve the performance of the LNT 14. The transmission 8
provides access to a range of operating points for producing a
given power output. Among these operating points, exhaust
temperatures and flow rates may vary significantly. By selecting an
appropriate operating point, the performance of the LNT 14 can
often be improved without undue consequences in terms of fuel
economy, emissions, or engine performance.
[0052] Step 103 involves making a search among operating points
that provide the currently required power level in order to find
one that improves LNT performance. In evaluating each operating
point, at least the effect of exhaust flow rate on performance of
the LNT 14 is considered. Generally the effects of both the exhaust
flow rate and the exhaust temperature are considered. Where the
engine 9 has a turbocharger, the temperature considered in making
this evaluation is preferably an exhaust temperature downstream of
a turbine, as opposed to upstream of the turbine. There is a
significant temperature drop in the exhaust as it passes the
turbine, and the degree of this drop depends on the turbine vane
position, the setting of which can vary from operating
point-to-operating point.
[0053] In addition to effects on LNT performance, various other
constraints and biases may be included in the operating point
selection of step 103. An operating point selection may be biased
based on a fuel penalty measure or emissions rates. For example,
the operating point may be selected to minimize brake-specific fuel
consumption (BSFC) subject to a limit on brake-specific NOx
emission from the aftertreatment system 7. This would not provide
an optimal fuel economy operating point in the usual sense, but
rather would provide a minimal departure from an optimal fuel
economy operating point while enhancing the efficiency of the LNT
14 and the aftertreatment system 7. A brake-specific NOx emission
could consider both NOx production by the engine 9, which varies
with operating point, and the effects on efficiency of the LNT 14
and optionally on the efficiencies of other components of the
aftertreatment system 7. If one or more engine operating parameters
can be varied independently of power level-engine speed selection,
a search for an optimal operating point can include a search and
selection among possible values for these other engine operating
parameters.
[0054] In some cases, the operating point selections can be made in
advance. When operating point selections are made in advance, they
are typically referred to as operating point maps. An operating
point map gives the torque multiplication factor or an equivalent
setting, as a function of power level. The method 100 would use one
operating point map for step 101 and another operating point map
for step 103. Operating point selections will generally be made in
advance if they do not depend on variable conditions or feedback
control. Examples of conditions include ambient air temperature and
engine temperature. Feedback control could be provided based on
actual response of the LNT 14 to changes in operating point.
[0055] An operating point selection typically depends on several
variables not all of which vary linearly or monotonically with
engine speed. Any suitable approach can be used to address this
complexity. One approach uses table look-ups, wherein for a
particular situation such as reformer start-up, denitration, or
desulfation, preferred operating points at various power levels are
determined in advance by simulation and/or experiment. Another
approach relies on storing and retrieving only some data, such as
exhaust temperatures and compositions at various operating points.
This data can then be applied together with sensor data, ambient
air temperature for example, in a model. The model is evaluated at
several operating points to determine which best achieves the
desired result.
[0056] Whether operating points are selected in step 101 or step
103, the process 100 proceeds to step 104 wherein a check is made
whether to regenerate the LNT 14 to remove accumulated NOx. In
general, any suitable method can be used to control the timing of
regeneration in terms of selecting an endpoint for a lean phase
and/or an endpoint for a rich phase. Generally, a control method
will be designed to regenerate the LNT 14 in order to meet an
emission control criteria. The emission control criteria could
include one or more of a limit on NOx concentration in the treated
exhaust and a limited on brake-specific NOx emission rates.
[0057] One control method is based on feedback from the NOx sensor
23. In one example, when the exhaust NOx concentration exceeds a
critical value, regeneration begins and proceeds to a fixed
endpoint. In another example, regeneration begins when a
brake-specific NOx emission rate exceeds a critical value. A
brake-specific NOx emission rate can be based on data from the NOx
sensor 23 normalized with data from the engine 9.
[0058] If the endpoint of the lean phase is determined based on a
NOx emission rate or concentration, the degree of NOx saturation at
the beginning of regeneration will vary with operating state even
though no parameter of regeneration depends on operating state. For
example, if the power level and the exhaust flow rate increase, the
NOx emission rate and concentration will be higher for a given NOx
saturation. If the sulfur poisoning level is higher, this will also
increase the NOx emission at fixed NOx saturation, causing
regeneration based on NOx concentration to begin earlier.
[0059] The endpoint of regeneration can be based on an estimate of
NOx saturation in the LNT 14. Typically, this estimate is based on
reductant slip. Alternatively, regeneration can be calculated to
remove a fixed amount of NOx, whereby if the level of saturation at
the beginning of regeneration varies with operating state, the
level of saturation at the end of regeneration will also vary. A
fixed amount of NOx removal can be estimated, for example, based on
the length of the rich phase or the amount of reductant supplied
during the rich phase.
[0060] A possible difficulty with the foregoing methods is that the
LNT 14 may release a significant quantity of unreduced NOx at the
beginning of each rich phase. This release may cause a temporary
increase in the NOx emission rate, whereby a limitation on
instantaneous NOx emissions may be exceeded. Specifically, in the
simple control described above, regeneration may begin when the
outlet NOx concentration reaches a pre-specified value. At the
beginning of regeneration, there may be a NOx spike that tends to
cause the NOx concentration to rise even higher, exceeding the
pre-specified value and possibly exceeding a regulatory limit on
instantaneous NOx emissions. Avoiding the limit may lead to
over-designing the aftertreatment system 7 and initiating
regenerations in many cases well before they are actually required,
which can increase the overall fuel penalty for aftertreatment.
[0061] FIG. 5 illustrates an exemplary method 200 that addresses
this issue and provides a more systematic basis for selecting
points at which to begin and end a regeneration. The method 200
implements several of the inventors' concepts, including that of
making the beginning and ending points of denitration dependent on
an operating state. The beginning and ending points vary such that
the extent of denitration is less and a higher nitrogen loading
level is tolerated in vehicle operating states that require less
NOx-reducing activity from the LNT 14. By tolerating higher
nitrogen loading levels when less NOx-reducing activity is required
of the LNT 14, the use of reductants is expected to be more
efficient and the fuel penalty for regeneration is expected to be
less.
[0062] The operating state can be defined in any suitable fashion.
In one embodiment, the operating state is an engine operating state
and relates to engine power requirements or is a direct measure of
power demand. The engine operating state may relate to an
instantaneous state, e.g., current power demand, or a more enduring
state, e.g., city driving, highway driving, uphill driving,
downhill driving, accelerating, decelerating, etc. Typically, if
the engine 9 is kept near its peak fuel economy operating points,
its speed, and consequently the volumetric flow rate of the
exhaust, will be relatively low for all but the highest levels of
power demand. As indicated by Equation (1), LNT conversion
efficiency is generally high when the exhaust flow rate is low. The
LNT 14 will be designed to meet the demands of the peak flow rate
operating conditions that typically occur only during a small
fraction of the vehicle's operating cycle. With respect to other
engine operating states, where the exhaust flow rate is generally
lower, the LNT 14 will generally be over-designed.
[0063] One concept is to use this over-design to operate the LNT at
high NOx saturations. Operating at high NOx saturations may involve
allowing the NOx saturation to become high before beginning
regeneration. Operating at high NOx saturation may also involve
terminating a denitration while the level of NOx saturation in the
LNT 14 remains comparatively high.
[0064] FIG. 6 is a theoretical plot of NOx saturation in a LNT
through a series of regeneration cycles. The plot assumes that
exhaust conditions remain constant except for switching between
lean and rich phases. Over a series of lean and rich phases, the
NOx saturation in the LNT falls into a cyclic pattern. The pattern
has a maximum that occurs near the end of each lean phase and a
minimum that occurs at the end of each rich phase. The magnitude of
these minimum and maximum depend on the frequency and duration of
the regeneration cycle. A regeneration method as conceived by the
inventors will set the duration of a regeneration cycle and/or the
period between regenerations to increase the minimum, maximum,
and/or average saturation in a manner that depends on operating
states, whereby greater fuel economy can be achieved when lesser
demands are placed on the exhaust aftertreatment system
performance.
[0065] The exemplary method 200 uses a model with a constant
exhaust condition, whereby NOx saturations are projected to fall
into a pattern such as illustrated by FIG. 6. The model is used to
predict LNT saturations and NOx emissions over future cycles as a
function of regeneration parameters. These projections are used to
calculate optimal values for the regeneration parameters. The model
captures the effects of such factors as operating state, LNT
temperature, and LNT poisoning level. Other approaches could be
taken to capture these dependencies and achieve similar results.
The optimal parameter values calculated using the constant exhaust
condition assumption are later converted to a form in which they
are relatively insensitive to varying exhaust conditions.
[0066] The method 200 begins with step 201 in which an estimate is
made of peak and average power levels for the next lean phase. The
peak power estimate is used to set a minimum to which the
efficiency of the aftertreatment system will be allowed to drop at
any point during the lean/rich cycle. The average power level is
used in projecting exhaust flow rates, temperatures, and NOx
concentrations and to calculate fuel penalty and perform other
calculations relating to the determination of a "best" period
(interval between regenerations) and duration (length of
regenerations) subject to constraints, such as the constraint that
the efficiency of the aftertreatment system never drop below the
minimum determined with reference to the peak power estimate.
[0067] The peak power level can be chosen in any suitable fashion.
Examples of suitable approaches include making the peak equal to
the current power level, making the peak equal to the highest level
realized in a preceding period, such as three minutes, or one of
the forgoing numbers multiplied by a margin of safety, e.g., 1.15.
Another example is to select the peak according to a
characterization of the driving state, e.g., one peak power being
used for hill climbing or sustained acceleration and another, lower
peak power-being used for other conditions. The average power level
can be assumed to be the same as the peak power level or can be
selected in a similar fashion. For example, the average power level
can be assumed to be the average power level during a preceding
period or can be set equal to the current power level.
[0068] The next several steps constitute a process of searching
among possible values for period and duration for values that
optimize an objective function subject to the constraints. An
objective function can be a fuel penalty measure. A fuel penalty
measure generally includes a start-up penalty for heating the
reformer 12 and consuming oxygen stored in the LNT 14, an oxygen
consumption fuel penalty for consuming excess oxygen in the exhaust
during regeneration, a fuel penalty for producing reductant
consumed in reducing NOx adsorbed in the LNT 14, and a reductant
slip fuel penalty for producing reductant that passes through the
LNT 14 unconsumed.
[0069] The search begins in step 202 where initial guesses for the
optimal period and duration are made. Generally, a period and
duration determined from a previous application of the process 200
provides an appropriate guess.
[0070] Step 203 is guessing S.sub.NOx(0),the NOx saturation of the
LNT at the beginning of a lean phase in a cycle such as illustrated
in FIG. 6. This is the beginning of an iterative calculation for
S.sub.NOx(0), which ultimately depends on the currently selected
period and duration and the assumed average power level.
[0071] Step 204 performs an integration to calculate S.sub.NOx(t)
through a projected lean phase, and then a projected rich phase.
The calculation is performed using the assumed power level, and
consequent values of exhaust flow rate, temperature, and NOx
concentration, and any other factors taken into account in
projecting the rate of NOx uptake by the LNT 14. The LNT
temperature can be assumed to be the same as the exhaust
temperature.
[0072] In a preferred embodiment, the NOx uptake model has a
dependency on a measure of reversible sulfur poisoning of the LNT
14. Capturing this dependency is useful in control strategies
related to desulfation and also facilitates using more frequent
and/or extensive denitrations to compensate for sulfur poisoning.
An exemplary model also includes a factor related to irreversible
poisoning. The exemplary model gives the NOx uptake rate,
dS.sub.NOxIdt during the lean phase, as: d S NOx d t = ( 1 - e - k
.function. ( T , S Pois ) .times. ( 1 - ( S SOx + S NOx ) ) .times.
V F ) .times. FC NOx Y Nox ( 2 ) ##EQU2## where the first term in
parenthesis corresponds to Equation (1), C.sub.NOx is the
concentration of NOx in the exhaust, and Y.sub.NOx is the molar NOx
storage capacity of the LNT 14. S.sub.SOx is the fraction of active
sites that are sulfur poisoned and S.sub.Pois is the fraction of
catalyst that is irreversibly poisoned. The adsorption rate is
proportional to the number of unoccupied active sites. The rate
coefficient k of Equation (1) has been shown as a function of the
temperature, T, and irreversible poisoning. An exemplary function k
is plotted in FIG. 4, wherein the degree of aging corresponds to
the degree of poisoning. The detailed functionality can be
determined experimentally with systematic poisoning and
measurements at various temperatures. The initial values of
S.sub.SOx and S.sub.Pois can be zero or the values obtained at the
end of the last vehicle operating cycle. The values are updated in
other steps of the method 100 as described more fully below.
Irreversible catalyst deactivation and permanent loss of adsorption
capacity can be modeled separately, if desired. The details of the
model used are not critical.
[0073] In a preferred embodiment, a model for the NOx removal rate
during the rich phase depends on the NOx saturation. Such a model
can be used to capture the effect of NOx saturation on fuel penalty
and is useful in a strategy of operating at high NOx saturation,
when practical, to reduce fuel penalty. In the exemplary model the
NOx removal rate from the LNT 14 during the rich phase is given by:
d S NOx d t = - ( 1 - e - k red .function. ( T , S Pois ) .times. S
NOx .times. V F ) .times. FC red .alpha. .times. .times. Y Nox ( 3
) ##EQU3## where k.sub.red, which is a function of temperature and
catalyst poisoning, C.sub.red is the reductant concentration, and a
is a coefficient for the stoichiometry of the reduction reaction.
In equation (3), the effective rate of reduction is proportional to
the NOx saturation, S.sub.NOx. Step 204 involves integrating
Equations (2) and (3) through a lean and a rich phase.
[0074] In step 205, S.sub.NOx(tF), the saturation at the end of the
rich phase, is compared to S.sub.NOx(0), the saturation at the
beginning of the lean phase. At a steady state operating condition
such as illustrated by FIG. 6, the numbers will be equal. If these
numbers are significantly different, the model has not converged
and a new estimate for S.sub.NOx(0) is made in step 206 and the
calculation of step 204 is repeated. If the numbers are
approximately the same, the model has converged and the next series
of steps begun, in which the performance of the ammonia-SCR
catalyst 16 is predicted using a similar iterative procedure.
[0075] Step 207 is guessing S.sub.NH3(0), the NH.sub.3 saturation
of the ammonia-SCR 16 at the beginning of a lean phase. Step 208
performs an integration to calculate S.sub.NH3(t) through a
projected lean phase, and then a projected rich phase. The NOx and
ammonia concentrations entering the ammonia-SCR catalyst 16 during
the lean phase can be calculated from the model of the LNT 14 used
in step 204. The NOx concentration entering the ammonia-SCR reactor
16, C'.sub.NOx, is given during the lean phase is given by: C NOx '
= C NOx .times. e - k .function. ( T , S Pois ) .times. ( 1 - ( S
SOx + S NOx ) ) .times. V F ( 4 ) ##EQU4## The NH.sub.3
concentration entering the ammonia-SCR reactor 16, C'.sub.NH3, is
given during the rich phase by given by: C NH .times. .times. 3 ' =
C red .times. x NH .times. .times. 3 .alpha. .times. ( 1 - e - k
red .function. ( T , S Pois ) .times. S NOx .times. V F ) ( 5 )
##EQU5## wherein X.sub.NH3 is the fraction of NOx removed from the
LNT 14 that is reduced to NH.sub.3. As a first approximation, this
fraction may be assumed constant. X.sub.NH3 may actually depend on
such factors as temperature, NOx saturation in the LNT 14, and
degree of poisoning and a more sophisticated model may take into
account one or more of these dependencies.
[0076] An exemplary model for ammonia consumption by reactions with
NOx in the ammonia-SCR catalyst 15 is: d S NH .times. .times. 3 d t
= - ( 1 - e - k SCR .function. ( T ) .times. S NH .times. .times. 3
.times. V SCR F ) .times. FC NOx ' .alpha. SCR .times. Y NH .times.
.times. 3 ( 6 ) ##EQU6## where k.sub.SCR is a kinetic constant for
reaction in the ammonia-SCR catalyst 16, V.sub.SCR is the volume of
the SCR catalyst, .alpha..sub.SCR is a stoichiometric constant, and
Y.sub.NH3 is the ammonia storage capacity of the ammonia-SCR
catalyst 16. An exemplary model for ammonia storage by the
ammonia-SCR catalyst 15 is: d S NH .times. .times. 3 d t = ( 1 - e
- k NH .times. .times. 3 .function. ( T ) .times. ( 1 - S NH
.times. .times. 3 ) .times. V SCR F ) .times. FC NH .times. .times.
3 ' Y NH .times. .times. 3 ( 7 ) ##EQU7## where k.sub.NH3 is a
kinetic constant for ammonia adsorption. Equations (6) and (7) are
integrated in step 208. Step 209 is checking for convergence. Step
210 revises S.sub.NH3(0) if convergence has not yet been
achieved.
[0077] After convergence, step 211 evaluates the objective function
being optimized.
[0078] Typically, the objective function will primarily indicate
fuel penalty, although weight can be given to other factors.
Constraints are also evaluated in step 211. In this example, one of
the constraints relates to limiting NOx emission at peak power. To
test whether this requirement is met, an instantaneous NOx emission
is computed assuming a peak power condition occurs at the end of a
lean phase.
[0079] The exemplary models used to calculate S.sub.NOx and
S.sub.NH3 and to evaluate the objective function did not include a
NOx spike at the beginning of regeneration. The exemplary model can
be modified to account for the NOx spike, if desired. Whether or
not a NOx spike model is used to calculate S.sub.NOx and S.sub.NH3
and to evaluate the objective function, it is preferred that a
model for a NOx spike be used in calculating whether a peak
instantaneous NOx emission constraint is violated. Moreover,
whereas S.sub.NOx, S.sub.NH3, and the objective function are
preferably computed using best estimates, including a best estimate
for the size of the NOx spike if one is used, the NOx emission
constraint is preferably calculated using a worst case scenario,
i.e., a NOx spike of the largest total volume and occurring in the
quickest burst as experiments indicate could realistically occur.
Thus, a peak NOx emission constraint is preferably checked assuming
a peak power condition occurs while S.sub.NOx and S.sub.NH3 are at
the values predicted to occur at the end of the lean phase and also
assuming that a regeneration with a comparatively large NOx spike
begins shortly thereafter.
[0080] Any suitable numerical algorithm can be used to find the
period and duration that optimize the object function subject to
the constraints. Numerous such algorithms are widely documented and
readily ascertainable. Computer software that implements these
algorithms is commercially available. Examples of numerical
algorithms in this genre include steepest descent, Newton's method,
and quasi-Newton methods. Most suitable algorithms involve
numerically estimating derivatives. Accordingly, step 212 is
provided to calculate these derivatives. Numerically calculating a
derivative involves making a small perturbation in the variable and
observing its effect on a result.
[0081] Step 213 is testing for convergence of the numerical method.
If convergence has not been achieved, new selections for period and
duration are made in step 214 and the calculations are repeated.
Once convergence has been reached, the process advances to step 215
where certain transformation are made to the calculated period and
duration.
[0082] One transformation is to express the period and duration on
bases other than time, whereby the control method can adapt to
short term variations in operating conditions. For example, the
endpoint of a lean phase is preferably express in terms of an
amount of NOx supplied to the LNT 14, rather than a fixed period of
time. Using the conditions from step 204, the period selected at
the end of step 213 can be transformed into a target total engine
out NOx between regenerations. The transformation involves
multiplying the time-based period by F and C.sub.NOx. The target
can be compared to actual NOx supplied to the LNT 14 as estimated
from the engine 9's operating points or calculated using data from
a NOx sensor, such as the NOx sensor 22. An ideal amount of NOx to
the LNT 14 between regenerations is expected to be much less
variable than an ideal amount of time between regenerations.
[0083] Instead or time duration for regeneration, total reductant
supplied to the LNT 14 is preferably used to determine the endpoint
of regeneration. The amount to target is determined from the
time-based duration multiplied by F and C.sub.red. The target
amount can be compared to total reductant production as determined
from data used to manage the reformer 12 and the fuel injector 11.
An ideal amount of reductant to supply to the LNT 14 over the
course of a regeneration is expected to be much less variable than
an ideal time for a regeneration.
[0084] In using the parameters period and duration, it should also
be taken into account that the starting values of S.sub.NOx and
S.sub.NH3 are not the same as S.sub.NOX(0) and S.sub.NH3(0)
determined by the model. Preferably, estimates of actual S.sub.NOx
and S.sub.NH3 are continuously maintained. These estimated can be
used to correct the endpoint for the lean phase. For example, if
the endpoint for the lean phase is expressed in terms of amount of
NOx to the LNT 14 between regenerations, the endpoint can be
corrected for an amount of NOx that would take the LNT 14 from the
estimate of actual S.sub.NOx to S.sub.NOx(0) or vice versa.
Alternatively, the endpoint of the lean phase can be taken as
reached when the estimated value of S.sub.NOx reaches the value
calculated for the end of the lean phase. The endpoint of
regeneration can be taken as the point where the estimated value of
actual S.sub.NOx reaches the value calculated for the end of the
rich phase. Estimates for actual S.sub.NOx can be maintained by
integrating Equations 2 and 3 through lean and rich phases using
actual exhaust conditions.
[0085] It should be appreciated that the models and numerical
methods described in connection with the process 200 are exemplary
and that many other models and methods can be devised that operate
in accordance with the various concepts disclosed herein.
[0086] The regeneration parameters may be determined at any
suitable point in the process 100. In one example, they are
initially determined prior to step 101 and subsequently determined
in step 104. Regardless of where the parameters are determined, the
parameter relating to determining an endpoint for a lean phase is
applied in step 104. Step 104 is determining whether denitration is
required. FIG. 7 provides an exemplary process 300 that can be
applied at step 104.
[0087] The process 300 implements several concepts. One concept is
to begin regeneration based on either of two criteria being
satisfied. One criterion relates to a predetermined endpoint and
may be characterized in any appropriate terms, including for
example a fixed time interval, an amount of NOx produced by the
engine 9 or supplied to the LNT 14, or a target saturation of the
LNT 14. The other criterion relates to feedback control and
involves checking whether NOx emissions have exceeded a
pre-specified level.
[0088] Another concept implemented by the process 300 is to revise
an estimate of NOx saturation in the LNT 14 based on a NOx emission
level measured at the end of a lean phase. In one embodiment, the
estimate of NOx saturation is increased or decreased in order to
reconcile a difference between a predicted NOx concentration
downstream of the LNT 14 at the end of a lean phase and an
estimated value for that concentration.
[0089] Another concept is to adjust the endpoint for a lean or rich
phase in response to a change in operating state. In one example,
the change in operating state corresponds to an unanticipated
increase in power demand. If the operating state changes, new lean
and rich phase endpoints can be calculated, for example by the
process 200 using new values for peak and average power. In one
embodiment, a regeneration is initiated in response to an increase
of power demand.
[0090] The process 300 begins with step 301, which is determining
whether a predetermined endpoint for the lean phase has been
reached. This can involve a measure of time; a determination of the
amount of NOx that has been supplied to the LNT 14, or the NOx
saturation in the LNT 14, as previously explained in connection
with the process 200.
[0091] If the predetermined endpoint has not been reached, another
check is made in step 302. Step 302 is determining whether a NOx
concentration in the treated exhaust is too high. In this example,
the NOx concentration is the NOx concentration measured by the
sensor 23. The NOx concentration in the treated exhaust may be
considered too high, for example, if it meets or exceeds a value
predicted for the end of the lean phase or a somewhat higher value.
If the NOx concentration is not too high, a third check is
performed in step 303. This check is whether a vehicle operating
state has changed. For example, step 303 may check whether a peak
power has been exceeded. The peak power can be the same peak power
used in the process 200, or a different peak power. In any event,
if the vehicle operating state has changed, the endpoints for the
lean and rich phases are recalculated in step 304 and the various
checks repeated using the new values.
[0092] If step 301 determines the predetermined endpoint has been
reached, a check is made in step 305 whether the NOx emission
measured downstream of the LNT 14 was above expectation. The sensor
23 can be used for this purpose, although a sensor immediately
downstream of the LNT 14 might provide more accurate data for
revising the estimate of NOx saturation. If the sensor reading is
above the predicted value, the process assumes that the S.sub.NOx
is higher than expected. The predicted value is a value predicted
by the model of the process 200 using the saturations predicted for
the end of the lean phase and current exhaust conditions, such as
current temperature, flow rate, and engine-out NOx
concentration.
[0093] Step 306 increases the current estimate of S.sub.NOx in a
manner consistent with the NOx concentration reading. Step 306 can
also be reached through step 302 if the NOx emission level exceeds
a maximum before the predetermined end of the lean phase. One
example of a procedure for increasing S.sub.NOx is to use the model
of process 200 to calculate a value of S.sub.NOx, that would give
the measured NOx concentration. The current estimate for S.sub.NOx
can be revised to equal that value.
[0094] Step 307 checks whether the measured NOx concentration is
below expectation. If it is, the estimate for S.sub.NOx can be
decreased in step 308 much as it is increased in step 306. Whether
or not S.sub.NOx is increased, decreased, or left unchanged, a
regeneration is initiated if the criteria of step 301 or step 302
is met.
[0095] Before denitration actually begins, the fuel reformer 12 is
started in step 105. Starting the reformer 12 generally involves
heating the reformer 12. The reformer 12 can be heated in any
suitable fashion. If the reformer is warm enough, it can be heated
by injecting diesel fuel. In some operating conditions, however,
the exhaust may make the reformer 12 too cold and in any event the
reformer can be heated more quickly if it is warmer.
[0096] Another of the inventors' concepts is a process for starting
a reformer that begins by using a transmission to select operating
points for an engine to warm the exhaust and thus the reformer.
This process generally takes place after the engine and the rest of
the exhaust system have completed their initial warm-up, e.g. ten
or more minutes after the engine has been running continuously. The
operating point selections designed to warm the reformer generally
commence in response to an electronically generated command to
start the reformer, which may be the result of a command to
regenerate an aftertreatment device.
[0097] FIG. 8 is a flow chart for a process 400 that implements the
inventors' concept for starting the reformer 12. The process 400
begins in step 401, which is using the transmission 8 to select
operating points for the engine 9 to produce a hotter exhaust. As
illustrated by FIG. 9, there is generally a range of exhaust
temperatures among operating points producing a given power level.
By selecting an appropriate operating point, an exhaust temperature
can be increased without affecting the engine power output.
Depending on various factors including the staring point and the
power level, in some cases the reformer 12 can be heated by, for
example, at least about 40 .degree.C. by the shift in operating
point section. In a narrower group of cases, the reformer 12 can be
heated by at least about 80.degree. C. through the operation of the
transmission 8. The operating point selection is preferably based
on the exhaust temperature downstream of the turbine, as opposed to
upstream of the turbine. As in the case of selecting an operating
point for enhancing the efficiency of the LNT 14, various
constraints and biases may be included in an operating point
selection designed to increase the exhaust temperature to
facilitate reformer start-up. The operating point can be selected
dynamically, or determined in advance and encoded in an operating
point map.
[0098] In step 402, the engine 9 is operated for a period of time
with operating points selected to increase the exhaust temperature.
The exact operating point may vary over the period in response to
changes in engine power demand. High exhaust temperature operation
is typically maintained for a period from about 0.5 to about 10
seconds preferably from about 1 to about 5 seconds. During this
time, the reformer 12 warms. The thermal mass of the reformer 12
generally limits the warm-up rate. Preferably, the reformer 12 has
nearly the minimal thermal mass determined by its functional
requirements.
[0099] After the reformer 12 has been warmed by the exhaust, it may
still be below a temperature at which it can effectively operate
with a rich fuel-oxygen mixture (lambda less than 1.0). If so, in
step 403, the fuel injector 11 is actuated to initiate fuel
injection at a rate that leaves the exhaust lean (lambda greater
than 1.0). The fuel burns in the reformer 12, further heating the
reformer 12. A typical objective is to heat the reformer to at
least about 500.degree. C., preferably at least about 600.degree.
C., and still more preferably at least about 650.degree. C.
[0100] Once the reformer is heated to a satisfactory degree, the
feed to the reformer is made rich in step 106, whereupon the
reformer 12 begins to produce reformate and denitration or
desulfation of the LNT 14 begins. The feed can be made rich by any
suitable combination of fuel injection, engine intake air
throttling (where provided for), and exhaust gas recirculation
(EGR). The operating point of the engine may also be adjusted to
assist in making the feed rich, as discussed more fully below.
[0101] During rich operation, substantially all the oxygen present
in the exhaust is consumed while producing reformate. Regardless of
the actual sequence of reactions, the operation of the reformer can
be modeled by the following
0.684CH.sub.1.85+O.sub.2.fwdarw.0.684CO.sub.2+0.632H.sub.2O (8)
0.316CH.sub.1.85+0.316H.sub.2O0.316CO+0.608H.sub.2 (9)
0.316CO+0.316H.sub.2O.fwdarw.0.316CO.sub.2+0.316H.sub.2 (10)
wherein CH.sub.1.85 represents diesel fuel with a 1.85 ratio
between carbon and hydrogen. Equation (8) is exothermic complete
combustion which consumes all the excess oxygen in the exhaust.
Equation (9) is endothermic steam reforming. Equation (10) is the
water gas shift reaction, which is relatively thermal neutral and
is not of great importance in the present disclosure, as both CO
and H.sub.2 are effective for regeneration.
[0102] In an ideal situation, the extents of Equations (8) and (9)
are balanced whereby the reformer temperature remains constant, or
that failing, Equation (9) dominates and the reformer cools while
large quantities of reformate are produced. When the oxygen
concentration is relatively low, e.g., 5-10% or less, depending on
the reformer, this ideal can be approached. At higher oxygen
concentrations, however, it has generally been observed that the
endothermic reforming rate cannot be matched to the combustion
rate. The result is that reformate production efficiency falls off
and the reformer 12 heats uncontrollably. Eventually, the reformer
must be shutdown to prevent it, or downstream devices, from over
heating. Thus, the reformer 12 tends to be unstable at high oxygen
concentrations.
[0103] Addressing these issues, one of the inventors' concepts is
to use a transmission to shift an engine operating point in order
to reduce an exhaust oxygen concentration during operation of a
fuel reformer drawing oxygen from the exhaust. The shift in
operating point preferably brings the oxygen into a range wherein
the reformer can be operated to consume excess oxygen and produce
reformate while remaining at a constant temperature. The preferred
oxygen concentration will depend on the reformer catalyst and other
factors, however, in general the preferred oxygen concentration
will be less than about 10%, more typically less than about 8%.
[0104] FIG. 10 is an exemplary plot of air-fuel ratio as a function
of operating point. Several curves of constant power level
operation are shown. While the exhaust oxygen concentration also
depends on other factors, exhaust oxygen concentration is a strong
function of engine air-fuel ratio. FIG. 10 illustrates that while
the exact relationship between oxygen concentration and operating
point can be complex, it is general possible to influence the
oxygen concentration significantly through selection of engine
operating points.
[0105] The inventors recognize that reducing the oxygen
concentration during regeneration can be beneficial even when a
reformer is not used. In general, if regeneration of an
aftertreatment device requires a reducing atmosphere, fuel or
reductant must be consumed removing excess oxygen. Accordingly, one
of the inventors' concepts is to select engine operating points to
reduce the oxygen concentration in the engine's exhaust during
regeneration of an exhaust aftertreatment device. This concept is
generally useful for reducing the fuel penalty associated with the
regeneration of an aftertreatment device, such as an LNT.
[0106] While reducing the oxygen concentration in the exhaust can
have other benefits, the main objective is often to reduce fuel
penalty. The oxygen concentration in the exhaust can be a primary
factor in determining the fuel penalty for a regeneration, but
other factors can also be important. For example, the exhaust flow
rate may also be important due to its effect on reductant slip. The
break specific fuel consumption of the engine will also change with
operating point and may outweigh the benefits of reducing the
oxygen concentration beyond a certain degree.
[0107] Therefore, another of the inventors' concepts is to use a
transmission to select operating points for an engine during
regeneration of an aftertreatment device in order to reduce a fuel
penalty. The term regeneration is generally inclusive of
denitration and desulfation. In general, the fuel penalty will
include a contribution associated with consuming excess oxygen in
the exhaust. In one embodiment, the fuel penalty includes a
contribution relating to the cost of operating the engine 9 at
operating points apart from its minimum brake-specific fuel
consumption operating point. In another embodiment, the fuel
penalty includes a contribution relating to reductant slip during
regeneration. A preferred method of selecting operating points
chooses operating points that maximizes an objective function,
wherein the objective function is proportional to the time that
will be required to complete the regeneration under current
conditions times the sum of the oxygen consumption fuel penalty
rate and the fuel penalty rate associated with operating the engine
at a BSFC above the minimum BSFC for the current power level.
[0108] Regardless of the exhaust composition and flow rate, it is
expected that the incremental fuel penalty will increase over the
course of regeneration. The fuel penalty increases due to
decreasing concentration of the contaminant being removed. As the
contaminant concentration decreases, a progressively greater
fraction of the reductant is lost to reductant slip. This
phenomenon can be used to detect the endpoint of a regeneration.
The extent of regeneration is indicated by the reductant slip,
which can be measured, for example, by a lambda sensor. In
interpreting the sensor reading, a calculation is preferably made
to take into account the effects of other conditions, such as the
LNT temperature, and including at least the current exhaust flow
rate.
[0109] An ideal endpoint for a LNT regeneration involves a tradeoff
between such factors as the required LNT efficiency, penalties that
increase with regeneration frequency, such as a startup fuel
penalty, and the variation of fuel penalty with extent of
regeneration. The inventors recognize that the point of optimal
trade-off varies with the operating state of the engine and with
the degree of poisoning of the LNT. Accordingly, one of the
inventors' concepts is to vary the endpoint of a denitration or a
desulfation whereby the regeneration is more extensive when the
engine is in an operating state requiring higher LNT efficiency.
Another concept is to regenerate to a further extent when the LNT
is poisoned to a greater extent by factors not affect by the
regeneration. With respect to desulfation, the factors can be
irreversible poisoning or other irreversible loss of activity. With
respect to denitration, the factors can be to irreversible loss of
activity or reversible sulfur poisoning.
[0110] The process 200 provides an example where the extent of
denitration is determined in a manner that depends on both extents
of poisoning and on engine operating mode. The extent determined in
process 200 is implemented in step 106, where the denitration is
performed.
[0111] After denitration, step 107 checks whether desulfation is
required. FIG. 11 is a flow chart of an exemplary method 600, which
may be implemented in step 107 for determining whether to
desulfate. The method 600 begins with step 601, determining whether
the NOx emission level at the end of the previous lean phase was
above expectation, expectation being a value predicted by the model
used in the method 200. The prediction is made before correcting
S.sub.NOx to account for the difference, as described in method
300. If the NOx emission level was above expectation, the estimate
of S.sub.SOx is increased in step 602. The amount of the increase
can be an amount that would explain the difference between the
observed NOx emission and the predicted NOx emission, however it is
preferred that the increase be some fraction of that amount, e.g.,
25%, whereby the estimate of S.sub.SOx changes slowly.
[0112] Step 603 uses the pattern of change in S.sub.SOx to evaluate
whether the LNT 14 has become irreversibly poisoned. In particular,
if S.sub.SOx increases rapidly following a desulfation, or
following each of the last several desulfations, this can be taken
as an indication that desulfation is not as effective as
anticipated, which in turn can be treated as an indication of
irreversible loss of activity. If the pattern indicates the LNT 14
has become irreversibly poisoned, the estimate for S.sub.POIS is
increased by an appropriate amount in step 604.
[0113] Step 605 uses process 200 to obtain new values for the
regeneration parameters, period and duration. Step 606 determines
whether parameters meeting the constraints were found in step 605,
and if they were found whether they give satisfactory performance
for the aftertreatment system 7 in terms of an appropriate
performance measure, such as fuel penalty or period between
denitrations. If the constraints could not be met or the
performance of the aftertreatment system 7 using the parameters
determined in step 605 is not acceptable, then a desulfation is
commenced in step 607. Because the parameter determination in step
605 depends on the engine operating state and the degree or
irreversible poisoning of the LNT 14, the method 600 is an example
of a method in which the time to commence desulfation depends on
engine operating state and the degree or irreversible
poisoning.
[0114] The extent of desulfation also preferably depends on engine
operating state and/or the degree or irreversible poisoning. FIG.
12 is a flow chart of a method 500 in which the extent of
desulfation depends on the engine operating state. The method 500
begins with step 501, estimating peak and average power levels for
the period following the desulfation. This is similar to step 201
of the method 200, except in this case the estimates relate to a
more extended period. The higher the peak and average power levels,
the greater the degree of desulfaltion that will be required.
[0115] The next few steps provide an iterative method for
determining the required degree of desulfation. The required degree
of desulfation will provide the aftertreatment system 7 with
sufficient activity to meet an emission control target under the
current operating condition. Step 502 is guessing the target sulfur
saturation level. Step 503 is calculating optimal denitration
parameters assuming the target sulfur saturation. Step 504
determines whether the aftertreatment system 7 provides the minimum
required performance at the target sulfur saturation assuming the
optimal NOx regeneration parameters are used. Step 505 is used to
revise the target sulfur saturation until convergence is
reached.
[0116] Step 506 makes an adjustment to the target sulfur saturation
to provide an interval between desulfations. The adjustment could
involve a fixed additional amount of sulfur removal or a target
time between desulfation. A target time between desulfations could
be, for example, about 10 or about 30 hours. The adjustment to the
target sulfur saturation is made by estimating the amount of
additional sulfur removal that would give the target interval. This
involves estimating the rate at which the LNT 12 accumulates sulfur
under current conditions.
[0117] Step 507 is carrying out the desulfation. The progress of
desulfation can be determined in any suitable fashion. In one
embodiment, a model for the desulfation reactions is used to
estimate how much desulfation has occurred. In another embodiment,
a sensor is used to monitor the concentration of SO.sub.2
downstream of the oxidation catalyst 17. The measure concentrations
can be integrated to determine the total amount of sulfur
removal.
[0118] As discussed previously, the foregoing methods are most
effective in extending LNT life and improving fuel economy when
applied to an engine that can operate efficiently in a low speed
range through most of its operating cycle. This is true of most
diesel engines, although a CVT is generally needed to track on of
the optimal fuel economy curves plotted in FIG. 3.
[0119] Efficiency can be further improved if the engine can be
efficiently operated in a narrow and preferably low speed range
throughout its power range, as illustrated in FIG. 3. An engine
that lends itself to such operation can be produced by a typical
engine manufacturer given the objective of narrow speed range
operation. The manufacturer has the ability to modify parameters
such as the turbocharger operation, the fuel system
characteristics, the cam shaft shape, and the electronic controls
in order to make the engine's peak efficiency occur within a narrow
speed range throughout its operating power range. A CVT could then
be used to always keep the engine at operating points within a
narrow speed range and near the peak fuel economy curve, excepting
perhaps for special circumstances such as provided by some of the
foregoing methods. Preferably, the engine has a peak fuel economy
curve with a speed range of about 300 RPM or less, more preferably
about 200 RPM or less, and still more preferably about 100 RPM or
less. A preferred power generation system includes a narrow speed
range engine, a CVT, and an aftertreatment system, which may be any
conventional aftertreatment system or the aftertreatment system 7.
For any aftertreatment system design, the narrow speed range, and
the consequently narrowly varying exhaust flow rate, reduce the
system requirements and simplify reductant dosing.
[0120] In one embodiment, a narrow speed range power generation
system is used to reduce the catalyst loading, especially of a
precious metal catalyst, in the aftertreatment system. Preferably,
the power generation system 5 comprises an aftertreatment system 7
comprising a component with at least about 10% less catalyst than a
comparable component in a comparable aftertreatment system of a
power generation system using a conventional engine and
transmission and having the same brake-specific NOx emission rate.
More preferably, the component comprises at least about 20% less
catalyst, and still more preferably at least about 30% less
catalyst.
[0121] While the engine 9 is preferably a compression ignition
diesel engine, the various concepts of the invention are applicable
to power generation systems with lean-burn gasoline engines or any
other type of engine that produces an oxygen rich, NOx-containing
exhaust. For purposes of the present disclosure, NOx consists of NO
and NO.sub.2.
[0122] The transmission 8 can be any suitable type of automatic
transmission. The transmission 8 can be a conventional transmission
such as a counter-shaft type mechanical transmission, but is
preferably a CVT. A CVT can provide a much larger selection of
operating points than a conventional transmission and generally
also provides a broader range of torque multipliers. In general, a
CVT will also avoid or minimize interruptions in power transmission
during shifting. Examples of CVT systems include hydrostatic
transmissions; rolling contact traction drives; overrunning clutch
designs; electrics; multispeed gear boxes with slipping clutches;
and V-belt traction drives. A CVT may involve power splitting and
may also include a multi-step transmission.
[0123] A preferred CVT provides a wide range of torque
multiplication ratios, reduces the need for shifting in comparison
to a conventional transmission, and subjects the CVT to only a
fraction of the peak torque levels produced by the engine. This can
be achieved using a step-down gear set to reduce the torque passing
through the CVT. Torque from the CVT passes through a step-up gear
set that restores the torque. The CVT is further protected by
splitting the torque from the engine, and recombining the torque in
a planetary gear set. The planetary gear set mixes or combines a
direct torque element transmitted from the engine through a stepped
automatic transmission with a torque element from a CVT, such as a
band-type CVT. The combination provides an overall CVT in which
only a portion of the torque passes through the band-type CVT.
[0124] The fuel injector 11 can be of any suitable type. It can
inject the fuel co-current, cross-current, or counter-current to
the exhaust flow. Preferably, it provides the fuel in an atomized
or vaporized spray. The fuel may be injected at the pressure
provided by a fuel pump for the engine 9. Preferably, however, the
fuel passes through a pressure intensifier operating on hydraulic
principles to at least double the fuel pressure from that provided
by the fuel pump to provide the fuel at a pressure of at least
about 4 bar.
[0125] The lean-NOx catalyst 15 can be either an HC-SCR catalyst, a
CO-SCR catalyst, or a H.sub.2-SCR catalyst. Examples of HC-SCR
catalysts include transitional metals loaded on refractory oxides
or exchanged into zeolites. Examples of transition metals include
copper, chromium, iron, cobalt, nickel, cadmium, silver, gold,
iridium, platinum and manganese, and mixtures thereof. Exemplary of
refractory oxides include alumina, zirconia, silica-alumina, and
titania. Useful zeolites include ZSM-5, Y zeolites, Mordenite, and
Ferrerite. Preferred zeolites have Si:Al ratios greater than about
5, optionally greater than about 20. Specific examples of
zeolite-based HC-SCR catalysts include Cu-ZSM-5, Fe-ZSM-5, and
Co-ZSM-5. A CeO.sub.2 coating may reduce water and SO.sub.2
deactivation of these catalysts. Cu/ZSM-5 is effective in the
temperature range from about 300 to about 450.degree. C. Specific
examples of refractory oxide-based catalysts include
alumina-supported silver. Two or more catalysts can be combined to
extend the effective temperature window.
[0126] Where a hydrocarbon-storing function is desired, zeolites
can be effective. U.S. Pat. No. 6,202,407 describes HC-SCR
catalysts that have a hydrocarbon storing function. The catalysts
are amphoteric metal oxides. The metal oxides are amphoteric in the
sense of showing reactivity with both acids and bases. Specific
examples include gamma-alumina, Ga.sub.2O.sub.3, and ZrO.sub.2.
Precious metals are optional. Where precious metals are used, the
less expensive precious metals such as Cu, Ni, or Sn can be used
instead of Pt, Pd, or Rh.
[0127] In the present disclosure, the term hydrocarbon is inclusive
of all species consisting essentially of hydrogen and carbon atoms,
however, a HC-SCR catalyst does not need to show activity with
respect to every hydrocarbon molecule. For example, some HC-SCR
catalysts will be better adapted to utilizing short-chain
hydrocarbons and HC-SCR catalysts in general are not expected to
show substantial activity with respect to CH.sub.4.
[0128] Examples of CO-SCR catalysts include precious metals on
refractory oxide supports. Specific examples include Rh on a
CeO.sub.2--ZrO.sub.2 support and Cu and/or Fe ZrO.sub.2
support.
[0129] Examples of H.sub.2-SCR catalysts also include precious
metals on refractory oxide supports. Specific examples include Pt
supported on mixed LaMnO.sub.3, CeO.sub.2, and MnO.sub.2, Pt
supported on mixed ZiO.sub.2 and TiO.sub.2, Ru supported on MgO,
and Ru supported on Al.sub.2O.sub.3.
[0130] The lean-NOx catalyst 15 can be positioned differently from
illustrated in FIG. 1. In one embodiment, the lean NOx catalyst 15
is upstream of the fuel injector 11. In another embodiment the lean
NOx catalyst is downstream of the reformer 12, whereby the lean NOx
catalyst 15 can use reformer products as reductants. In a further
embodiment, the lean NOx catalyst 15 is well downstream of the LNT
14, whereby the lean NOx catalyst 15 can be protected from high
temperatures associated with desulfating the LNT 14.
[0131] A fuel reformer is a device that converts heavier fuels into
lighter compounds without fully combusting the fuel. A fuel
reformer can be a catalytic reformer or a plasma reformer.
Preferably, the reformer 12 is a partial oxidation catalytic
reformer. A partial oxidation catalytic reformer comprises a
reformer catalyst. Examples of reformer catalysts include precious
metals, such as Pt, Pd, or Ru, and oxides of Al, Mg, and Ni, the
later group being typically combined with one or more of CaO,
K.sub.2O, and a rare earth metal such as Ce to increase activity. A
reformer is preferably small in size as compared to an oxidation
catalyst or a three-way catalyst designed to perform its primary
functions at temperatures below 500.degree. C. A partial oxidation
catalytic reformer is generally operative at temperatures from
about 600 to about 1100.degree. C.
[0132] The NOx adsorber-catalyst 14 can comprise any suitable
NOx-adsorbing material. Examples of NOx adsorbing materials include
oxides, carbonates, and hydroxides of alkaline earth metals such as
Mg, Ca, Sr, and Be or alkali metals such as K or Ce. Further
examples of NOx-adsorbing materials include molecular sieves, such
as zeolites, alumina, silica, and activated carbon. Still further
examples include metal phosphates, such as phosphates of titanium
and zirconium. Generally, the NOx-adsorbing material is an alkaline
earth oxide. The adsorbant is typically combined with a binder and
either formed into a self-supporting structure or applied as a
coating over an inert substrate.
[0133] The LNT 14 also comprises a catalyst for the reduction of
NOx in a reducing environment. The catalyst can be, for example,
one or more precious metals, such as Au, Ag, and Cu, group VIII
metals, such as Pt, Pd, Ru, Ni, and Co, Cr, Mo, or K. A typical
catalyst includes Pt and Rh, although it may be desirable to reduce
or eliminate the Rh to favor the production of NH.sub.3 over
N.sub.2. Precious metal catalysts also facilitate the adsorbant
function of alkaline earth oxide adsorbers.
[0134] Adsorbant and catalysts according to the present invention
are generally adapted for use in vehicle exhaust systems. Vehicle
exhaust systems create restriction on weight, dimensions, and
durability. For example, a NOx adsorbant bed for a vehicle exhaust
systems must be reasonably resistant to degradation under the
vibrations encountered during vehicle operation.
[0135] An adsorbant bed or catalyst brick can have any suitable
structure. Examples of suitable structures may include monoliths,
packed beds, and layered screening. A packed bed is preferably
formed into a cohesive mass by sintering the particles or adhering
them with a binder. When the bed has an adsorbant function,
preferably any thick walls, large particles, or thick coatings have
a macro-porous structure facilitating access to micro-pores where
adsorption occurs. A macro-porous structure can be developed by
forming the walls, particles, or coatings from small particles of
adsorbant sintered together or held together with a binder.
[0136] The ammonia-SCR catalyst 16 is a catalyst effective to
catalyze reactions between NOx and NH.sub.3 to reduce NOx to
N.sub.2 in lean exhaust. Examples of SCR catalysts include oxides
of metals such as Cu, Zn, V, Cr, Al, Ti, Mn, Co, Fe, Ni, Pd, Pt,
Rh, Rd, Mo, W, and Ce, zeolites, such as ZSM-5 or ZSM-11,
substituted with metal ions such as cations of Cu, Co, Ag, Zn, or
Pt, and activated carbon. Preferably, the ammonia-SCR catalyst 16
is designed to tolerate temperatures required to desulfate the LNT
14.
[0137] The particulate filter 13 can have any suitable structure.
Examples of suitable structures include monolithic wall flow
filters, which are typically made from ceramics, especially
cordierite or SiC, blocks of ceramic foams, monolith-like
structures of porous sintered metals or metal-foams, and wound,
knit, or braided structures of temperature resistant fibers, such
as ceramic or metallic fibers. Typical pore sizes for the filter
elements are about 10 .mu.m or less. Optionally, one or more of the
LNT 14, the lean-NOx catalyst 15, or the ammonia-SCR catalyst 16 is
integrated as a coating on the DPF 13.
[0138] The DPF 13 is regenerated to remove accumulated soot. The
DPF 13 can be of the type that is regenerated continuously or
intermittently. For intermittent regeneration, the DFP 13 is
heated, using a reformer 12 for example. The DPF 13 is heated to a
temperature at which accumulated soot combusts with O.sub.2. This
temperature can be lowered by providing the DPF 13 with a suitable
catalyst. After the DPF 13 is heated, soot is combusted in an
oxygen rich environment.
[0139] For continuous regeneration, the DPF 13 may be provided with
a catalyst that promotes combustion of soot by both NO.sub.2 and
O.sub.2. Examples of catalysts that promote the oxidation of soot
by both NO.sub.2 and O.sub.2 include oxides of Ce, Zr, La, Y, and
Nd. To completely eliminate the need for intermittent regeneration,
it may be necessary to provide an additional oxidation catalyst to
promote the oxidation of NO to NO.sub.2 and thereby provide
sufficient NO.sub.2 to combust soot as quickly as it accumulates.
Where regeneration is continuous, the DPF 13 is suitably placed
upstream of the reformer 12. Where the DPF 13 is not continuously
regenerated, it is generally positioned as illustrated in FIG. 1 or
a point downstream. An advantage of the position illustrated in
FIG. 1 is that the DPF 13 buffers the temperature between the
reformer 12 and the LNT 14.
[0140] The clean-up catalyst 17 is preferably functional to oxidize
unburned hydrocarbons from the engine 9, unused reductants, and any
H.sub.2S released from the NOx adsorber-catalyst 13 and not
oxidized by the ammonia-SCR catalyst 15. Any suitable oxidation
catalyst can be used. A typical oxidation catalyst is a precious
metal, such as platinum. To allow the clean-up catalyst 17 to
function under rich conditions, the catalyst may include an
oxygen-storing component, such as ceria. Removal of H.sub.2S, where
required, may be facilitated by one or more additional components
such as NiO, Fe.sub.2O3, MnO.sub.2, CoO, and CrO.sub.2.
[0141] The invention as delineated by the following claims has been
shown and/or described in terms of certain concepts, components,
and features. While a particular component or feature may have been
disclosed herein with respect to only one of several concepts or
examples or in both broad and narrow terms, the components or
features in their broad or narrow conceptions may be combined with
one or more other components or features in their broad or narrow
conceptions wherein such a combination would be recognized as
logical by one of ordinary skill in the art. Also, this one
specification may describe more than one invention and the
following claims do not necessarily encompass every concept,
aspect, embodiment, or example described herein.
* * * * *