US20020134075A1

US20020134075A1 - Method for operation of a nox storage catalyst in internal combustion engines

Info

Publication number: US20020134075A1
Application number: US09/914,468
Authority: US
Inventors: Wilhelm Polach; Bernd Hupfeld; Thomas Wahl; Frank Brenner
Original assignee: Individual
Current assignee: Robert Bosch GmbH
Priority date: 1999-12-29
Filing date: 2000-12-29
Publication date: 2002-09-26
Also published as: EP1163431B1; JP2003519317A; WO2001049985A1; EP1163431A1; KR20010102422A; DE50004565D1; DE19963624A1; CN1342244A

Abstract

A method for the regeneration of an NOx store in the exhaust gas of internal combustion engines is presented. The NOx store takes up NOx from the exhaust gas when there is an oxygen excess therein and regenerates by outputting nitrogen when there is an oxygen deficiency in the exhaust gas. In the method, a reversal is made alternately between first phases having oxygen excess in the exhaust gas and second phases having oxygen deficiency. In the method, the oxygen deficiency in the exhaust gas is generated to regenerate the storage catalytic converter via a specific mass of fuel excess in the exhaust gas ahead of the NOx store. This mass is to be held constant.

Description

STATE OF THE ART

The invention relates to the operation of an NOx storage catalytic converter in internal combustion engines which are operated with a lean air/fuel mixture. Examples of such internal combustion engines are lean operated spark ignition engines or diesel engines.

NOx storage catalytic converters are used for the exhaust-gas after treatment in the combustion of lean air/fuel mixtures. The NOx storage catalytic converters store the nitrogen oxide, which is emitted in lean engine operation, in a first operating phase whose length is in the order of magnitude of minutes. In a second shorter operating phase having a length in the range of seconds, an emptying of the storage takes place via the supply of exhaust gas with a reducing agent to the storage catalytic converter.

The storage capability of the NOx catalytic converter is dependent upon charge and reduces continuously. If the first phase takes too long, then unwanted nitrogen oxide emissions occur. A second phase which takes too long has increased HC emissions and CO emissions as a consequence.

Accordingly, the problem is present to undertake the change between the two phases so that neither increased NOx emissions nor HC emissions and CO emissions occur.

In this connection, it is known from DE 197 398 48 to model the particular degree of the charge of the NOx store with NOx. The NOx entry into the store is determined from operating modes of the engine such as intake air mass flow and mixture composition. There is a change from the first phase into the second phase when a specific degree of charge is reached. The degree of charge reduces in the second phase and is likewise modeled or a termination of the second phase takes place when an exhaust-gas probe rearward of the storage catalytic converter signalizes a complete regeneration.

The modeling in one or both phases requires a very high complexity with respect to computing and thereby imposes high requirements on the engine control. Furthermore, the catalytic converter changes with respect to its storage and converting performance because of the deterioration process.

In some vehicle applications, and especially for diesel vehicles having NOx catalytic converters, the regeneration is preferably achieved by injecting fuel into the exhaust-gas system ahead of the catalytic converter. In this variation, the transition from lean to rich and the total regeneration process is especially critical because the varying flow conditions make the appropriate metering for generating a homogeneous distribution of the reducing agent in the exhaust gas ahead of the catalytic converter difficult. This transition phase appears not to be capable of being modeled with a reasonable complexity under the changing flow conditions in the exhaust gas.

In view of this background, the invention relates to the problem of controlling the change between the two phases.

This problem is solved with the combination of features of the claim.

ADVANTAGES

An advantage of the invention lies in the significantly more unified conditions of the regeneration phase.

As a further advantage, a significant reduction in computation complexity in the control of the charge and regeneration of the catalytic converter results.

As another advantage, a simple possibility results for checking the exhaust-gas treating system as well as in the method-conditioned adaptation of the control strategy to a performance of the catalytic converter caused, for example, by deterioration.

In the following, an embodiment of the invention is explained in greater detail. Reference is made to DE 1 97 298 48 for the disclosure of the technical background.

FIG. 1 shows the technical background in which the invention develops its effect; [0014]
FIG. 2 shows time-dependent traces of various signals; [0015]
FIG. 3 shows a modified structure for realizing the invention; and, FIG. 4 shows an embodiment in the form of a flowchart.[0016]
In detail, FIG. 1 shows an [0017] internal combustion engine 1 having an NOx storage catalytic converter 2, exhaust- gas probes 3 and 4, a control apparatus 5, a fuel metering device 6 and various sensors (7, 8, 9) for load L and rpm n as well as other operating parameters of the engine as may be required such as temperature, throttle flap position, et cetera.
From the above-mentioned and possibly further input signals, the control apparatus forms, inter alia, fuel metering signals with which the fuel metering means [0018] 6 is driven. The fuel metering means 6 can be configured for a so-called intake manifold injection as well as for a gasoline direct injection into the combustion chambers la of the individual cylinders. The variation of the mixture composition can take place via a change of the injection pulse widths with which the fuel metering means is driven. The essence of the method of the invention relates, in this environment, primarily to the control apparatus 5 and the exhaust-gas probe 4 mounted rearward of the catalytic converter.
FIG. 2 presents in FIG. 2[0019] a the change in the mixture composition “lambda” ahead of the catalytic converter (line 2 a) in combination with the signal US of the rearward exhaust-gas probe 4 (line 2 b) and the NOx concentration (line 2 c) rearward of the catalytic converter. The rearward exhaust-gas probe can, for example, be realized as an oxygen measuring sensor, a hydrocarbon sensor (HC sensor), as a carbon dioxide sensor (CO sensor) or as a nitrogen oxide sensor. What is shown is the signal of an oxygen sensor which supplies a high signal level when there is an oxygen deficiency and a low signal level when there is an oxygen excess.
In a first phase Ph[0020] 1 from t=0 to t=60, the engine is operated with a lambda greater than one, that is, with an air excess. The low signal level of the rearward probe (line 2 b) indicates that an air or oxygen excess is present also rearward of the catalytic converter. At time point t=60, the mixture composition is reversed from lambda greater than one to lambda less than one, that is, oxygen deficiency. Shortly thereafter, approximately at time point t=62, the rearward sensor 4 reacts to the oxygen deficiency with an increase of its signal from a low level to the high level.
As shown in FIG. 2, the time point t=62 can, for example, be determined by the signal of the rearward probe exceeding the threshold value. [0021]
The illustrated change of the mixture composition leads to the situation that the engine emits hydrocarbons and carbon monoxide as reducing agents. Alternatively to the emission of exhaust-gas components, which act in a reducing manner, the reducing agent can be supplied from a [0022] supply tank 11 via a valve 12 to the exhaust gas forward of the catalytic converter. The valve 12 is driven by the control apparatus 5. The engine can then be operated continuously with a lean mixture.
A corresponding modification of the structure of FIG. 1 is shown in FIG. 3. [0023]
In the method according to the invention, the regeneration phase is not modeled via computations and therefore is held variable. In lieu thereof, a predetermined constant mass of fuel is introduced into the exhaust-gas system ahead of the catalytic converter in each case for regeneration. The storing phase is then adapted in its duration to the regeneration phase. Defective adaptations are determined by an exhaust-gas probe mounted rearward of the catalytic converter and are reduced by influencing the length of the storing phase. For this purpose, the storing phase is shortened when an exhaust-gas probe signals no adequate change of the concentration of an exhaust-gas component toward the end of the regeneration phase. If such a change occurs, however, too early, then the storing phase is lengthened. [0024]
The advantage of the essentially more unified conditions of the regeneration phase results in that only the mass flow of the fuel, which is to be injected, needs to be adapted to the exhaust-gas mass flow in order to generate a rich exhaust gas having a specific desired lambda value. [0025]
The advantage of a significantly reduced complexity as to computation results from the possible omission of a modeling of the total fuel mass necessary for complete regeneration of the NOx store. [0026]
The further advantage of a simple possibility for checking the exhaust-gas treating system results as follows: if the storing times, which adjust when executing the method of the invention, deviate too greatly from plausible pregiven values, then a malfunction is present. [0027]
The further advantage of the method-conditioned adaptation of the control strategy to a catalytic converter performance, which is caused, for example, by deterioration, results as follows: if the quantity, which is to be stored in the NOx store in the storing phase, exceeds the storage capacity, which reduces because of the deterioration thereof, this is noted by a reaction of the exhaust-gas probe in the next regeneration phase and is considered in the control strategy. [0028]
An embodiment of the method of the invention is shown in FIG. 4. [0029]
[0030] Step 1 represents an engine operation with a lean mixture. The NOx, which is emitted by the engine in this operating phase, it taken up by the storage catalytic converter.
The degree of charge of the storage catalytic converter is computed in [0031] step 2 from operating parameters of the engine such as known, for example, from DE 197 39 848.
If the degree of the charge reaches a threshold value SW—NOx, the control apparatus triggers a regeneration of the storage catalytic converter. This is shown in [0032] steps 3 and 4.
What is essential to the invention is that this takes place with a predetermined mass of reducing agent. In the embodiment of FIG. 3, the predetermined mass of reducing agent can be metered from the [0033] tank 11 into the exhaust gas forward of the storage catalytic converter via the controllable valve 12. In the embodiment of FIG. 2, the predetermined mass of reducing agent is generated in the exhaust gas by a rich engine operation. For example, all fuel metering signals, which are specific for the normal engine operation with lean mixture, can be increased in a predetermined manner until the sum of the increases of the fuel metering signals corresponds to the desired fuel mass predetermined for the regeneration.
Lean operation again takes place when this fuel or reducing agent mass has been metered. The reaction of the [0034] rearward probe 4 is evaluated as to the regeneration toward the end of the regeneration phase. If the rearward probe is an oxygen measurement sensor, then its signal US can be compared with a threshold value (step 5).
If the signal does not reach the threshold, this means that no oxygen deficiency has arisen rearward of the catalytic converter at the end of the regeneration. The reducing agent quantity has then not been adequate to completely regenerate the NOx storage catalytic converter. As a consequence, and different from the state of the art, it is not the reduction agent quantity which is increased, but the storing phase is shortened. The illustrated example achieves the shortening via a reduction of the threshold value SW—NOx in [0035] step 6. If, in contrast, the reaction of the rearward probe is too pronounced (which, for example, can be determined in step 5 by the threshold value being exceeded), a lengthening of the storing phase takes place in step 7 via an increase of the threshold value SW—NOx.

Claims

1. method for regenerating an nox store in the exhaust gas of internal combustion engines which takes up Nox from the exhaust gas for an oxygen excess therein and which regenerates by discharging nitrogen for an oxygen deficiency in the exhaust gas, in which method there is a reversal between first phases having oxygen excess and second phases having oxygen deficiency in the exhaust gas, characterized in that the oxygen deficiency in the exhaust gas is generated for regenerating the storage catalytic converter via a specific mass of fuel excess, which is to be held constant, in the exhaust gas forward of the Nox store: