DE19915793A1 - Process for the desorption of a nitrogen oxide adsorber of an exhaust gas cleaning system - Google Patents

Abstract

The invention relates to a method for performing desorption operational phases for a nitrogen oxide adsorber of an exhaust gas purification system. In said method, the ratio of air to exhaust gas which is fed to the nitrogen oxide adsorber during each desorption operational phase is maintained in the rich-mixture range. According to the invention, said exhaust gas to air ratio is set to an initial rich-mixture value which can be predetermined at the beginning of the desorption operational phase and is increased during the course of the desorption operational phase from said initial value to a predeterminable, final rich-mixture value which is attained at the latest by the end of the desorption operational phase. The method can be used, for example, for desorbing a nitrogen oxide adsorber in a system for purifying the exhaust gas of an internal combustion engine in a lean-mixture driven motor vehicle.

Description

The invention relates to a method for performing of desorption operating phases in a nitrogen oxide adsorber Emission control system, the air ratio of the Nitrogen oxide adsorber supplied exhaust gas during each Desorption operating phase is kept in the rich range.

Such a method is particularly for exhaust gas purification plants against lean-operated motor vehicle internal combustion engines reversible, which contain a nitrogen oxide adsorber to the in Ma operating phases of the internal combustion engine mostly increasingly in the exhaust gas temporarily store existing nitrogen oxides adsorptively and there by reducing nitrogen oxide emissions. During this ad sorption phases, the nitrogen oxides from the nitrogen oxide adsorber pri mär bound as nitrate, the adsorber is usually a catalyzed contains effective material and in this case also as Nitrogen oxide adsorber catalyst is called. The one Nitric oxide storage capacity is limited from time to time the implementation of regeneration phases, in the present case desorpti ons operating phases, necessary during which the in Ni adsorbed nitrogen oxides are desorbed again.

It is known, when changing from adsorption operation to desorption operation, to change from the lean exhaust gas atmosphere mostly prevailing during adsorption operation to a rich, chemically reducing exhaust gas atmosphere, ie the exhaust gas air ratio from a value above the stoichiometric value one, present as a lean value referred to quickly change to a value below the stoichiometric value, here referred to as fat value. Since the air ratio is also referred to as the lambda value, this rapid change in the exhaust gas atmosphere from lean to rich is also generally referred to as “lambda jump”. Desorption processes of this type with a lambda jump and the resulting exhaust gas emission behavior of the nitrogen oxide adsorber and an automotive internal combustion engine producing the exhaust gases are described, for example, in the publication MS Brogan et al., Evaluation of NO _x Stora ge Catalysts as an Effective System for NO _x Removal from the Ex haust of Leanburn Gasoline Engines, SAE Technical Paper Series 952490, 1995. The fundamental investigations presented there report in particular on desorption processes in which the exhaust gas air ratio is kept at a predefinable, constant fat value during a respective desorption operating phase, which is maintained over a more or less long period of a few seconds to minutes becomes.

Other methods of the type mentioned that with a Lambda jump when changing from adsorption operation to desorption operation are described in the published patent applications EP 0 598 917 A1 and EP 0 636 770 A1. In EP 0 598 917 A1 especially suggested for desorption on a fat value of the exhaust air ratio to change, as well as the The slope of the associated lambda jump from the temperature of the Exhaust gas and thus the nitrogen oxide adsorber determined depending becomes. There are also methods in these two publications described for detection when regeneration of the nitrogen oxide adsorbers is necessary. This can e.g. B. with the signal one Nitrogen oxides arranged downstream of the nitrogen oxide adsorber sors or via a calculation of the nitrogen oxide adsorber and the comparison of the calculated amount with a maximum permissible nitrogen oxide storage quantity,  why the exhaust air ratio downstream using a lambda probe Wards of the nitrogen oxide adsorber can be detected.

In the present type of desorption of the nitrogen oxide adsorber occur as difficulties above all that at the beginning of the Desorption operating phase for so-called nitrogen oxide breakthroughs if the desorption operation is not sufficient in time much reducing agent is started, and towards the end of the desorp tion can lead to undesirable slippage of reducing agents. The latter represents an undesirable emission of reducing agent with the exhaust gas if the desorption operation is not correct is terminated early, during which such reducing agent z. B. in Form of unburned hydrocarbons and carbon monoxide or from urea metered externally into the exhaust line or There is ammonia in the exhaust gas. The reducing agent slip affects flows both the fuel consumption for the combustion source as well as pollutant emissions negative. It gets bigger the longer the desorption operating phase over the actually emergency agile duration and the smaller the exhaust air ratio is the desorption operating phase during this period.

The invention is a technical problem of providing based on a desorption process of the type mentioned at the beginning, with which the nitrogen oxide adsorber with a relatively small stick oxide breakthrough and reducing agent slip phenomena regene can rieren.

The invention solves this problem by providing a Desorption process with the features of claim 1 this procedure, the exhaust air ratio at the beginning of the Desorption operating phase to a predetermined minimum An Catch fat value set and from there in the course of the desorpti operating phase to a definable, at the latest towards the end the final fat value present in the desorption operating phase ben. This means that the final fat value in fat, i.e. H. sour Low-lambda lambda range closer to the stoichiometric value one is as the initial fat value. The initial hiring  a relatively rich exhaust air ratio has the desired Consequence that when changing from adsorption to regeneration drive provided as much reducing agent as quickly as possible which causes nitrogen oxide breakthrough symptoms, i.e. H. un desired nitrogen oxide emissions due to slow provision of sufficient reducing agent when reaching full Reduce the nitrogen oxide load of the nitrogen oxide adsorber or let it be avoided entirely. The transition to a less fat one Exhaust gas ratio at the latest towards the end of the desorption drive phase has the advantage that an undesirable reduction medium slip at the transition to the next adsorption phase can be minimized or avoided entirely. Because this Reduk agent slip that occurs when the desorption operation phase upright over its actually necessary length of time is obtained and then no longer sufficient nitrogen oxide desor beers that bind the reducing agents is all the more ge wrestler, the closer the rich exhaust air ratio during this time space is at the stoichiometric value.

In a further developed according to claim 2, the Initial fat value depending on the operating status of that too cleaning exhaust gas emitting combustion source variable given. This can, for. B. the setting of a fat From gas air ratio in the given operating state of combustion voltage source can be avoided, what with an internal combustion engine otherwise affect its smoothness and cause soot formation could do things.

It turns out that the setting of the initial fat value and of the final fat value within the further education of the Er Finding value ranges specified in claim 3 at favorable Er results.

In a further developed according to claim 4 method for the respective desorption operating phase is more relevant based on this Determined a theoretical minimum desorption time, and the exhaust gas air ratio is then at most for a time  kept at the minimum initial fat level that this theoretical minimum desorption time minus one performance readjustment duration that is required for this exhaust air ratio change for the power output of the combustion source le neutral to carry out, d. H. without the exhaust air ratio change from an undesirable change in the power output the combustion source, e.g. B. from an internal combustion engine given torque is accompanied. This approach takes into account the fact that such a service neutral Exhaust air ratio change, which usually corresponds to a appropriate change in the operating parameters of the combustion source thing is not possible instantaneously, but instead the said Power adjustment period is required.

Furthermore, the method further developed in this way the initial, most rich exhaust gas air ratio over egg ne as long as possible, what the to Desorption of the nitrogen oxide adsorber is the total time required keeps short, without on the other hand the danger that the Desorption mode with this rich exhaust air ratio too long ge is maintained and a noticeable reducing agent slip caused. Because the desorption operation with the strongest rich, initial exhaust air ratio only for the maintain theoretical minimum desorption time in each Case for the desorption at the end of the previous adsorp tion operating phase in the nitrogen oxide adsorber quantity is needed.

In a further embodiment of this measure, according to an Say 5 provided a safety period by which the exhaust gas air ratio earlier than according to the theoretical minimum desorp duration and the performance adjustment period required from mi nominal initial fat level is raised to ensure that when the final fat value is reached desorbed, to reduce ornamental nitrogen oxides are present, so that no noticeable reduction Onsmittelschlupf occurs. The security period is before preferably depending on the operating state of the combustion source  chosen variably, so that the desorption for the resp can run optimally in terms of fuel consumption.

Advantageous embodiments of the invention are in the drawing are shown and are described below. Here demonstrate:

Fig. 1 is a schematic diagram of time-dependent representation of the exhaust air-fuel ratio during a Desorptionsbe operating phase with an initial lambda jump and allmähli cher increase to a less rich exhaust air-fuel ratio,

Fig. 2 is a diagram of the exhaust gas air ratio as a function of time to illustrate a desorption process while maintaining the initial rich exhaust gas air ratio for a calculated period of time and

Fig. 3 to 6 exhaust gas ratio time diagrams for demonstration Lich other different ways of adjusting the exhaust air ratio during a desorpti onsbetriebsphase.

In the diagrams shown in FIGS . 1 to 6, various process implementations for carrying out desorption operating phases in a nitrogen oxide adsorber of an exhaust gas cleaning system are illustrated, which differ in particular in terms of the time course of the exhaust air ratio λ selected during the desorption operation, which is shown schematically as a function of time. It is assumed without restriction of generality that before the time t _{a at} the beginning of the respective desorption operation phase shown ge and after the time t _{e at} the end of this desorption operation phase, the combustion source, for. B. a motor vehicle internal combustion engine, is operated lean, so that there is a correspondingly lean exhaust gas air ratio λ _M , which is above the stoichiometric value one. In these operating phases, the nitrogen oxide adsorber is in adsorption mode, in which it temporarily stores the nitrogen oxides, which accumulate in the exhaust gas, mostly in nitrate form.

To initiate the desorption operating phase, the exhaust air ratio λ is reduced from this lean value λ _M to a certain _minimum initial fat value λ _min . This lambda jump is implemented in a conventional manner, for. B. by corre sponding enrichment of the air / fuel mixture burned by the combustion source and / or a reducing agent injection in the exhaust tract upstream of the nitrogen oxide adsorber. The desorption operation is then carried out for a corresponding period of time t _d = t _e - t _a , during which the exhaust gas air ratio λ is kept in the rich range, ie below the stoichiometric value one. Then it goes back to the normal Ma gerbetrieb the combustion source and consequently from desorption to adsorption of the nitrogen oxide adsorber. During the desorption operating phase, the exhaust gas air ratio λ is raised from the initial minimum fat value λ _min to an end fat value λ _end , which is at the latest at the end time t _e of the desorption operating phase and is still in the rich range, but significantly closer to the stoichiometric value one than the initial minimum fat value λ _min .

At the beginning t _{a of} the desorption operation, on the one hand, as much as possible a reduction in the exhaust air ratio λ, ie the provision of as much reducing agent as possible, is desirable in order to avoid any nitrogen oxide breakthroughs when switching from adsorption to desorption operation. On the other hand, the lowering of the exhaust gas air ratio λ is limited by the fact that too much a lowering causes the risk of soot formation and, in the case of an internal combustion engine, impairs its smoothness. The initial minimum fat value λ _min is therefore selected as a function of the specific circumstances of the given combustion source and its current operating point, it being shown that the minimum fat value λ _{min is} preferably in the range between approximately 0.6 and approximately 0.7. A very quick setting of this minimum air ratio λ _min leads to a large reductant flow and in this way to the desired minimization of any nitrogen oxide breakthroughs.

The further selection of the exhaust air ratio after the desorption operation has started is based on the following considerations. On the one hand, maintaining the minimum fat value λ _min during the desorption operation enables regeneration of the nitrogen oxide adsorber as quickly as possible and results in only a minimal increase in fuel consumption. On the other hand, if the desorption operation takes too long for control-technical reasons, in particular due to process-related idle times, the reducing agent slip that occurs occurs when the exhaust gas air ratio at this point in time is still very rich. The re duction agent slip affects the total emissions negatively, since z. B. unburned hydrocarbons and carbon monoxide can be emitted as pollutants.

It is therefore favorable to increase the exhaust gas air ratio from the initial minimum fat value λ _min to the final fat value λ _end closer to the stoichiometric value at the latest towards the end of the desorption operating phase, for example at the latest when the critical period of the air ratio change from lean to rich has been overcome, during which a nitrogen oxide breakthrough can occur. Fig. 1 shows a possible time course λ _{1 of} the air ratio, which takes these considerations into account and in this case has a concave curve shape. The final fat value λ _end is chosen in any case so that it is still far enough from the stoichiometric value one in the fat range to supply the nitrogen oxide adsorber with sufficient reducing agents and to allow the regeneration to proceed completely and quickly. It can be seen that a final fat value λ _end in the range between approximately 0.85 and approximately 0.95 leads to particularly satisfactory results, although small or larger final fat values λ _end are also suitable in principle. If the regeneration phase now lasts a little too long for control-related reasons, the resulting slippage of reducing agent remains relatively small, since the final air ratio λ _end, which is closer to the stoichiometric value, does not contain as much reducing agent in the exhaust gas during this period when setting the minimum fat value λ _min .

Fig. 2 illustrates a strategy for adjusting the exhaust air ratio λ during the desorption operating phase, in which the exhaust air ratio λ is not raised as early as in the example of FIG. 1 from the minimum initial fat value λ _min towards the final fat value λ _end . Rather, the minimum exhaust gas air ratio λ _{min is maintained} for a suitable, longer period of time, which keeps the additional fuel consumption comparatively low. This procedure is based on the calculation of a theoretical minimum regeneration period t _tm , that is, depending on the selected course of the exhaust gas air ratio λ, at least the period of time to be expected until the substantially complete regeneration of the nitrogen oxide adsorber. This calculation of the theoretical minimum regeneration duration can be done in one of the ways familiar to the expert, as z. B. are described in the aforementioned prior art. It is possible, for example, by modeling the raw nitrogen oxide emissions of the combustion source and the adsorption and desorption behavior of the nitrogen oxide adsorber used. If necessary, the end of the respective desorption operating phase can be measured, e.g. B. determined on the basis of the signal of a lambda probe arranged downstream of the nitrogen oxide adsor and thus an adaptation of the computing model used to the actually measured conditions can be provided.

At the beginning of a desorption operating phase, the minimum air ratio λ _{min that is} possible under the condition of an undisturbed further operation of the combustion source is set in the method illustrated in FIG. 2, and at the same time the theoretical minimum _{regeneration duration} t _tm for the upcoming desorption operating phase is calculated. In order to regulate the change from the minimum air ratio λ _min to the air ratio λ _end at the end of the desorption operating phase for the combustion source in a power-neutral manner, in the case of an internal combustion engine torque-neutral, a certain power readjustment period t _{reg is} required, i.e. in the case of an internal combustion engine a minimum duration of one corresponds to a certain minimum number of work cycles. So that the desorption operating phase can be achieved with as little additional fuel consumption as possible, the minimum air ratio λ _{min is maintained} for the theoretically necessary minimum regeneration time t _tm minus this power readjustment time t _reg for the lambda change. In the event of changes in the air ratio λ during desorption compared to the assumptions made when calculating the theoretical minimum regeneration duration t _tm , the latter is continuously adapted by means of a corresponding recalculation. The course of the exhaust gas air ratio λ resulting from this procedure during the desorption operation phase is represented in FIG. 2 by a characteristic curve λ ₂ . As can be seen from this, after the final fat value λ _{end has been reached,} the desorption operation is maintained for a certain period of time t _s , during which it can be checked whether the desorption of the nitrogen oxide adsorber has actually been completely carried out before the desorption operation phase is then over gear to lean operation of the combustion source is ended and the nitrogen oxide adsorber works again in the adsorption operation.

In order to avoid that a possible slight miscalculation of the minimum regeneration period t _tm leads to a delay in reaching the final fat value λ _end and thus to a slippage of reducing agent, provision is preferably made to advance the start of the increase in the exhaust gas air ratio by a safety period t _c , ie the minimum air ratio λ _min is maintained for a period t _m = t _tm -t _reg -t _c , which corresponds to the calculated minimum regeneration duration t _tm minus the power readjustment period t _reg and minus the safety period t _c . The safety period t _c is specified as an applicable size, depending on the specific conditions of the combustion source, in a motor vehicle engine z. B. depending on the engine conditions and driving condition, can be selected variably. The increase in the exhaust gas air ratio λ taking place according to this variant is represented in FIG. 2 on the basis of a dash-dotted characteristic curve λ ₃ .

The procedure illustrated in FIG. 2, which makes use of a calculated minimum regeneration period, ensures a minimization of the reducing agent slip at the end of the desorption operating phase and a fuel consumption-optimal course of the desorption of the nitrogen oxide adsorber.

In addition to the concave curve profiles of the exhaust gas air ratio λ shown in FIGS. 1 and 2, the latter can also be raised from the minimum fat value λ _min to the final fat value λ _end according to other functional time dependencies. Some related possibilities are shown in Figs. 3 to 6.

In the example of FIG. 3, there is a sudden, one-step increase in the exhaust air ratio λ according to a stair curve λ ₄ . Accordingly, the exhaust gas air ratio λ is kept for an initial time t ₁ at the minimum fat value λ _min before it is abruptly raised to the final fat value λ _end and is held there for the remaining time t _{2 of} the desorption operating phase.

In the example of Fig. 4, the increase is of Abgasluftver holds isses λ in accordance with a multi-stage step function λ ₅ in meh reren steps, until it at a certain time t _z during Desorptionsbetriebsphase the final fat value λ has _end reached and there until the end of the desorption operation is held.

Fig. 5 illustrates a method example in which the exhaust gas air ratio λ is linearly raised according to a straight line λ ₆ from the minimum initial fat value λ _min to the final fat value λ _end .

Fig. 6 shows a method example in which the exhaust air ratio λ is increased during the desorption operating phase according to a convex curve λ ₇ from the initial fat value λ _min to the final fat value λ _end .

The various exemplary embodiments described above make it clear that a nitrogen oxide adsorber can be regenerated in an advantageous manner by the process according to the invention by quickly reducing the exhaust air ratio to a minimum fat value at the beginning of the desorption process, so that sufficient reducing agent is immediately available to avoid a nitrogen oxide breakthrough, and in the further course of the desorption operating phase the air ratio is raised to a final value λ _end closer to the stoichiometric value, so that even if the desorption time that is actually required is slightly exceeded, no significant reduction of the reducing agent occurs.

Claims (5)

1. A method for performing desorption operating phases in a nitrogen oxide adsorber of an exhaust gas cleaning system, in particular for a motor vehicle internal combustion engine, in which

• the air ratio (λ) of the exhaust gas supplied to the nitrogen oxide adsorber is kept in the rich range during the respective desorption operating phase,

characterized in that

• - The exhaust gas air ratio (λ) at the beginning of the desorption operation phase is set to a predeterminable minimum initial fat value (λ _min ) and from there during the desorption operation phase to a predefinable end fat value (λ _end ) present at the latest towards the end of the desorption operation phase becomes.

2. The method according to claim 1, further characterized in that the initial fat value (λ _min ) is variable depending on the operating state of the combustion source emitting the exhaust gas to be cleaned variable. 3. The method of claim 1 or 2, further characterized in that the initial fat value (λ _min ) in the range between about 0.6 and about 0.7 and / or the end fat value (λ _end ) in the range between about 0 , 85 and approximately 0.95 lying down. 4. The method according to any one of claims 1 to 3, further characterized in that

• - A theoretical minimum desorption duration (t _tm ) for the respective desorption _{operating phase is} determined on the basis of empirical and / or modeled data about the nitrogen oxide emission behavior of the combustion source emitting the exhaust gas and about the adsorption and desorption behavior of the nitrogen oxide adsorber, and
• - The exhaust gas air ratio (λ) is kept at the minimum initial fat value (λ _min ) for a maximum of a period of time that corresponds to the theoretical minimum desorption time determined minus a power readjustment time (t _reg ) that is required to neutralize this exhaust gas air ratio change for the output of the combustion source perform.

5. The method according to claim 4, further characterized in that the exhaust gas air ratio (λ) is kept for a period (t _m ) on the mini paint initial fat value (λ _min ), the theoretical minimum _{desorption time} determined (t _tm ) minus the lei post-adjustment period (t _reg ) and minus a specifiable safety period (t _a ).