DE10023060A1 - Determining aging state of nitrogen oxides storage catalyst arranged in exhaust gas stream of IC engine comprises calculating theoretical regeneration - Google Patents

Abstract

Determining the aging state of at least one nitrogen oxide (NOx) storage catalyst arranged in the exhaust gas stream of an IC engine comprises calculating the theoretical regeneration time of the catalyst using a first data set, and measuring lambda values of the gas fed to the catalyst. Determining the aging state of at least one NOx storage catalyst arranged in the exhaust gas stream of an IC engine comprises calculating the theoretical regeneration time of the catalyst using a first data set; introducing the regeneration phase to a start time point at which the first lambda value of the gas fed to the catalyst is reduced to a regeneration value of less than 1; measuring a second lambda value downstream of the catalyst and measuring the actual time span between the start time point and the time point at which the second lambda value exceeds a termination value of less than 1; and forming the difference time value between the theoretical regeneration time and the actual time span. Preferred Features: The regeneration value is immediately increased from the initial regeneration phase for all coming regeneration phases. The regeneration value is increased by a value which is a function of the difference time value.

Description

The invention relates to a method for determining the aging state of a NO _x storage catalytic converter according to the preamble of claim 1 and a method for carrying out the NO _x regeneration by means of a NO _x storage catalytic converter according to the preamble of claim 2.

From EP 580 389 A1 and EP 0585 900 A1, so-called NO _x storage catalysts for use in the exhaust gas flow of lean-burn internal combustion engines are known. The lean operation of internal combustion engines has the advantage that fuel consumption can be significantly reduced, but the disadvantage is that a relatively large amount of nitrogen oxides (NO _x ) is present in the raw exhaust gas during lean operation. In order to solve this problem, it is known to arrange a NO _x storage catalytic converter in the exhaust gas flow of the internal combustion engine, on the surface of which the nitrogen oxides are deposited and chemically bound, for example according to the following reaction:

2BaO + O ₂ + 4NO ₂ → 2Ba (NO ₃ ) ₂

This significantly reduces the amount of nitrogen oxides in the exhaust gas flow from the catalyst. Such a storage catalytic converter must be regenerated at certain intervals since its storage capacity is exhausted. This happens because the internal combustion engine is operated rich for a certain period of time, so that there are unburned hydrocarbons and carbon monoxide in the raw exhaust gas stream, that is between the internal combustion engine and the NO _x catalyst. The storage catalytic converter is regenerated by means of these reducing agents, the nitrogen oxides or the corresponding nitrates being reduced to nitrogen by the reducing agents present, for example with the following reaction:

Ba (NO ₃ ) ₂ + 5CO → BaO + N ₂ + 5CO ₂

Ideally, these reducing agents are fully implemented, so that in the Final exhaust gas neither nitrogen oxides nor unburned during the regeneration phase Hydrocarbons or carbon monoxide.

So far, the regeneration of the NO _x storage catalytic converter has been carried out as follows: At a certain point in time, the combustion in the internal combustion engine is switched to rich, and the lambda value in the raw exhaust gas flow is thereby reduced to a predetermined value which is at or below 1. Efforts are made to keep this value as small as possible, that is to make the reducing agent flow as large as possible during the regeneration phase in order to achieve the shortest possible regeneration times. During the regeneration phase, the lambda value in the final exhaust gas flow is measured. The regeneration is stopped as soon as the lambda value in the final exhaust gas flow falls below a predetermined termination limit value, which is less than 1, that is to say when reducing agents pass through the NO _x storage cat. This is called a breakthrough of reducing agents.

However, this procedure has disadvantages, since it has been found that the aging of the NO _x storage catalytic converter leads to an increasing slowdown in the reduction reactions. This can result in the lambda value in the final exhaust gas stream falling below the termination limit value during the reduction phase, although the regeneration process has not yet been completed, that is, while nitrogen oxides or nitrates are still in the NO _x storage catalytic converter. This leads to the regeneration not being completed, which leads to a further deterioration in the state of the NO _x storage catalytic converter. This means that the regeneration control or its parameters, which lead to good results with new storage catalytic converters, are only conditionally suitable or even unsuitable for aged catalysts.

Starting from this prior art, it is an object of the invention to provide a method with which the aging state of a NO _x storage catalytic converter can be determined.

This object is achieved according to a method with the features of claim 1 solved.  

For this purpose, a theoretical regeneration time Δt _th is first calculated from a set of engine status data and from the lambda value of the raw exhaust gas flow. It is necessary to know the NO _x mass stored in the storage cat, which results from the course of certain values such as speed, fuel consumption and lambda value since the end of the last regeneration phase, the engine operating point, the target lambda specification in front of the storage cat and the catalyst temperature. Depending on the engine operating point and the set lambda value, the HC and CO mass flow can be approximately modeled during the regeneration phase. With knowledge of the reducing agent flows, the stoichiometric NO _x discharge can be approximately calculated, taking into account that the oxygen stored in the catalytic converter is first consumed. The O ₂ storage capacity depends essentially on the catalyst temperature. If the amount of nitrogen oxides stored at the beginning of the regeneration phase is known and if the stoichiometric NO _x discharge is known approximately, the theoretical regeneration time can be approximately calculated.

The time period between the initiation of the Regeneration process and the time at which the lambda value of the Final exhaust gas flow falls below a predetermined termination limit. These two Times are compared in a third step. The resulting one Time difference is a measure of the aging of the catalyst.

Another object of this invention is to provide a method for carrying out the NO _x regeneration which, even with aged catalysts, provides satisfactory results with regard to the emission behavior.

This object is achieved with a method having the features of claim 2.

First, as in the method of claim 1, the time difference between theoretical regeneration time and the time at which the lambda value in the Final exhaust flow falls below the termination limit. As soon as this time difference exceeds a certain level, the lambda value in the raw exhaust gas flow during the Regeneration phase increased from the original value. This is synonymous with a reduction in the mass flow of reducing agent. The Regeneration phase is thereby extended, it is still on  low-emission vehicle operation guaranteed. The increase in the lambda value in Raw exhaust gas flow can happen immediately or only during the next regeneration cycle.

The invention will now be explained in more detail with reference to the drawings. Show it:

Fig. 1 is a schematic representation of the motor unit, control electronics and NO _x storage catalytic converter,

Fig. 2 is a schematic illustration of the operational phases of a lean burn internal combustion engine,

Fig. 3a, 3b gradients of the lambda signals during the regeneration phase.

In Fig. 1, a motor unit with control electronics and the associated exhaust system is symbolically shown. A motor unit 10 is connected to control electronics 20 via a bidirectional data line. The control electronics 20 receive information such as speed, engine temperature, etc .; The control electronics 20 specifies parameters such as ignition timing, mixture ratio, etc. A NO _x storage catalytic converter 40 is arranged in the exhaust gas flow of the engine unit 10 . Between the engine unit 10 and NO _x storing catalyst 40, that is, in Rohabgasstrom, a first lambda probe 31 is arranged, the lambda signal λ _a is passed to the control electronics 20th The first lambda probe 31 is preferably a broadband lambda probe. A second lambda probe 32 is arranged downstream of the NO _x storage catalytic converter and likewise transmits its lambda value λ _out to the control electronics 20 . A grade rule lambda probe is usually sufficient here. The control electronics 20 also receive information about the temperature of the storage catalytic converter 40 by means of a temperature sensor 35 .

In FIG. 2, the idealized waveforms of the two lambda values are shown _and λ _out during the two operating phases "lean" and "fat" of a lean burn internal combustion engine λ. During the lean phases MP, both lambda values are of the same size and have a value <1. The lean phases typically last from one to a few minutes. From the engine status data such as temperature, speed, mixture ratio, etc., the control electronics 20 calculates the amount of nitrogen oxides generated since the last regeneration phase. Since ideally all nitrogen oxides generated are stored in the NO _x storage catalytic converter, the control electronics can thus determine when the catalytic converter is full. As soon as the catalyst is full, the regeneration phase is initiated. For this purpose the motor to rich operation is changed, the value of the first lambda signal that is λ is lowered below _a first The reducing agents present in the raw exhaust gas stream are converted almost stoichiometrically in the storage catalytic converter, so that the second lambda value λ _off has the value 1 during the regeneration operation, that is to say during the rich phase FP. Towards the end of the regeneration phase, the second lambda value λ _off drops because there are no longer enough nitrogen oxides to convert all the reducing agents. If the second lambda value below λ _from an abort value, the regeneration is canceled and the engine returns to lean operation. The fat phases typically last a few seconds.

Since it is desirable to keep the consumption-unfavorable fat-phase as short as possible, the first lambda value is λ _a while this fat phase (= regeneration) selected as low as possible. These relatively low values of λ _a while but fat phase lead with increasing aging of the NO _x storage problems. Due to various aging processes and sometimes due to increasing sulfur poisoning of the catalyst, the chemical processes taking place during regeneration are impeded and run more slowly than with a new catalyst. As a result, the second lambda value λ _off behaves differently with an aged catalytic converter than with a new one. This situation is shown in Fig. 3a. The solid curve represents the second lambda value during regeneration in the case of a new catalytic converter. This curve is referred to here as λ _out1 . At the beginning of the regeneration phase λ _aus1 falls to the value 1 and remains on a plateau here. Towards the end of the regeneration phase is lowered λ _OFF1 and falls at time t ₁ the termination value λ _A. This falling below the limit value is used as an abort criterion for ending the regeneration phase and the engine is switched back to lean operation immediately or with a predetermined delay. λ _a and therefore also _from λ rise to a predetermined lean operation value λ _M. As long as the catalytic converter is new, the actual time period Δt _ist = t ₁ -t ₀ corresponds approximately to the theoretically calculated regeneration time Δt _th . The situation changes, however, when the reduction reactions are slower due to aging. This situation is also shown in Fig. 3a. The dashed curve shows the course of the second lambda value λ _off with an aged catalytic converter; the curve is designated here with λ _aus2 . All other operating parameters correspond to the above. At the beginning of the regeneration phase at time t ₀ , the value λ _aus2 also initially drops to 1 and remains here on a plateau. Because of the now slower regeneration reactions, however, there is a breakthrough of the reducing agent at time t ₂ and the second lambda value λ _aus2 drops below the termination value λ _A. If this point in time is used as the point in time for the termination, the regeneration is stopped far too early, that is to say when considerable amounts of nitrogen oxides are still stored in the storage catalytic converter. For a better understanding, however, the situation is shown here how λ _aus2 behaves if the regeneration is not stopped here. λ _off2 drops slowly over a longer period of time and at the end of the actual regeneration time λ _off2 then drops sharply, similar to a new catalytic converter.

The difference time value Δt between the theoretical regeneration time Δt _th and the actual time period Δt _is therefore a measure of the aging state of the catalytic converter. This difference time value Δt can now be used, for example, to adapt the operating parameters to the aging state of the catalytic converter, as the following example shows:
One possible measure for adapting the operating parameters to the aging condition of the catalytic converter is to increase the regeneration value λ _R during the rich phase. This happens as follows: As soon as it has been determined that the actual time period Δt _is less than the theoretical regeneration time Δt _th by more than a predetermined amount (see the explanations above and Fig. 3a), the regeneration value λ _{R2 is increased} compared to the previous regeneration value λ _R1 , This can happen immediately or only during the next regeneration phase. This situation is shown in Fig. 3b. An increase in the regeneration value λ _R means a reduction in the reduction mass flow, which takes account of the now slower reactions. As a result, the second lambda value, here λ _off3 , remains on a plateau at λ = 1 for a long time and does not drop until the end of the regeneration _phase . At time t ₃ , the dropout value λ _{A is} undershot and the end of regeneration is initiated.

These parameters are now maintained until the actual time period Δt _ist again deviates from the theoretical regeneration time Δt _th by more than a predetermined amount, then the regeneration value λ _{R is increased} again. The regeneration value λ _R can be raised by a predetermined amount or determined depending on the difference time value Δt.

Claims (6)

1. A method for determining the aging state of at least one NO _x storage catalytic converter arranged in the exhaust gas flow of an alternately lean and richly operated internal combustion engine using a first data set of engine state data and a lambda value of a lambda probe arranged in the exhaust gas flow of the NO _x storage catalytic converter, with the following steps:

• Calculating the theoretical regeneration time (Δt _th ) of the NO _x storage catalytic converter using the first data set,
• - initiating the regeneration phase at a starting time point (t _0), wherein the first lambda value (λ _a) of the NO _x storage catalytic converter exhaust gas supplied is decreased to a regeneration value (λ _R) <1,
• - Measuring the second lambda value (λ _off ) downstream of the NO _x storage catalytic converter and measuring the actual time period (Δt _ist ) that has elapsed between the start time (t ₀ ) and the time at which the second lambda value (λ _off ) has an abort value ( falls below λ _A ) <1,
• - Forming the difference time value (Δt) between the theoretical regeneration time (Δ _th ) and the actual time period (Δt _ist ).

2. Method for carrying out the NO _x regeneration of an NO _x storage catalytic converter arranged in the exhaust gas flow of an alternately lean and richly operated internal combustion engine using a first data set of engine status data and the lambda value of a lambda probe arranged in the exhaust gas flow of the NO _x storage catalytic converter, with the following steps:

• Calculating the theoretical regeneration time (Δt _th ) of the NO _x storage catalytic converter using the first data set,
• - initiating the regeneration phase at a starting time point (t _0), wherein the first lambda value (λ _a) of the NO _x storage catalytic converter exhaust gas supplied is decreased to a regeneration value (λ _R) <1,
• - Measuring the second lambda value (λ _off ) downstream of the NO _x storage catalytic converter and measuring the actual time period (Δt _ist ) that has elapsed between the start time (t ₀ ) and the time at which the second lambda value (t ₂ ) has an abort value ( falls below λ _A ) <1,
• Formation of the difference time value (Δt) between the theoretical regeneration time (Δt _th ) and the actual time period (Δt _ist ),
• - Increase the regeneration value (λ _R ) when the difference time value (Δt) exceeds a predetermined value.

3. The method according to claim 2, characterized in that the regeneration value (λ _R ) is raised immediately. 4. The method according to claim 2, characterized in that the regeneration value (λ _R ) is increased from the next rain generation phase for all upcoming regeneration phases. 5. The method according to any one of claims 2 to 4, characterized in that the regeneration value (λ _R ) is increased by a predetermined amount, preferably in predetermined steps. 6. The method according to claim 2, characterized in that the regeneration value (λ _R ) is increased by an amount which is a function of the difference time value (Δt).