DE102006009935A1 - Method for operating nitric oxide reduction catalytic converter involves determining ammonia level in upper temperature range and adding reduction agent which is ceased in lower temperature range - Google Patents

Abstract

The method involves adding as reduction agent ammonia in an adjustable rate to the exhaust gas. . In a first upper temperature range of the catalytic converter (3) the ammonia level therein is determined through adding the conversion rate of ammonia and/or nitric oxide and a supply rate of ammonia to adjust the dosing rate of the reduction agent so that a predeterminable ideal value is reached for the ammonia level. In a second lower temperature range the addition of the reduction agent is ceased, and in a third middle temperature rate the dosing rate of the reduction agent is raised at least temporarily by a predetermined amount. In the third temperature range the anticipated future path of the temperature of the catalytic converter is estimated.

Description

The invention relates to a method for operating a nitrogen oxide reduction catalyst with ammonia storage capability, the exhaust gas is fed to a predominantly lean internal combustion engine, in which as the reducing agent ammonia (NH ₃ ) or a substance from which ammonia can be released, is added to the exhaust gas with an adjustable metering

From the DE 101 26 456 A1 A method for removing nitrogen oxides (NO _x ) from the exhaust gas of a lean-burn internal combustion engine by means of a NH ₃ -containing reducing agent is known, in which a nitrogen oxide reduction catalyst is used with ammonia storage capability. On the output side of the catalyst, a sensor is provided which detects the ammonia slip of the catalyst. An addition of the reducing agent is controlled on the basis of the detected ammonia slip.

From the EP 0 554 766 A1 a method is known for operating a nitrogen oxide reduction catalytic converter with ammonia storage capacity connected to a predominantly lean-burn internal combustion engine, in which a pulsed addition of NH ₃ takes place. The NH ₃ addition is interrupted again after its start only when the amount of NH ₃ stored in the catalyst has reached a certain upper threshold value.

A difficulty of the known methods is that at low temperatures the nitrogen oxide reduction catalyst is inactive and / or a required treatment of a NH ₃ -releasing reducing agent is no longer possible. Therefore, especially at low temperatures, the reduction of nitrogen oxide is limited or not possible.

task the invention is to provide a method for operating a nitrogen oxide reduction catalyst, which improved especially at low temperatures Nitrogen oxide turnover allows.

These The object is achieved by a method having the features of the claim 1 solved.

In the method according to the invention, an ammonia level in the nitrogen oxide reduction catalyst is determined in a first, upper temperature range of the nitrogen oxide reduction catalyst, in particular by summation and temporal integration of at least one conversion rate of NH ₃ and / or NO _x reacted in the nitrogen oxide reduction catalyst and a feed rate of NH ₃ fed to the nitrogen oxide reduction catalyst, and the metering rate the reducing agent is determined and adjusted such that a predefinable setpoint for the ammonia level in the nitrogen oxide reduction catalyst is at least approximately achieved. In a second, lower temperature range of the nitrogen oxide reduction catalyst, the addition of reducing agent is stopped. In a third, middle temperature range of the nitrogen oxide reduction catalyst, which directly adjoins the first upper temperature range and the second lower temperature range, the metering rate of the reducing agent determined analogously to the first temperature range is increased at least temporarily by a predeterminable amount. In this way, at low temperatures stored in the catalyst NH ₃ for NO _x reduction is available. As a result, NO _x reduction is possible even at low temperatures.

As a reducing agent is primarily urea, ammonium carbamate, ammonium formate or a similar, capable of cleaving NH ₃ capable substance in question. The reducing agent can be added to the exhaust gas in pure form or as an aqueous solution and / or as an aerosol-like mist at an addition point of a corresponding exhaust gas line arranged in particular upstream of the nitrogen oxide reduction catalyst. The addition point is preferably formed in the manner of a metering valve with one or more nozzle openings. The nitrogen oxide reduction catalyst is preferably designed as a classic SCR catalyst, in particular based on tungsten oxide or vanadium pentoxide or as a zeolitic catalyst. It is known that these types of catalysts can store considerable quantities of NH ₃ , with stored NH ₃ acting as a reaction partner for a reduction of nitrogen oxides supplied to the exhaust gas. The following is a simplified discussion of a catalyst. In this sense, the NH ₃ level is understood to mean the amount of NH ₃ stored in the catalyst bed. Preferably ₃ filling level is used as a relative value NH, which indicates the stored NH ₃ amount in respect to the maximum storable under the current conditions NH ₃ amount.

To determine the NH ₃ level, for example, a sensor arranged in the catalyst bed and sensitive to NH ₃ may be used. However, it is preferably vorgese hen to determine the NH ₃ level by calculation, which can be used on sensory variables such as a NH ₃ -slip and / or an output side of the catalyst detected NO _x concentration. To determine the NH ₃ level, an electronic unit is preferably provided which can access stored data, preferably in the form of characteristic curves, characteristic diagrams and programs. Based on the data and additional input signals regarding relevant state variables of the catalyst and the exhaust gas as well as operating data the internal combustion engine, such as temperature, oxygen content of the exhaust gas, exhaust gas mass flow, engine load and speed, there is a model-based computational determination of the NH ₃ level in the catalyst. It is advantageous if the electronics unit can communicate with a conventionally provided anyway electronic engine control, so that it can be accessed on all operationally relevant variables. Physically, the electronics unit can also be an integral part of the electronic engine control. However, it can also be designed as part of a dosing unit for the reducing agent.

According to the invention, the NH ₃ level of the catalyst is determined by summation and temporal integration of at least one conversion rate of NH ₃ and / or NO _x reacted in the catalyst and a feed rate of the catalyst with the exhaust gas supplied to the reducing agent. The integration preferably takes place after the summation of the values relevant for balancing with regard to the NH ₃ level. Due to the procedure according to the invention, an ongoing determination of the current NH ₃ level is made possible by continuously performing balancing of the variables influencing the level. In order to further improve the accuracy of the level detection, additional factors can be taken into account in the balancing. In particular, consideration can also be given to reducing agent losses by desorption of NH ₃ desorbed from the catalyst and of side reactions, such as, for example, a reduction agent direct oxidation. The influence of side reactions can also be included in the conversion rate.

It is advantageous if, in order to determine the NH ₃ level, maps or characteristic curves are provided with data, which have been advantageously determined in advance, for the respective catalyst. These may relate to a NO _x or NH ₃ conversion rate, an NH ₃ desorption rate and optionally further influencing factors, such as, for example, a rate of direct oxidation. It is also expedient to provide a setting range with an upper and a lower threshold value for the filling level to be set NH ₃ so that a setting of the NH ₃ is carried -Füllstands in the catalyst in this area. The threshold values are expediently predetermined in advance such that, overall, an optimal NO _x conversion results therewith with at the same time minimal or negligible NH ₃ slip.

On the basis of the available information, the metering rate of the reducing agent is determined and adjusted in a first, upper temperature range in such a way that a specifiable target value for the NH ₃ level in the catalyst is achieved as far as possible. The first temperature range is chosen according to the type of catalyst used so that the catalyst has a high efficiency in this temperature range. To set the dosing rate, for example, a dosing unit controls corresponding actuators such as pumps, valves and the like. The dosing unit is preferably designed as a controller, which can adjust the NH ₃ - level of the catalyst to the predetermined target value. Dosing rate is understood to mean the amount of reducing agent added per unit of time to the exhaust gas, which can result from averaging over a more or less extended time interval. Reducing agent addition may be accomplished by pulsed on-off switching as well as by variable adjustment within a continuous range of values, or by adjusting an amount of added equipment within a given time period.

The Upper limit of the second, lower temperature range is preferably based on a temperature limit for the catalyst activity and / or one for sufficient treatment of the reducing agent required Temperature set. By the procedure according to the invention, in the second Temperature range the catalyst no reducing agent can supply thus avoiding that not or insufficiently processed and thus inactive reducing agent reaches the catalyst. this could there its surface block and lead to reduced efficacy. moreover is avoided because of a lack of catalyst activity not reacted reducing agent discharged with the exhaust gas to the environment becomes.

The third, middle temperature range lies between the first and the second temperature range and represents a transition region with regard to the catalyst activity or a treatment of the reducing agent. Also in the third temperature range, preference is given to adjusting the reducing agent addition analogously to the procedure explained for the first temperature range. According to the invention, however, it is additionally provided to increase the metering rate of the reducing agent thus determined at least temporarily by a predeterminable amount. Thus, at least temporarily, the NH ₃ level can be increased beyond the otherwise intended level. In this way it is possible to react to the particular conditions present in the third temperature range.

In an embodiment of the method, at least in the third temperature range of the nitrogen oxide reduction catalyst, an estimate of him to waiting future course of the temperature of the nitrogen oxide reduction catalyst made. In this way, a statement can be made as to whether the catalyst temperature evolves from the critical transition region for efficacy and / or reductant treatment toward the first temperature region with improved catalyst action or towards the second temperature region with degraded effect. Thus, the addition of reducing agent can be optimally adapted to the expected temperature development of the catalyst.

In Another embodiment of the method is a temperature of the nitrogen oxide reduction catalyst continuously and the estimate of the expected future Temperature trend based on a current temperature value and / or at least one previously determined temperature value. The temperature determination is advantageously carried out on the basis of sensory recorded temperature values before and / or after or in the catalyst. The future Temperature course is preferably by extrapolation already calculated temperature values. In the extrapolation calculation can while more information, such as the operating state the internal combustion engine are included.

In Another embodiment of the method is used to estimate the expected future course the temperature of the nitrogen oxide reduction catalyst, a heat model for the Used nitric oxide reduction catalyst. In this way, the reliability the estimate be improved.

In Another embodiment of the method becomes an operating profile for the temperature of the nitrogen oxide reduction catalyst and / or at least one the catalyst temperature influencing operating parameters of Internal combustion engine determined and based on the operating profile and a recorded course for the operating parameter and / or the catalyst temperature the estimate for the expected future Course of the temperature of the nitrogen oxide reduction catalyst made. That way, the reliability can be the extrapolation of the catalyst temperature can be further improved. additionally can the extrapolation length be enlarged.

In a further embodiment of the method, the increased metering rate of the reducing agent is set when the estimate for the expected future course of the temperature of the nitrogen oxide reduction catalyst shows that the temperature of the nitrogen oxide reduction catalyst falls within a predefinable time interval from the current time in the second, lower temperature range. In this way, the NH ₃ level is at least temporarily raised before the reductant supply is stopped. As a result, NH ₃ stored in the catalyst is increasingly available, so that a reducing agent dosing interval can be at least partially bridged. The catalyst has an improved activity due to the increased NH ₃ level even at the low temperatures of the second, lower temperature range. Therefore, nitrogen oxides contained in the exhaust gas can be reacted even at low temperatures.

In a further refinement of the method, the extent of the increase in the metering rate is set as a function of the expected time of falling below the upper limit of the second temperature range and / or the lowest value of the temperature of the nitrogen oxide reduction catalytic converter to be expected in the time interval from the current time point. In this way, the NH ₃ level can be optimally adapted to the expected catalyst temperature.

advantageous embodiments The invention is illustrated and illustrated in the drawings described below. Here are the above and to be explained below Features not only in the specified feature combination, but also usable in other combinations or in isolation, without to leave the scope of the present invention.

there demonstrate:

1 a schematic block diagram of an internal combustion engine with associated exhaust system,

2 a diagram for illustrating the NH ₃ storage capacity of a nitrogen oxide reduction catalyst,

3 a diagram to illustrate the temperature dependence of the NH ₃ storage capability,

4 a diagram to illustrate the dependence of NO _x conversion and NH ₃ -sleep from the NH ₃ level of a nitrogen oxide reduction catalyst,

5 a diagram to illustrate the dependence of NO _x conversion and Reduktionsmittelaufbereitungsgrad of the temperature,

6 a schematic block diagram representation of an advantageous embodiment of a unit for NH ₃ level determination and

7 a time chart illustrating a preferred approach in determining a catalyst temperature.

1 shows an example of a schematic block diagram of an internal combustion engine 1 with associated emission control system. The internal combustion engine 1 is preferably designed as an air-compressing internal combustion engine, hereinafter referred to simply as a diesel engine. That from the diesel engine 1 Exhaust gas emitted is from an exhaust pipe 2 taken and flows through a formed as a nitrogen oxide reduction catalyst exhaust gas catalyst 3 , Input and output side of the catalyst 3 are temperature sensors 4 . 5 in the exhaust pipe 2 arranged. Further, in the exhaust pipe 2 upstream of the first temperature sensor 4 a metering valve 13 arranged to deliver a reducing agent in the exhaust gas. The supply of the metering valve 13 with the reducing agent takes place from a container 11 , Without limiting the generality, it is assumed below that the reducing agent is an aqueous urea solution. In the hot exhaust gas, the actually effective reducing agent NH _{3 is} released by thermolysis and / or hydrolysis from the urea, which selectively acts with respect to the reduction of the nitrogen oxides contained in the exhaust gas. Accordingly, the nitrogen oxide reduction catalyst 3 preferably formed as a classic SCR full catalyst based on V ₂ O ₅ / WO ₃ or as a zeolitically coated SCR supported catalyst. Downstream of the catalyst 3 is also a sensitive to NH ₃ and / or NO _x exhaust gas sensor 6 arranged, with the help of residual portions of NH ₃ and / or NO _x can be detected in the exhaust gas.

The temperature sensors 4 . 5 as well as the dosing valve 13 are via control or signal lines 12 with a central control unit 7 connected. The control unit 7 is also via another line 10 with the diesel engine 1 connected. About this line 10 receives the control unit 7 Information about operating state variables of the diesel engine 1 , This can be, for example, information about the delivered torque or the speed. The control unit 7 preferably comprises a computing unit and a memory unit and an input-output unit, which is not shown in detail. This is the control unit 7 able to perform complex signal processing operations and the operation of the diesel engine 1 and the exhaust system to detect and control or regulate. Necessary maps for this purpose are preferably stored in the memory unit, wherein an adaptive adaptation of the maps may be provided. The maps relate mainly to the relevant state variables of the exhaust gas, such as mass flow, raw emissions, temperature as a function of the operating state variables of the diesel engine 1 such as load, speed, air ratio, etc. Furthermore, maps for the relevant state variables of the catalyst 3 , such as nitrogen oxide conversion, NH ₃ storage capability, and the like.

For particularly regulated control of the metering valve 13 and for adjusting an addition of the urea solution is a metering unit 8th provided, which here exemplifies part of the control unit 7 is, however, can also be designed as an independent unit. The same applies to a level detection unit 9 which mainly determine the in the catalyst 3 stored NH ₃ amount is used, which will be discussed in more detail below.

It is envisaged that the in 1 sketched attachment further here for clarity not shown components or may have. For the diesel engine 1 Here are examples of charging units, exhaust gas recirculation units or fuel injectors, etc. to call. In the exhaust pipe 2 For example, additional components such as an additional oxidation catalyst, a particulate filter, exhaust gas sensors, etc. may be included.

The following will be with reference to the 2 to 4 various parameters of a typical SCR catalyst explained. It is in 2 a diagram illustrating the NH ₃ storage capability is shown.

The diagram of 2 represents a typical time course c _NH3 (t) of NH ₃ concentrations c _NH3 when the SCR catalyst is charged with NH _3. It is assumed that an NH ₃ -free SCR catalyst is stored under isothermal conditions at time t1 with an exhaust gas inlet flow of predetermined and constant in time size and NH ₃ input concentration is applied, which by the track 20 is reproduced. In accordance with its NH ₃ storage capacity, the SCR catalyst increases in time in the time interval between t1 and t2 in decreasing amounts to NH ₃ . Accordingly, the NH ₃ concentration in the exhaust gas stream leaving the SCR catalyst remains below the input concentration, which is due to the trace 21 is reproduced. At time t2, the SCR catalyst is saturated, which is why it can no longer store any further NH ₃ and the trace 21 in the track 20 opens. The NH ₃ level has then reached the maximum value of 100%. The amount of NH ₃ stored by the SCR catalyst, which represents the NH ₃ storage capacity under the appropriate conditions, is determined by the size of the area 22 between the two tracks 20 . 21 represents.

The NH ₃ storage capacity is primarily temperature dependent, which is due to the in 3 represented diagram is shown. This is the track 30 a typical curve of the temperature-dependent NH ₃ storage capacity Sp _NH3 (T). The NH ₃ storage capacity Sp _NH3 (T) is, as the Dia gram of 3 can be seen at low temperatures T comparatively large and decreases at high temperatures T, about above 300 ° C. In addition, there is a dependence on the gas flow rate, which is not shown in detail.

In this context it is assumed hereinafter that the NH ₃ indicating filling level of the SCR catalyst, the stored amount of NH ₃ with respect to the maximum storable under the respective conditions NH ₃ amount according to the illustrated ratios.

An important aspect related to the properties of a typical SCR catalyst relates to the dependence of the NO _x conversion on the NH ₃ level. In 4 , is through the track 40 this dependence is shown schematically. In comparison, is through the track 41 the dependence of the NH ₃ slip S _NH3 reproduced from the NH ₃ filling level. As can be seen, increases with increasing NH ₃ level F of NO _x conversion U _NOx (F) with flatter slope continuously up to a maximum value, which is determined essentially by the gas flow rate and the temperature. This means that from a certain value for the filling level F of the NH ₃ NO _x conversion U _NOx can not be increased by a further storage of NH ₃ in the catalyst. Rather, it increases, as if through the track 41 shown, the NH ₃ -slip S _NH3 . When setting an optimum for the respective conditions value for the NH ₃ level F, the consideration of these facts of importance.

An important aspect in connection with the achievable NO _x reduction relates to the temperature dependence of a degree of purification of the urea used. In 5 this is through the track 51 shown schematically. As can be seen, the degree of preparation A _Red is low at low temperatures. This is the result of an incomplete thermolysis or hydrolysis, which, however, is decisive for a release of the actually effective reducing agent NH ₃ . However, as the temperature rises, starting from low values, the treatment level A _Red rapidly increases to a maximum value and remains almost constant as the temperature increases further. For comparison, there is the track 50 the temperature dependence of the NO _x conversion U of a typical _NOx SCR catalyst in a NH ₃ slip predetermined size.

According to the invention specified the NO _x conversion U _NOx and the purification degree A _Red characterizing temperature ranges depending on the reductant used and the type of catalyst used. An optionally open top, upper temperature range T1 is preferably set so that in this temperature range T1, a high NO _x conversion U _NOX and an at least approximately maximum treatment level A _{Red are} given. An optionally downwardly open second, lower temperature range T2 is preferably set so that in this temperature range, a low NO _x conversion U _NOx and a low degree of conditioning _Red A are given. A third, average temperature range T3 lies between the first temperature range T1 and the second temperature range T2 and directly adjoins these temperature ranges T1, T2. Accordingly, the third temperature range T3 represents a transition temperature range in which at least partially a mean NO _x conversion U _NOx or _{degree of} preparation A _{Red is} given. To determine which of the temperature ranges T1, T2, T3 is currently relevant, the output signal of the first temperature sensor 4 and / or the second temperature sensor 5 evaluated. However, it may alternatively or additionally not in 1 represented, arranged in the catalyst interior temperature sensor can be used. A model-based temperature determination is also possible.

The in the 2 to 5 dependencies shown schematically are advantageously determined in advance for the SCR catalyst to be used and stored as curves or maps. On this can the control unit 7 access, so the condition of the catalyst 3 can be comprehensively determined for each operating state. It is preferably provided to provide a map set for an un-aged new state and another map set for a defined aging state, preferably in accordance with a predetermined limit aging. From suitable data, such as frequency and height of temperature peaks, catalyst aging can be determined and an aging factor can be determined in terms of limiting aging. This makes it possible to supply the reducing agent to the current state of aging of the catalyst 3 adapt.

Hereinafter, referring to 6 a preferred procedure for the determination and adjustment of the NH ₃ level explained in more detail. It is assumed that the temperature of the catalyst 3 in the first, upper temperature range T1.

In 6 is an advantageous embodiment of the level detection unit 9 schematically in block diagram Darge presents. Only the parts are shown, which are used to determine the NH ₃ level of the catalyst 3 be used. The level detection unit 9 receives input quantities E, which in addition to the current aging factor AF Exhaust gas state variables such as the exhaust gas temperature, the NO _x content and the exhaust gas mass flow relate. The level detection unit 9 has map sets 60 . 61 . 62 for the reaction with NO _x , the direct reaction with oxygen and the desorption rate of NH ₃ stored in the catalyst. The relevant data are determined according to the available input variables. The values for the reaction with NO _x, the direct reaction with oxygen and by the NH ₃ slip S _NH3 desorption rate are given together with the NH ₃ feed rate Z to a summation element 63 fed, which sums the relevant quantities with the correct sign. In this way, an accounting for the sizes is possible, which essentially determine the amount of NH ₃ stored in the catalyst. The sum value becomes an integrator 64 whose output is the current NH ₃ level F of the catalyst 3 represents. In addition, further output quantities A are determined from the available data. These are above all quantities relating to the NO _x content of the catalyst 3 Outflowing exhaust gas, the NH ₃ -slip S _NH3 and the possibly due to the heat of reaction or due to heat losses changed exhaust gas temperature. The NH ₃ feed rate Z of the catalyst 3 is largely determined by the over the metering valve 13 delivered urea dosing. If NH ₃ is added directly instead of urea, the metering rate corresponds to the NH ₃ feed rate. Preferably, there are losses due to wall absorption or wall deposition in the exhaust pipe 2 , incomplete preparation and the like.

From the determined NH ₃ level F, it is again determined whether an addition of urea solution should take place or the dosing rate is calculated. This is preferably done by the controller of the dosing unit 8th to which the NH ₃ level F can be fed as a controlled variable. It is envisaged to regulate the NH ₃ level between a respective predeterminable lower and upper threshold. The thresholds are set so that the highest possible NO _x conversion U _NOx and a predeterminable NH ₃ slip S _NH3 arise between them.

Is the temperature of the catalyst 3 in the middle temperature range T3, the NH ₃ feed rate Z or the urea metering rate is also determined temporarily in accordance with the procedure explained above. According to the invention, however, an additional estimate is provided for an expected temperature profile of the catalyst in this case 3 make. Depending on the result of the estimation, this procedure is maintained or, on the other hand, a higher urea dosing rate or an increased NH ₃ target level is set. Hereinafter, referring to 7 a preferred procedure explains this.

In 7 is a schematic diagram with an exemplary time course for a temperature T of the catalyst 3 shown. It is assumed that at equidistant times t _n a determination of the catalyst temperature, for example by evaluation of the signals of the temperature sensors 4 . 5 is made. The temperature determination can take place, for example, at intervals of a few seconds. In the case illustrated, t _{0 represents} a fictitious current time. Accordingly, the times t_ ₁ , t _{-2 represent} past points in time. The determined temperatures result in a curve branch 70 illustrated temperature profile. It should be noted here that the temperatures determined at the time t ₀ , t _-1 , t _-2 fall within the third temperature range T3, at which the temperature ranges T1 and T2 follow upwards or downwards in accordance with the above explanations , It is assumed here that in the time interval t_ ₂ <t <t _0, an addition of urea into the exhaust gas is set according to a dosing rate was as they would have resulted from the presence of falling within the temperature range T1 temperatures.

According to the invention, an estimate for the expected future course of the temperature is carried out on the basis of the temperatures determined at the times t ₀ , t_ ₁ , t_ ₂ . This estimate leads to a through the branch 71 reproduced extrapolated temperature profile of the past temperature profile (Kurvenast 70 ) and extends over a time interval .DELTA.t of preferably predeterminable size. If the estimate that, as shown in the present case, the catalyst temperature T in the time interval .DELTA.t falls in the second, lower temperature range T2, the metering rate is increased by a predetermined amount. This increased dosing rate is maintained until it is determined that the temperature T determined for the current time falls within the second, lower temperature range T2. At this point, the addition of urea is stopped. According to the estimate, this is the case at time t ₁ . However, measurement may result in a correction of this estimated time. The dosage is resumed when the temperature returns to the third temperature range T3.

As regards the extent of the increase in the metering rate, the increase is preferably carried out only to the extent that an associated NH ₃ slip does not exceed a predetermined limit value. However, this limit can be raised at least temporarily above a usual value of, for example, 10 ppm. Furthermore, additionally or alternatively, the current NH ₃ level can be taken into account. The increase in the metering rate is preferably selected so that a predeterminable NH ₃ limit level of, for example 95 is not exceeded. When the NH ₃ level exceeds the NH ₃ level, a prohibition of an increased metering rate can also be provided. In this context, it is further advantageous if the slope of an estimated temperature decrease is taken into account. If, for example, the estimate indicates that the catalyst temperature T dips very deeply into the second temperature range T2 in a comparatively short period of time (eg 5 s), then only a slight increase in the metering rate is preferably set. If a comparatively large negative temperature gradient of about minus 0.5 K / s or steeper is expected, then a ban for an increased metering rate can also be provided. It can advantageously further include a consideration of the distance .DELTA.T of time _t0 detected temperature from the upper limit of the second temperature range T2 to be considered.

With regard to the estimation method used for the expected future temperature profile, various possibilities are considered which can increase the reliability of the estimation. Preferably, the interval length .DELTA.t for the prediction is linked to relevant parameters. For example, statistical statements regarding the probabilities for an application can be obtained for this purpose. These can result, for example, from a regression coefficient for an interpolated curve, which is defined by the temperature values determined at the times t ₀ , t _-1 , t _-2 . The number of past times considered for an estimate may be increased to ten or more. Depending on the dynamics of the current engine operating state, the times may be selected narrower or farther apart.

It may be further provided, the temperature estimate in addition or alternatively based on a heat model for the Make catalyst. The heat model preferably describes the thermal behavior of the catalyst. For this can Energy balances concerning a heat input and a heat emission via the exhaust gas, a heat dissipation over a Catalyst outer wall and other thermo-technical Phenomena become. Therefor required data such as for the exhaust gas mass flow and its time course are usually available anyway. Further Data such as heat transfer coefficients can be found in be kept in a data store.

Preferably Several operating profiles are determined and stored and the Course of the corresponding operating parameter or the catalyst temperature compared with the company profiles.

Of the Operating parameter is preferably a size that in the vehicle anyway recorded or measured. It can be a current operating point the combustion device characterizing parameters, for example a speed and / or an injection quantity of fuel and / or the vehicle speed act. The operating profile includes while expedient and frequency distributions of the respective operating parameter. The operating profile also expediently contains statistical transition probabilities for individual Operating conditions, how they over to occur over time.

A further increase the reliability of Temperature curve estimation can by consideration typical operating profiles for the catalyst temperature and / or engine or exhaust parameters be achieved. For this, the corresponding sizes are preferably continuously recorded and history patterns determined. The gradient patterns can with stored operating profiles are compared. It can the Operating profiles assigned probabilities for their occurrence will.

In In this context, it is also possible from the in the vehicle collected information to determine a driving condition, the the determination of the operating profile is taken into account. In an advantageous manner Way the driving condition includes a division into different, the driving condition characterizing classes that control the operation of the Represent vehicle. desired Class divisions can doing so using logical links, experience based assignments or neural networks. A possible For example, classes can be classified as low-load (Idle mode / congestion), part load (city traffic) include. To the division Driving conditions in different classes information can be used derived from engine management parameters, such as the Engine speed and the time of fuel injection. As well can In the transmission management detected variables, such as a current gear ratio and switching thresholds and the times of switching between the individual gear positions are used for it. For example found that of a low gear ratio in such with higher translation is changed, so can typically with this event connected and stored temperature changes used become.

Claims (7)

A method of operating a nitrogen oxide reduction catalyst with ammonia storage capability, the exhaust of a predominantly lean plants nen internal combustion engine ( 1 in which ammonia or a substance from which ammonia can be liberated is added to the exhaust gas at an adjustable metering rate as the reducing agent, and in which - in a first, upper temperature range (T1) of the nitrogen oxide reduction catalyst ( 3 ) an ammonia level (F) in the nitrogen oxide reduction catalyst ( 3 ) in particular by summation and temporal integration of at least one conversion rate of in the nitrogen oxide reduction catalyst ( 3 ) converted ammonia and / or nitrogen oxide and a feed rate (Z) from the nitrogen oxide reduction catalyst ( 3 ) supplied ammonia is determined and the metering rate of the reducing agent is determined and adjusted such that a predefinable setpoint for the ammonia level in the nitrogen oxide reduction catalyst ( 3 ) is at least approximately reached, - in a second, lower temperature range (T2) of the nitrogen oxide reduction catalyst ( 3 ) the addition of reducing agent is stopped and - in a third, medium temperature range (T3) of the nitrogen oxide reduction catalyst ( 3 ), which directly adjoins the first temperature range (T1) and the second temperature range (T2), the dosing rate of the reducing agent determined analogously to the first temperature range (T1) is increased at least temporarily by a predeterminable amount. A method according to claim 1, characterized in that at least in the third temperature range (T3) of the nitrogen oxide reduction catalyst ( 3 ) an estimate of an expected future course of the temperature (T) of the nitrogen oxide reduction catalyst ( 3 ) is made. A method according to claim 2, characterized in that a temperature (T) of the nitrogen oxide reduction catalyst ( 3 ) is continuously determined and the estimate of the expected future course of the temperature (T) on the basis of a current temperature value and / or at least one previously determined temperature value is made. A method according to claim 2 or 3, characterized in that for the estimation of the expected future course of the temperature (T) of the nitrogen oxide reduction catalyst ( 3 ) a heat model for the nitrogen oxide reduction catalyst ( 3 ) is used. Method according to one of claims 2 to 4, characterized in that an operating profile for the temperature of the nitrogen oxide reduction catalyst ( 3 ) and / or at least for one the temperature (T) of the nitrogen oxide reduction catalyst ( 3 ) influencing operating parameters of the internal combustion engine ( 1 ) is determined and based on the operating profile and a recorded course for the operating parameter and / or the temperature (T) of the nitrogen oxide reduction catalyst ( 3 ) the estimate for the expected future course of the temperature (T) of the nitrogen oxide reduction catalyst ( 3 ) is made. Method according to one of claims 2 to 5, characterized in that the increased dosing rate of the reducing agent is set when the estimate for the expected future course of the temperature (T) of the nitrogen oxide reduction catalyst ( 3 ) shows that the temperature (T) of the nitrogen oxide reduction catalyst falls within a predeterminable time interval (Δt) from the current time (t ₀ ) in the second, lower temperature range (T2). A method according to claim 6, characterized in that the degree of increase in the dosing rate in dependence on the expected time of falling below the upper limit of the second temperature range (T2) and / or the lowest value of the time interval (At) from the current time (t ₀ ) expected temperature (T) of the nitrogen oxide reduction catalyst ( 3 ) is set.