HK1234464B - A method for aging determination, a method for ash detection, control device and internal combustion engine - Google Patents

Description

Method for determining aging, method for identifying ash, control device, and internal combustion engine

Technical Field

The invention relates to a method for determining the aging of an oxidation catalytic converter in an exhaust gas aftertreatment system of an internal combustion engine, a method for detecting ash in a particle filter in an exhaust gas aftertreatment system of an internal combustion engine, a control device, and an internal combustion engine.

Background Art

Exhaust gas aftertreatment systems for internal combustion engines typically include a particle filter, which is configured to clean the exhaust gas emitted by the internal combustion engine from soot particles (sometimes referred to as carbon black particles). The loading of the particle filter is determined based on a loading model by measuring the differential pressure drop across the particle filter. If passive regeneration of the particle filter due to nitrogen dioxide in the exhaust gas is insufficient, for example when the internal combustion engine is operated for a long time in the low-load range or at idle, active regeneration of the particle filter is performed (particularly depending on the determined loading), during which the soot particles are oxidized. It has been shown that the differential pressure level across the particle filter increases slowly over the service life of the exhaust gas aftertreatment system. This increase is caused by two independent effects: firstly, increasing ash deposits in the particle filter—components that cannot be burned but instead remain permanently in the particle filter and, through their inclusion (sometimes referred to as clogging) in the particle filter, increase the differential pressure level. Secondly, aging of the oxidation catalyst, located upstream of the particle filter in the direction of the exhaust gas flow, leads to a reduced formation of nitrogen dioxide in the exhaust gas. Nitrogen dioxide is typically used as an oxidizing agent in the particle filter to oxidize soot particles even in operating states of the internal combustion engine in which the exhaust gas temperature is insufficient to promote oxidation due to the residual oxygen concentration remaining in the exhaust gas. The decreasing nitrogen dioxide concentration associated with increasing aging of the oxidation catalyst leads to reduced regeneration, so that the differential pressure level increases because the particle filter no longer regenerates to the same extent as when the oxidation catalyst is new.

The two effects (i.e., increased ashing of the particle filter on the one hand and aging of the oxidation catalytic converter on the other) cannot be separated from each other using differential pressure measurement alone. Therefore, it is generally not possible to determine from the differential pressure increase whether only increased ashing of the particle filter is occurring or whether the oxidation catalytic converter is also showing aging effects. Consequently, only an inadequate correction of the loading model can be achieved due to the increase in differential pressure, as it is not possible to distinguish between the contributions of the two effects to the differential pressure increase. However, in order to consistently reliably determine the loading of the particle filter with soot particles, the loading model must take into account both the degree of particle filter ashing and, on the one hand, the regeneration of the particle filter.

Conventional methods for ash detection are performed at operating points of the internal combustion engine where the aging of the oxidation catalyst has no effect on the soot burn rate (sometimes referred to as soot burn rate) in the particle filter. This is typically the case at the rated power of the internal combustion engine and at high exhaust gas temperatures, where soot particles are oxidized by the residual oxygen concentration in the exhaust gas. Obviously, this method does not yield any information about the aging of the oxidation catalyst, as the aging of the oxidation catalyst has no effect on the soot burn rate at the operating points under consideration. Consequently, the matching of the loading model to the slowly rising differential pressure level inevitably remains inaccurate.

Summary of the Invention

The object of the present invention is to provide a method for determining the aging of an oxidation catalytic converter in which the aforementioned disadvantages do not occur. Furthermore, the object of the present invention is to provide a method for detecting ash in a particle filter in which the aforementioned disadvantages do not occur and in which, in particular, a distinction can be made between the effects of ash formation of the particle filter on the one hand and the effects of aging of the oxidation catalytic converter on the other hand. Furthermore, the object of the present invention is to provide a control device and an internal combustion engine in which the aforementioned disadvantages likewise do not occur.

This object is achieved by creating a method for determining the aging of an oxidation catalyst in an exhaust gas aftertreatment system of an internal combustion engine. The method comprises the following steps: determining a soot combustion rate of a particle filter of the exhaust gas aftertreatment system. Adapting a function with at least one adaptation parameter to the determined soot combustion rate as a function of at least one variable, wherein the value of the adaptation parameter for observing the soot combustion rate as a function of the variable is dependent on the aging of the oxidation catalyst. The aging of the oxidation catalyst is determined based on the value of the at least one adaptation parameter determined by the adaptation function. It has been recognized that when observing the soot combustion rate as a function of a variable for which the value of the adaptation parameter of the function to be adapted is dependent on the aging of the oxidation catalyst, the soot combustion rate, and thus the regeneration rate of the particle filter, can be used to determine the aging of the oxidation catalyst. Thus, plotting the soot combustion rate against the variable (in a mathematically formulated manner) yields a function that is parametrically dependent on the aging of the oxidation catalyst. For this purpose, the soot combustion rate is preferably determined at an operating point or operating range of the internal combustion engine in which such a parametric relationship is present. Because ashing has no effect on the soot combustion rate (which, on the contrary, depends only on the parameters of the oxidation catalyst's aging state), it is possible to determine the oxidation catalyst's aging in this way. This allows for the separation of the effects described above, which in turn leads to improved ashing detection and a better adaptation of the loading model to the actual current state of the exhaust gas aftertreatment system. Aging determination can also be used for demand-based replacement of the oxidation catalyst. This eliminates the need for preventive replacement, which brings cost advantages, particularly reduced maintenance costs.

The method is preferably carried out for an exhaust gas aftertreatment system in which a particle filter is arranged downstream of an oxidation catalyst, so that the oxidizing agent, in particular nitrogen dioxide, generated in the oxidation catalyst can be used for regeneration of the particle filter. Alternatively or additionally, the method is preferably carried out for an exhaust gas aftertreatment system in which the particle filter is catalytically coated with an oxidation catalyst material. In this case, an oxidation catalyst upstream of the particle filter can be omitted.

The term “oxidation catalytic converter” therefore generally denotes here and hereinafter a device for catalyzing oxidation reactions in exhaust gases, in particular a separate oxidation catalytic converter and/or a corresponding catalytic coating of an element of an exhaust gas aftertreatment system, in particular a particle filter.

An embodiment of the method is preferred, which is characterized by determining the aging of the oxidation catalyst by comparing at least one adaptation parameter with at least one previously determined characteristic value. If the function to be adapted has only one adaptation parameter, this adaptation parameter is preferably compared with a threshold value or limit value. It is possible to store a plurality of threshold values or limit values, each of which is characteristic of a specific aging level of the oxidation catalyst, thereby enabling a very precise determination of the aging of the oxidation catalyst (depending on the specific number and density of the threshold values or limit values). If a function with more than one adaptation parameter is used, these are preferably compared with a parameter set stored in a characteristic map to determine the aging of the oxidation catalyst. Preferably, the threshold value or limit value or the characteristic map is fixedly implemented or stored for a specific exhaust gas aftertreatment system or a specific internal combustion engine, particularly preferably in a control unit configured to carry out the method, for example, in a control unit of the internal combustion engine.

Also preferred is an embodiment of the method that is distinguished by adapting the function to the soot combustion rate as a function of the particle filter temperature and the nitrogen oxide concentration in the exhaust gas. Thus, in this case, the soot combustion rate is observed as a function of the particle filter temperature, on the one hand, and the nitrogen oxide concentration in the exhaust gas, on the other hand, and the function is adapted to the soot combustion rate by varying at least one adaptation parameter. It is recognized that the soot combustion rate depends on the aging state of the oxidation catalyst within a temperature range in which particle filter regeneration occurs substantially or even completely by oxidation of soot using nitrogen dioxide as a reducing agent, preferably ranging from at least 250° C. to a maximum of 450° C. The soot combustion rate depends on the temperature in the particle filter, in particular the temperature of the exhaust gas in the particle filter, and on the other hand, the nitrogen oxide concentration in the exhaust gas, wherein the nitrogen oxide concentration indicates how much nitrogen dioxide is available as a reducing agent. The term "nitrogen oxide concentration" herein and hereinafter always refers to the total nitrogen oxide concentration, which is the sum of the concentrations of nitrogen monoxide and nitrogen dioxide. It should be noted that the nitrogen oxides produced by the internal combustion engine upstream of the oxidation catalyst consist approximately only of nitrogen monoxide, which is oxidized to nitrogen dioxide in the oxidation catalyst (particularly depending on the aging state of the oxidation catalyst). The nitrogen oxide sensor preferably used only measures the total nitrogen oxide concentration. If the soot combustion rate is now observed as a function of temperature on the one hand and nitrogen oxide concentration on the other hand, it is shown that the soot combustion rate is parametrically dependent on the proportion of nitrogen dioxide at the nitrogen oxide concentration, which in turn depends on the aging state of the oxidation catalyst. Therefore, by adapting the function when at least one adaptation parameter is changed, it is possible to obtain information about the aging of the oxidation catalyst. That is, in an aged oxidation catalyst, the oxidation of nitrogen monoxide to nitrogen dioxide occurs with reduced efficiency, so that the aged oxidation catalyst produces less nitrogen dioxide than an oxidation catalyst in its new state. Therefore, in general, the nitrogen dioxide production in the oxidation catalyst also decreases as the oxidation catalyst gradually ages. This can be easily understood based on at least one adaptation parameter.

An embodiment of the method is also preferred, which is distinguished by adapting a linear function to the soot combustion rate. Alternatively or additionally, it is possible to adapt a quadratic or cubic polynomial to the soot combustion rate. If a linear function is adapted to the soot combustion rate, the linear function has exactly one adaptation parameter. A quadratic polynomial has two adaptation parameters. Finally, a cubic polynomial has three adaptation parameters. Alternatively or additionally, it is also possible to adapt a higher-order polynomial to the soot combustion rate, which then has correspondingly more adaptation parameters. Regarding the dependence of the soot combustion rate on temperature, it has been determined that this dependence can typically be well described by a cubic polynomial. However, it is also possible that a quadratic polynomial or even a linear function is sufficient for a sufficiently accurate description. The soot combustion rate typically depends linearly on the nitrogen oxide concentration. The functions mentioned here are generally very simple functions that can also be easily adapted to the soot combustion rate (preferably using the least spring squares method). This is very fast and possible with low computing power. The functions mentioned here simultaneously provide sufficiently high accuracy so that more complex functions are not necessary, which are more time-consuming and more expensive in terms of computing power during adaptation.

The temperature of the particle filter is preferably determined by measurement. In particular, a first temperature sensor is preferably arranged upstream of the particle filter, with a second temperature sensor being arranged downstream of the particle filter. The particle filter temperature is then calculated as the average value of the two temperature measurements upstream and downstream of the particle filter. The temperature measurement can alternatively be performed using a temperature sensor that directly measures the temperature in the particle filter and is suitably arranged on the particle filter. The nitrogen oxide concentration is preferably determined (as already indicated) using a nitrogen oxide sensor, which is preferably arranged directly downstream of the internal combustion engine and preferably also upstream of the oxidation catalyst. Alternatively, the nitrogen oxide concentration can be determined from a map, in particular as a function of at least one operating parameter of the internal combustion engine.

Also preferred is an embodiment of the method that is distinguished by the following steps: determining the soot burn rate by determining the differential pressure drop across the particle filter. Determining the loading of the particle filter from the differential pressure according to a loading model. Furthermore, determining the soot input into the particle filter (particularly the soot input rate). Finally, determining the soot burn rate from the loading and the soot input. This makes it possible to determine the soot burn rate both simply and precisely.

The differential pressure across the particle filter is preferably measured using a differential pressure sensor, wherein a first measuring point of the differential pressure sensor is arranged directly upstream of the particle filter and a second measuring point is arranged directly downstream of the particle filter. Alternatively, it is possible to calculate the differential pressure as the difference in measured values from two pressure sensors, wherein the first pressure sensor is arranged directly upstream of the particle filter and the second pressure sensor is arranged directly downstream of the particle filter. The difference in the measured values of the first and second sensors then provides the expected differential pressure drop across the particle filter in a manner similar to a measurement performed using a differential pressure sensor.

The loading of the particle filter is preferably determined in a time-dependent manner using a loading model. The loading model is preferably ash-corrected, that is, in particular adapted to the current ash state of the particle filter. In this way, the loading model is designed to always be able to calculate the loading of the particle filter with soot particles from the differential pressure with the highest possible accuracy, without errors caused by ash.

The soot input into the particle filter is preferably calculated as a soot input rate from at least one operating parameter of the internal combustion engine. At least one operating parameter is used that characterizes conditions in the combustion chamber of the internal combustion engine that are important for soot formation. The operating parameter is preferably selected from the group consisting of fuel injection quantity, injection time, lambda probe measurement value, exhaust gas recirculation rate, throttle valve position, and internal combustion engine speed. The soot input rate is preferably calculated in a control unit or determined based on one or more performance maps, wherein the control unit is configured to carry out the method. In particular, the soot input rate is preferably calculated in a controller of the internal combustion engine. A model is used that describes the soot input rate as a function of at least one operating parameter of the internal combustion engine. Preferably, a plurality of operating parameters are used to calculate the soot input rate.

It has been shown that the soot input rate is preferably calculated independently of the differential pressure across the particle filter. Thus, the loading of the particle filter on the one hand and the soot input rate on the other hand are determined independently of one another using two different independent models.

The temporal development of the particle filter's loading depends on the soot input rate, on the one hand, and on the soot burn rate, on the other. In particular, the time-dependent derivative of the loading is derived as the sum of the soot input rate and the soot burn rate. Therefore, it is possible to calculate the soot burn rate from the particle filter's loading, on the one hand, and the soot input rate, on the other. If the loading and the soot input rate are determined using independent models, a very precise calculation of the soot burn rate results, since no redundant information is used, but rather complementary information.

An embodiment of the method is preferred, which is distinguished by calculating the soot combustion rate using a Kalman filter. The Kalman filter offers a particularly advantageous option for calculating the soot combustion rate using low computing power, preferably in real time, from the load on the one hand and the soot input rate on the other. In addition to the load on the particle filter and the soot input rate, the Kalman filter preferably also includes an unfiltered measured value of the differential pressure across the particle filter. In this case, the Kalman filter also provides the filtered progression of the differential pressure and the filtered load on the particle filter, determined from the filtered differential pressure progression, for the soot combustion rate. Furthermore, the Kalman filter preferably includes at least one error estimate that enables error correction in the Kalman filter. The error estimate specifically considers whether the internal combustion engine is operating at that moment (i.e., in which the pressure differential value may not be effective), whether the incoming measured values, particularly the incoming differential pressure values, have a reliable evaluation range, and/or the extent of dependent measurement noise. Preferably, errors in the differential pressure measurement, particularly those that depend on the operating point of the internal combustion engine, are incorporated into the error estimate of the Kalman filter. If the error estimate indicates that a reliable calculation of the soot combustion rate based on the variables fed into the Kalman filter is not currently possible, an extrapolation of the previously calculated soot combustion rate is preferably performed. This is readily possible with the Kalman filter, as the Kalman filter can assume internal states, which sometimes do not take into account currently fed values. This makes error correction with the aid of the Kalman filter possible without any problems.

Alternatively, an embodiment of the method is preferred in which the soot combustion rate is calculated by loading and back-integrating the soot input rate. This may require less computing power than a Kalman filter, but error correction is rarely possible.

Also preferred is an embodiment of the method that is distinguished by evaluating the temporal development of the soot combustion rate. Fault detection is performed based on this temporal development. This makes it possible to distinguish different fault states from the typically occurring, slow aging of the oxidation catalyst. For example, it is possible to detect a sudden change in the soot combustion rate, from which it is possible to infer damage to or removal of the oxidation catalyst, or a sensor fault. It is also possible to use a specific form of change in the soot combustion rate to identify a specific fault. In contrast, a slow, continuous change in the soot combustion rate is usually expected to occur due to conventional thermal aging and/or poisoning of the oxidation catalyst.

Particularly preferred is an embodiment of the method that is distinguished by evaluating the temporal evolution of at least one adaptation parameter as the temporal evolution of the soot combustion rate. This approach is distinguished by a particularly low computational effort, since only a small amount of data must be observed, possibly even only the temporal evolution of a single parameter value. This makes it very easy to distinguish suddenly occurring fault states from normal, slow aging of the oxidation catalyst.

The object is also achieved by creating a method for detecting ash in a particle filter of an exhaust gas aftertreatment system. Here, the differential pressure drop across the particle filter is detected. This is preferably performed using a differential pressure sensor or two sensors, with the first sensor being arranged upstream of the particle filter and the second sensor being arranged downstream of the particle filter, with the difference in the measured values of the two sensors being formed as the differential pressure, as described in more detail above. Using the embodiments of the method described above, the aging of an oxidation catalytic converter of the exhaust gas aftertreatment system is determined. A loading model for the particle filter is adapted to the aging of the oxidation catalytic converter. This includes not adapting the loading model if it is determined that the oxidation catalytic converter is not aged or is not further aged. Furthermore, it is possible that, when the method is initially executed using a like-new oxidation catalytic converter, the aging of the oxidation catalytic converter is not determined using the method described above, but rather the method is initially initialized using values corresponding to no aging of the oxidation catalytic converter. However, in subsequent execution of the method, the aging determination is performed as described above. The loading of the particle filter is determined based on a modified loading model (which, if applicable, includes an unchanged loading model if the oxidation catalytic converter is not aged). Finally, the particle filter ashing is determined from the loading on the one hand and the differential pressure on the other. Since the loading model can now be adapted to the aging of the oxidation catalyst, it is possible to explicitly take this effect into account and thus separate it from the effects of the particle filter ashing. This can be determined without difficulty, because after correcting the loading model for the aging of the oxidation catalyst, any remaining deviations between the calculated loading value and the loading value actually to be expected based on the differential pressure value can be attributed to the ashing of the particle filter.

An embodiment of the method is preferred, which is distinguished by the fact that the loading model (in particular following the determination of the particle filter ashing) is adapted to the particle filter ashing. The loading model is then corrected not only with respect to the aging of the oxidation catalytic converter but also with respect to the particle filter ashing, so that a very precise loading value for the particle filter can be calculated using the loading model.

The loading model is preferably adapted to the aging of the oxidation catalyst and/or the ashing of the particle filter by correcting the measured differential pressure value as a function of the aging and/or ashing and then using the measured differential pressure value in corrected form as the basis for the loading calculation by the loading model. Alternatively or additionally, it is possible to adapt the target value or the characteristic curve for the differential pressure used within the loading model to the aging of the oxidation catalyst and/or the ashing of the particle filter. Furthermore, it is also possible, alternatively or additionally, to adapt the functional relationship between the loading of the particle filter and the differential pressure used within the loading model to the aging of the oxidation catalyst and/or the ashing of the particle filter.

Furthermore, an embodiment of the method is preferred, which is distinguished by initializing the method using an initial value for the ashing. It is preferably possible to use an initial value in a new state of the particle filter and/or the exhaust gas aftertreatment system, which initial value represents a characteristic of absent or no ashing, for example a value of 0. Alternatively or additionally, it is possible to determine the ashing according to known ashing detection algorithms, in particular for initializing the method, in particular at operating points at which aging of the oxidation catalyst has no influence, preferably at the rated load of the internal combustion engine and at an exhaust gas temperature at which the particle filter is completely regenerated by the residual oxygen content of the exhaust gas. In this way, ashing detection (in particular for initializing the method) can be performed at least once independently of determining the aging of the oxidation catalyst.

Also preferred is an embodiment of the method in which the ashing is repeatedly, preferably periodically, determined according to the known ashing detection method described above. The values thus obtained for the ashing of the particle filter are then used for error correction or adaptation of the ashing detection method according to the invention.

Finally, preferred is an embodiment of the method in which the method is performed repeatedly. It is particularly preferred that the method be performed continuously during operation of the internal combustion engine for ash detection, thereby continuously determining the aging of the oxidation catalyst and the ashing of the particle filter. Accordingly, the loading model is also continuously adapted to the aging of the oxidation catalyst and the ashing of the particle filter. Thus, accurate, current values for the loading of the particle filter are always available. Alternatively, it is possible to perform the method with time interruptions, preferably at predetermined time intervals. In this case, the loading model is preferably adapted to the ashing of the particle filter and the aging of the oxidation catalyst at predetermined times. This results in less computational effort than when the method is performed continuously. In this respect, the embodiment of the method is economical and less computationally intensive. The predetermined time intervals are preferably selected so that, despite only performing the method point by point, sufficiently accurate loading values are ensured using the modified loading model.

Finally, the object is also achieved by providing a control device configured to carry out the method according to the invention for determining the aging of an oxidation catalytic converter in an exhaust gas aftertreatment system of an internal combustion engine and/or the method according to the invention for detecting ash in a particle filter of an exhaust gas aftertreatment system. The combined device achieves the advantages already explained in conjunction with the method.

The control device is preferably designed as a control unit (Engine Control Unit – ECU) of the internal combustion engine. Alternatively, it is possible to design the control device as a separate control device that is specifically configured to perform aging determination and/or ash detection. In this case, it is possible for the control device to also take on other tasks.

By configuring the hardware structure of the control device in a suitable manner, the control device is preferably configured to carry out the method. Alternatively, it is possible to load a computer program product into the control device, the computer program product having commands based on which, when the computer program product is executed on the control device, at least one of the methods for determining aging and detecting embers in the control device can be carried out.

Finally, the object is also achieved by providing an internal combustion engine having a control device according to one of the above-described embodiments. In combination with the internal combustion engine, the advantages already explained in connection with the control device or in connection with the method are thereby achieved.

The internal combustion engine is preferably configured as a reciprocating engine. In a preferred embodiment, the internal combustion engine is used to drive particularly heavy land vehicles or water vehicles, such as mine cars, trains (where the internal combustion engine is used in locomotives or propulsion vehicles (Triebwagen)) or ships. It is also feasible to use the internal combustion engine to drive vehicles for defense, such as armored vehicles. Preferably, embodiments of the internal combustion engine are also static, such as for static energy supply in emergency power operation, continuous load operation or peak load operation, wherein the internal combustion engine preferably drives a generator in this case. Static applications of the internal combustion engine for driving auxiliary equipment (such as fire pumps on offshore drilling platforms) are also feasible. In addition, it is feasible to use the internal combustion engine within the scope of transporting fossil raw materials and especially fuels (such as oil and/or gas). It is also feasible to use the internal combustion engine in the industrial field or in the construction field, such as in construction machinery or construction machinery, such as in cranes or excavators. The internal combustion engine is preferably configured as a diesel engine or a gasoline engine.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention is explained in more detail below with reference to the drawings.

FIG1 shows a schematic diagram of an embodiment of an internal combustion engine;

FIG2 shows a schematic diagram of an embodiment of a method for aging determination;

FIG3 shows a diagram of the dependence of the soot combustion rate on the aging parameters of the oxidation catalyst, and

FIG4 shows a schematic diagram of an embodiment of a method for ash identification.

DETAILED DESCRIPTION

FIG1 shows a schematic diagram of an exemplary embodiment of an internal combustion engine 1 with an exhaust gas aftertreatment system 3. The exhaust gas aftertreatment system 3 has a particle filter 5 and (as viewed in the direction of the exhaust gas flow) an oxidation catalytic converter 7, which is arranged separately, upstream of the particle filter 5. Exhaust gas flows from a schematically illustrated engine region 9 through the exhaust gas aftertreatment system 3 toward an exhaust gas emission system (not shown in FIG1 ), the exhaust gas flowing from left to right in FIG1 and first passing through the oxidation catalytic converter 7 and then through the particle filter 5. Additionally or alternatively to the separate oxidation catalytic converter 7, it is possible for the particle filter 5 to have a catalytic coating that serves as an oxidation catalytic converter.

In order to detect the differential pressure dropping across the particle filter 5, a differential pressure sensor 11 is provided in the exemplary embodiment shown here, which has a first measuring point 13 upstream of the particle filter 5 and a second measuring point 15 downstream of the particle filter 5. The differential pressure sensor 11 is designed to measure the differential pressure between the first measuring point 13 and the second measuring point 15.

To determine the temperature of the particle filter 5, in particular the temperature of the exhaust gas passing through the particle filter 5, a first temperature sensor 17 is arranged upstream of the particle filter 5 and a second temperature sensor 19 is arranged downstream of the particle filter 5. The first temperature sensor 17 and the second temperature sensor 19 are designed to detect the exhaust gas temperature at their respective locations within the exhaust gas aftertreatment system 3. The temperature sensors 17, 19 are preferably operatively connected to a control device 21, which is designed to calculate the temperature of the particle filter 5, preferably as an average value of the temperature values detected by the temperature sensors 17, 19.

A nitrogen oxide sensor 23 is arranged upstream of the oxidation catalytic converter 7 in the exhaust gas aftertreatment system 3. This nitrogen oxide sensor 23 is configured to detect the total nitrogen oxide concentration in the exhaust gas flowing through the exhaust gas aftertreatment system 3. The term "total nitrogen oxide concentration," or simply "nitrogen oxide concentration," refers to the sum of the concentrations of nitrogen monoxide and nitrogen dioxide in the exhaust gas. Alternatively, it is possible that the nitrogen oxide sensor 23 is configured to specifically detect the concentration of nitrogen monoxide in the exhaust gas. This, in the position of the nitrogen oxide sensor 23 shown in FIG. 1 , results at most in a small deviation in the measured value from the measured value of a sensor that is not specifically sensitive to nitrogen oxide concentration. This indicates that nitrogen monoxide is essentially formed upstream of the oxidation catalytic converter 7 and is initially partially oxidized to nitrogen dioxide in the oxidation catalytic converter 7. Therefore, the nitrogen oxide concentration at the position of the nitrogen oxide sensor 23 shown in FIG. 1 essentially corresponds to the nitrogen monoxide concentration in the exhaust gas.

The control device 21 is operatively connected to the nitrogen oxide sensor 23, to the temperature sensors 17, 19, and to the differential pressure sensor 11. In this way, in particular, the measured values of the various sensors can be transmitted to the control device 21 for evaluation. If, in another embodiment, two pressure sensors are used instead of the differential pressure sensor 11 (one of which is arranged upstream of the particle filter 5 and the other downstream of the particle filter 5), the control device 21 is preferably operatively connected to both pressure sensors and is configured to calculate the differential pressure from the measured values of the two pressure sensors.

Furthermore, the control device 21 is configured to carry out a method for determining the aging of the oxidation catalytic converter 7 according to one of the embodiments described above and below. Alternatively or additionally, the control device 21 is preferably configured to carry out a method for detecting ash in the particle filter 5 according to one of the embodiments of the ash detection method according to the invention described above and explained in greater detail below.

FIG2 shows a schematic diagram of an embodiment of a method for determining the aging of an oxidation catalytic converter 7 . In a first method step S01 , the soot combustion rate of the particle filter 5 is determined. Input values for this first method step S01 include a differential pressure 25 dropping across the particle filter 5 , preferably determined by means of a differential pressure sensor 11 ; a load 27 on the particle filter 5 calculated from the differential pressure according to a load model; and a soot input rate 29 , preferably calculated by a control unit 21 based on at least one operating parameter of the internal combustion engine 1 . It is possible to use only the load 27 and the soot input rate 29 as input values for step S01 . In this regard, it is not absolutely necessary to explicitly use the differential pressure 25 for the first step S01 . Instead, it may be sufficient to calculate the load 27 from the differential pressure 25 according to the load model in a preceding step. However, if the differential pressure 25 is explicitly introduced into the first step S01, it can be smoothed (sometimes referred to as filtered) and/or filtered, so that a smoothed and/or filtered differential pressure 25' is generated as a result of the first step S01. In the same way, a smoothed and/or filtered load 27' is preferably also generated as a result of step S01, in particular a load 27' calculated based on the smoothed and/or filtered differential pressure 25'.

In step S01, the soot combustion rate 31 is calculated from the input values, in particular from the charge 27 and the soot input rate 29. This is supplied as an input value to the second step S02 of the method.

The first step S01 is preferably performed using a Kalman filter. This method is particularly suitable for real-time calculations, in order to enable reliable and accurate calculation of the soot burn rate quickly and with as little computing power as possible. Alternatively, it is possible to calculate the soot burn rate from the load 27 and the soot input rate 29 by back integration in the first step S01.

In particular, when a Kalman filter is used in step S01, an error estimate 33 is preferably entered into the Kalman filter as an additional input value. It is possible for more than one error estimate 33 to be entered into the first step S01. The error estimate 33 preferably takes into account errors in the differential pressure sensor 11 or, in general, in the measurement of the differential pressure, wherein this error is typically dependent on the operating point of the internal combustion engine 1. In particular, the error in the measurement of the differential pressure is typically greater in the instantaneous state of the internal combustion engine 1 than in a static operating point. Additionally or alternatively, the error estimate 33 preferably allows for a weighting of the input values entered into the first step S01, wherein, depending on the expected validity or accuracy of the input values, it is sometimes possible not to calculate the soot combustion rate 31 directly from the currently available input values, but rather to extrapolate the soot combustion rate 31 based on previously calculated values. This is particularly possible when, in first step S01, a Kalman filter is used in which the calculation of the soot combustion rate 31 can be shifted to internal states even with currently severely defective or less effective input values. This again preferably takes place in the instantaneous operating state of the internal combustion engine 1.

In second step S02, in addition to the soot combustion rate 31, the temperature 35 in the particle filter 5, preferably detected by means of temperature sensors 17 , 19 , and the nitrogen oxide concentration 37 in the exhaust gas, preferably detected by means of nitrogen oxide sensor 23 , are entered as further input variables. In second step S02, the soot combustion rate 31 is now estimated as a function of the temperature 35 and the nitrogen oxide concentration 37 , wherein a function is adapted to the soot combustion rate 31 by varying at least one adaptation parameter. The least spring squares method or another suitable method is preferably used to adapt the function. A preferably linear function, a quadratic polynomial, or a cubic polynomial is adapted to the soot combustion rate 31. It is therefore possible to vary more than one adaptation parameter to adapt the function to the soot combustion rate 31.

If a function with only one adaptation parameter is used, then in the second step S02, the value for the adaptation parameter is precisely generated from the adaptation of the function to the soot combustion rate 31. Otherwise, a set of values for different adaptation parameters is generated from the second step S02. These cases are considered together here, with the schematic representation that the adaptation parameter 39 is generated as a result of the second step S02. The adaptation parameter 39 can be a single value or a set or group of different values.

In the third step S03 of the method, the adaptation parameter 39 is evaluated, and the aging of the oxidation catalytic converter 7 is determined based on the adaptation parameter 39. Furthermore, the adaptation parameter 39 is preferably compared with at least one characteristic value, particularly preferably with a plurality of threshold values or limit values, or with a characteristic map. Accordingly, the aging 41 of the oxidation catalytic converter 7 is generated as a result of the third step S03. This aging 41 can be used for further methods, in particular for ash detection or ash correction of the particle filter 5.

In the third step S03, the temporal development 43 of the soot combustion rate is preferably also evaluated, in which the temporal development of the adaptation parameter 39 is particularly calculated. In the fourth step S04, based on the temporal development 43 of the soot combustion rate, it is preferably possible to perform fault detection and, in particular, distinguish normal, slow aging of the oxidation catalytic converter 7 from sudden changes caused, for example, by oxidation catalytic converter damage or, however, also by sensor failures. Particularly preferably, based on the specific course of the temporal development 43, it is possible to distinguish between different specific faults. In this regard, the fourth step S04 preferably results in a status 45 of the exhaust gas aftertreatment system 3, which indicates whether the exhaust gas aftertreatment system 3 is operating flawlessly or whether a fault, such as a damage to the oxidation catalytic converter 7 or a sensor failure, is present. The status 45 can then be used, for example, to output a warning to the operator of the internal combustion engine 1 and/or to initiate measures to overcome the fault. It is also possible to shut down the internal combustion engine 1 based on the evaluation of the status 45 in order to prevent damage or destruction of the internal combustion engine 1.

FIG3 shows a schematic and line diagram of the soot combustion rate, showing its dependence on the age of the oxidation catalyst. For a simpler illustration, the soot combustion rate is not plotted as a function of both the nitrogen oxide concentration and the exhaust gas temperature T, but rather solely as a function of the exhaust gas temperature T, resulting in a more intuitive, two-dimensional view. Here, a solid curve 47 plots the soot combustion rate against the exhaust gas temperature T in the particle filter 5 at a constant nitrogen oxide concentration, with the course of the solid curve 47 corresponding to the new state of the oxidation catalyst 7. The effects of oxidation catalyst aging 7 are shown here by arrow P, with a dashed curve 49 corresponding to the corresponding course of the soot combustion rate as a function of the exhaust gas temperature T in the particle filter 5 at a constant nitrogen oxide concentration for an aged oxidation catalyst 7. The aging of the oxidation catalyst 7 is evident in the fact that nitrogen monoxide is only still converted to nitrogen dioxide to a reduced extent, so that the ratio of the nitrogen dioxide concentration to the total nitrogen oxide concentration decreases with increasing aging of the oxidation catalyst. Correspondingly (as indicated by the arrow P), the soot combustion rate also decreases or, plotted against temperature, shows a flatter course at a constant nitrogen oxide concentration.

The soot combustion rate is preferably adapted using a linear function, a quadratic polynomial or a cubic polynomial, wherein in particular a cubic polynomial of the form:

Here, [NO _x ] is the nitrogen oxide concentration, T is the temperature of the exhaust gas in the particle filter 5 , and T ₀ is an offset parameter (offset parameter) which takes into account that soot combustion hardly occurs at a certain temperature. The offset parameter T ₀ is the temperature below which the soot combustion rate R is approximately zero. The offset temperature T ₀ is preferably from at least 200° C. to at most 300° C., particularly preferably 250° C. The parameters a, b, and c are adaptation parameters that are varied within the scope of the method in order to adapt the function according to equation (1) to the course of the soot combustion rate.

If a quadratic polynomial is used instead of a cubic polynomial according to equation (1), the third term with the parameter c is preferably simply omitted compared to equation (1). In this respect, only the parameters a, b can then be varied. If a linear function is used, this function preferably only has the first term of equation (1), so that only the adaptation parameter a is varied.

If the function according to equation (1) is adapted to the soot combustion rate, values for the adaptation parameters a, b, c are generated from the adaptation, which are then used further as a value set or as adaptation parameters 39 in the third step S03 according to FIG. 2 in order to determine the aging 41 of the oxidation catalytic converter 7 and/or to determine the temporal progression 43 of the soot combustion rate 31.

As is readily apparent from FIG. 3 , the exhaust gas temperature in the particle filter 5 is a variable for which the values of the adaptation parameters a, b, and c depend on the age of the oxidation catalyst 7. This applies similarly to the nitrogen oxide concentration [NO _x ], and in particular to the combination of the temperature in the particle filter 5 on the one hand and the nitrogen oxide concentration [NO _x ] on the other. Therefore, the values of the adaptation parameters a, b, and c characterize the ageing state of the oxidation catalyst 7, making it possible to readily detect this ageing state based on the corresponding values. This is possible in any operating state of the internal combustion engine 1 in which the exhaust gas temperature in the particle filter 5 is within a range in which soot combustion occurs at least substantially, preferably completely, through the reaction of soot particles with nitrogen dioxide as an oxidant. The temperature range that satisfies this condition is preferably at least 150°C to a maximum of 500°C, particularly preferably at least 250°C to a maximum of 450°C.

In the exhaust gas temperature range of the particle filter 5 (i.e., in which regeneration is essentially independent of the nitrogen dioxide concentration, particularly at the rated power of the internal combustion engine and high exhaust gas temperatures (especially above 450°C)), the aging determination outlined here is barely possible or impossible. However, it is possible to perform ash detection using conventional methods in the corresponding operating states, since the regeneration rate is independent of the aging of the oxidation catalyst. It is particularly preferred to combine the method according to the present invention for determining the aging of the oxidation catalyst with conventional methods for ash detection during operation of the internal combustion engine 1. This allows for the acquisition of complementary information about the state of the exhaust gas aftertreatment system 3, which increases the accuracy of both the ash detection and the aging determination of the oxidation catalyst.

4 shows a schematic diagram of an embodiment of the method for ash detection according to the present invention, in the form of a flow chart. In a first step S11, the differential pressure drop across the particle filter 5 is detected. In a second step S12, the aging of the oxidation catalyst 7 is determined using one of the above-described method embodiments.

In a third step S13 , the loading model is adapted to the aging 7 of the oxidation catalytic converter.

In a fourth step S14 , the loading of the particle filter 5 is determined based on the modified loading model, wherein in a fifth step S15 , the ashing of the particle filter is determined from the loading calculated based on the modified loading model on the one hand and the differential pressure on the other hand.

Finally, the loading model is preferably adapted to the determined ashing 5 of the particle filter in a sixth step S16 .

The method is preferably carried out repeatedly, so that after the sixth step S16 the method begins again in the first step S11 .

It is possible, in particular during the first operation, when using a new or cleaned particle filter 5 to remove soot, to initialize the method with initial values for the ash content, so that in this respect, the ash content is not then determined in the fifth step S15, but rather the initial values are used. Alternatively or additionally, it is possible (preferably at regular intervals) to perform ash detection according to known methods in operating states in which aging of the oxidation catalyst does not have an impact on soot combustion, wherein preferably in the fifth step S15, values for the ash content determined within the scope of conventional ash detection methods are then used instead of the values determined there within the scope of the ash detection method proposed herein. This can, if necessary, increase the accuracy of the method overall.

In particular, it is preferably possible to carry out ash detection according to conventional methods in operating states in which aging of the oxidation catalytic converter is irrelevant for soot combustion, and to carry out the method in the form proposed here according to the invention in all other operating states.

Overall, it has therefore been shown that with the aid of the method, the control device and the internal combustion engine proposed here it is possible to separate the effects of the ashing of the particle filter 5 on the one hand and the aging of the oxidation catalytic converter 7 on the other hand on the increase in the differential pressure level across the particle filter 5 and thereby to produce, in particular, an improved prediction of the aging determination of the oxidation catalytic converter 7 on the one hand and of the loading of the soot particle filter 5 on the other hand.

Claims (14)

1. A method for determining the aging of an oxidation catalyst (7) in an exhaust aftertreatment system (3) of an internal combustion engine (1), comprising the following steps: - Obtain the soot combustion rate (31) of the particulate filter (5) of the exhaust gas after-treatment system (3); - A function with at least one matching parameter (39) is matched to the soot combustion rate (31) by at least one variable, the value of which depends on the aging of the oxidation catalyst (7), and - The aging of the oxidation catalyst (7) is determined based on the value of the matching parameter (39) obtained by matching the function. The exhaust gas temperature in the particulate filter (5) is within a range in which soot combustion occurs at least substantially through the reaction of soot particles with an oxidant. 2. The method according to claim 1, characterized in that the aging of the oxidation catalyst (7) is determined by comparing the at least one matching parameter (39) with at least one predetermined characteristic value. 3. The method according to any one of the preceding claims, characterized in that the function is matched with the soot combustion rate depending on the temperature (T) of the particulate filter (5) and the concentration of nitrogen oxides in the exhaust gas. 4. The method according to claim 1 or 2, characterized in that a linear function, a quadratic polynomial, or a cubic polynomial is matched with the soot combustion rate. 5. The method according to claim 1 or 2, characterized in that the soot combustion rate (31) is obtained by the following steps: - Obtain the differential pressure (25) that decreases beyond the particulate filter (5); - The loading (27) of the particulate filter (5) is obtained from the differential pressure (25) according to the loading model; - Obtain the soot input (29) entering the particulate filter (5), and - Obtain the soot combustion rate (31) from the loading (27) and the soot input (29). 6. The method according to claim 5, characterized in that the soot combustion rate (31) is obtained by means of a Kalman filter. 7. The method according to claim 1 or 2, characterized in that the temporal progress (43) of the soot burning rate (31) is evaluated, wherein fault identification is performed based on the temporal progress (43) of the soot burning rate (31). 8. The method according to claim 7, characterized in that the progress (43) of the at least one matching parameter (39) over time is evaluated. 9. A method for identifying ash in a particulate filter (5) of an exhaust gas aftertreatment system (3), comprising the following steps: - Detect the differential pressure drop across the particulate filter (5); - The aging of the oxidation catalyst (7) of the exhaust gas aftertreatment system (3) is determined by the method according to any one of claims 1 to 8; - To match the loading model with the aging of the oxidation catalyst (7); - The loading (27) of the particulate filter (5) is determined according to the modified loading model, and - The ashification of the particulate filter (5) is determined from the loading (27) and the differential pressure (25). 10. The method according to claim 9, characterized in that the loading model is matched with the ashification of the particulate filter (5). 11. The method according to any one of claims 9 and 10, characterized in that the method is initialized using the initial value for ashification of the particulate filter (5). 12. The method according to any one of claims 9 and 10, characterized in that the method is performed repeatedly. 13. A control device for an internal combustion engine (1), configured to perform the method according to any one of claims 1 to 8 and/or to perform the method according to any one of claims 9 to 12. 14. An internal combustion engine (1) having a control device (21) according to claim 13.