EP1530674A2

EP1530674A2 - Internal combustion engine and method for operating an internal combustion engine comprising a fuel regulating device

Info

Publication number: EP1530674A2
Application number: EP03792327A
Authority: EP
Inventors: Ekkehard Pott; Rudolf Krebs; Axel Lang; Oliver Kirstein; Frank Wittig
Original assignee: Volkswagen AG
Current assignee: Volkswagen AG
Priority date: 2002-08-22
Filing date: 2003-08-15
Publication date: 2005-05-18
Also published as: WO2004018858A3; WO2004018858A2; DE10239258A1

Abstract

The invention relates to an internal combustion engine (1) comprising an exhaust system which is provided with a pre-catalyst device (3) and a main catalyst device (4) arranged downstream of said pre-catalyst device (3), and a fuel regulating device (11) for regulating the mixture according to the signal emitted by at least one oxygen sensor arranged in the exhaust system. According to the invention, a first oxygen sensor (5) is arranged downstream of the pre-catalyst device (3) and upstream of the main catalyst device (4), the signal of said sensor being supplied to the fuel regulating device (11) for regulating the mixture. The invention also relates to a method for operating one such internal combustion engine.

Description

Internal combustion engine and method for operating an internal combustion engine with a fuel control device

The invention relates to an internal combustion engine and a method for operating an internal combustion engine according to the preambles of the independent claims.

Combustion engines with exhaust systems are known from the prior art which comprise at least one pre-catalytic converter close to the engine and at least one main catalytic converter arranged downstream. To control the exhaust gas composition, a lambda probe is usually arranged upstream of the pre-catalytic converter and another lambda probe or a NOx sensor with an oxygen measuring device is arranged downstream of the main catalytic converter.

A broadband lambda probe is usually installed in front of the pre-catalytic converter and a step response lambda probe behind the main catalytic converter. With such a probe configuration, mixture control and regulation is possible in such a way that a deviation of the actual mixture composition from a target mixture composition is detected via the front probe and the detected deviation is converted into a control intervention of a mixture pre-control. The front probe is arranged comparatively close to the internal combustion engine, so that deviations from the target composition can be quickly recognized and corrected. The signal from the further lambda probe or the NOx sensor with oxygen measuring device arranged downstream of the main catalytic converter is used for fine control. In particular, the lambda = 1 point of the front probe is precisely calibrated by the signal of the rear probe in lambda = 1 operation. The control deviation to achieve a target lambda value is included in the control of the mixture deviation via the front probe.

A disadvantage of the conventional probe configuration is that the probe of a high thermal and mechanical which is arranged upstream of the precatalyst is close to the engine Exposure to high exhaust gas temperatures and pulsating exhaust gas flows. The exhaust gas flow also leads to inaccuracies in the measurement of the mixed exhaust gas composition of all engine cylinders, so that an increased effort is required when positioning the probe upstream of the pre-catalyst. Since the flow conditions upstream of the pre-catalytic converter are dependent on the exhaust gas mass flow and the exhaust gas temperature, this is also only a compromise design.

A separate pre-catalyst diagnosis is not possible with the described probe configuration according to the state of the art and the usual method of measuring the oxygen storage capacity (OSC) of the pre-catalyst, since the oxygen storage capacity of the overall system is tested. For a differentiated diagnosis of the exhaust system, including a separate pre-catalytic converter diagnosis, as is required to meet strict exhaust gas limit values, an additional lambda probe between the pre-catalytic converter and the main catalytic converter is necessary with the result of increased costs and higher application costs.

The object of the present invention is therefore to create an internal combustion engine with an exhaust system having at least one pre-catalyst device and at least one main catalyst device arranged downstream of the pre-catalyst device, mixture control taking place as a function of a signal from an oxygen sensor arranged in the exhaust gas, in which the probe arrangement-specific disadvantages of the prior art be avoided. It is also an object of the present invention to provide a method for operating such an internal combustion engine. According to the invention, the objects are achieved by the features of the independent claims.

According to the invention, a first oxygen sensor is arranged, for example, downstream of the first pre-catalytic converter relative to the internal combustion engine and upstream of the main catalytic converter, the signal of which is fed to the fuel control device for mixture control. Such an installation position of the first oxygen sensor gives significant advantages over the conventional sensor arrangement. An improved probe flow is ensured for the mixed exhaust gas of all cylinders. A lower effective cross-sensitivity of the oxygen sensor to hydrocarbons, hydrogen and ammonia due to a lower pollutant concentration downstream of the pre-catalyst is also to be noted. The conversion of pollutants on the pre-catalytic converter leads to an increase in the temperature of the catalytic converter and exhaust gas, but since the exhaust gas system is not adiabatic at least in this area, the additionally generated energy is at least partially dissipated again, so that overall with a maximum comparable thermal load compared to an arrangement of the oxygen sensor upstream of the pre-catalyst is to be expected.

According to the invention, the exhaust system is also controlled or regulated to maintain specified pollutant emissions as a function of the signal from a sensor arranged downstream of the pre-catalyst device, preferably an oxygen or NOx sensor. Compared to the conventional sensor configuration, in which the first oxygen sensor is arranged upstream of the first precatalyst, in the probe configuration according to the invention an improved flow of probe and lower cross sensitivity of the first oxygen sensor and thus a higher functional reliability of the exhaust system are achieved.

Furthermore, according to the invention, for the separate determination of a damage state of the pre-catalyst device as a function of the signal from the first oxygen sensor, an oxygen storage capacity of the pre-catalyst device is determined, the value of an oxygen concentration calculated upstream of the pre-catalyst device being used. In addition to the advantages already mentioned, which result from the installation position of the first oxygen sensor downstream of the pre-catalyst device, there is a decisive advantage in the possibility of separately determining the state of damage of the pre-catalyst device without installing an additional second sensor.

According to the invention, it is also provided that to determine an oxygen storage capacity of the pre-catalyst device upstream of the pre-catalyst Predefined lambda value excitation takes place in the exhaust gas and an associated reaction downstream of the pre-catalyst is detected by means of the first oxygen sensor. Compared to the method for determining the oxygen storage capacity of a catalytic converter known from the prior art, for example DE 43 389 17 A1, the lambda value excitation in the exhaust gas is not regulated, but controlled, according to the invention, so that an additional oxygen sensor upstream of the precatalyst device is dispensed with can.

Further features and advantages of the invention result not only from the associated patent claims - alone or in combination - but also from the following description of preferred exemplary embodiments in conjunction with the associated drawings.

The drawings show:

Figure 1 is a representation of sensor configurations in exhaust systems

Figure 2 is a schematic diagram of the known from the prior art

enforcement system

Figure 3 is a schematic diagram of a management control according to the invention

FIG. 4 shows a representation of the mixture control according to the prior art in the case of a lambda probe arranged upstream of a precatalyst

FIG. 5 shows a mixture control in a lambda probe arranged downstream of a pre-catalyst

FIG. 6 shows a mixture control according to the invention FIG. 7 shows a further mixture control according to the invention with a premature lean breakthrough downstream of a pre-catalyst

Figure 8 is an illustration of a mixture control according to the invention with an extended rich phase downstream of a pre-catalyst

FIG. 9 shows a lambda value excitation and an assigned lambda reaction for determining an oxygen storage capacity according to the invention

FIG. 10 shows a further lambda value excitation and an assigned lambda reaction

FIG. 11 shows a further lambda value excitation and an assigned lambda reaction

FIG. 12 shows a representation of sensor configurations known from the prior art for double-flow exhaust systems

FIG. 13 shows a representation of sensor configurations according to the invention for double-flow exhaust systems.

The illustration of sensor configurations in exhaust gas systems shown in FIG. 1 shows, for sensor configurations A, B and C, an internal combustion engine 1, 1A, 1 B with a downstream exhaust gas system 2, 2A, 2B, which has a pre-catalytic converter 3, 3A, 3B and a main catalytic converter 4, 4A, 4B.

In the configuration A known from the prior art, a first oxygen sensor 5A is arranged upstream of the pre-catalytic converter 3A and a second sensor 6A, for example an oxygen sensor or a NOx sensor, is arranged downstream of the main catalytic converter. In configuration B of FIG. 1, a separate diagnosis of the pre-catalyst 3B, another oxygen sensor 7B is arranged downstream of the pre-catalyst 3B.

In configuration C, an internal combustion engine 1 according to the invention is shown with an exhaust system 2, in which a first oxygen sensor 5 is arranged downstream of the pre-catalyst device 3. The internal combustion engine 1 is preferably a direct-injection gasoline engine or a diesel engine. The internal combustion engine 1 is preferably capable of stratified charging. The pre-catalyst device 3 is designed as a 3-way catalyst or as an oxidation catalyst and can also comprise a plurality of individual catalysts. The first oxygen sensor 5, which is arranged downstream of the pre-catalyst device 3 according to the invention, is preferably a broadband lambda probe, which enables the oxygen concentration or the lambda value of the exhaust gas to be recorded relatively quickly, but generally only roughly.

The main catalytic converter device 4 arranged downstream of the oxygen sensor 5 can be a 3-way catalytic converter, a NOx storage catalytic converter or an oxidation catalytic converter and can also comprise a plurality of individual catalytic converters. The sensor 6 arranged downstream of the main catalytic converter device 4 is preferably designed as a step response lambda probe, as a NOx sensor with oxygen measurement device or as a multifunction sensor.

The installation position of the oxygen sensor 5 upstream of the precatalyst has some disadvantages, as has already been explained above.

The configuration C of FIG. 1 has an installation position of the first oxygen probe that is optimized compared to the prior art, which is subject to lower loads there and moreover allows the damage condition of the pre-catalyst device to be determined separately in a relatively simple manner. For optimized operation of the internal combustion engine according to the invention, however, it is advisable to take into account that the changed installation location of the first oxygen sensor downstream of the pre-catalyst device results in a longer exhaust gas run length has and thus a possibly excessive inertia of the controlled system of the mixture control.

For this purpose, FIG. 2 shows a basic illustration of the guide control known from the prior art, corresponding to the sensor configuration 1A. The signal from the oxygen sensor 5A, preferably a broadband lambda probe, is used to detect deviations of an actual lambda value from a target lambda value specified by a fuel control device 11.

Since, due to the construction, such broadband probes react relatively quickly, but also have relatively high error deviations and, in particular, accurate detection of the lambda = 1 point is difficult, the control intervention of the oxygen sensor 5A is corrected by the amount of the sensor 6A arranged downstream of the main catalyst device 4A , Around the lambda = 1 point, an accuracy in the alcohol range can be achieved with a step response probe in particular. The signal corrected by the amount of the step response probe is fed to a control unit 8, which in particular controls injection valves 9 and / or a throttle valve 10. Since the exhaust gas run length is short in the control section belonging to the first control loop, fast, stable control can be achieved with little effort. To control or regulate the exhaust system, the control unit 8 comprises an exhaust gas control device.

FIG. 3 shows a basic illustration of the guidance control for an internal combustion engine according to the invention, corresponding to sensor configuration C in FIG. Corresponding modules are provided with the same reference numerals as in FIG. 2. Since the first oxygen sensor 5 is arranged downstream of the pre-catalyst device 3, there is an increased inertia of the controlled system of the first control circuit and thus a lower stability. In contrast to the control system known from the prior art, the oxygen storage capacity of the pre-catalyst device 3 also contributes to the inertia of the controlled system. According to the invention, the volume of the pre-catalyst device is limited to reduce the activity of the controlled system and to achieve a predetermined stability limit of the mixture control. The relative volume of the pre-catalyst device 3 is preferably limited to a value between 0.7 and 0.3 of the engine displacement. Values of 0.6, 0.45 and 0.35 for the relative volume are particularly preferred.

Additionally or alternatively, a maximum volume-independent oxygen storage capacity of the pre-catalyst device 3 is selected to achieve a predetermined stability limit of the mixture control. The maximum oxygen storage capacity is preferably limited to a value between 2.1 grams and 0.4 grams. Values of 2 grams, 1 gram and 0.5 grams of oxygen are particularly preferred. Such values are preferred, in particular, for a pre-catalyst device 3 which has been conditioned for 4 hours at 600 degrees Celsius and a lambda value = 0.98. If the pre-catalyst device 3 comprises several catalysts, the corresponding maximum values apply to the totality of the catalysts, i.e. for the entire catalyst volume or the total oxygen storage capacity of the individual catalysts.

Regardless of a limitation of the relative volume or the maximum oxygen storage capacity of a pre-catalytic converter 3, it can be expected in stoichiometrically operated operating states without large load and speed fluctuations that the control interventions via the first oxygen sensor are only slight and the oxygen stored in the pre-catalytic converter 3 only marginally is loaded and unloaded. In such operating states, as is known in the prior art, a constant lambda control can take place via a probe 6 connected downstream of the main catalyst device 5. In dynamic operating states, such as a start-up phase, simple strategies for mixture control can be used with an oxygen sensor, which is arranged upstream of the pre-catalyst, as in the prior art. A deviation between an actual and a target lambda value is converted into a possibly filtered control intervention. Such a control strategy is illustrated in FIG. 4 for such a system with a low inertia of the controlled system. The upper curve shows a lambda value measured in front of a pre-catalytic converter, while the lower curve describes a control intervention depending on the deviation from the target value. A positive control intervention leads to an enrichment, a negative control intervention leads to a leaner exhaust gas. At a time TO, a starting process begins with a resulting enrichment of the exhaust gas. At time T1, the fuel control device intervenes for the first time through a lean-out control intervention, since there is an excessive deviation of the actual lambda value from the target lambda value over an excessively long period of time. At the point in time T2, the control intervention is intensified since there is no discernible tendency to lean the exhaust gas. Only at the later point in time T3 does the measured enrichment of the exhaust gas decrease, so that the intervention of the controller can be withdrawn. At the time T4, the control intervention is ended because the actual lambda value and the target lambda value match.

FIG. 5 shows a comparable scenario in an internal combustion engine 1 according to the invention with an oxygen sensor 5 arranged upstream of a pre-catalyst device 3. The lambda value that can be measured in this sensor configuration downstream of the pre-catalyst device 3 is shown as a strongly drawn curve, the non-measurable lambda value upstream of the pre-catalyst device 3 as a thinly drawn curve. Again at the time TO, a start-up phase begins, which is accompanied by a longer enrichment of the lambda value upstream of the pre-catalyst device, which leads to a control intervention only at a time T1. However, the control loop here is not matched to the extended exhaust gas runtime and the oxygen storage capacity of the pre-catalyst device 3, so that the controller intervenes too weakly and then reacts to the delayed consequences of the lambda value measured downstream of the pre-catalyst device 3 with increased control interventions by which the system is made to vibrate. These vibrations are only corrected at a time T2. According to the invention, the extended exhaust gas runtime and the oxygen storage capacity of the pre-catalyst device 3 are taken into account in the form of the control intervention in order to avoid this behavior, which does not allow a favorable exhaust gas quality to be expected. According to the invention, the control intervention is shown in the form of a decreasing enrichment or emaciation of the exhaust gas. The maximum value and decay rate are dependent on the size of the measured lambda value deviation, ie the difference between a lambda setpoint and an actual lambda value determined from the oxygen signal of the first oxygen sensor. There is also a dependency on the leaning or enrichment speed of the exhaust gas measured downstream of the pre-catalyst device 3, the engine operating point, in particular the exhaust gas mass flow and an exhaust gas recirculation rate, and on parameters such as the exhaust gas or catalyst temperature and the respective oxygen storage capacity of the catalyst device 3. The oxygen storage capacity can be modeled or determined according to a method described below. Thus, even in dynamic operation, a certain amount of oxygen remains in the pre-catalyst device 3 and compensates for the buffering of the controlled system by the oxygen storage capacity of the catalyst device.

A scenario analogous to FIG. 5 is illustrated in FIG. The curve designation is the same as in FIG. 4, however the control intervention is modified in accordance with the above. If mixture enrichment takes place from a point in time TO, this does not initially lead to any change in the lambda value of the exhaust gas downstream of the pre-catalyst device, since the enrichment is buffered by an oxygen release due to the stored oxygen. However, the lambda value of the exhaust gas upstream of the pre-catalyst device shows a deviation immediately after the time TO. Downstream of the pre-catalyst device, a richness is measured from time T1 and, according to the invention, at time T2 the lambda value is trimmed towards the lean by a predetermined amount. The size of this amount is preferably selected as a function of the exhaust gas leaning rate in the time interval T1 to T2. Then, with a predeterminable leaning rate, at which the engine operating point, the exhaust gas mass flow, the catalyst temperature and / or the recognized oxygen storage capacity is taken into account, the exhaust gas continues to be leaned out until a maximum deviation from the target value is measured at the oxygen sensor at time T4. Compared to the measured thinning rate, the decay rate of thinning is slower, linear or degressive.

FIG. 7 illustrates a scenario in which lean exhaust gas is measured 3 as the lean out of the control intervention downstream of the pre-catalyst device 3. In this case, the emaciation is discontinued at time T6 and replaced by a fading enrichment. FIG. 8 illustrates a scenario in which the measured lambda value downstream of the pre-catalytic converter device 3 runs insufficiently or not at all in the lean direction during the decay of a lean-out control intervention. In this case, the leaning of the exhaust gas is continued with a constant lean control intervention until the measured lambda value falls below a predetermined deviation threshold from the setpoint lambda by the time T7. After time T7, the lean-out rule intervention is withdrawn. The lambda value in advance of the pre-catalyst device 3 goes back from the time T8 to the desired value. The control intervention is ended at time T9.

Since the lean breakthrough indicates a reduced oxygen storage capacity of the pre-catalyst device 3, the oxygen storage capacity entering into the form of the control intervention can be reduced by an amount dependent on the time and the amount of the lean opening and can be used in subsequent control interventions. Furthermore, it can generally be assumed that increasing control interventions are associated with a reduced oxygen storage capacity, since then the buffering due to the oxygen storage capacity is reduced. Accordingly, a change in the oxygen storage capacity of the pre-catalyst device 3 can be determined from the average amount of the control intervention. A change in the oxygen storage capacity can also be derived from inadequate emaciation of the exhaust gas. The arrangement according to the invention of the first oxygen sensor downstream of the pre-catalytic converter 3 allows a separate determination of a damage state of the pre-catalytic converter 3 without any additional effort for sensors. For this purpose, a calculated value of an oxygen concentration upstream of the pre-catalyst device 3 is compared with the value determined from the signal of the first oxygen sensor and an oxygen storage capacity is determined therefrom, which is correlated in a manner known per se with a damage state of the pre-catalyst device. For this purpose, the control unit 8 expediently comprises a diagnostic device.

As is also known per se from the prior art, a lambda value excitation is generated in the exhaust gas in order to determine an oxygen storage capacity, particularly upstream of the pre-catalyst device 3, and an associated reaction is detected downstream of the pre-catalyst device. In contrast to the prior art, for example DE 43 389 17 A 1, the lambda value excitation upstream of the pre-catalyst device 3 is not regulated, but rather is controlled.

A lambda wobble with a predetermined amplitude and frequency is preferably generated as lambda excitation. In FIG. 9, the controlled lambda value upstream of the pre-catalyst device 3 (dotted line), the lambda value downstream of the pre-catalyst device 3 with a high oxygen storage capacity of the pre-catalyst device 3 (solid line) and the lambda value downstream of the pre-catalyst device 3 for a pre-catalyst device 3 shown with a low oxygen storage capacity (dashed line). As can be seen in FIG. 9, in the case of a pre-catalyst device 3 with a low oxygen storage capacity, the assigned reaction is faster and the curve shape is more similar to the stimulating wobble.

An alternative or additional method is shown in FIG. 10, in which the wobble frequency is reduced over a predetermined period of time. It can be seen that the assigned reaction downstream of the pre-catalyst device 3 occurs with a low oxygen storage capacity (dashed line) even at relatively high wobble frequencies and the maximum possible Reach amplitude. In contrast, with a high oxygen storage capacity (solid line), assigned reactions are only achieved at a relatively low wobble frequency, with their amplitude increasing only slowly.

In the embodiment of the method according to FIG. 11, the wobble amplitude is increased over time at a constant wobble frequency. The lambda value downstream of the pre-catalytic converter 3 follows with a high oxygen storage capacity with a delay and with a relatively low amplitude (solid line), while a high oxygen storage capacity leads to a rapid subsequent reaction with a relatively high amplitude (dashed line).

In all of the wobble processes shown, the wobble parameters can be varied continuously or narrowly. Furthermore, an evaluation of response times of the first oxygen sensor to a jump excitation from a slightly rich lambda value, for example lambda = 0.96, to a slightly lean lambda value, for example lambda = 1, 04, or vice versa, can be evaluated. A high oxygen storage capacity corresponds to a long response time, while a low oxygen storage capacity corresponds to a short response time.

Combustion engines with exhaust systems with a plurality of exhaust lines are also known from the prior art, in which pairs of probes are arranged upstream of the precatalysts of each exhaust line and, if appropriate, further pairs of probes downstream of the main catalysts of each exhaust line. FIG. 12A shows such a sensor configuration for a double-flow exhaust system without the possibility of a separate diagnosis of the pre-catalyst. FIG. 12B also shows a sensor configuration for a double-flow exhaust system, in which a separate diagnosis of the pre-catalysts can be carried out by means of a third pair of probes.

The method according to the invention can also be used for exhaust systems which have a plurality of exhaust lines. For this purpose, FIG. 13A shows a sensor configuration in an internal combustion engine 1 with an exhaust system that has two exhaust lines 12, 13. Each exhaust line 12, 13 has a pre-catalyst 14 and a main catalytic converter 15 arranged downstream of the pre-catalytic converter 14. An oxygen sensor 16, 17 is arranged downstream of each pre-catalytic converter 14 and upstream of each main catalytic converter 15, so that a pair of sensors is arranged downstream of the pre-catalytic converters 14. A sensor 17 is also arranged downstream of each main catalytic converter 15, so that a pair of sensors is also arranged downstream of the main catalytic converters 15. The determination of a damage condition of a pre-catalytic converter can be carried out separately for each exhaust line.

FIG. 13B shows a sensor configuration for an exhaust system with an exhaust gas junction 18 downstream of the pre-catalysts 14. First oxygen sensors 19 are arranged upstream of the exhaust gas junction 18 and downstream of the pre-catalysts 14. A main catalytic converter 20 is arranged downstream of the exhaust gas junction 18. An oxygen sensor 21 is in turn arranged downstream of the main catalytic converter 20. In this case too, the damage state of the individual pre-catalysts 14 can be determined separately. As in the exemplary embodiments described above, the pre-catalysts and main catalysts can be replaced by pre-catalyst devices and main catalyst devices with more than one catalyst without leaving the scope of the invention. This also applies to the sensors used.

Claims

Patent claims

1. internal combustion engine (1) with an exhaust system, which has a pre-catalyst device (3) and a downstream of the pre-catalyst device (3) arranged main catalyst device (4), and with a fuel control device (11) for mixture control depending on the signal of at least one in the exhaust gas arranged oxygen sensor, characterized in that a first oxygen sensor (5) is arranged downstream of the pre-catalyst device (3) and upstream of the main catalyst device (4), the signal of which is supplied to the fuel control device (11) for mixture control.

2. Internal combustion engine according to claim 1, characterized in that an exhaust gas control device for controlling or regulating the exhaust system to comply with predetermined pollutant emission limits as a function of the signal of the first oxygen sensor (5) is provided.

3. Internal combustion engine according to claim 1 or 2, characterized in that a diagnostic device for determining a damage state of the pre-catalyst device (3) is provided, of which depending on the signal of the first oxygen sensor (5) and a calculated value of an oxygen concentration upstream of the pre-catalyst device ( 3) an oxygen storage capacity of the pre-catalyst device (3) is determined and the determined oxygen storage capacity is correlated with a state of damage of the pre-catalyst device (3).

4. Internal combustion engine according to one of the preceding claims, characterized in that a sensor (6), in particular a step response lambda probe or a NOx sensor with an oxygen measuring device, is arranged downstream of the main catalyst device (4), the signal of the fuel control device (11), can be fed to the exhaust gas control device and / or the diagnostic device.

5. Method for operating an internal combustion engine (1) with an exhaust system, which has at least one pre-catalyst device (3) and at least one main catalyst device (4) arranged downstream of the pre-catalyst device (3), and with a fuel control device (11) for mixture control as a function of the signal of an oxygen sensor arranged in the exhaust gas, characterized in that a first oxygen sensor (5) is arranged downstream of the pre-catalyst device (3) and upstream of the main catalyst device (4), the signal of which is supplied to the fuel control device (11) for mixture control.

6. The method according to claim 5, characterized in that to achieve a predetermined stability limit of the mixture control, the relative volume of the pre-catalyst device (3) is selected to be smaller than a predetermined maximum relative volume, in each case based on the engine displacement.

7. The method according to claim 6, characterized in that the maximum relative volume in a range between 0.3 and 0.7, in particular at 0.6, 0.45 or 0.35, is selected.

8. The method according to at least one of claims 5 to 7, characterized in that a maximum volume-independent oxygen storage capacity of the pre-catalyst device (3), preferably in a range between 2.1 grams and 0.4 grams, is selected to achieve a predetermined stability limit of the mixture control ,

9. The method according to at least one of claims 5 to 8, characterized in that for mixture control a difference between a lambda target value and an actual lambda value determined from the oxygen signal of the first oxygen sensor is determined and as a function of this actual target value difference a control intervention from an additive and / or multiplicative immediate correction value and optionally a post-correction value that subsequently decays is formed.

10. The method according to claim 9, characterized in that the immediate correction value and / or possibly the post-correction value is selected as a function of an engine operating point, an exhaust gas mass flow, a pre-catalyst device temperature or an oxygen storage capacity of the pre-catalyst device.

11. The method according to claim 9 or 10, characterized in that, if necessary, the post-correction value is selected as a function of the actual setpoint difference.

12. The method according to at least one of claims 9 to 11, characterized in that an expected value of the actual setpoint difference is calculated and that the sign of the immediate correction value and possibly the post-correction value is reversed if the actual setpoint difference decreases faster than determined by the expected value.

13. The method according to at least one of claims 9 to 12, characterized in that an expected value of the actual setpoint difference is calculated and, if the actual setpoint difference decreases more slowly than determined by the expected value, the value of immediate and / or if necessary post-correction value is frozen.

14. The method according to at least one of claims 9 to 13, characterized in that an oxygen storage capacity of the pre-catalyst device (3) is determined from the time course of the actual setpoint difference of the sensor signal within a predetermined time interval.

15. The method according to at least one of claims 9 to 14, characterized in that the average amount of the control intervention is determined and from an oxygen storage capacity of the pre-catalyst device (3) is determined.

16. The method according to at least one of claims 5 to 15, characterized in that the signal of the first oxygen sensor (5) is used to control or regulate the exhaust system to comply with predetermined pollutant emission limits.

17. The method according to at least one of claims 5 to 16, characterized in that the signal of the first oxygen sensor (5) is used to control or regulate the exhaust system to comply with predetermined pollutant emission limits.

18. The method according to at least one of claims 5 to 16, characterized in that the pre-catalyst device (3) comprises at least one 3-way catalyst.

19. The method according to at least one of the preceding claims 5 to 18, characterized in that the pre-catalyst device (3) comprises at least one oxidation catalyst.

20. The method according to at least one of the preceding claims 5 to 19, characterized in that the main catalyst device (4) comprises at least one NOx storage catalyst.

21. The method according to at least one of the preceding claims 5 to 20, characterized in that the main catalyst device (4) comprises at least one 3-way catalyst.

22. The method according to at least one of claims 5 to 21, characterized in that the first oxygen sensor (5) is a broadband lambda probe.

23. The method according to at least one of claims 5 to 22, characterized in that at least one second sensor is arranged downstream of the main catalytic converter (4), the signal of which is used for regulating the mixture or for regulating or controlling the exhaust system.

24. The method according to claim 23, characterized in that the second sensor (6) is a step response lambda probe

25. The method according to claim 23, characterized in that the second sensor (6) is a NOx sensor with an oxygen measuring device.

26. The method according to at least one of claims 5 to 25, characterized in that for separately determining a damage state of the pre-catalyst device (3) depending on the signal of the first oxygen sensor (5) and a calculated value of an oxygen concentration upstream of the pre-catalyst device (3) The oxygen storage capacity of the pre-catalyst device (3) is determined and the determined oxygen storage capacity is correlated with a damage state of the pre-catalyst device (3).

27. The method according to at least one of claims 5 to 25, characterized in that for determining an oxygen storage capacity of the pre-catalyst device (3) upstream of the pre-catalyst device (3), a predetermined controlled lambda value excitation takes place in the exhaust gas and an associated reaction in the exhaust gas downstream of the pre-catalyst device ( 3) is detected by means of the first oxygen sensor.

28. The method according to claim 27, characterized in that the lambda excitation is generated by a lambda wobble with a predetermined amplitude and frequency, and that an amplitude of the lambda value is determined from the assigned reaction in the exhaust gas and from the ratio of the two amplitudes the oxygen storage capacity of the pre-catalyst device (3) is calculated.

29. The method according to claim 27, characterized in that the frequency of the lambda value wobble is increased continuously or narrowly and the oxygen storage capacity of the pre-catalyst device (3) is determined as a function of the amplitude of the oxygen sensor signal.

30. The method according to claim 27, characterized in that the amplitude of the lambda value wobble is raised continuously or narrowly and the amplitude peak of the oxygen sensor indicates the oxygen storage capacity of the pre-catalyst device (3).

31. The method according to claim 27, characterized in that a jump from a slightly rich lambda value, in particular lambda = 0.96, to a slightly lean lambda value, in particular lambda = 1, 04, is carried out and that the response times of the first oxygen sensor.

32. The method according to any one of claims 5 to 31, characterized in that the exhaust system has a plurality of exhaust lines (12, 13), and each exhaust line has at least one pre-catalyst device (14), at least one main catalyst device (15, 20) downstream of the pre-catalyst devices (14) ) is arranged such that an exhaust gas junction (18) is provided downstream of the pre-catalyst devices (14) and that a first oxygen sensor (16) is provided downstream of each pre-catalyst device (14), possibly upstream of the exhaust gas junction (18).

33. The method according to claim 33, characterized in that a sensor (17, 21), preferably a step response lambda probe or a NOx sensor with an oxygen measuring device, is provided downstream of each main catalyst device (15, 20).