CN114233499A

CN114233499A - Method for adapting the mixture formation of an internal combustion engine having a dual fuel metering device

Info

Publication number: CN114233499A
Application number: CN202111049584.8A
Authority: CN
Inventors: A·派勒斯; D·格伦巴赫
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 2020-09-09
Filing date: 2021-09-08
Publication date: 2022-03-25
Also published as: DE102020211318A1

Abstract

The invention relates to a method for adapting the mixture formation of an internal combustion engine (115) having a dual fuel metering device, for which an intake-pipe-based fuel metering and a direct fuel metering are carried out in a mixing mode, characterized in that a first adaptation value (PFI) for an intake-pipe injection (PFI) is obtained in the mixing modefacLamAdpn _PFI) And a second adaptation value for direct injection (GDI) (GDI)facLamAdpn _GFI) And offset value for injection quantity (ratLamAdpn _ges) Wherein (c) is determined according to the first and second adaptation valuesfacLamAdpn _GFI；facLamAdpn _PFI) To be provided withAnd the offset value (ratLamAdpn _ges) To adapt and diagnose the mixture formation.

Description

Method for adapting the mixture formation of an internal combustion engine having a dual fuel metering device

Technical Field

The invention relates to a method for adapting the mixture formation of an internal combustion engine having a dual fuel metering device, and to a computing unit and a computer program for carrying out the method.

Background

One possible method for fuel injection in gasoline motors is intake manifold injection, which is increasingly replaced by direct fuel injection. The latter method achieves a significantly better fuel distribution in the combustion chamber and thus a better power yield with less fuel consumption.

Furthermore, there are gasoline motors with a combination of intake pipe injection and direct injection, that is to say a so-called dual system. This is advantageous precisely in respect of increasingly stringent emission requirements or emission limit values, since intake manifold injections, for example, in the medium load range, result in better emission values than direct injections. In the full load range, the direct injection, for example, reduces so-called knocking.

DE 10256906 a1 discloses a method for regulating an air/fuel mixture in an internal combustion engine (1), in which the injected fuel quantity is corrected for adapting the air/fuel mixture ratio. The method according to the invention allows the correction of the error in the injected fuel quantity and the error in the delivered air quantity to be adapted separately. When fuel is injected multiple times with multiple injection events for one combustion event in a cylinder (5) of the internal combustion engine (1), a correction for the amount of fuel injected is carried out for each of these injection events.

DE 102015216119 a1 discloses a method and a device for operating a combustion motor with dual intake-pipe-based and direct fuel metering, wherein the intake-pipe-based and the direct fuel metering are carried out in a mixed mode, and wherein, in particular, in the variable mixed mode mentioned, an intake-pipe-based and directly-metered fuel distribution is carried out (315) on the basis of a reduction of the noise emissions (310) caused by the fuel metering and/or the fuel combustion.

Disclosure of Invention

The object of the present invention is to provide a solution for adapting the mixture formation of an internal combustion engine having a dual fuel metering device, wherein the mixture formation of the dual fuel injections is adapted and diagnosed.

In a first aspect, a method for adapting a mixture formation of an internal combustion engine having a dual fuel metering device is described, for which an intake-pipe-based fuel metering and a direct fuel metering are carried out in a mixing mode, wherein a first adaptation value for an intake-pipe injection and a second adaptation value for a direct injection and an offset value for an injection quantity are determined in the mixing mode, wherein the mixture formation is adapted and diagnosed as a function of the first and second adaptation values and the offset value.

The method has the following particular advantages, namely: the mixture formation is adapted during the double spraying process. In other words, it is not necessary to switch off one of the two fuel metering devices, so that the internal combustion engine can be operated with an optimum injection profile over the entire operating cycle. This is advantageous precisely in respect of increasingly stringent emission requirements or emission limit values, since better combustion and thus better emission values can be achieved by individual adaptation of the two injection paths.

In a preferred embodiment of the method, an adapted release for the mixture formation is provided if a stable or quasi-stable operating state for the internal combustion engine is present. This is advantageous because accurate measurements can thus be carried out for the method.

Furthermore, a stable or quasi-stable state exists if the relative air quantity does not change more than a predeterminable threshold value over a predeterminable period of time.

Furthermore, the first adaptation value for the intake pipe injection, the second adaptation value for the direct injection, and the offset value can be obtained from the relative air amount, injection amount, and division factor.

It is particularly advantageous if the relative air quantity is divided into a relative air quantity of the intake manifold pressure path and a relative air quantity of the direct injection path as a function of the division factor, wherein the values obtained during the present injection event for the relative air quantities, the injection quantity, the division factor, the relative air quantity of the intake manifold pressure path and the relative air quantity of the direct injection path are stored as averages in a characteristic map. This allows the method to be implemented on the controller in a resource-saving manner.

Advantageously, an n-dimensional system of linear equations is solved, which is formed by the respective n mean values for the injection quantity, the relative air quantity of the intake manifold pressure path and the relative air quantity of the direct injection path, and the first adaptation value for the intake manifold injection, the second adaptation value for the direct injection and the offset value for the injection quantity to be determined, wherein n corresponds to the number of regions in the combined characteristic curve.

It is also advantageous if the average value of the injection quantity, the relative air quantity of the intake manifold pressure path and the relative air quantity of the direct injection path in the region of the combined characteristic curve is formed from at least three measured values for each region of the combined characteristic curve. Thereby ensuring that the database provides sufficient quality for the method.

Furthermore, a defect for the direct injection is identified if the first adaptation value exceeds or falls below a predefinable threshold value band. This has the particular advantage that: it is possible to individually perform a diagnosis for the direct injection when intake-pipe-based and direct fuel metering is performed simultaneously. In this way, during the diagnosis, the intake manifold injection or the direct injection need not be switched off, so that an optimum injection profile can be operated over the entire operating cycle. This has the result that better emission values can be achieved and thus that predefined emission limit values can be adhered to.

Furthermore, a defect for the intake pipe injection is identified if the second adaptation value exceeds or falls below a predefinable threshold value band. This has the particular advantage that: it is possible to individually perform a diagnosis for the intake pipe injection when intake pipe-based and direct fuel metering is performed simultaneously. In this way, during the diagnosis, the intake manifold injection or the direct injection need not be switched off, so that an optimum injection profile can be operated over the entire operating cycle. This has the result that better emission values can be achieved and thus that predefined emission limit values can be adhered to.

In a further aspect, the invention relates to an apparatus, in particular a controller, and a computer program which is set up, in particular programmed, to carry out one of the methods. In yet another aspect, the invention relates to a machine-readable storage medium on which the computer program is stored.

Drawings

The invention is described in detail below with the aid of embodiments and with reference to the accompanying drawings. Here:

FIG. 1 shows a schematic diagram of a cylinder of an internal combustion engine;

fig. 2 shows a diagram for describing a method for performing mixture formation for an internal combustion engine having a dual fuel metering device.

Detailed Description

Fig. 1 schematically shows a cylinder 102 of the internal combustion engine 100 in an exemplary manner. The cylinder 102 has a combustion chamber 103 that is expanded or contracted by the movement of a piston 104. The internal combustion engine can be in particular a gasoline motor.

The cylinder 102 has an inlet valve 105 for introducing air or a fuel-air mixture into the combustion chamber 103. Air is fed through the air inlet pipe 106, which is part of the air supply system, where the fuel injectors 107 are located. The sucked air is put into the combustion chamber 103 of the cylinder 102 through the inlet valve 105. A throttle valve 112 in the air supply system is used to regulate the necessary mass flow of air into the cylinders 102. Furthermore, an air mass sensor 99 for measuring the relative air mass is arranged in the intake pipe 106, in particular upstream of the throttle valve 112. This can be, in particular, a hot film air flow meter (HFM).

The internal combustion engine can be operated during intake pipe injection. During this intake pipe injection, fuel is injected into the intake pipe 106 by means of the fuel injector 107, so that there is formed an air-fuel mixture which is put into the combustion chamber 103 of the cylinder 102 through the inlet valve 105.

Furthermore, the internal combustion engine can be operated during direct injection. For this purpose, the fuel injector 111 is arranged on the cylinder 102 for injecting fuel directly into the combustion chamber 103. In such direct injection, the air-fuel mixture required for combustion is formed directly in the combustion chamber 103 of the cylinder 102. The cylinder 102 is furthermore provided with an ignition mechanism 110 for generating an ignition spark in the combustion chamber 103 in order to initiate combustion.

After combustion, combustion exhaust gas is discharged from the cylinder 102 through an exhaust pipe 108. The discharge process is performed according to the opening of the outlet valve 109 also arranged on the cylinder 102. The inlet and

outlet valves

105, 109 are opened and closed for performing a four-stroke operation of the internal combustion engine 100 in a known manner. By means of the lambda sensor 123, the lambda value of the exhaust gas in the exhaust pipe 108 can be determined.

The internal combustion engine 100 can be operated with direct injection, with intake manifold injection or in hybrid operation. This allows the selection of the respectively optimum operating mode for operating the internal combustion engine 100 as a function of the current operating point. Therefore, when the internal combustion engine 100 is operated at a low speed and a low load, the internal combustion engine can be operated in, for example, an intake pipe injection operation, and when the internal combustion engine is operated at a high speed and a high load, the internal combustion engine can be operated in a direct injection operation. However, it is expedient in a large operating range to operate the internal combustion engine 100 in a mixed mode in which the fuel quantity to be supplied to the combustion chamber 103 is delivered in portions by intake manifold injection and direct injection.

Furthermore, a computing unit is provided which is designed as a controller 115 for controlling the internal combustion engine 100. The controller 115 can operate the internal combustion engine 100 in direct injection, intake manifold injection, or hybrid operation. Furthermore, the controller 115 can also acquire the value of the λ sensor 123.

The controller 115 actuates both the injection valves in the intake manifold, with which the amount of fuel to be delivered to the internal combustion engine is specified, and the injection valves in the cylinders. In this case, the required fuel quantity is set by a lambda controller integrated in controller 115, in particular as a function of the motor load and the required lambda value, wherein the basic control is preferably performed by an adaptable pilot controller contained in the lambda controller. For this purpose, the output signal of the pilot control is added to the output signal of the lambda controller. The pilot control determines the fuel quantity, in particular by means of the motor load. The correlation between the motor load and the fuel quantity to be predefined is preferably stored in a combined characteristic curve in the control unit 115. Due to system drift, the correlation between the motor load and the fuel quantity to be predefined may vary. To compensate for this, an adaptation period is provided within the scope of the mixture adaptation, in which adaptation period the correlation is relearned in the preliminary control unit.

During the mixture adaptation, systematic errors of the fuel-air mixture are corrected by means of the preferred adaptation means and the adaptation values obtained therefrom. In this case, different types of errors can occur which lead to a deviation of the mixture. Errors in the determination of the air quantity delivered to the internal combustion engine have a multiplicative effect on the fuel metering, whereas errors due to leakage air effects or to a delay in the start of the injection valve have a superimposed effect. The multiplied error is particularly significant in the medium load range of the internal combustion engine 100, whereas the superimposed error prevails at low loads. Accordingly, according to known methods, the adaptation to the fuel metering is preferably carried out in the medium load range with respect to the multiplied error and in the low load range with respect to the superimposed error. Since the multiplied errors also have an effect in the low load range and the superimposed errors also have an effect in the medium load range, the adaptation is carried out alternately in the two load ranges until a sufficiently stable adaptation of the pilot control device is present.

In particular, the internal combustion engine 100 can be designed as an internal combustion engine having one or more motor blocks. In particular, an individual intake manifold injector can be provided in each air supply section for a cylinder. Likewise, each air supply section of each motor cylinder can have its own throttle for regulating the inflowing air. Furthermore, the internal combustion engine 100 can comprise a plurality of exhaust pipes, wherein a lambda sensor for measuring the air-fuel ratio is provided in each exhaust pipe.

Preferably, the internal combustion engine 100 is an internal combustion engine having 2, 3, 4, 6 or 8 cylinders.

The adaptation value is calculated on the basis of the relative fuel quantity in the injection pathratMFu _gesThe calculation chain of (2). The following calculation chain is generated assuming that the pre-control factor and the target air-fuel ratio that is not stoichiometric are ignored:

whereinratMFu _gesIs the relative amount of the fuel,ratMAiris the relative amount of air that is present,facLamis the lambda adjustment factor for the air-fuel ratio,facLamAdpnis an adapted lambda control factor for the air-fuel ratio andratLamAdpnis the adapted relative air-fuel ratio.

The lambda control factor is formed here on the basis of the signal of the lambda sensor 123.

In a system where the adaptation is completed, the lambda adjustment factorfacLamShould be neutral, that is to say the possible mixture deviations are compensated for by the adaptation values. Accordingly, this lambda adjustment factorfacLamThe value 1 is assumed and is not further considered. If the two adaptation values should now be obtainedfacLamAdpnAnd the offset valueratLamAdpn，Two must be observed in correspondence therewithAnd (4) an equation. The following system of equations is generated:

。

this system of equations can now be converted to the following form:

。

obtained by means of a variant:

。

this corresponds to the general writing for a system of linear equations of the form:

。

in order to improve the adaptation to the entire operating range of internal combustion engine 100, the adaptation value can be determined from both two measured values and also from n measured values:

。

there is thus an overdetermined system of equations that is solved as follows:

。

in order to be able to compensate for time effects in the injection system or in the air system and to be able to calculate functions more resource-efficiently on the control unit 115, the average values in the different load-speed ranges (1 to n) are used instead of the individual measured values for the calculation:

。

in systems with dual injection system, i.e. with injection into the intake pipe in the cylinder and direct injection of PFI, GDI, the relative fuel quantities are measuredratMFu _gesTo two injection paths GDI, PFI:

whereinratMFu _gesIs the total relative amount of fuel,ratMFu _PFIis the relative fuel quantity of a fuel low pressure system (PFI) andratMFu _GDIis the relative fuel quantity of the high pressure fuel system (GDI), the latter two being for intake pipe pressure injection and direct injection, respectively.

Accordingly, the entire air path can be virtually divided by the following equation:

whereinratMAir _gesIs the relative amount of air that is present,ratMAir _PFIis the relative air quantity of the fuel low-pressure system (PFI) andratMAir _GDIis the relative amount of fuel in a high pressure fuel system (GDI).

Since the gasoline motor is an air-conducting system, the desired division factor (division factor) is first calculated on the air path and the necessary fuel quantity is then obtained for the respective injection path:

，

wherein the division factor is applied

. In the forward path, the respective fuel quantity is then determined therefrom:

whereinratMFu _PFIIs the amount of fuel for the intake pipe pressure injection PFI,facLamAdpn _PFIis an adaptation value for the intake pipe pressure injection PFI,ratLamAdpn _PFIis an offset value for the intake pipe pressure injection,ratMFu _GDIis the amount of fuel used for direct injection of GDI,facLamAdpn _GDIis an adaptation value for direct injection GDI andratLamAdpn _GDIis an offset value for direct injection.

An overdetermined system of equations can now be formed for this correlation and solved using the previously described scheme. Since the two offset corrections also enter the calculation chain and do not correspond to the division factorfacSpltSo that the two offset corrections can first be combined into a uniform adaptation valueratLamAdpn _gesAnd the following system of equations is generated:

wherein

Is the average value of the amount of fuel,

is an average value of the relative air amounts of the intake pipe pressure PFI,

is an average of the relative air amounts for direct injection of GDI,facLamAdpn _PFIandfacLamAdpn _GDIis an adaptation value andratLamAdpn _gesare common offset values.

Thereby coming directly from the dividing operationFactor adaptation value for in-line adaptationfacLamAdpn _PFIAndfacLamAdpn _GDIfor both injection modes. For offset adaptation, only one common adaptation value is usedratLamAdpn _gesAnd (5) standby. If the GDI is then, for example, operated purely, the individual offset for the GDI path can be determined, since in this case no injection takes place from the PFI path. For this purpose, a single equation and thus a single operating point are sufficient for determining:

。

first the offset value generated from this equationratLamAdpn _GDINot only air errors but also fuel errors may be involved due to physical properties. The sum of the air error and the physical fuel error is also included inratLamAdpn _gesThe method comprises the following steps:

。

therefore, the true physical offset error of the PFI path can be obtained from these two equations:

。

accordingly, in pure PFI operation, the physical offset error of the GDI path can be obtained:

。

once the two physical errors of the fuel path are obtained, the relative air error can be obtained:

。

once the three errors are obtained, the adaptation value can be matched to the errors and the following correlation used in the forward path of fuel determination:

wherein

And thus:

。

fig. 2 shows a functional diagram for illustrating the method. In a first step 500 the release conditions for the method are checked. If a stable or quasi-stable operating mode or operating point for the internal combustion engine 100 is determined, a release for the adaptation method is given.

A stable or quasi-stable operating mode or operating point for the internal combustion engine 100 is present, for example, if the rotational speed and/or the mass air flow and/or the motor torque and/or the accelerator pedal position do not change substantially within a predetermined time interval. For the method described, it is advantageously checked whether a relative air quantity is present in a predetermined time interval: (ratMAir) Small variations of (a). If the relative amounts of air (c:)ratMAir) Does not exceed a predefinable threshold value in the time interval, it can be assumed that a stable or quasi-stable operating mode or operating point for the internal combustion engine 100 is present and a release is given. Another necessary criterion for the method is the division factorfacSpltBetween one and zero. The division factor specifies a division for the quantity of fuel injected by the intake manifold injection valve and the direct injection valve. The advantage of said method compared to the known prior art methods is that the intake manifold injection can be run simultaneouslyThe intake pipe injection path and the direct injection path are diagnosed at the time of injection and direct injection. It is not necessary to cut off any path for the diagnosis (invasive test).

Thus, if a stable or quasi-stable operating state is identified, a release for continuing the method in step 510 is given. Another release condition is preferably that the current motor temperature exceeds a predeterminable temperature. The motor temperature is preferably obtained by the controller 115 via a temperature sensor or via a temperature model based on sensor values.

In step 510, the relative amount of air will be usedratMAirRelative injection amountratMFugesCurrent value of and the division factorfacSpltRead in and stored in the controller 115. Preferably, the relative injection quantities can be replacedratMFugesUsing the necessary injection amountratMFunotWherein the necessary injection amountratMFunotThe injection quantity required to achieve the target lambda predefined by the lambda controller.

In this case, the division factor is determined from one or more characteristic curves in the control unit 115 as a function of the current motor operating pointfacSplt。The parameters are read in mainly with each fuel injection.

The variables acquired in step 510 for carrying out the respective injection process form a respective current database. Preferably with said division factorfacSpltRelative air quantity for direct injection path GDI is also obtainedratMAirRelative air quantities of GDI and intake pipe injection path PFIratMAirParameters of PFII. For a division factor of onefacSpltIn other words, the injection operation is performed entirely by the intake pipe injection (PFI). And for a division factor of zerofacSpltFor values between zero and one, the known split operation is involved. Here, the injection takes place via the two injection paths. The virtual relative air volumeratMAir _PFI 、 ratMAir _GFIBy means of the followingThe formula is obtained in the controller 115:

。

then, in step 520, the current database is updatedratMFu _Ges 、ratMAir _Ges 、ratMAir _PFI 、ratMAir _GDI 、 facSpltFor mixture adaptation and storage in a combined characteristic curveK ₁In (1). Preferably according to the relative air quantities obtainedratMAir _GesAnd the current segmentation factorfacSpltTo obtain a characteristic curve capable of being combinedK ₁And then storing the database. The number of regions can be freely configured and is defined in the controller 115 in the preparation phase of the application phase as a function of the load region and the division factor. The combined characteristic curveK ₁Relative injection quantities are included hereratMFu _Ges 、Relative air volumeratMAir _GesAnd a division factorfacSpltThe parameter (c) of (d). Preferably as the relative amount of airratMAir _GesCan also use the relative air quantity of the intake pipe pressure path PFI or the direct injection path GDIratMAir _PFI、ratMAir _GDI. The combined characteristic curveK ₁Preferably having at least three zones.

In step 530, the characteristic curve obtained in step 510 and having the characteristic curve already stored therein is storedK ₁The data base of the values in (a) is calculated as the respective average value

、

、

、

And stored. In the initial state for the controller 115, there are not yet sufficient measurements for the combined characteristic curve to be used in this caseK ₁To generate an average value. In this case, therefore, a check is made in the control unit 115 as to whether a predeterminable number of measurements have been made per region. If each region falls below a predetermined number of fewer than three measurements, the method continues from the beginning in step 500. If the number of measurements that can be specified is exceeded, the method is continued in step 540.

Subsequently, in step 540, the characteristic curve is combinedK ₁Obtaining the offset value based on the databaseratLamAdpn _gesAnd separately obtains adaptation values for the intake pressure injection path PFIfacLamAdpn _PFIAnd adaptation value for direct injection path GDIfacLamAdpn _GDI. For this purpose, the following system of equations is solved:

。

in this case, a uniform offset value is obtainedratLamAdpn _gesAnd a first adaptation value for the direct injection path GDIfacLamAdpn _GDIAnd a second adaptation value for the intake pipe injection path PFIfacLamAdpn _PFI。

The first and second adaptation values are processed in step 550facLamAdpn _GDI、facLamAdpn _PFIAnd offset valueratLamAdpn _gesTo the injection model calculated on the controller 115. The injection model here uses the first and second adaptation values and the offset value to determine an adapted mixture formation and to carry out an injection process in the next injection process.

Checked in step 560, the first adaptation value for the direct injection GDIfacLamAdpn _GDIWhether it is in a predeterminable threshold range S₁Within. If the first adaptation valuefacLamAdpn _GDIAbove or below the threshold band S₁There are drawbacks to the direct-injection GDI. In the case of a defect being determined, the error coordinator can additionally carry out different substitution reactions. The threshold band S₁Is defined by a predefinable upper threshold value and a lower threshold value. Preferably, a malfunction indicator light in the instrument panel is activated or the vehicle is switched into emergency operation. The method then continues in step 570.

Checked in step 570, is the second adaptation value for the intake pipe injection PFIfacLamAdpn _PFIWhether it is in a predeterminable threshold range S₂Within. If the second adaptation valuefacLamAdpn _PFIAbove or below the threshold band S₂There is a drawback to the intake pipe injection PFI.

In the case of a defect being determined, the error coordinator can additionally carry out different substitution reactions. The threshold band S₂Is defined by a predefinable upper threshold value and a lower threshold value. Preferably, a malfunction indicator light in the instrument panel is activated or the vehicle is switched into emergency operation.

The method then starts from the beginning in step 500.

Claims

1. Method for adapting the mixture formation of an internal combustion engine (115) having a dual fuel metering device, for which an intake-pipe-based fuel metering and a direct fuel metering are carried out in a mixing mode, characterized in that a first adaptation value (PFI) for an intake-pipe injection (PFI) is determined in the mixing modefacLamAdpn _PFI) And a second adaptation value for direct injection (GDI) (GDI)facLamAdpn _GFI) And offset value for injection amount of (ratLamAdpn _ges) Wherein (c) is determined according to the first and second adaptation valuesfacLamAdpn _GFI；facLamAdpn _PFI) To be provided withAnd the offset value (ratLamAdpn _ges) To adapt and diagnose the mixture formation.

2. The method according to claim 1, characterized in that an adapted release for the mixture formation is given if there are stable or quasi-stable operating conditions for the internal combustion engine (115).

3. Method according to claim 2, characterized in that if the relative air quantities in a predeterminable time period (c) are (d)ratMAir _ges) If the change in the value of (c) does not exceed a predefinable threshold value, a stable or quasi-stable state exists.

4. Method according to any one of the preceding claims, characterized in that (a) is calculated from the relative amounts of airratMAir) Injection amount (b)ratMFu _ges) And a division factor (facSplt) To obtain a first adaptation value (PFI) for the intake pipe injection (PFI)facLamAdpn _PFI) And a second adaptation value for the direct injection (GDI) (GDI)facLamAdpn _GDI) And an offset value ofratLamAdpn _Ges）。

5. The method of claim 4, wherein (c) is determined based on the segmentation factor (cfacSplt) To adjust the relative air quantity (ratMAir) Relative air quantity divided into intake pipe pressure Path (PFI) ((ratMAir _PFI) And relative air volume of direct injection path (GDI) (GDI)ratMAir _GDI) Wherein the relative amounts of air (A), (B), (C) and (C) are usedratMAir) The injection amount (c)ratMFu _ges) Of a division factor of (facSplt) Relative air quantity of the intake pipe pressure Path (PFI) ((ratMAir _PFI) And relative air volume of direct injection path (GDI) ((ratMAir _GDI) The values obtained during the current injection operation are stored as average values in a characteristic curve (c)K ₁) Within.

6. The method of claim 5, wherein n for each injection quantity (A), (B), and (C) is given

) Relative air quantity of the intake pipe pressure Path (PFI) ((

) And relative air volume of direct injection path (GDI) (GDI)

) And a first adaptation value (PFI) to be determined for the intake pipe injection (PFI)facLamAdpn _PFI) Second adaptation value for direct injection (GDI) ((GDI))facLamAdpn _GFI) And offset value for injection quantity (ratLamAdpn _ges) Is solved by a linear system of equations of n dimensions, where n corresponds to the combined characteristic curve: (K ₁) The number of regions in (a).

7. Method according to claim 6, characterized in that for said injection quantity (C:)

) Relative air quantity of the intake pipe pressure Path (PFI) ((

) And relative air volume of direct injection path (GDI) (GDI)

) Is determined by the combined characteristic curve (a)K ₁) At least three measurements per region.

8. The method according to any of the preceding claims, characterized in that if the first adaptation value (C & (R) is not equal tofacLamAdpn _GDI) Exceeding or falling below a predefinable threshold band (S ₁) Then a defect for the direct injection (GDI) is identified.

9. The method according to any of the preceding claims, characterized in that if the second adaptation value (C & (R) is not available, (B) is not availablefacLamAdpn _PFI) Exceeding or falling below a predefinable threshold band (S ₂) A defect for the intake pipe injection (PFI) is identified.

10. A computing unit (115) which is set up to carry out the method according to one of the preceding claims.

11. Computer program which, when executed on a computing unit (115), causes the computing unit (115) to carry out the method according to any one of claims 1 to 9.

12. A machine readable storage medium having stored thereon a computer program according to claim 11.