DE102020110396A1 - Method for the detection of defective injection nozzles of an internal combustion engine - Google Patents

Abstract

The invention relates to a method for detecting defective injection nozzles (INi) for supplying fuel to the combustion chambers (BKi) of a vehicle internal combustion engine with one or more cylinder banks (ZB), a respective cylinder bank (ZB) having several cylinders each with one formed therein Combustion chamber (BKi) and at least one injection nozzle (INi) and wherein the combustion chambers (BKi) of a respective cylinder bank (ZB) are supplied with a common air mass flow and a common exhaust gas flow is discharged from the combustion chambers (BKi) of a respective cylinder bank (ZB) to a lambda probe will. In the method according to the invention, a standard deviation value (oi) for each injection nozzle (INi) and a total leakage flow (L0, sum) are determined. A number of successive test steps (TSj) are carried out for a respective cylinder bank (ZB) when the internal combustion engine is idling, with the reciprocal of the standard deviation value (oi) determined for the respective injection nozzle (INi) for the individual injection nozzles in a respective test step (TSj) is set, with the exception of one of the injection nozzles (INi), for which a predetermined classification parameter (a) is selected. During the test steps (TSj), measurements of the lambda value of the exhaust gas flow and measurements of the supplied air mass flow are carried out. After the number of test steps (TSj) has been carried out, a distribution key (oi) for each injection nozzle (INi) and a residual leakage value (ERL0, sum) are determined. The determination of distribution factors (vi) for the respective injection nozzles (INi) and the residual leakage value (ERL0, sum) is carried out by means of a computer-aided solution of a matrix calculation, which includes an equation for a respective test step (TSj), which contains the distribution key (oi) and the residual leakage value (ERL0, sum) as a function of the reciprocal norm deviation values (oi) set in the respective test step (TSj) and the classification parameter (a), and a lambda value (λreal, k) that is valid for the respective test step (TSj) and derived from the measurements of the lambda value ) describes. The distribution factors (vi) for the respective injection nozzles (INi) can then be determined from this.

Description

The invention relates to a method for detecting defective injection nozzles for supplying fuel to the combustion chambers of an internal combustion engine, in particular in a motor vehicle, and a corresponding engine test device for detecting defective injection nozzles.

Internal combustion engines include injection nozzles for supplying fuel to the individual combustion chambers in the cylinders. The injection nozzles ensure that the desired mixture quality (air-fuel ratio) is suitably set according to the requirements of engine operation.

Injector nozzle defects in internal combustion engines are often difficult to identify when the injectors are installed. As a result, in the event of malfunctions in an internal combustion engine, which could be caused by defective injection nozzles, these nozzles are often replaced on suspicion. In many cases, this means that injection nozzles are incorrectly exchanged without a defect. As a consequence, repeated repairs are necessary. Furthermore, the motor manufacturer's warranty costs increase due to unnecessary repairs.

From the EP 3 194 750 B1 a method for detecting defective ice injection nozzles of an internal combustion engine is known, with which a standard deviation value for each injection nozzle and a total leakage flow can be determined, the standard deviation value for a respective injection nozzle describing a deviation of the fuel mass flow generated by the respective injection nozzle from a standard operating value of the respective injection nozzle and the total leakage flow describes the fuel mass flow caused by leakage of all injection nozzles of the respective cylinder bank. In the event that at least one standard deviation value for a respective injection nozzle lies outside a predetermined value range, a first injection nozzle defect is detected in the respective cylinder bank in the form of an injection quantity deviation of at least one injection nozzle. In the event that the total leakage flow is greater than a predetermined threshold value, a second injector effect is detected in the respective cylinder bank in the form of a leak from at least one injector.

With the help of this method, it is thus possible to identify injection nozzle effects with a statement about the type of defect without the injection nozzles having to be removed from the internal combustion engine. In the event that a leak is detected, this is offered in EP 3 194 750 B1 The method described also provides the possibility of determining which of the injection nozzles have a leak. For this purpose, the method makes use of the knowledge that overly rich injection mixtures and thus leaks can be detected via a connection between the uneven running and the mixture injection. A disadvantage of this procedure is a limited evaluation accuracy, since the underlying mathematical model only inadequately reflects reality. In addition, the detection of a leak based on the uneven running only offers an indirect correlation between defect and symptom.

The object of the invention is to create a method and a corresponding engine test device with which an improved selective possibility is created to determine which of the injection nozzles have a leak and to quantify the leakage.

This object is achieved by a method according to claim 1 and an engine test device according to the features of claim 12. Developments of the invention are given in the dependent claims.

The method according to the invention serves to identify defective injection nozzles for feeding fuel into the combustion chambers of an internal combustion engine, in particular in a motor vehicle. The internal combustion engine has one or more cylinder banks, each cylinder bank comprising several cylinders, each with a combustion chamber formed therein and at least one injection nozzle. In a preferred embodiment, exactly one injection nozzle is provided in each combustion chamber. A common air mass flow is fed to the combustion chambers of a respective cylinder bank. A common exhaust gas flow is also discharged from the combustion chambers of a respective cylinder bank. Each cylinder bank is also provided with at least one detection sensor system for the mixture quality of the combined exhaust gas flows, such as a lambda probe.

In the context of the method according to the invention, a standard deviation value for each injection nozzle and a total leakage flow are determined, the standard deviation value for a respective injection nozzle describing a deviation of the fuel mass flow generated by the respective injection nozzle from a standard operating value of the respective injection nozzle and the total leakage flow describing the fuel mass flow caused by leakages of all Injectors of the respective cylinder bank is caused. The standard deviation value for each injection nozzle and the total leakage flow are, for example, compared with the one in the EP 3 194 750 B1 described method determined and made available for further processing.

As part of the further method, a number of successive test steps are carried out for a respective cylinder bank when the internal combustion engine is idling. The number of test steps is greater than the number of cylinders in the respective cylinder bank. This is necessary because otherwise the system of equations described below cannot be solved unambiguously. In a respective test step, the reciprocal of the standard deviation value determined for the respective injection nozzle is set for the individual injection nozzles, with the exception of one of the injection nozzles for which a previously determined classification parameter, which represents a quality measure for the internal combustion engine in relation to the total leakage flow, is set. During the test steps, measurements of the lambda value of the exhaust gas flow removed from the cylinder bank (e.g. by means of a lambda probe) and measurements or estimates of the air mass flow supplied with the cylinder bank are carried out. Setting the reciprocal of the standard deviation value determined for the respective injection nozzle is understood to mean a deliberate trimming of the injector injection time (and thus the amount of fuel injected). If a factor value of 1 is set, the injector operation is unaffected compared to the specification of the engine control. If a factor value of <1 is set, the injector runs leaner and allows less fuel to flow through. If a factor value> 1 is set, the injector runs richer and allows more fuel to flow through. With the help of the measurement of air mass (inward) and exhaust gas lambda (outward), the masses can be balanced. The classification parameters are selected in particular in such a way that the equation system described later has a unique solution. The term lambda value (also referred to as the combustion air ratio) is known per se and describes the air-fuel ratio in relation to the fuel-specific stoichiometric air-fuel ratio.

After the number of test steps has been carried out, a distribution key for each injector and a residual leakage value are determined. The distribution key for a respective injection nozzle describes a proportion of the leakage proportion of the total leakage flow caused by the respective injection nozzle. The residual leakage value describes the difference between the total leakage flow and the sum of the leakages from the injection nozzles.

The determination of distribution factors for the respective injection nozzles and the residual leakage value takes place in such a way that a matrix calculation is solved with the aid of a computer, which includes an equation for a respective test step, which the distribution key and the residual leakage value depending on the reciprocal norm deviation values set in the respective test step and the classification parameter, and describes a lambda value that is valid for the respective test step and derived from the measurements of the lambda value.

The method according to the invention has the advantage that it can be detected in a simple manner from the known standard deviation values for each injection nozzle and the total leakage flow which injection nozzle has which proportion of the total leakage flow. The determination of the leakage flow per injection nozzle by means of the implementation of a number of successive test steps and the solution of a system of equations can be carried out with great precision using an engine test device.

According to an expedient embodiment, the distribution keys each represent a percentage factor which, when multiplied by the total leakage flow, delivers the leakage fuel mass flow through the respective injection nozzle.

According to an expedient embodiment, the distribution factors for the respective injection nozzles or the absolute distribution leaks determined from the distribution factors are output via an interface. At the interface it can then be determined, e.g. by a service technician, which of the injection nozzle or nozzles actually has or have a defect that requires replacement. Distribution factors or the absolute distribution leaks determined from the distribution factors for the respective injection nozzles can be output, for example, via a user interface. However, an output can also be understood to mean storage of the corresponding information in a digital memory that can be evaluated at a later point in time.

In a further refinement, a respective equation of the system of equations comprises the residual leakage value adjusted for a reciprocal correction factor determined in advance, the correction factor being dependent on the quality measure for the internal combustion engine in relation to the total leakage flow. The correction factor, as well as the classification parameter, can be determined from the residual leakage value by a classification procedure, preferably upstream of the method.

In particular, the classification parameter and the correction factor are determined by a comparison with the leakage flow, with, if the total leakage current is less than a first limit value, the classification parameter is set to a first classification parameter value and the correction factor is set to a first correction factor, and if the total leakage current is greater than a second limit value, the classification parameter to a second classification parameter value and the correction factor to a second Correction factor value can be set. In particular, the first limit value results from the difference between a limit value for a permissible leakage of an injector and a predefined standard deviation, which is multiplied by the number of cylinders. The second limit value results in particular from the sum of a limit value for a permissible leakage of an injector and a predefined standard deviation, which is multiplied by the number of cylinders. The classification parameters and correction factors specified for the respective class can be determined through previous tests (best practice procedure).

According to a further expedient refinement, the classification parameter is set to a third classification parameter value which is determined by interpolation between the first and the second classification parameter value. The correction factor is preferably set to a third correction factor value which is determined by interpolation between the first and the second correction factor value when the total leakage current is greater than the first limit value and less than the second limit value. In other words, if the total leakage current is between the first and the second limit value, the classification parameter and the correction factor are not specified to a fixed value, but rather an interpolation is carried out, the values then used for the classification parameter and the correction factor being mutually exclusive determined by interpolation between the first and the second classification parameter value or correction factor value. In order to be able to carry out this interpolation, for an internal combustion engine whose "quality measure" in relation to the total leakage current is neither in the first class (the total leakage current is less than the first limit value) nor in the second class (i.e. the total leakage current is greater than the second Limit value) falls, metrologically the classification parameter and the correction factor value for the first class and then the classification parameter and the correction factor value for the second class are determined and the interpolation described is then carried out.

The internal combustion engine is therefore classified with regard to its quality on the basis of the total leakage current. The classification enables the classification parameter and the correction factor to be determined on the basis of the limit value for a permissible leakage of an injector and a predetermined standard deviation. The limit value for the permissible leakage and the associated permissible standard deviation can be specified and obtained in the development process for the injection nozzles. It is useful to start from the premise of a Gaussian distribution of the leakage flows present with different injection nozzles. The distribution is expediently limited to the limit value ± 5 * σ (standard deviation), which results in a coverage width that forms the basis for determining the correction factor and the classification parameter.

According to a further useful embodiment, the reciprocal standard deviation values for the respective test steps are set in such a way that a test step in which at least one injection nozzle generates a fuel mass flow with a reciprocal standard deviation value is followed by a test step in which this injection nozzle generates a fuel mass flow according to the previously determined classification parameter.

wobei i = 1, ..., M die Einspritzdüsen aus der Anzahl von M Einspritzdüsen der jeweiligen Zylinderbank indiziert;
wobei gv_i die im entsprechenden Testschritt eingestellten Gemischfaktoren in der Form einer Gemischvertrimmung repräsentieren, welche ein prozentualer Faktor ist, der durch Multiplikation mit dem von der jeweiligen Einspritzdüse im Normalbetrieb des Verbrennungsmotors erzeugten Kraftstoffmassenstrom den tatsächlichen Kraftstoffmassenstrom der jeweiligen Einspritzdüse liefert;
wobei alle cb_i auf den Wert 1 gesetzt sind oder wobei ein jeweiliges cb_i ein Gemischanpassungsparameter für eine jeweilige Einspritzdüse ist, der den durch die jeweilige Einspritzdüse bei deren Ansteuerung generierten Kraftstoffmassenstrom zur Erreichung von Laufruhe des Verbrennungsmotors anpasst und von einer Motorsteuerung eingestellt wird;
wobei 0_i der Normabweichungswert für eine jeweilige Einspritzdüse ist, der einen prozentualen Faktor darstellt, der durch Multiplikation mit dem Normbetriebswert der jeweiligen Einspritzdüse den durch die jeweilige Einspritzdüse erzeugten Kraftstoffmassenstrom liefert;
wobei L_st · λ_soll ein gewünschtes und für alle Brennkammern identisches Luft-Kraftstoff-Verhältnis bei einer Kraftstoffverbrennung in der Brennkammer ist und dabei L_st das stöchiometrische Luft-Kraftstoff-Verhältnis und λ_soll einen gewünschten Lambdawert der Kraftstoffverbrennung in der Brennkammer repräsentiert;
wobei MSHFM der für den jeweiligen Testschritt gültige Luftmassenstrom der gesamten Zylinderbank ist;
wobei L_0,sum der Gesamtleckagestrom aller Einspritzdüsen dieser Zylinderbank ist;
wobei λ_real,k der für den jeweiligen Testschritt gültige Lambdawert ist.In a particularly preferred variant of the method according to the invention, a respective determining equation of the system of equations reads as follows: $${\displaystyle \sum _i\left(G{v}_i\cdot c{b}_i\cdot O_i\right)}+\frac{{\mathrm{L.}}_{st}\cdot {\lambda }_{sOll}\cdot \mathrm{M.}}{\mathrm{M.}\mathrm{S.}H\mathrm{F.}\mathrm{M.}}\cdot {\mathrm{L.}}_{\mathrm{0,}sum}=\frac{\mathrm{M.}}{{\lambda }_{real,k}}$$

where i = 1, ..., M indicates the injection nozzles from the number of M injection nozzles of the respective cylinder bank;
whereby gv _i represent the mixture factors set in the corresponding test step in the form of a mixture trimming, which is a percentage factor which, by multiplication with the fuel mass flow generated by the respective injection nozzle during normal operation of the internal combustion engine, delivers the actual fuel mass flow of the respective injection nozzle;
where all cb _{i are set} to the value 1 or where a respective cb _{i is} a mixture adjustment parameter for a respective injection nozzle, which adapts the fuel mass flow generated by the respective injection nozzle when it is activated to achieve smoothness of the internal combustion engine and is set by an engine controller;
where 0 _{i is} the standard deviation value for a respective injection nozzle, which represents a percentage factor which, by multiplication with the standard operating value of the respective injection nozzle, delivers the fuel mass flow generated by the respective injection nozzle;
where L _st · λ _{should have} a desired air-fuel ratio that is identical for all combustion chambers a fuel combustion is in the combustion chamber and L _{st is} the stoichiometric air-fuel ratio and λ _{should represent} a desired lambda value of the fuel combustion in the combustion chamber;
whereby MSHFM is the air mass flow of the entire cylinder bank valid for the respective test step;
whereby L _{0, sum} is the total leakage flow of all injectors in this cylinder bank;
whereby λ _{real, k} is the lambda value valid for the respective test step.

In a further preferred variant, the system of equations processed in the method is solved using a matrix calculation. This ensures a robust solution to the equations. In order to obtain an exact and unambiguous solution of the system of equations, the mixture factors for the respective test steps are preferably set in such a way that after running through the test steps each injector does not generate a fuel mass flow at least once (ie is switched off), at least once generates a fuel mass flow that is greater than one is the fuel flow generated by the respective injection nozzle in normal operation of the internal combustion engine (i.e. the nozzle injects too rich), and at least once generates a fuel mass flow which is smaller than a fuel mass flow generated by the respective injection nozzle in normal operation of the internal combustion engine (i.e. the nozzle injects too lean a). Additionally or alternatively, the mixture factors for the respective test steps are preferably set in such a way that at least one test step exists in which all injection nozzles generate a fuel mass flow that corresponds to a fuel mass flow generated by the respective injection nozzle during normal operation of the internal combustion engine (i.e. the nozzles inject untrimmed, gv _i = 1) and / or that a test step in which at least one injection nozzle does not generate a fuel mass flow is followed by a test step in which each injection nozzle generates a fuel mass flow. In this way it is ensured that the individual injection nozzles run through all sections of their injection quantity characteristic curve, so that a reliable detection of injection nozzle defects is guaranteed.

According to a further expedient embodiment it is provided that in a case in which the numerical solution results in negative distribution factors for the respective injection nozzles, the relevant distribution factor is set to 0 (“zero”) for the correct balancing of the leakage flows. Alternatively, it can be provided that the method described herein is carried out for changed values of the classification parameter and the correction factor in order to avoid negative distribution factors.

In addition to the method described above, the invention also relates to an engine test device for detecting defective injection nozzles for feeding fuel into the combustion chambers of an internal combustion engine. The engine test device is set up to carry out the method according to the invention or one or more preferred variants of the method according to the invention. The engine test device can, for example, be an external engine test device or, if necessary, also be integrated in the motor vehicle.

The invention also relates to a motor vehicle with an internal combustion engine and injection nozzles for feeding fuel into the combustion chambers of the internal combustion engine, the motor vehicle comprising the engine test device described above.

An exemplary embodiment of the invention is described in detail below with reference to the accompanying drawing.

• 1 ein Ablaufdiagramm, welches eine Ausführungsform des erfindungsgemäßen Verfahrens verdeutlicht.

Show it:

• 1 a flowchart which illustrates an embodiment of the method according to the invention.

The method described below is based on a simple physical modeling of individual injection nozzles (hereinafter also referred to as injectors) in a cylinder bank of an internal combustion engine of a motor vehicle. The procedure is based on that in the EP 3 194 750 B1 described procedure. With this modeling, mathematical equations are set up in which the error patterns of the injectors act as unknown variables. The equations are filled with measurement data based on a number of test steps, which can then be used to solve for the unknown variables. The result of this procedure is the existence of a standard deviation value for each individual injector as well as a common total leakage flow for all cylinders of the cylinder bank of the internal combustion engine, which form the prerequisite for the computer-aided implementation of the method according to the invention for determining the leakages of a respective combustion chamber of the internal combustion engine.

Before details of the measurement sequence according to the invention are explained, the physical modeling on which the invention is based is first described in accordance with FIG EP 3 194 750 B1 described.

In the context of the modeling, an internal combustion engine is considered which has at least one cylinder bank with M cylinders and combustion chambers formed therein. A cylinder bank is characterized by a common air mass flow that is supplied to all the cylinders of the bank and that flows over an air mass meter (eg a hot film air mass meter) is recorded, as well as by a common exhaust gas flow of all cylinders of the bank, the lambda value of which is determined by a common lambda probe for all cylinders. Without loss of generality, an internal combustion engine with a single injector per cylinder of the cylinder bank is considered below.

The theoretical fuel mass flow ṁ _{K, th} in kg / h, which is generated by a respective injector of the cylinder bank when it is activated, can be described as follows:

Here designated MSHFM the air mass flow supplied to the cylinder bank, M corresponds to the number of cylinders or injectors (e.g. M = 4), L _st is the stoichiometric air-fuel ratio during combustion in the corresponding combustion chamber and λ _soll is the desired target lambda value (also referred to as the combustion air ratio) of the combustion.

In the following, the term fuel flow is used equivalent to the term fuel mass flow. The real fuel flow of an injector results from the theoretical fuel flow as follows:

Here, o denotes the deviation from the characteristic curve of the injector in the event of a corresponding defect. o is a percentage factor that represents the deviation of the real fuel flow compared to the theoretical fuel flow. For o = 1, the real fuel flow corresponds to the theoretical fuel flow, so that the corresponding injector does not have a characteristic curve defect. If o <1, the corresponding injector generates a mixture that is too lean with too little fuel compared to the pilot control. Correspondingly, if o> 1, the injector generates a mixture that is too rich with too much fuel.

gv = 1 bedeutet, dass der Injektor unvertrimmt einspritzt, gv < 1 bedeutet, dass der Injektor mager einspritzt, und gv > 1 bedeutet, dass der Injektor fett einspritzt. In der hier beschriebenen Ausführungsform wird ein Verbrennungsmotor betrachtet, bei dem der Kraftstoffstrom ferner über einen prozentualen Anpassungsfaktor in der Form eines sog. Cylinder-Balancing-Faktors adaptiert wird. Dieser Faktor wird durch die Motorsteuerung bestimmt und dient mittels Zylindergleichstellung dazu, eine hohe Laufruhe des Verbrennungsmotors zu erreichen.The real fuel flow just described occurs during normal operation of the internal combustion engine, that is to say in an operation in which no test intervention is carried out on the engine. As part of the test sequence described in more detail below, this real fuel flow can be trimmed, which is achieved by a suitable setting of a percentage factor gv, which is also referred to below as mixture trimming. The trimmed, real fuel flow of the injector is as follows:

gv = 1 means that the injector injects untrusted, gv <1 means that the injector injects lean, and gv> 1 means that the injector injects richly. In the embodiment described here, an internal combustion engine is considered in which the fuel flow is also adapted via a percentage adaptation factor in the form of a so-called cylinder balancing factor. This factor is determined by the engine control and, by means of cylinder synchronization, is used to achieve extremely smooth running of the internal combustion engine.

cb = 1 bedeutet, dass keine Adaption des Kraftstoffstroms des Injektors stattfindet, cb < 1 bedeutet, dass der Kraftstoffstrom abgemagert wird, und cb > 1 bedeutet, dass der Kraftstoffstrom angefettet wird. Im Rahmen der Modellierung des Injektors wird ferner eine reale Injektorleckage betrachtet, welche einen Leckagestrom L₀ von über eine Leckage austretendem Kraftstoffstrom darstellt. Der Leckagestrom ist in kg/h angegeben.Taking into account the cylinder balancing factor, which is referred to as cb in the following, the real trimmed fuel flow of an injector results as follows:

cb = 1 means that there is no adaptation of the fuel flow of the injector, cb <1 means that the fuel flow is leaned, and cb> 1 means that the fuel flow is enriched. In the context of the modeling of the injector, a real injector leakage is also considered, which represents a leakage flow L ₀ of a fuel flow exiting via a leakage. The leakage flow is given in kg / h.

Overall, the real fuel flow of an injector can be modeled as follows:

The fuel flow ṁ _{K, th has been} replaced by the expression of the above equation (1).

Without loss of generality, it is assumed that the internal combustion engine should burn stoichiometrically, that is, λ _should = 1. This results in the real fuel flow taking into account the trimming and the cylinder balancing factor as follows:

In the following, the mass flows of the individual injectors of the corresponding cylinder bank are now considered, so that the index i is introduced, which designates a corresponding injector or the assigned cylinder or the assigned combustion chamber in the cylinder bank.

_{For a cylinder bank, an actual lambda value λ real is determined} via a lambda probe for all mass flows of the cylinders that are brought together in an exhaust gas line. Due to the conservation of mass, the real mass flows present at the lambda probe result as follows:

Here, ṁ _{K, real, i} denotes the fuel flow of a respective injector and ṁ _{L, i} denotes the air mass flow supplied to a respective cylinder. Here and in the following, the sums over the index i denote a summation over the cylinders of the cylinder bank.

_{The lambda value λ real} results from the above mass flows at the lambda probe as follows:

By inserting the fuel flow from equation (2) above into equation (3), the following lambda value results at the lambda probe: $${\lambda }_{real}=\frac{\mathrm{M.}\mathrm{S.}H\mathrm{F.}\mathrm{M.}}{{\mathrm{L.}}_{st}\cdot \left(\frac{\mathrm{M.}\mathrm{S.}H\mathrm{F.}\mathrm{M.}}{\mathrm{M.}\cdot {\mathrm{L.}}_{st}}\cdot {\displaystyle \sum _i\left(G{v}_i\cdot c{b}_i\cdot O_i\right)}+{\displaystyle \sum _i{\mathrm{L.}}_{\mathrm{0,}i}}\right)}$$

gilt.This equation also takes into account that

is applicable.

Via a transformation and the use of the expression L _{0, sum} = ∑L _{0, i} , λ _real i can be written as follows: $${\lambda }_{real}=\frac{\mathrm{M.}}{\left({\displaystyle \sum _i\left(G{v}_i\cdot c{b}_i\cdot O_i\right)}+\frac{{\mathrm{L.}}_{st}\cdot \mathrm{M.}}{\mathrm{M.}\mathrm{S.}H\mathrm{F.}\mathrm{M.}}\cdot {\mathrm{L.}}_{\mathrm{0,}sum}\right)}$$

Changing equation (4) results in the following basic equation, which is used in the context of the embodiment described here for the detection of defective injection nozzles: $${\displaystyle \sum _i\left(G{v}_i\cdot c{b}_i\cdot O_i\right)}+\frac{{\mathrm{L.}}_{st}\cdot \mathrm{M.}}{\mathrm{M.}\mathrm{S.}H\mathrm{F.}\mathrm{M.}}\cdot {\mathrm{L.}}_{\mathrm{0,}sum}=\frac{\mathrm{M.}}{{\lambda }_{real}}$$

As described in more detail below, the above equation (5) is used for several test steps with differently set mixture trims gv _i from which different lambda values follow. N test steps are considered, where N is greater than the number M. the cylinder is. In this way a system of equations with N equations of the type (5) is obtained, from which the unknowns in the form of the M deviations from the characteristic curve o _i as well as the total leakage current L _{0, sum} are determinable. The air mass flow MSHFM in this system of equations is the air mass flow measured and averaged over the entire test sequence of the individual test steps, which is assumed to be approximately constant. The lambda value used below λ _{real, k} is the one for the respective test step k (k = 1, ..., N) certain lambda value area and represents the mean value of the individual lambda values measured in the period of the respective test step. The individual cylinder balancing factors cb _i can be read out from the engine control unit.

According to the above, a linear system of N equations results from the above equations, which in matrix notation reads as follows: $${\mathrm{M.}}_{Gv}\cdot V_O=V_{\lambda }$$

M _gv denotes the specified test matrix, which defines the structure of the test and has the dimension N x (M + 1).

Without loss of generality, a cylinder bank with M = 4 cylinders is shown below considered. The test matrix is then an N x 5 matrix, which is as follows: $${\mathrm{M.}}_{Gv}=\left(\begin{array}{ccccc}G{v}_{1.1}& G{v}_{1.2}& G{v}_{1.3}& G{v}_{1.4}& 1\\ G{v}_{2.1}& G{v}_{2.2}& G{v}_{2.3}& G{v}_{2.4}& 1\\ G{v}_{3.1}& G{v}_{3.2}& G{v}_{3.3}& G{v}_{3.4}& 1\\ \cdots & \cdots & \cdots & \cdots & 1\\ G{v}_{N\mathrm{,1}}& G{v}_{N\mathrm{,\; 2}}& G{v}_{N\mathrm{,\; 3}}& G{v}_{N\mathrm{,\; 4}}& 1\end{array}\right)$$

A line to match k relates to the kth test step and a corresponding column i to cylinder i. The values gv _{k, i} thus provide the correspondingly set trimmings of the individual cylinders i in the respective test steps k designated. The trimming is achieved by adapting the injection time of the respective injection nozzle accordingly. gv _{k, i} = 0 means that the corresponding injector does not inject any fuel and is therefore switched off. In contrast, gv _{k, i} <1 means that the corresponding injector injects leaner, while gv _{k, i} > 1 means that the corresponding injector injects more richly, gv _{k, i} = 1 represents the untrimmed state of the injector without external trim intervention The terms leaner, richer and undershot each refer to the state of the injection as it would be provided by the pilot control of the engine control unit without test intervention.

As can be seen from the explanations below, the last column of the above matrix M _{gv includes the total leakage current} in the linear system of equations. In a specific example, _{a second matrix line (1.3 0 1.3 1.3 1) in the above matrix M gv} means that in the second test step cylinders 1, 3 and 4 are set 30% richer and cylinder 2 is switched off .

In the above matrix equation (6), the vector V _{0 represents} the defect vector of the injectors of the cylinder bank under consideration, according to which the solution is to be performed. This vector is as follows: $$V_O=\left(\begin{array}{c}{\mathrm{<figure-callout\; id="1"\; label="E.\; O"\; filenames="DE102020110396A1\_0030.png"\; state="\{\{state\}\}">E.</figure-callout>}}_O\\ {\mathrm{E.}}_{O2}\\ {\mathrm{E.}}_{O3}\\ {\mathrm{E.}}_{O\mathrm{4th}}\\ {\mathrm{E.}}_{\mathrm{L.}O,sum}\end{array}\right)=\left(\begin{array}{c}c{\mathrm{<figure-callout\; id="1"\; label="b"\; filenames="DE102020110396A1\_0030.png"\; state="\{\{state\}\}">b</figure-callout>}}_1\cdot O_1\\ c{b}_2\cdot O_2\\ c{b}_3\cdot O_3\\ c{b}_{\mathrm{4th}}\cdot O_{\mathrm{4th}}\\ {\mathrm{L.}}_{st}\cdot \raisebox{1ex}{$\mathrm{4th}$}\!\left/ \!\raisebox{-1ex}{$\mathrm{M.}\mathrm{S.}H\mathrm{F.}\mathrm{M.}$}\right.\cdot {\mathrm{L.}}_{\mathrm{0,}sum}\end{array}\right)$$

Here, E _oi denotes error values of the respective injectors i and E _{Lo, sum} the sum of the error value of the total leakage current. As already mentioned above, o ₁ to o ₄ denote the corresponding deviations from the characteristic curve of the respective injector, a value of the characteristic curve deviation greater than 1 means that the injector is injecting too rich, a value of 1 means that the injector is injecting properly, and one value less than 1 means that the injector is injecting too lean. As also mentioned above, designated L _{0, sum} the sum of the leakages of all injectors on a bank in kg / h. Furthermore, the factors cb ₁ to cb _{4 are} the corresponding cylinder balancing factors that can be read from the engine control unit. L _st is the stoichiometric air-fuel ratio already defined above, which, depending on the chemical fuel composition, is between 14 and 16 kg / kg, for example. MSHFM is the air mass flow measured and determined over the entire test process via the hot film air mass meter.

In the above matrix equation (6), V _λ denotes the vector from the lambda values associated with the test steps λ _{real, k} . This vector is as follows: $$V_{\lambda }=\left(\begin{array}{c}\mathrm{4th}/{\lambda }_{real\mathrm{,1}}\\ \mathrm{4th}/{\lambda }_{real\mathrm{,\; 2}}\\ \mathrm{4th}/{\lambda }_{real\mathrm{,\; 3}}\\ \cdots \\ \mathrm{4th}/{\lambda }_{real,N}\end{array}\right)$$

die obige Gleichung (6) wie folgt umgeformt: $$M_{gv}{}^T\cdot M_{gv}\cdot V_o=M_{gv}{}^T\cdot V_{\lambda }$$

In the context of the embodiment described here, a matrix operation is carried out with the aid of the transposed matrix ${\mathrm{M.}}_{Gv}^T$

the above equation (6) transformed as follows: $${\mathrm{M.}}_{Gv}{}^T\cdot {\mathrm{M.}}_{Gv}\cdot V_O={\mathrm{M.}}_{Gv}{}^T\cdot V_{\lambda }$$

_{From this, the vector V 0} is determined using a matrix operation with an inverse matrix: $$V_O={\left({\mathrm{M.}}_{Gv}{}^T\cdot {\mathrm{M.}}_{Gv}\right)}^{-1}\cdot \left({\mathrm{M.}}_{Gv}{}^T\cdot V_{\lambda }\right)$$

Die einzelnen Kennlinien der Injektoren und der Gesamtleckagestrom bilden die Ausgangsinformationen zur Bestimmung eines Verteilungsschlüssels θ_i für jeden Injektor IN_i sowie eines Restleckagewerts ERL_0,sum. Mit Hilfe der einzelnen Kennlinienabweichungen o_i wird eine Konzentrationsmatrix M_L gebildet, wobei diese analog der Testmatrix M_gv aufgebaut wird. Die Konzentrationsmatrix M_L für Leckagen ist eine Matrix der Dimension N x (M+1), wobei nachfolgend ohne Beschränkung der Allgemeinheit eine Zylinderbank mit M = 4 Zylindern betrachtet wird. Die Konzentrationsmatrix M_L ist dann eine N x 5 Matrix, die wie folgt lautet: $$M_L=\left(\begin{array}{ccccc}a& 1/o_2& 1/o_3& 1/o_4& 1\\ 1/o_1& a& 1/o_3& 1/o_4& 1\\ 1/o_1& 1/o_2& a& 1/o_4& 1\\ 1/o_1& 1/o_2& 1/o_3& a& 1\\ 1/o_1& 1/o_2& 1/o_3& 1/o_4& 1\end{array}\right)$$

From this vector, the individual deviations from the characteristic curve and the total leakage current can be determined as follows, taking into account the above equation (7): $$\begin{array}{c}{\mathrm{<figure-callout\; id="1"\; label="O"\; filenames="DE102020110396A1\_0030.png"\; state="\{\{state\}\}">O</figure-callout>}}_1=\raisebox{1ex}{${\mathrm{E.}}_{O1}$}\!\left/ \!\raisebox{-1ex}{$\mathrm{<figure-callout\; id="1"\; label="c\; b"\; filenames="DE102020110396A1\_0030.png"\; state="\{\{state\}\}">c</figure-callout>}{b}_{}{}_1$}\right.\\ \cdots \\ O_i=\raisebox{1ex}{${\mathrm{E.}}_{Oi}$}\!\left/ \!\raisebox{-1ex}{$c{b}_i$}\right.\\ \cdots \\ {\mathrm{L.}}_{\mathrm{0,}sum}=\raisebox{1ex}{${\mathrm{E.}}_{\mathrm{L.}\mathrm{0,}sum}$}\!\left/ \!\raisebox{-1ex}{$\left({\mathrm{L.}}_{st}\cdot \mathrm{4th}/\mathrm{M.}\mathrm{S.}H\mathrm{F.}\mathrm{M.}\right)$}\right.\end{array}$$

The individual characteristics of the injectors and the total leakage flow form the initial information for determining a distribution key θ _i for each injector IN _i as well as a residual leakage value ERL _{0, sum} . With the help of the individual deviations from the characteristic o _i a concentration _{matrix M L is} formed, this being built up analogously to the test matrix M _gv . The concentration _{matrix M L} for leaks is a matrix of dimensions N x (M + 1), with a cylinder bank with M = 4 cylinders being considered below without loss of generality. The concentration _{matrix M L} is then an N x 5 matrix, which reads as follows: $${\mathrm{M.}}_{\mathrm{L.}}=\left(\begin{array}{ccccc}a& 1/O_2& 1/O_3& 1/{\mathrm{<figure-callout\; id="1"\; label="O\; 4th"\; filenames="DE102020110396A1\_0030.png"\; state="\{\{state\}\}">O</figure-callout>}}_{\mathrm{4th}}& 1\\ 1/O_1& a& 1/O_3& 1/{\mathrm{<figure-callout\; id="1"\; label="O\; 4th"\; filenames="DE102020110396A1\_0030.png"\; state="\{\{state\}\}">O</figure-callout>}}_{\mathrm{4th}}& 1\\ 1/{\mathrm{<figure-callout\; id="1"\; label="O"\; filenames="DE102020110396A1\_0030.png"\; state="\{\{state\}\}">O</figure-callout>}}_1& 1/O_2& a& 1/{\mathrm{<figure-callout\; id="1"\; label="O\; 4th"\; filenames="DE102020110396A1\_0030.png"\; state="\{\{state\}\}">O</figure-callout>}}_{\mathrm{4th}}& 1\\ 1/{\mathrm{<figure-callout\; id="1"\; label="O"\; filenames="DE102020110396A1\_0030.png"\; state="\{\{state\}\}">O</figure-callout>}}_1& 1/O_2& 1/O_3& a& 1\\ 1/{\mathrm{<figure-callout\; id="1"\; label="O"\; filenames="DE102020110396A1\_0030.png"\; state="\{\{state\}\}">O</figure-callout>}}_1& 1/O_2& 1/O_3& 1/O_{\mathrm{4th}}& 1\end{array}\right)$$

A line to match k relates to the kth test step and a corresponding column i to cylinder i. Thus, the values 1 / o _i and a classification parameter a, the corresponding reciprocal deviations from the characteristic curve of the individual cylinders i in the respective test steps k designated. The classification parameter a represents a quality measure of the internal combustion engine in relation to the total leakage flow L _{0, sum} and is determined by a classification.

The classification is made by comparing the total leakage flow L _{0, sum} with a first limit value GW1 'and a second limit value GW2'. The first limit value GW1 'results from the difference between a limit value GW for a permissible leakage of an injector IN _i and a predetermined standard deviation σ, which is multiplied by the number of cylinders, ie GW1 '= number of cylinders * (GW - 5 * σ). The multiplier 5 was determined to be preferred, but it can also be chosen differently. Is the total leakage current L _{0, sum} below the first limit value GW1 ', an internal combustion engine of a first class A1 is present. Is the total leakage current L _{0, sum} above the second limit value GW2 ', which is the sum of the limit value GW for the permissible leakage of an injector IN _i and the predefined standard deviation σ, which is multiplied by the number of cylinders (ie GW2 '= number of cylinders * (GW + 5 * σ)), the internal combustion engine is in a second class A3. For classes A1 and A3, the classification parameter a has been determined with the help of a best practice procedure. For the first class A1, for example, a = 1, for the second class A3, for example, a = p _L0 . Here, p _{L0 is} a so-called coverage range, which results from limiting the distribution to the limit value GW + - 5 * σ, namely to p _L0 = 10 ^-6 .

This classification method also determines a further required correction factor f, the correction factor f also being determined by a best practice method. If the internal combustion engine falls into the first class A1, then f is defined as f = P _L0 ² . If the internal combustion engine falls into the second class A3, the correction factor f is set to f = p _L0 .

In the event that the total leakage flow L _{0, sum} lies between the first and the second limit value GW1 'or GW2' (third class A2), the classification parameter a and the correction factor f are interpolated using the values for a and f specified for classes A1 and A3 In this case, interpolation is the total leakage current L _{0, sum} .

If the classification parameter a is known, the concentration matrix M _L can be completely covered in accordance with the above form. The concentration matrix M _L controls the injectors in the k test steps IN _i according to the procedure described above. In every test step k are the deviations from the characteristic, ie the component defects of the injectors IN _i due to the reciprocal deviation from the characteristic o _i compensated.

As can be seen from the explanations, the last column of the concentration matrix M _L includes an error term in the system of equations. Furthermore, due to the diagonal arrangement of the classification factors a in the rows k the concentration matrix M _L ensures that the rows are linearly independent of one another, which improves the solvability of the system of equations.

ist ein Leckverteilungsvektor V_L dargestellt, wobei θ_i die Verteilungsschlüssel für den Gesamtleckagestrom L_0,sum und damit der gesuchte Lösungsraum sind. ERL_0,sum stellt einen Fehlerterm dar, damit die Gleichung numerisch stabil ist. Der Korrekturfaktor f wird als Reziprokwert mit dem Fehlerterm ERL_0,sum multipliziert. ERL_0,sum stellt einen numerischen Fehlereinfluss der Restleckage dar. Er dient dazu, zusammen mit dem Korrekturfaktor f die Modellgenauigkeit zu skalieren. Der Korrekturfaktor f bestimmt sich aus der oben beschriebenen Klassifikation.In the following equation $$V_{\mathrm{L.}}=\left(\begin{array}{c}{\mathrm{<figure-callout\; id="1"\; label="\theta "\; filenames="DE102020110396A1\_0030.png"\; state="\{\{state\}\}">\theta </figure-callout>}}_1\\ {\theta }_2\\ {\theta }_3\\ {\theta }_{\mathrm{4th}}\\ {ƒ}^{-1}\cdot \mathrm{E.}\mathrm{R.}{\mathrm{L.}}_{\mathrm{0,}sum}\end{array}\right)$$

a leakage distribution vector V _{L is} shown, where θ _{i is} the distribution key for the total leakage flow L _{0, sum} and are therefore the solution space we are looking for. ERL _{0, sum} represents an error term in order for the equation to be numerically stable. The correction factor f is multiplied _{as a reciprocal value with the error term ERL 0, sum.} ERL _{0, sum} represents a numerical error influence of the residual leakage. It is used to scale the model accuracy together with the correction factor f. The correction factor f is determined from the classification described above.

The lambda values λ _{real, k} for the measured value vector are within each measurement step k measured with the associated lambda probe, as described above. $$V_{\lambda }=\left(\begin{array}{c}\mathrm{4th}/{\lambda }_{real\mathrm{,1}}\\ \cdots \\ \cdots \\ \mathrm{4th}/{\lambda }_{real\mathrm{,\; 4}}\\ \mathrm{4th}/{\lambda }_{real\mathrm{,\; 5}}\end{array}\right)$$

According to the above, a linear system of equations with N equations results, which in matrix notation is as follows: $${\text{M.}}_{\text{L.}}*{\text{V}}_{\text{L.}}={\text{V}}_{\mathrm{\lambda }}$$

M _{L is} the concentration matrix, V _{L is} the distribution _{vector and V λ is} the vector with the lambda values determined from the test steps λ _{real, k} . Equation (11) can be solved for the leakage distribution vector V _L to $$V_{\mathrm{L.}}={\left({\mathrm{M.}}_{\mathrm{L.}}{}^T\cdot {\mathrm{M.}}_{\mathrm{L.}}\right)}^{-1}\cdot \left({\mathrm{M.}}_{\mathrm{L.}}{}^T\cdot V_{\lambda }\right)$$

The distribution factors v _i by determining the distribution key θ _i from the resolved leakage _{distribution vector V L} : $$v_i=\frac{{\mathrm{\Theta }}_i}{{\displaystyle {\sum }_i\left|{\mathrm{\Theta }}_i\right|}}$$

In this equation (13) gives the distribution factor v _i the proportional leakage per injector or cylinder again.

Should the numerical solution described above result in negative distribution factors v _i result, a respective distribution factor is required for the correct balancing of the leakage flows v _i set to "0". Alternatively, the procedure described above could be carried out with slightly changed values for the classification factor a and the correction factor f, until there are no negative distribution factors v _i result.

kann dann der zylinderindividuelle Leckagestrom L_0,i ermittelt werden, wobei sich dieser aus dem Produkt des jeweiligen Verteilungsfaktors v_i und dem Gesamtleckagestrom L_0,sum ergibt.By the equation below $${\mathrm{L.}}_{O,i}=v_i\cdot {\mathrm{L.}}_{O,sum}$$

the cylinder-specific leakage flow L _{0, i} can then be determined, this being derived from the product of the respective distribution factor v _i and the total leakage flow L _{0, sum} results.

1 summarizes the essential steps of the described method in a flowchart. The starting point is a cylinder bank E.g. with M operating chambers BK, each with an injection nozzle IN _i per combustion chamber BK _i is provided. This starting point is in 1 labeled ST. The first step is to determine the deviation from the characteristic o _i of each injector INi and the total leakage flow L _{0, sum} of the cylinder bank (step S1). The determination of the individual deviations from the characteristic curve as well as the total leakage current can be carried out with the in the EP 3 194 750 B1 procedures described.

In a second step S2, the classification parameter a and the correction factor f are determined on the basis of a comparison of the total leakage current L _{0, sum} with two differently selected limit values LV1 'and LV2'. Depending on whether the total leakage flow L _{0, sum} is less than the first limit value, greater than the second limit value or between the first and the second limit value, values for the classification parameters a and the correction factor f predetermined by best practice methods are selected.

As part of the test sequence, the reciprocal deviations from the characteristic curve are determined in a step S3 in respective test steps TSj (j = 1,..., N, where N> M) o _i the injectors IN _i set and the lambda value of the cylinder bank E.g. determined. At the same time, the classification parameter a is used in the concentration matrix used for this purpose.

According to step S4, a distribution factor is determined v _i for each injector IN _i as well as a residual leakage value ERL _{0, sum} through the computer-aided solution of a matrix calculation. By multiplying by the total leakage flow L _{0, sum} the cylinder-specific leakage flow L _{0, i} results.

List of reference symbols

E.g.

Cylinder bank

BKi

Combustion chamber

M.

Number of injectors

INi

Injector

gvi

Mixture trimmings

k

Test step index

N

Number of test steps

oi

Deviation from characteristic

L0, sum

Total leakage flow

MSHFM

Air mass flow

λreal, k

Lambda value valid in the respective test step

WB

Range of values

TH

Threshold

F.

function

LUR

Uneven running

Ä

Cylinder combustion air ratio

vi

Distribution factor

Θi

Distribution key

QUOTES INCLUDED IN THE DESCRIPTION

This list of the documents listed by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent literature cited

• EP 3194750 B1 [0004, 0005, 0009, 0028, 0029, 0074]

Claims (14)

Method for detecting defective injection nozzles (IN _i ) for feeding fuel into the combustion chambers (BK _i ) of an internal combustion engine, in particular in a motor vehicle, the internal combustion engine having one or more cylinder banks (ZB), with a respective cylinder bank (ZB) having several cylinders each with a combustion chamber (BK _i ) formed therein and at least one injection nozzle (INi) and wherein the combustion chambers (BK _i ) of a respective cylinder bank (ZB) are supplied with a common air mass flow and from the combustion chambers (BK _i ) of a respective cylinder bank (ZB ) a common exhaust gas flow is discharged, whereby - a standard deviation value (o _i ) for each injection nozzle (IN _i ) and a total _{leakage flow (L 0, sum} ) are determined, the standard deviation value (o _i ) for a respective injection nozzle (IN _i ) being a Deviation of the fuel mass flow generated by the respective injection nozzle (IN _i ) from a standard operating value of the respective injection nozzle e (IN _i ) describes and the total _{leakage flow (L 0, sum} ) describes the fuel mass flow that is caused by leakages of all injection nozzles (IN _i ) of the respective cylinder bank (ZB); - For a respective cylinder bank (ZB) in idle mode of the internal combustion engine, a number of successive test steps (TSj) is carried out, which is greater than the number of cylinders of the respective cylinder bank (ZB), in a respective test step (TSj) for the individual Injection nozzles the reciprocal of the standard deviation value (o _i ) determined for the respective injection nozzle (INi) is set, with the exception of one of the injection nozzles (IN _i ), for which a previously determined classification parameter (a), which is a quality measure for the internal combustion engine in relation to represents the total _{leakage flow (L 0, sum} ), is set, and measurements of the lambda value of the exhaust gas flow discharged from the cylinder bank (ZB) and measurements of the air mass flow supplied to the cylinder bank (ZB) are carried out during the test steps (TSj); - After the number of test steps (TSj) has been carried out, a distribution key (o _i ) for each injection nozzle (IN _i ) and a residual leakage value (ERL _{0, sum} ) are determined, the distribution _{key (o i} ) for a respective injection nozzle (INi) being one The proportion of the leakage proportion caused by the respective injection nozzle (IN _i ) in the total leakage flow (L _{0, sum} ) describes and the residual leakage value (ERL _{0, sum} ) describes a difference between the total _{leakage flow (L 0, sum} ) and the sum of the leakage proportions of the injection nozzles (IN _i ) describes; - the determination of distribution _{factors (v i} ) for the respective injection nozzles (IN _i ) and the residual leakage value (ERL _{0, sum} ) takes place in such a way that a matrix calculation is solved with the aid of a computer, which includes an equation for a respective test step (TSj) which contains the Distribution key (o _i ) and the residual leakage value (ERL _{0, sum} ) depending on the reciprocal norm deviation values (o _i ) set in the respective test step (TSj) and the classification parameter (a), and a valid and off for the respective test step (TSj) _{describes the lambda value (λ real, k} ) derived from the measurements of the lambda value. Procedure according to Claim 1 , characterized in that the distribution keys (o _i ) each represent a percentage factor which, when multiplied by the total _{leakage flow (L 0, sum} ), delivers the leakage fuel mass flow through the respective injection nozzle (IN _i). Procedure according to Claim 1 or 2 , characterized in that the distribution _{factors (v i} ) for the respective injection nozzles (INi) or the _{absolute distribution leaks determined from the distribution factors (v i} ) are output via an interface. Method according to one of the preceding claims, characterized in that a respective equation of the system of equations comprises the residual leakage value (ERL _{0, sum} ) adjusted by a previously determined reciprocal correction factor (f), the correction factor (f) depending on the quality measure for the internal combustion engine in Relation to the total leakage current (L _{0, sum} ). Method according to one of the preceding claims, characterized in that the classification parameter (a) and the correction factor (f) are determined by a comparison with the total leakage flow (L _{0, sum} ), wherein - if the total leakage flow (L _{0, sum} ) is less than a first limit value (GW1 '), the classification parameter (a) is set to a first classification parameter value and the correction factor (f) is set to a first correction factor value; - if the total _{leakage current (L 0, sum} ) is greater than a second limit value (GW2 '), the classification parameter (a) is set to a second classification parameter value and the correction factor (f) is set to a second correction factor value. Procedure according to Claim 5 , characterized in that the first limit value (GW1 ') is derived from the difference between a limit value (GW) for a permissible leakage of an injector (INi) and a given standard deviation (σ), which is multiplied by the number of cylinders. Procedure according to Claim 5 , characterized in that the second limit value (GW2 ') results from the sum of a limit value (GW) for a permissible leakage of an injector (INi) and a predetermined standard deviation (σ) which is multiplied by the number of cylinders. Method according to one of the Claims 5 until 7th , characterized in that the classification parameter (a) is set to a third classification parameter value which is determined by interpolation between the first and the second classification parameter value, and the correction factor (f) is set to a third correction factor value which is determined by interpolation between the first and the second correction factor _{value is determined if the total leakage current (L 0, sum} ) is greater than the first limit value (GW1 ') and less than the second limit value (GW2'). Method according to one of the preceding claims, characterized in that the reciprocal standard deviation values (o _i ) for the respective test steps (TSj) are set in such a way that a test step (TSj) in which at least one injection nozzle (IN _i ) generates a fuel mass flow with reciprocal Standard deviation value (o _i ) is generated, a test step (TSj) follows, in which each injection nozzle (IN _i ) generates a fuel mass flow according to the previously determined classification parameter (a). Method according to one of the preceding claims, characterized in that a respective equation for determining the standard deviation values (o _i ) of the total _{leakage flow (L 0, sum} ) and the system of equations is as follows: $${\displaystyle \sum _i\left(G{v}_i\cdot c{b}_i\cdot O_i\right)}+\frac{{\mathrm{L.}}_{st}\cdot {\lambda }_{sOll}\cdot \mathrm{M.}}{\mathrm{M.}\mathrm{S.}H\mathrm{F.}\mathrm{M.}}\cdot {\mathrm{L.}}_{\mathrm{0,}sum}=\frac{\mathrm{M.}}{{\lambda }_{real,k}}$$

where i = 1, ..., M _{indicates the injection nozzles (IN i} ) from the number of M injection nozzles (IN _i ) of the respective cylinder bank (ZB); where gv _{i represent the mixture factors (gv i} ) set in the corresponding test step (TSj) in the form of a mixture trimming, which is a percentage factor which, by multiplying with the _{fuel mass flow generated by the respective injection nozzle (IN i} ) in normal operation of the internal combustion engine, the actual fuel mass flow Supplies fuel mass flow to the respective injection nozzle (IN _i ); where all cb _{i are set} to the value 1 or where a respective cb _{i is} a mixture adjustment parameter for a respective injection nozzle (IN _i ), which adjusts the fuel mass flow generated by the respective injection nozzle (INi) when it is activated in order to achieve smooth running of the internal combustion engine; where 0 _i is the standard deviation value (o _i) for a respective injection nozzle (IN _i), representing a percentage factor, which provides by multiplication with the standard operating value of the respective injection nozzle (IN _i) the fuel mass flow generated by the respective injection nozzle (IN _i) ; where L _st · λ _{soll is} a desired air-fuel ratio that is identical for all combustion chambers (BK _i ) during fuel combustion in the combustion chamber (BK _i ) and L _{st is} the stoichiometric air-fuel ratio and λ _{soll is} a desired lambda value represents fuel combustion in the combustion chamber; where MSHFM is the air mass flow (MSHFM) valid for the respective test step (TSj); where L _{0, sum is} the total leakage flow (L _{0, sum} ); where λ _{real, k is the lambda value (λ real, k} ) valid for the respective test step (TSj). Procedure according to Claim 10 , characterized in that the system of equations is solved using a matrix calculation. Engine test device for detecting defective injection nozzles (IN _i ) for feeding fuel into the combustion chambers (BK _i ) of an internal combustion engine, in particular in a motor vehicle, the internal combustion engine having one or more cylinder banks (ZB), with a respective cylinder bank (ZB) having several cylinders each with a combustion chamber (BK _i ) formed therein and at least one injection nozzle (IN _i ) and the combustion chambers (BK _i ) of a respective cylinder bank (ZB) being supplied with a common air mass flow and from the combustion chambers (BK _i ) of a respective cylinder bank ( For example, a common exhaust gas flow is discharged, the engine test device being set up to carry out a method in which: a standard deviation value (o _i ) for each injection nozzle (IN _i ) and a total _{leakage flow (L 0, sum} ) are determined, the standard deviation value (o _i ) for a respective injection nozzle (IN _i ) a deviation of the by the respective injection nozzle üse (IN _i) describes fuel mass flow generated by a standard operating value of the respective injection nozzle (IN _i) and the total leakage flow (L _{0, sum)} describes the fuel mass flow of the respective cylinder bank (ZB) is caused by leaks of all injectors (IN _i); - For a respective cylinder bank (ZB) in idle mode of the internal combustion engine, a number of successive test steps (TSj) is carried out, which is greater than the number of cylinders of the respective cylinder bank (ZB), in a respective test step (TSj) for the individual injectors of the reciprocal value of the set for the respective injection nozzle (iN _i) calculated standard deviation value (o _i), except for one of the injection nozzles (iN _i) for which a pre certain classification parameters (a), of a quality measure for the internal combustion engine in reference is represented on the total _{leakage flow (L 0, sum} ), and wherein during the test steps (TSj) measurements of the lambda value of the exhaust gas flow discharged from the cylinder bank (ZB) and measurements of the air mass flow supplied to the cylinder bank (ZB) are carried out; - after the number of test steps (TSj) has been carried out, a distribution key (o _i ) for each injection nozzle (IN _i ) and a residual leakage value (ERL _{0, sum} ) are determined, with the distribution _{key (o i} ) for a respective injection nozzle (IN _i ) describes a proportion of the leakage proportion of the total leakage flow (L _{0, sum} ) caused by the respective injection nozzle (IN _i ) and the residual leakage value (ERL _{0, sum} ) describes a difference between the total _{leakage flow (L 0, sum} ) and the sum of the leakage proportions of the injection nozzles ( IN _i ) describes; - The determination of the distribution _{factors (v i} ) for the respective injection nozzles (IN _i ) and the residual leakage value (ERL _{0, sum} ) is carried out in such a way that a matrix calculation is solved with the aid of a computer, which includes an equation for a respective test step (TSj) that contains the Distribution key (o _i ) and the residual leakage value (ERL _{0, sum} ) depending on the reciprocal norm deviation values (o _i ) set in the respective test step (TSj) and the classification parameter (a), and a valid and off for the respective test step (TSj) _{describes the lambda value (λ real, k} ) derived from the measurements of the lambda value. Motor test facility according to Claim 12 , characterized in that the engine test device for performing a method according to one of the Claims 2 until 11 is set up. Motor vehicle with internal combustion engine and injection nozzles (IN _i ) for feeding fuel into the combustion chambers (BK _i ) of the internal combustion engine, characterized in that the motor vehicle has an engine test device according to Claim 12 or 13th includes.

Images