阅读说明：本技术 用于分析流体的方法和用于执行该方法的设备 (Method for analyzing a fluid and device for carrying out the method ) 是由 A·希尔切海因 G·波尔滕 J·霍夫施泰特尔 L·奥伯林 S·戈特利布 于 2021-03-22 设计创作，主要内容包括：本发明涉及一种用于分析流体(40)的方法,该流体从内燃机(10)的空间(23)、尤其是燃烧室流入流体引导件(42)中,其中内燃机(10)具有至少一个用于输送燃料(54)的机构(44)、特别是喷射阀,其中借助于探头(48)、尤其是拉姆达探头进行分析,流体引导件(42)中的流体(40)作用于探头上,其特征在于,在内燃机(10)发动期间进行分析,并且作用于探头(48)上的流体(40)不受受控输送的燃料(54)的影响。(The invention relates to a method for analyzing a fluid (40) flowing from a space (23), in particular a combustion chamber, of an internal combustion engine (10) into a fluid guide (42), wherein the internal combustion engine (10) has at least one means (44), in particular an injection valve, for delivering fuel (54), wherein the analysis is carried out by means of a probe (48), in particular a lambda probe, on which the fluid (40) in the fluid guide (42) acts, characterized in that the analysis is carried out during the start-up of the internal combustion engine (10) and in that the fluid (40) acting on the probe (48) is not influenced by the fuel (54) delivered in a controlled manner.)

1. Method for analyzing a fluid (40) flowing from a space (23), in particular a combustion chamber, of an internal combustion engine (10) into a fluid guide (42), wherein the internal combustion engine (10) has at least one means (44), in particular an injection valve, for delivering a fuel (54), wherein the analysis is carried out by means of a probe (48), in particular a lambda probe, on which the fluid (40) in the fluid guide (42) acts, characterized in that the analysis is carried out during the cranking of the internal combustion engine (10) and in that the fluid (40) acting on the probe (48) is not influenced by the fuel (54) delivered under control.

2. The method according to claim 1, characterized in that the analysis of the fluid (40) for the leakage of fuel (54) is ended before the fluid (40) flowing out of the space (23) and enriched with the controlled delivery of the fuel (54) reaches the probe (48).

3. Method according to claim 2, characterized in that the analysis of the fluid (40) for the amount of leakage of fuel (54) is ended after the controlled delivery of fuel (54) into the space (23) has started.

4. Method according to any one of the preceding claims, characterized in that the analysis is carried out before the delivery of fuel (54) by means of a mechanism (44) operating the internal combustion engine (10), in particular before the delivery into the space (23) or into the fluid guide (42).

5. Method according to one of the preceding claims, characterized in that the signal of the probe (48) is evaluated during the analysis and in this case the fraction of the fuel in the fluid (40), in particular the fraction of unburned or partially burned fuel, is determined.

6. Method according to one of the preceding claims, characterized in that for a plurality of current probe measurement values, preferably all current probe measurement values, of the probe (48), a difference (dL) is determined for each current probe measurement value (L) and a reference value (Lref), wherein a plurality of current probe measurement values are determined, preferably within a measurement window, and a plurality of the differences (dL) are added to a current integrator value (Lint), and after a defined number of differences, the integrator value (Lint) is compared with an error threshold in order to determine whether the mechanism (44) is faulty.

7. Method according to one of the preceding claims, characterized in that in the analysis, by evaluating the signal of the probe (48), no leakage or an allowable leakage is deduced and the operation of the internal combustion engine (10) is continued.

8. Method according to one of the preceding claims, characterized in that in the analysis, by evaluating the signal of the probe, an impermissibly high leakage amount is deduced.

9. Method according to any one of the preceding claims, characterized in that the signal profile (Lio, Lnio) of the probe (48) is compared with a reference value (Lref, w) and in the event that after a certain time (tschw) has elapsed after the start of the measurement window (tF1) an inadmissibly high leakage quantity and thus a malfunction of the mechanism (44) is concluded in the event that the deviation (Ldiff) exceeds a threshold value (lshw).

10. Method according to any of the preceding claims 1 to 10, characterized in that the signal profile (Lio, Lnio) of the probe (48) is compared with a reference value (Lref, w) and in the case of a deviation (Ldiff) exceeding a threshold value (lscw) before a certain time (tschw) has elapsed after the start of the measurement window (tF1), it is concluded that the mechanism (44) is normal.

11. The method according to any one of the preceding claims, characterized by determining an earliest start of signal evaluation (tF1) of the probe (48) and/or a latest end of signal evaluation (tF2) of the probe (48).

12. Method according to any one of the preceding claims, characterized in that the earliest start (tF1) of the signal evaluation of the probe (48) is related to the stationary rotational position of the drive shaft (19) of the internal combustion engine (10), in particular the earlier the next exhaust valve of a combustion chamber opens the earlier the start is.

13. Method according to any of the preceding claims, characterized in that the start of the signal evaluation (tF1) of the probe (48) starts at the latest with the start of the expected rise in the lambda value (1).

14. The method according to any of the preceding claims, characterized in that the probe (48) is heated to its working temperature (t48) before the start of the analysis (tF 1).

15. Method according to any one of the preceding claims, characterized in that the preset first downtime of the internal combustion engine (10) at least reaches a certain minimum duration following the start of the analysis (tF 1).

16. Method according to any one of the preceding claims, characterized in that a probe signal representative of the fluid (40) with unburned fuel is associated with a specific cylinder (13), which in turn is associated with a specific mechanism (44) associated with the cylinder (13).

17. Method according to any of the preceding claims, characterized in that a) in the case of being below a threshold value of the integrator value (Lint) and below a preset first stop time, or b) in the case of exceeding the threshold value (Lschw) after the time (tschw) has elapsed and below a preset second stop time of the internal combustion engine (10), a reference value (Lint, dyn; tschw, dyn).

18. An arrangement having a plurality of components and a control and/or regulating device (50) of an internal combustion engine (10), characterized in that the plurality of components and the control and/or regulating device are designed for carrying out the method according to any one of the preceding claims.

19. A computer program (52) comprising commands which, when run in the control and/or regulation device (50) of an apparatus according to claim 18, cause the apparatus to carry out the steps of the method according to any one of claims 1 to 14.

Technical Field

The present invention relates to a method for analyzing a fluid and to an apparatus for performing the method.

Background

The requirement for diagnosing a fuel system for an internal combustion engine, in particular a gasoline engine operating according to the otto principle or an externally ignited engine, leads to a more specific analytical method. For example, analytical methods are proposed so that restrictions in emissions can be observed. For this purpose, for example, an identification device can be used which identifies an unsealed injector and thus an uncontrolled quantity of fuel reaching a part of the internal combustion engine (injector leak identification). Such an identification device should identify an event in which the fuel is undesirably/uncontrollably moved out into the combustion chamber (leakage into the combustion chamber) via the valve seat of the injector. The cause of such errors may be, for example, impurities in the production of the injector, wherein particles from the machining process accumulate, for example, in the valve seat of the injector and lead to leaks. This may be the case in injectors having an outwardly opening nozzle (so-called a-nozzle). In principle, leaks can also occur even in magnetically actuated multi-port valve injectors (MLVs).

Depending on the magnitude of such a leakage, the leakage can lead to different false reactions of the internal combustion engine. Such internal combustion engines generally do not perform significantly if they are operated in normal driving operation of the motor vehicle. This is because even very large leaks overall represent only a relatively small proportion of the total amount of fuel injected into the cylinder individually. For example, the fraction is much less than 3%. The error reactions mentioned here include the following ranges: from emissions worsening due to engine effects imperceptible to the driver to combustion misfires caused by over-enrichment of the mixture in the combustion chamber during start-up or in phases following start-up. Furthermore, in direct injection engines, poor starting conditions are caused by high-pressure build-up delays, particularly in start-stop driving conditions.

In the prior art, typically two or more so-called signal paths, which are independent of one another, are regarded as a method of detecting a leakage of a high-pressure injection valve. The first signal path evaluates the time profile of the rail pressure by means of a built-in high-pressure sensor. However, this method reacts sensitively to any leakage in the high-pressure system. The method thereby also reacts to internal leaks in the high-pressure pump, through which the compressed fuel flows back into the low-pressure circuit. Therefore, there is no outward discharge of fuel, here for example into the combustion chamber of the cylinder, associated therewith. For this reason, the method is sufficiently accurate (sensitive) only when the leakage value of the injector is very high. In addition, this method is not suitable for associating a leak with each cylinder, so that cylinder-selective evaluation is not possible. As an additional and independent method, the so-called combustion stability of the individual cylinders can be evaluated or estimated. The german patent application published by the german patent and trademark office under the reference No. 102019208018 discloses an application for this process. As already mentioned, a combustion misfire occurs only in the first combustion after or after a start, since the fuel introduced into the combustion chamber leads at least initially to an excessively rich mixture as a result of the longer standstill phase of the internal combustion engine. By means of the evaluation of the combustion stability in the start/restart phase, an error analysis can be carried out, by means of which a cylinder-specific fault can be evaluated.

The above-described over-enrichment of the combustion chamber mixture leads to a signal characteristic of the exhaust gas probe or lambda probe that changes with respect to a normal system (error-free system) during the rotational speed increase and also during the restart phase. Thus, the evaluation of the signal curve of the exhaust probe or the lambda probe can be used as a further independent method.

From the prior art, for example from the publication DE 2216705 a1 of the german patent office, a method is additionally and principally known which makes it possible to analyze the composition of a fluid (exhaust gas) in a fluid guide (exhaust pipe) by means of a lambda probe and to change the exhaust gas composition by means of a control circuit.

Disclosure of Invention

According to a first aspect of the invention, a method is proposed for analyzing a fluid flowing from a space of an internal combustion engine, in particular a combustion chamber, into a fluid guide, for example into an exhaust pipe, an exhaust guide, or an exhaust guide from a so-called exhaust valve. The internal combustion engine also has at least one means, in particular a valve, for delivering fuel. The analysis is carried out by means of a probe, in particular in the form of a so-called lambda probe. During the method, fluid in the fluid guide acts on the probe. The fluid is analyzed during the start-up of the internal combustion engine and when the mechanism for delivering fuel is closed. In other words, the analysis is carried out during the start of the internal combustion engine and with the aid of a fluid which is not influenced by the controlled delivery of fuel after the start of the start, since the fluid is not influenced by the fuel after the start of the start. Controlled addition of fuel, possibly controlled addition, is performed prior to the start of cranking. Cranking with an internal combustion engine is expressed herein: the shaft of the internal combustion engine, in particular the drive shaft, for example the crankshaft, is driven in rotation by an auxiliary drive. During cranking, the internal combustion engine may not be started autonomously. The auxiliary drive can be, for example, an electric motor which directly drives a shaft of the internal combustion engine (crankshaft starter) or can be driven, for example, by means of an interposed transmission, for example, a transmission formed by a ring gear and a pinion, by means of a conventional starter, for example. As mentioned, the fluid must not be affected by the controlled delivery of fuel during the method proposed herein. This may mean, for example, that the respective means of the internal combustion engine (for example, injectors) are not actuated and thus do not deliver fuel, in particular into the space (combustion chamber) or the fluid guide.

The corresponding method has the advantages that: in the analysis of the fluid, it is preferably possible to exclude the fluid from being damaged by the fuel which has just entered (in a controlled manner) the fluid guide. Thus, when the method is performed, it can be determined accordingly that the fuel traces in the fluid are from leaking/non-sealing mechanisms. In this method, the start-up process is thereby evaluated in terms of the lambda signal curve or the probe signal curve during the start-up/restart phase. In this way, only the start-up process is evaluated in which only an injector leak has a significant effect on the signal profile of the probe or causes a corresponding effect on the criterion derived therefrom. Within the scope of the method sequence, the limitation to a specific suitable state of the system is target-oriented. For example, the following system states (system is an internal combustion engine or a subset thereof) may be considered, which prevail when the internal combustion engine is switched off or stopped: examples of such states are probe/lambda values, tank ventilation rates, crankcase ventilation states and other states.

Tank and crankcase ventilation affect the lambda signal, since the fuel fraction from the tank and crankcase ventilation is still contained in the intake manifold when the internal combustion engine is at rest. At the start of the internal combustion engine, not only air but also an unknown air/fuel mixture is supplied to the lambda probe in this case. This then yields a measurement value similar to an unsealed injector.

According to a further aspect of the invention, the analysis of the leakage of fuel to the fluid is ended after the controlled delivery of fuel into the space has started, but before the probe is reached by the fuel-rich fluid flowing out of the space. As is known to those skilled in the art of internal combustion engines, the delivery of fuel into a space or combustion chamber typically occurs, for example, shortly before the so-called ignition OT (top dead center, ZOT) and possibly shortly after the ZOT. For example, a 180 degree angle (degrees of crankshaft or degrees of drive shaft) will not be transmitted from the ZOT until the amount of fuel combusted in this space or the corresponding fluid reaches a position where a probe on the fluid guide can signal behind it. This enables a particularly long analysis period. According to another embodiment, it is provided that the evaluation is carried out before the fuel delivery by means of a device (for example an injection valve) which actuates the internal combustion engine, in particular before the fuel is delivered into the space or into the fluid guide. This results in: during the analysis, the controlled introduction of the fuel portion into the fluid can be reliably prevented.

During the analysis, the signal of the probe is evaluated and the fraction of fuel in the fluid, in particular the fraction of unburned or partially burned fuel, is determined. Unburned fuel may normally be regarded herein as fuel leaving the mechanism uncontrolled (leakage), and partially burned fuel may be attributed to incomplete scavenging. If, during the evaluation, no leakage or permissible leakage is inferred from the signals of the evaluation probe, it is proposed to continue the operation of the internal combustion engine, since no malfunction of the mechanism is indicated according to the standard mentioned immediately above. Accordingly, the exhaust gas composition is expected to meet the specifications. If, during the analysis, an impermissibly high leakage amount is inferred by evaluating the signals of the probe, it is possible, for example and advantageously, to infer that the mechanism or mechanisms or a mechanism (i.e., for example, one or more of the injectors or valves) of the internal combustion engine is/are defective. From the conclusion that a high leakage amount is not permissible as a result of the evaluation of the probe signals, the operation of the internal combustion engine can either be continued or alternatively not continued. For example, the continued operation of the internal combustion engine means that the internal combustion engine can still be operated reliably, and the internal combustion engine therefore outputs normal power in order to continue moving the motor vehicle and thus can be driven into the workshop, for example for inspection and possible maintenance purposes. If, for example, an impermissibly high leakage quantity is inferred by evaluating the probe signal during the analysis and it is determined that the internal combustion engine is not operating further, this can be associated, for example, with the fact that further damage is present at the internal combustion engine and the machine operation is regarded as impermissible. However, another reason or possible additional or alternative reasons for not operating the internal combustion engine after a fault has been determined may also be, for example, an inadmissible exhaust gas component which requires a forced repair before the internal combustion engine continues to operate. According to one aspect of the invention, during the evaluation, the signal curve of the probe is compared with a reference signal curve by evaluating the probe signal, and if this deviation is exceeded, an impermissibly high leakage rate and thus a malfunction of the mechanism is inferred. The signal profile of the probe can be an arrangement of individual signal values of the probe, which signal values describe one or more, in particular continuous, signal profiles of the probe in relation to one another. For example, the signals each determine the degrees of the crankshaft or the degrees of the drive shaft, such that a value, for example a lambda value, is provided for each degree of the crankshaft or degree of the drive shaft. The reference signal curve corresponds in principle to the signal curve of the probe, but this is a special signal curve. The reference signal profile is, for example, a model signal profile determined beforehand at an internal combustion engine of the same type, which has a fully functional engine. In this context, for example, the permissible deviations from the reference signal profile are predetermined. The reason for such permissible deviations can be, for example, that measurement errors or evaluation errors are typically not avoided and that manufacturing errors or manufacturing deviations and manufacturing tolerances are not only not excluded, but are also in principle also to be expected in the components involved in the method. Due to this tolerance, a deviation can be obtained, which can therefore be used to infer a reliable operating state of the internal combustion engine. If the deviation set in this way or its threshold value is exceeded after a specific operating time of the method or analysis method, the prediction can likewise conclude that an impermissibly high leakage and thus a mechanical failure has occurred. For example, mention may be made here of: if, for example, a deviation of 0.1 is exceeded approximately 300 milliseconds after the start of the evaluation method by the sensor or the lambda value, an impermissibly high leakage quantity and thus a malfunction of the mechanism can be concluded. According to a further aspect of the invention: in order to determine that a high leakage amount may not be permissible, the earliest start of the signal evaluation of the probe and/or the latest end of the signal evaluation of the probe are determined. The earliest start of the signal evaluation of the probe may therefore not be before the operating temperature of the probe is reached, for example, since otherwise incorrect results and therefore incorrect conclusions can be drawn therefrom. For example, if the probe mentioned here, in particular the lambda probe, does not have a normal operating temperature, there is typically an erroneous process here. For this reason, for example, the lambda probe may be required to be heated to its operating temperature first by means of a probe heating device. The reaching of the operating temperature can then, for example, indicate the earliest start of a meaningful/corresponding signal evaluation of the probe. In this case, the sensor is preferably arranged in the form of a sensor, which is arranged in the flow path of the fluid, and which is connected to the sensor in a manner known per se. This finally incorrectly determined fuel fraction makes the method proposed here meaningless. As an earlier start of the signal evaluation of the probe, the drive shaft of the internal combustion engine can be correlated, for example, with a stationary rotational position. In particular, this is relevant, for example, for so-called exhaust valves of combustion chambers. In the context of this method, the earlier the next exhaust valve of the combustion chamber opens after the internal combustion engine has been shut down, the earlier the start can be observed. Thus, if for example the opening point in time of the next exhaust valve is imminent, the furnace from the combustion chamber is discharged as early as possible into the fluid guide and guided to the probe by means of gas switching triggered by an event movement of the drive shaft of the internal combustion engine caused by the piston movement. If, in contrast, it is proposed in a single-cylinder horizontal engine that the exhaust valve only starts to open after a half-turn of the drive shaft, for example, then the start of the signal evaluation of the probe should accordingly be the latest, calculated from the start of the rotational movement of the drive shaft. Such a procedure enables a significant so-called time window or the beginning thereof, so that the signal evaluation is only carried out at technically meaningful points in time, since a meaningful evaluation is possible at all. In other words, if the signal evaluation of the probe technically foresees that the next exhaust valve has already started at a point in time at which the exhaust valve is opened after half a crankshaft revolution, the start signal evaluation is too early for this at least half a crankshaft revolution. But the method can still be started early.

According to another aspect it is proposed: the start of signal evaluation of the probe (preferably at the latest) begins with the start of the expected rise in lambda value. The term "expected rise of the lambda value" is applied here, for example, to the already proposed reference signal curve. It is therefore determined that, in the case of mechanisms which are not sealed, i.e. leak, to be expected or which have already been determined, the corresponding value of the signal of the probe (starting from the same boundary conditions, i.e. for example the start of the evaluation) is measurably reached subsequently, for example a probe value of lambda 1.1 is reached. This may be, for example, 160 milliseconds.

According to another aspect of the proposed method it is proposed that: the probe is heated to its operating temperature before the analysis begins, since otherwise a faulty result is determined and a faulty probe value is summed. Furthermore, it can be advantageous to comply with at least a predefined idle time (stop phase ) (first idle time) of the internal combustion engine by means of the start of the evaluation. The reason for this is that the leakage amount is ultimately the result of, for example, a time-dependent process. Thus, the following would be the case: a faulty, i.e. leaking, component (injector, valve) flows through a leakage quantity that is just still permissible within, for example, 6 hours, while a determination of the leakage quantity discharged during, for example, a 12 hour shut-down time (after the first shut-down time) indicates a damage to the respective component.

The method proposed here is also used to determine which of the individual cylinders a leakage of the mechanism can be associated with or in which cylinder a leakage of the mechanism is associated. The flow of fluid in the fluid guide consists of material (air and combustion residues and possibly fresh leaking propellant) which leaves the individual cylinders one after the other at start-up. The stop position of the internal combustion engine is known and stored by means of the start after the last operating phase, or is determined by means of the start, in which position the respective piston is located. The proper discharge of the material from the cylinder is determined from the dynamics of the propeller, the point in time at which the respective discharge valve is opened, the distance of the pipe connection of the fluid guide between the discharge valve and the position of the probe from the discharge valve, and the known or determined flow rate. If the signal of the probe is associated with a specific material in the cylinder, it can be inferred from the probe information, which identifies the faulty mechanism, that the respective mechanism is faulty in which cylinder. Accordingly, a method is proposed in which a probe signal representing a fluid with unburned fuel is associated with a particular cylinder and is then associated with a particular mechanism associated with that cylinder.

Furthermore, an arrangement is proposed which has a plurality of components and a control and/or regulating device of an internal combustion engine, which are designed to carry out the method according to the features and aspects described above. Furthermore, a computer program is proposed, which comprises commands which, when executed in a device control and/or regulation apparatus as described above, cause a device to carry out the steps of the method set forth.

Drawings

The invention is explained in more detail on the basis of the drawings briefly described below. The figures show:

figure 1 shows a schematic view of an internal combustion engine,

figure 2 outputs a schematic diagram of the reason for the proposed method,

fig. 3 shows a first diagram which describes curves of different characteristics at the start of a start when the internal combustion engine is running. In this case, different curves (lambda values) of the probe signals are shown for two different system states.

Fig. 4 shows a further diagram with a variable curve during the starting process, wherein the diagram relates to an internal combustion engine, the system state of which is error-free,

fig. 5 shows a diagram corresponding in principle to the diagram in fig. 3, wherein the system status is faulty,

fig. 6 schematically shows a view of the flow of the method.

Detailed Description

Fig. 1 shows an internal combustion engine 10 in a strongly schematic manner. The internal combustion engine 10 here shows a cylinder 13, wherein the internal combustion engine 10 can have not only one cylinder 13, but also, for example, two, three, four, five, six or more cylinders 13 in a known manner. A piston 15 is arranged in the cylinder, which piston is coupled to a drive shaft 19 via a rotary joint and a connecting rod 17, which are not illustrated in detail here. The drive shaft 19 shown here schematically can be designed as a crankshaft, for example. The cylinder 13 is closed above the piston 15 by a cylinder head 21. The cylinder head 21 and the piston 15 define a space 23 therebetween, which is typically referred to as a combustion chamber. The space 23 serves in a known manner for receiving the mixture and transmits the resulting pressure increase in the space 23 as a corresponding force via the piston 15 and the connecting rod 17 to the drive shaft 19, in order for it to drive, for example, a motor vehicle in a likewise known manner or, in the case of a stationary machine of the internal combustion engine 10, a power plant (for example, a motor generator).

In the context of the combustion process just described, the need for fresh air to flow into the combustion chamber is symbolized here by the arrow 25. The fresh air 25 is here conducted via an inlet line 27 to the space 23 (intake stroke). In order for fresh air 25 to flow into space 23, fresh air 25 must pass through intake valve 29. The inlet valve 29 has a so-called valve seat which is introduced into the cylinder head 21 and has a valve closure 31 interacting therewith. During the combustion process in the space 23, the valve closure 31 and thus the intake valve 29 is closed (working stroke). In order to leave the space 23 with the burnt gas mixture (exhaust) in a known exhaust stroke following the working stroke, the exhaust valve 33 is opened in a known manner in time. The exhaust valve 33 also has a valve seat and a valve closure 37 interacting therewith. After passing through exhaust valve 33, the exhaust gas, referred to herein as fluid 40, flows into fluid guide 42. Here, the fluid 40 is likewise indicated by an arrow symbol.

The fluid guide 42 mentioned here may also be referred to as an exhaust pipe, an exhaust guide. Also shown in this embodiment is a mechanism 44, which may also be referred to as a valve or injection valve or high pressure injection valve or injector. The means 44 are provided for injecting fuel into the space 23, so that the fuel is mixed with the fresh air present there and can be combusted into exhaust gases after the ignition process has started. To initiate the ignition process, an ignition device 46 (e.g., a spark plug) may be positioned at the preferred cylinder head 21 such that the ignition device may ignite the mixture of fresh air 25 and fuel. The probe 48 is arranged in the mentioned fluid guide 42, so that the probe can analyze the composition of the fluid 40 by means of other technical means, or react normally to an inflow, an outflow or a cross-flow of the fluid 40. The probe 48 mentioned here can be a so-called lambda probe. The probe 48 is connected to a control and/or regulating device 50, so that the signals generated by the probe 48 can be evaluated by the control and/or regulating device 50. For example, the measured values of the other sensors 55, 56, 57, 58 also act on the control and/or regulating device 50. A computer program 52 is shown in a symbolic view, which comprises commands that, when executed in the control and/or regulating device 50 of the apparatus, perform the method described below or the method steps thereof.

The reason for the proposed method is schematically shown in fig. 2. As described at the outset, if the mechanism 44 fails such that unintentional/uncontrolled fuel 54 escapes into the space 23 via the valve seat of the mechanism 44 (leakage into the combustion chamber), this causes an increasingly large portion of the content of the space 23 to be occupied by the fuel 54. The symbols here show: this leakage amount of the fuel 54 accumulates as droplets (liquid) on the surface of the piston 15. It is possible here to: the amount of fuel 54 escaping from the means 44 does not accumulate at all as droplets, but is for example dissolved in the air or fresh air located in the space 23. However, a mixture of the two (liquid and gas) may be formed in the space 23.

The method proposed here allows the detection of this leakage and thus the detection of a malfunction of the mechanism (valve).

Fig. 3 shows a diagram depicting various measured and/or calculated variables. The X-axis shows time in seconds. Here, a time equal to zero seconds is initially not associated with an event. Via this time course, various curves are shown, which can be associated overall with two different states of the internal combustion engine 10. For example, two curves n19nio, n19io are thus shown. Curve n19io describes the speed curve of internal combustion engine 10, the state of the internal combustion engine and more particularly here mechanism 44 being normal, and thus free of leaks. The curve n19nio shows the speed curve of the drive shaft 19 of the internal combustion engine 10, wherein the mechanism 44 is not normal, i.e. in this case leaks. As can be recognized here well are: the speed curve runs very differently, in particular in the region of the idle speed, starting from approximately 86.8 seconds at time t. Although the curve n19io stretches relatively uniformly, the curve n19nio shows some abrupt speed changes or more abrupt amplitudes up or down. In addition, two curves Lio and Lnio are shown in the figure. For the times shown here, these two values each describe a curve of the value of lambda (L) determined by the probe 48 in the fluid guide 42. As can be recognized here well: both curves deviate from the beginning of the measurement window from tF1 except for a small region. The other curve (i.e., curve ni) indicates at which point in time the next injection can be performed in the other cylinder 13 of the internal combustion engine 10. Each step shown here represents another occurrence of injection. As can be recognized in the two curves Lnio and Lio, the amplitudes are very different.

The signal curve Lio shows that the lambda signal rises very early according to a thin line. This means that the internal combustion engine 10 must have a high air content in the intake air/fuel mixture. The signal curve Lnio will only show a clearly later rise on decay (starting from reference tF 1). This means that the air taken in must be contaminated (after a long shut-down) by the fuel fraction from the leakage.

The signal curves Lio, Lnio of the probe 48 are compared with reference values Lref, w and, if the deviation Ldiff exceeds a threshold value lscw after a certain time tschw has elapsed after the start of the measurement window tF1, an impermissibly high leakage quantity and thus a malfunction of the means 44 is inferred. In this example, the signal curve Lio exceeds the threshold value Lschw before a certain time tschw has elapsed after the start of the measurement window tF1, and the signal curve Lnio exceeds the threshold value Lschw after a certain time tschw has elapsed after the start of the measurement window tF 1. This means, on the other hand, that the signal profiles Lio, Lnio of the probes 48 are compared with the reference values Lref, w and that the mechanism 44 is concluded to be normal if the deviation Ldiff exceeds the threshold value lscw before a certain time tschw has elapsed after the start of the measurement window tF 1.

Fig. 4 shows a further illustration of the various curves recorded during the operation of the internal combustion engine 10. Since the underlying internal combustion engine 10 is operating perfectly, the speed curve n19 shown here is the normal speed curve of the internal combustion engine 10. In addition, a straight horizontal line is shown as a reference line for the reference value lambda L ═ 1.0. Furthermore, a plot of the measured lambda L is shown. The reference value Lref is shown here as a further curved contour (straight line, horizontal run). This value Lref is preferably stored as a reference value at the start of engine start. At the beginning of the measurement process corresponding to the current lambda value, the result is clearly open, i.e. at this point in time it is not yet known: whether it is Lio or Lnio. Which here begins at time tF 1. Within the scope of the defined measuring steps, the current lambda value L is measured, for example, in each case as a function of the crankshaft or shaft revolutions (degrees of crankshaft angle) that have already passed, and finally combined in succession to a curve/polygon/step line L. Within the measurement window, starting from time tF1, the difference between the current probe measurement value L and the reference value Lref mentioned is determined for preferably each measurement value of lambda L. By means of a further measurement step, a new difference value is formed from the current probe measurement value and the reference value. In the first measurement, the first difference value determined corresponds to the first sum. Each further difference value is added to the respective previously found sum value (integral of the difference between the current probe measurement value and the reference value). At the end of the measurement window, here at time tF2, the total value of the integration method (integrator value) is then checked whether the limit, here referred to as the error threshold, has been reached or exceeded. In this example, the integrator value Lint reaches a total value of 4.77 at the time tF2 (end of the measurement window) with the end of the integration method. If the integrated value Lint does not exceed a value of 3.0, for example, in this case, the result would be: in the context of this method, a fault is detected or inferred in the means 44. As can also be recognized in this case: the method ends after the injection activity of the mechanism 44 into the space 23 has started.

For example, the measured values are determined with the aid of each synchronization grid (i.e., in synchronization with the ignition). In the case of an internal combustion engine 10 having four cylinders, at an interval of 180 ° KW, and in the case of an internal combustion engine having six cylinders, at an interval of 120 ° KW.

In the method described here by means of integrating the difference between the current probe measurement value and the reference value, this can optionally also be done via averaged measurement values. The marker (integrator value) calculated from the signal curve of the probe 48 is then compared with the measured pressure drop of the so-called rail in the so-called standstill phase of the internal combustion engine 10. I.e. the method may be checked for plausibility by an additional (not absolutely necessary) review check. Thus, if the integral value indicates a malfunction of the mechanism 44 because it does not exceed the threshold value, this can be confirmed by a corresponding curve of the pressure drop of the rail, if necessary. Accordingly, the rail pressure drop due to failure and leakage of the mechanism 44 is stronger than under normal conditions. If one or both of the indications mentioned here (integrated lambda value, pressure drop in the rail) deduces an impermissible leak, further diagnostic measures can be taken to confirm the indicated error situation by other methods if necessary. For example, in conjunction with fig. 4 and 3, two times tF1, tF2 are illustrated, at which the length of the measurement window is determined. In order to determine the measurement window or its length in this way, it is important to take into account all relevant influencing factors, which can be derived from the different stop and start states. Thus, for example, the measurement window can be tracked in a so-called dynamic manner. I.e. depending on the so-called engine stop position, the start and end and the length of the measurement window can thus be determined differently, for example. Other influencing variables may be the current intake cam profile (in particular in the case of an adjustable intake camshaft), the state of the drive train (here, for example, in relation to the clutch state), the selected shift situation (for example, idling or engaged particular gear), or also the starting strategy. The so-called start-up strategy is for example as follows: the selection of a driving program, as in the present vehicle, can be represented, for example, by a particularly economical mode (Eco) or a more sporty mode, or, for example, by a particularly comfortable mode. Furthermore, for the selection of the measurement window, the probe properties, in particular the so-called dynamic behavior of the broadband probe, should also be taken into account.

For one or more of the integration methods described, the signal behavior can also be evaluated based on the absolute value and gradient of the lambda signal curve.

Fig. 5 shows another example for evaluating the start behavior of internal combustion engine 10. The curves shown here allow the identification: the device 44 causes a leak. The flow shown in fig. 5 corresponds exactly to the flow shown in fig. 4. At the beginning of the measurement window tF1, the current lambda value L is determined at a preferably regular interval (for example once per stroke of the internal combustion engine 10). For this lambda value L, the difference dL from the reference lambda value Lref is determined and the integral value of the lambda difference value, which is represented by the curve Lint, is determined by means of the integration method already described. With the end of the measurement window at time tF2, a diagnostic value, i.e. here an integrator value of 2.0, can be determined. This value is less than the threshold value, which is not numerically named here (but can be 4.0, for example), which separates a fully functional internal combustion engine 10 with a fully functional mechanism 44 from an incompletely functional internal combustion engine 10 with a regularly leaky mechanism 44. The internal combustion engine 10 diagnosed here is therefore not normal.

The proposed method is schematically illustrated in fig. 6. The method then begins with step S1, for example, in a further step to ensure that the probe 48 is ready, step S2. For this purpose, it may be necessary, for example, to heat the probe 48 to the operating temperature by means of a probe heating device in a step not shown here beforehand. Starting in the following step S3: the internal combustion engine 10 is started or increased in speed by the auxiliary drive machine already mentioned, so that the drive shaft 19 and other components of the propeller, such as the connecting rod and the piston 15, and finally also valves, such as the inlet valve 29 and the outlet valve 33, start moving. Then, it is achieved by the above-mentioned starting movement of the vent valve 33 that the fluid 40 can be discharged from the space 23 into the fluid guide 42 in step S4, and thus the ready probe 48 can recognize the properties of the fluid 40 and provide a corresponding signal to the control and/or regulating device 50, S5. In another step S6, the probe characteristics or physical state thereof are evaluated to determine the current lambda value L, step S6. For example, in the context of the evaluation, the proposed integration method can be considered, for example. Alternatively, it is possible (see also fig. 3): the absolute value of the lambda value, which exceeds or does not exceed the lambda value L, here for example 1.1, is evaluated. The following can then be used here, for example, as a rule: which time point between the limits tF1 and tF2 of the measurement window exceeds the threshold. If this is essentially at an intermediate or specific point in time (for example) in the region of the measurement window, it can be concluded that the means 44 is in a normal, i.e. non-leaking state. If the mentioned threshold value, here lambda 1.1, is reached, for example, after approximately 90% of the total time of the measurement window has elapsed or at a specific point in time in the region of the measurement window, i.e. significantly later than with a normal mechanism 44, it is possible to conclude that the mechanism 44 has failed due to a leak.

Since the method proposed here should operate in such a way that the fluid 40 should not be influenced by the controlled delivery of the fuel 54, it can be proposed according to the method: analyzing the fluid 40 for a leak of the fuel 54 ends, for example, after the controlled delivery of the fuel 54 into the space 23 begins, but at the latest before the fluid 40 flowing out of the space 23 reaches the probe 48. In particular, it is proposed that: the analysis is carried out before the delivery of the fuel 54, the instrument into the space 23 or the fluid guide 42 by the means 44 for operating the internal combustion engine 10. In the analysis, the signal of the probe 48 is evaluated and the proportion of in particular unburned or partially burned fuel in the fluid is determined. If, in the analysis, no leakage or an allowable leakage is inferred from the signals of the evaluation probe 48, the operation of the internal combustion engine 10 is continued according to a variant of the method. In the context of the method, an impermissibly high leakage rate can be inferred from the evaluation of the signal of the probe 47 during the analysis. In this case, the possible consequence is that the internal combustion engine 10 continues to operate. An alternative variant may be that the internal combustion engine 10 does not continue to operate. According to the described integrator method or the described lambda threshold method, it is proposed that: the signal curve of the probe is compared with a reference signal curve and, if a deviation threshold is exceeded, an impermissibly high leakage quantity and thus a malfunction of the means 44 is inferred. According to a further step of the method: the earliest beginning tF1 of the signal evaluation of the probe 48 and/or the latest ending tF2 of the signal evaluation of the probe 48 are determined. The end of the signal evaluation is referred to herein as the result of the correlation, in which the end of the signal evaluation is understood to be for the purpose of an impermissibly high leakage rate. The signal evaluation of the probe 48 continues in the usual manner, which is however usually caused by purging air.

The earliest start of signal evaluation tF1 of the feeler 48 and thus the start of the measuring window can be correlated to the stationary rotational position of the drive shaft 19 of the internal combustion engine 10. This applies in particular in the following respect: i.e. the earlier the next exhaust valve 33 of the crank drive of the internal combustion engine 10 opens the combustion chamber or space 23, the earlier the start tF1 is determined. Thus, if the next opening point in time of the outlet valve 33 is imminent, the start of the measurement window tF1 may be determined from the earlier point in time. However, as is also evident when considering the process in internal combustion engine 10, this start of the measurement window can also be set later. Typically, the exhaust valve 33 is opened after the so-called ignition OT, for example around a 120-degree crankshaft or driveshaft position. However, this generally occurs first with the piston 15 moving downward, so that in the somewhat cold state of the internal combustion engine 10, the fluid 40 initially still does not leave the space 23. Instead, at this point in time, the fluid 40 is first still drawn from the fluid guide (exhaust branch) back into the space 23. Only with the passage through the so-called bottom dead center between the working stroke and the exhaust stroke is the following possible: i.e., the fluid 50 moves from the space 23 into the fluid guide 42 with the amount of leakage of the fuel 54. If the gas quantity reaches the probe 48, the latest start tF1 of the measurement window should be set (or shortly before) in the case of using the integrator method. Thus, for example, it is proposed: in the case of the integrator method, the start of the signal evaluation of the probe 48 and thus the start of the measurement window tF1 begins at the latest with the start of the expected rise in the lambda value L. In the mentioned method with threshold determination, it is finally important for the determination of whether the means 44 are normal that a specific threshold value is not reached at a preset point in time. That is, the measurement window may also start after a preset point in time and not exceed the threshold, immediately determining that the mechanism 44 is normal. The probe 48 must be heated, i.e., to its operating temperature, before a trusted analysis can begin. Thus, the operating temperature of the probe 48 can be reached in the method flow in a timely manner before the analysis begins. Furthermore, with the start of the analysis tF1, the preset stop time (stop phase, first stop time) of the internal combustion engine 10 reaches at least a certain minimum duration. In particular, it is proposed within the scope of the method to start the evaluation during the first exhaust stroke (gas exchange) of the cylinder 13.

Determining the signal curve L (fig. 3) in conjunction with a short idle time (second idle time) of the internal combustion engine 10 cannot be attributed to a mechanical failure, since the time for a large or indirectly measurable escape of fuel 54 is too short due to a leak. Short downtimes are understood to mean, for example, time periods in the second range (for example 5s or 30s or 100s) as occur, for example, in start-stop systems. That is, the first downtime is much greater than the second downtime. The first downtime may for example comprise an interval between two trips or an interval of several hours, while the second downtime occurs during one trip. The signal curve L shows: the lambda signal rises very early according to the thick line. This means that: the internal combustion engine 10 must have a low air fraction or a high fraction of fuel 54 in the exhaust air-fuel mixture. It is concluded that the relatively high fraction of fuel 54 does not come from mechanism 44 due to the short downtime (second downtime), but from another source. This further source may be, for example, the fuel fraction in a so-called oil sump of the internal combustion engine 10. In particular in older vehicles and/or vehicles which have not been subjected to an oil change for a long time, the fuel fraction (oil-fuel mixture) in the oil sump can be determined. A part of this fuel fraction can flow from the crankshaft housing of the internal combustion engine 10, for example past piston rings that have been worn more heavily, into the space 23.

It is proposed for this case to use dynamic reference values for different programs.

Although, for example, a threshold value of 4.0 for the integrator value has been proposed so far, which separates a fully functional internal combustion engine 10 with a fully functional mechanism 44 from a non-fully functional internal combustion engine 10 with a defective leakage mechanism 44, a dynamically adapted reference value should be suitable for this case. By integrating the value Ldiff as already described, for this case (short downtime) an integrator value is determined, which is suitable as a dynamically adapted reference value (here dynamically adapted integrator value Lint, dyn ═ 2.3).

Alternatively, for this case, the time tschw may also be changed as a dynamically adapted reference value. For example, a time tschw, dyn is set at which the signal curve L must reach the threshold value lscw at the latest in order to identify the internal combustion engine 10 as fully functional. The time tschw, dyn can be formed here as the sum of the measured time tgem at which the signal curve L actually reaches the threshold value Lschw and the additional time dt.

Accordingly, method steps are provided in which a) the reference value, i.e. the integrator value Lint, dyn or the time tschw, dyn, is dynamically adapted until the signal curve L reaches the threshold value lshw if the threshold value of the integrator value Lint is undershot and undershoots a preset first stop time, or b) if the threshold value lshw is exceeded after the time tschw and undershoots a preset second stop time of the internal combustion engine 10.