Computer-implemented method for identifying purges in fuel cell

Abstract

The invention relates to a computer-implemented method for detecting a purge (22) induced by means of an anode purge valve (20) in a fuel cell (10), comprising the following steps: detecting an actuation phase (26) of the anode purge valve (20) for inducing the purge (22); monitoring a first measured variable (28), which is dependent on the pressure (30) that varies in the event of a purge (22); monitoring a second measured variable (32), which influences the pressure (30) independently of the purge (22); determining whether the second measured variable (32) has changed before or during a monitoring time period (34) if the first measured variable (28) has changed during and / or after the control phase (26) within the monitoring time period (34); and detecting the occurrence of a purge (22) under a criterion (K3), that is, the second measured variable (32) does not change before or within the monitoring time period (34).

Description

Technical Field

[0001] The technology disclosed in the present invention relates to a computer-implemented method for identifying purge in a fuel cell. In addition, the present invention also relates to a data processing device, a computer program and a computer-readable storage medium. Background Art

[0002] A fuel cell comprises an anode, a membrane and a cathode. The anode and the cathode may each have a flow channel. Usually, a fuel such as hydrogen flows through the flow channel of the anode, while an oxidant such as air or oxygen flows through the flow channel of the cathode. These flow channels distribute the fuel and the oxidant over the membrane. The chemical reaction in the fuel cell converts the reaction energy of the fuel and the oxidant into electrical energy and thermal energy. Fuel exhaust gas is produced here. In particular, water is produced in fuel cells using hydrogen as fuel.

[0003] Fuel exhaust gases or water and possibly other byproducts can diffuse into the flow channels of the anode and there affect the efficiency and service life of the fuel cell. Therefore, a purge, especially a hydrogen purge, is generally performed in the anode. Summary of the invention

[0004] The object of the technology disclosed in the present invention is to reduce or eliminate at least one disadvantage of the prior known solutions or to propose an alternative solution. In particular, a preferred object is to improve the efficiency and service life of the fuel cell in dynamic operation, that is, when there are driving dynamic parameters, in particular acceleration and / or braking. Other preferred objects may result from the advantageous effects of the technology disclosed in the present invention.

[0005] These objects are achieved by the subject matter of the independent claims. The dependent claims indicate preferred embodiments.

[0006] According to one aspect, the technology disclosed in the present invention provides a computer-implemented method for identifying a purge induced by an anode purge valve in a fuel cell, the method comprising the steps of:

[0007] - Check the anode purge valve for the control phase to initiate the purge;

[0008] - monitoring a first measured variable which is correlated with the pressure change due to the purge, if the purge was successfully initiated, i.e. the purge actually took place;

[0009] - monitoring of a second measured variable which influences the pressure in addition to the purge, that is to say independently of the purge actually taking place;

[0010] - if the first measured variable changes during the control phase and / or during the monitoring period after the control phase, determining whether the second measured variable has changed significantly before or during the monitoring period, i.e. by a value greater than or equal to a predetermined value; and

[0011] - Successful initiation of purging or actual occurrence of purging is detected under the criterion that the second measured variable has not changed significantly in the above-mentioned sense before or during the monitoring period.

[0012] In other words, the disclosed technology is concerned with detecting whether an initiated purging, preferably anode purging, is actually occurring. This detection is to be made possible in particular during dynamic driving operation, where it is recognized that the pressure monitored by the first measured variable may also change due to the dynamic driving operation, so that purging may not be detected.

[0013] By means of the method, it is possible to determine during dynamic driving operation, in which the second measured variable may change, whether the anode purge valve is correctly opened or closed during actuation phases.

[0014] For this purpose, the control device of the fuel cell system can determine during the control phase of the anode purge valve whether the anode purge valve is actually switched and / or how effectively water is discharged from the system, for example. Otherwise, the detection of the purge must be realized, for example, by additional sensors, for example by plausibility testing of a hydrogen sensor in the exhaust gas duct.

[0015] The idea of ​​a preferred embodiment is to continuously monitor a first measured variable which is related to the pressure change caused by the purging, in particular during dynamic system operation. For example, the first measured variable is the anode pressure. It is therefore preferred that the monitoring of the first measured variable and / or the monitoring of the second measured variable is continuously performed. For monitoring, one or more pressure sensors present in the anode can be used. As a result, no additional sensor devices are required.

[0016] If the purge is not actually or effectively performed for an extended period of time (depending on the exhaust or water production of the fuel cell or fuel cell stack, this could be anywhere from 30 seconds to 10 minutes), this could result in significant degradation or damage to the fuel cell or fuel cell stack, or even complete system failure, which would require replacement of the fuel cell or the entire fuel cell stack.

[0017] Preferably, the method further comprises the following steps:

[0018] - The actual occurrence of a purge is identified under an additional criterion which can be designed in different embodiments according to the following description.

[0019] Preferably, the method further comprises the step of: if a change occurs in the first measured variable during the monitoring period, determining the magnitude of the change. Preferably, then the additional criterion is that the magnitude is greater than or equal to a predetermined magnitude.

[0020] Preferably, the method further comprises the following steps:

[0021] If the first measured variable changes within the monitoring period, a curve characteristic of the change is determined.

[0022] Preferably, then, the additional criterion is that the curve feature corresponds to a predetermined feature. Such a predetermined feature can have a tolerance range or tolerance band, so that there is no need for exact conformity with a single curve course. The predetermined feature can, for example, specify an upper and a lower course, between which the tolerance band lies, within which the curve feature must lie in order to detect conformity.

[0023] Preferably, the predetermined characteristic comprises a slope of change and / or a delimitation of change in a previous time course. Thus, a transition or relative change to a previous time course is monitored according to an additional criterion, such as a relative change exceeding a predetermined percentage.

[0024] Preferably, the additional criterion is that the magnitude is greater than or equal to a predetermined magnitude and that the curve characteristic matches the predetermined characteristic, ie a combined monitoring is performed.

[0025] Preferably, the first measured variable is the pressure in the flow channel of the fuel cell or (for example in a combined measurement) comprises the pressure.

[0026] Preferably, the flow channel is arranged at the anode of the fuel cell.

[0027] Preferably, the first measured variable is the pressure difference between the inlet pressure and the outlet pressure in the flow channel or (eg in a combined measurement) comprises this pressure difference. For example, the first measured variable may be the pressure difference between the anode inlet pressure and the anode outlet pressure.

[0028] The method can also be implemented with just one pressure sensor, and the pressure sensor is not necessarily installed in the anode. One or more pressure sensors can also be located in the flow channel outside the anode, up to the ambient air. For example, the method can be implemented with a medium-high pressure sensor. However, the preferred implementation with one or more anode sensors has the smallest time difference with respect to the initiation of the purge and is therefore the simplest to use.

[0029] Preferably, the second measured variable is the pressure at the anode of the fuel cell or (for example in a combined measurement) includes this pressure. As the pressure, a set pressure can also be detected, which the regulator can set for generating electrical power according to the required driving power. The reason for this is based on the realization that the course of the pressure in the anode can be assumed to reliably follow the set pressure. The second measured variable can also be the current in the fuel cell or (for example in a combined measurement) include this current.

[0030] Preferably, the second measured variable is the consumption of fuel or hydrogen in the fuel cell or (in the case of a combined measurement) comprises this consumption.

[0031] The "combined measurement" mentioned here means that the second measured variable can also be composed of a combination of multiple measured variables. A measured variable can also be pre-processed by short-time averaging.

[0032] Preferably, the determination of whether the second measured variable has changed by a value greater than or equal to a predetermined value before or during a monitoring period begins 10-1000 milliseconds, preferably 50-250 milliseconds, more preferably 90-110 milliseconds before the monitoring period. This can be achieved, for example, by periodic monitoring that is independent of the control phase.

[0033] Preferably, the monitoring period begins with the start of the manipulation phase.

[0034] Preferably, the monitoring period ends with the end of the control phase or within 1000 milliseconds after the control phase, preferably within 500 milliseconds, more preferably within 250 milliseconds.

[0035] If, during the purging check, it is detected that purging has not taken place, an error signal can be generated as a reaction to this, or the operation of the motor vehicle can be restricted.

[0036] Preferably, the method further comprises the following steps:

[0037] - The detection of the purge is averaged over several control phases. This prevents a single failed or non-performed purge from leading to false signals or limitations.

[0038] In a preferred embodiment of the method, three evaluation criteria may be examined, namely:

[0039] K1) whether it is recognized that the value of the change in the first measured variable during the monitoring period is greater than or equal to a predetermined value, for example, whether a sufficient pressure drop is recognized (absolute height change of a pressure event); K2) whether the curve characteristics of the change in the first measured variable during the monitoring period are consistent with predetermined characteristics, for example, whether the pressure drop is consistent with a normal purge (high dynamics and clear delimitation of pressure events); and K3) the second measured variable has changed in value by less than a predetermined threshold value before or during the monitoring period, that is, there is no external influence on the pressure.

[0040] Rapidly changing external influences can be, for example, a pressure change at the anode (very common), or a step change in the hydrogen consumption, which corresponds to a step change in the current (extremely rare).

[0041] All criteria K1), K2) and K3) can preferably be checked over the entire control phase or monitoring phase, wherein changes in the first measured variable may also occur with a slight delay due to the influence of liquid water. The second measured variable, i.e. the external influence, can even be checked before the start of the control phase, for example within the last 100 milliseconds, so that an event immediately preceding this does not trigger a "false positive" detection.

[0042] The result can be evaluated as follows: a) If at least one criterion K1), K2) or K3) does not reach significance in the test, that is, at least one criterion K1), K2) or K3) is not met, this is evaluated as an unsuccessful control phase without purging. b) If all three criteria K1), K2) and K3) are met, this is evaluated as a successful control phase with purging.

[0043] According to another aspect, the technology disclosed in the present invention provides a data processing device, which includes a device for executing a method according to one of the above embodiments. The data processing device can be implemented by a controller of a motor vehicle or a combination of multiple interconnected controllers. The device can be provided with at least one microprocessor and / or at least one microcontroller. The at least one sensor described can also be configured to implement the system.

[0044] According to another aspect, the technology disclosed in the present invention provides a motor vehicle having an anode purge valve in a fuel cell and having a data processing device of an embodiment.

[0045] According to another aspect, the technology disclosed in the present invention provides a computer program, which includes program instructions. When the program instructions are executed by a microprocessor, the program instructions cause the microprocessor to execute the method according to one of the above embodiments.

[0046] According to another aspect, the technology disclosed in the present invention provides a computer-readable storage medium, on which the above-mentioned computer program is stored. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Now the technology disclosed by the present invention is explained with the help of the accompanying drawings. Shown in the figure:

[0048] Figure 1 is a schematic diagram of a computer-implemented method for identifying purges according to an embodiment of the present invention;

[0049] Figure 2 is a schematic diagram of a computer-implemented method for identifying purge according to another embodiment of the present invention;

[0050] Figure 3 is based on Figure 2 An example of the evaluation of the method;

[0051] Figure 4 is based on Figure 2 Another example of the evaluation of the method;

[0052] Figure 5 and Figure 6 is based on Figure 2 Examples of measurements of the method; and

[0053] Figure 7 is an example of an alternative measurement approach. DETAILED DESCRIPTION

[0054] In the following Figures 1 to 7 In the description of the illustrated embodiments, the same reference numerals are used for features that are identical and / or at least similar in their design and / or function. If these features are not described in detail again, their design and / or function are identical to the design and / or function of the features described above.

[0055] A computer-implemented method is described below with reference to a fuel cell 10 (not shown in greater detail). The fuel cell can be installed in a motor vehicle and operated during driving.

[0056] The fuel cell 10 comprises an anode 12, a membrane 14 and a cathode 16. The anode 12 and the cathode 16 can each have a flow channel 18. The flow channel 18 in the anode 12 has an anode purge valve 20 for performing a purge 22. The anode purge valve 20 can be opened and closed by a control device 24 of the fuel cell 10. The opening and closing of the anode purge valve 20 is preferably performed in a periodic manner.

[0057] During the purge, the anode purge valve 20 is opened for a certain period of time and then closed again. The purge 22 can discharge fuel waste gas from the chemical reaction in the fuel cell 10 from the flow channel 18 of the anode 12 .

[0058] The following references Figure 1 A computer-implemented method for identifying a purge 22 according to one embodiment of the present invention is described.

[0059] The method comprises step S11:

[0060] Testing the anode purge valve 20 for initiating the actuation phase 26 of the purge 22 .

[0061] In other words, the control device 24 can control the anode purge valve 20 during the control phase 26 and provide a command or a signal for opening or closing to the anode purge valve 20. The control phase 26 is detected in the method.

[0062] The method further comprises step S12:

[0063] A first measured variable 28 is monitored, which is a function of the pressure 30 which has changed as a result of the purging 22 .

[0064] The first measured variable 28 is dependent on the pressure 30 which changes due to the purge 22. For example, the pressure 30 in the anode 12 can change during the purge 22. The first measured variable 28 also changes due to the purge 22. The first measured variable 28 is monitored, preferably continuously.

[0065] The method further comprises step S13:

[0066] A second measured variable 32 is monitored, which in addition to or independently of the purge 22 also influences the pressure 30 .

[0067] In other words, second measured variable 32 influences pressure 30 independently of purging 22 . Second measured variable 32 can represent an influence on pressure 30 that is not caused by purging 22 .

[0068] Second measured variable 32 may change significantly during dynamic driving operation. Dynamic driving operation is understood here to mean, for example, that the driver influences the operation of fuel cell 10 through his driving style.

[0069] For example, the second measured variable 32 may be the pressure at the anode 12 of the fuel cell 10 and / or the current in the fuel cell 10 and / or the fuel consumption in the fuel cell 10 .

[0070] In particular, the second measured variable 32 can be a set pressure in the fuel cell 10 that is influenced by the driver. The set pressure can rise or fall continuously during dynamic driving operation, because a corresponding electrical power demand is drawn from the fuel cell, for example, by acceleration and / or braking of the vehicle. This results in the set pressure being set accordingly by a regulator (known from the prior art) in order to generate the required electrical power.

[0071] The detection of the control phase 26 , the monitoring of the first measured variable 28 and the monitoring of the second measured variable 32 can be carried out in parallel or sequentially.

[0072] The method further comprises steps S14 and S15:

[0073] If first measured variable 28 changes during control phase 26 and / or within monitoring period 34 after the control phase (step S14 ), it is determined whether second measured variable 32 has changed by more than a threshold value before or within monitoring period 34 (step S15 ).

[0074] In other words, the change in first measured variable 28 is determined within a time period in which first measured variable 28 is monitored and which extends during and / or after control phase 26 .

[0075] Monitoring period 34 can correspond, for example, to control phase 26 , but this is not mandatory. The change in first measured variable 28 can also be monitored with a time delay relative to control phase 26 .

[0076] Furthermore, it is checked whether second measured variable 32 has changed significantly in the above sense before or during monitoring period 34. As already mentioned, a significant change in second measured variable 32 may be caused by dynamic driving operation.

[0077] A significant change in second measured variable 32 is preferably understood to mean a change in second measured variable 32 such that an influence of purge 22 on pressure 30 can no longer be distinguished from an influence of second measured variable 32 on pressure 30 .

[0078] In order to determine a significant change in the second measurement variable 32, a threshold value 35 can be defined, for example, which the change in the second measurement variable 32 must exceed. The threshold value 35 can be an absolute threshold value or a threshold value relative to an initial value. For example, the threshold value 35 can be exceeded when the relative change in the second measurement variable 32 reaches a certain percentage or when the relative change in the second measurement variable 32 reaches N times (where N can be any value greater than 1).

[0079] In order to determine a significant change in the second measured variable 32, the second measured variable 32 may also be averaged in a moving average domain. If the value of the second measured variable 32 averaged in the time average domain exceeds a threshold value 35, it may be determined that the second measured variable 32 has changed significantly.

[0080] Other possibilities for determining a significant change in second measured variable 32 are conceivable, for example by ascertaining the slope of the change in second measured variable 32 .

[0081] In the next step S16 of the method:

[0082] - purge 22 is detected under criterion K3, ie the second measured variable 32 has not changed significantly before or during the monitoring period 34,

[0083] It is checked whether the second measured variable 32 has or has had an influence on the pressure 30 before or during the monitoring period 34. This is regarded as evidence that a significant change in the second measured variable 32 has a causal effect on the change in the first measured variable 28. This is therefore evaluated as a control phase 26 without purging 22.

[0084] Under criterion K3, i.e., if second measured variable 32 does not change significantly before or during monitoring period 34, it is concluded that the opening and closing of anode purge valve 20 is causally related to the change in first measured variable 28. This is therefore evaluated as a control phase 26 with purge 22.

[0085] Therefore, the evaluation of whether the anode purge valve 20 is correctly opened or closed during the control phase 26 excludes these times corresponding to dynamic driving operation. If these times are included, it cannot be reliably determined whether the anode purge valve 20 is correctly opened or closed during dynamic driving operation.

[0086] Preferably, for example, threshold value 35 for determining a significant change in second measured variable 32 is defined such that an influence of second measured variable 32 on pressure 30 can be ruled out with high probability when the change in second measured variable 32 is below threshold value 35. Thus, purging 20 can also be detected with high probability.

[0087] If, on the other hand, second measured variable 32 exceeds the threshold value, an influence of second measured variable 32 can no longer be ruled out with a high probability, so that even in the case of actual purging 20 , purging 20 is preferably not detected.

[0088] Thus, the risk of a false recognition of a purge 20 can be reduced, although a purge 20 has not actually occurred. The risk of damage to the fuel cell 10 or even a complete failure is thus also reduced.

[0089] Figure 2 A schematic diagram of a method according to another embodiment of the present invention is shown.

[0090] For a better overview, Figure 2 Steps S11, S12 and S13 are not shown in:

[0091] - detecting the anode purge valve 20 for initiating the control phase 26 of the purge 22 (step S11 );

[0092] - monitoring a first measured variable 28 which is related to the pressure 30 which has changed due to the purging 22 (step S12 ); and

[0093] - Monitoring of a second measured variable 32 influencing the pressure 30 in addition to the purge 22 (step S13). However, according to Figure 2 The method may include the above steps.

[0094] In accordance with Figure 2 In the method, the additional criteria K1 and K2 are connected before the criterion K3. However, any other order of the criteria K1, K2, K3 is also possible and the checking of the criteria K1, K2, K3 can be carried out in parallel or sequentially in time. In addition, some embodiments of the invention are also possible in which only one of the additional criteria K1 or K2 can be implemented.

[0095] according to Figure 2 The method comprises steps S14 and S17:

[0096] If a change in first measured variable 28 occurs within monitoring period 34 (step S14 ), a magnitude 36 of the change in first measured variable 28 is determined (step S17 ).

[0097] In step S18 , it is then checked whether the magnitude 36 is greater than or equal to a predetermined magnitude 38 .

[0098] If the magnitude 36 is not greater than or equal to the predetermined magnitude 38 , this is evaluated as a control phase 26 without purging 22 .

[0099] In addition, the method further comprises step S19:

[0100] - Determining a curve characteristic 40 of the change in first measured variable 28 .

[0101] The curve characteristic 40 can be, for example, a slope of the change of the measured variable 28. The curve characteristic 40 can also be a smoothed function or an average function of the change of the first measured variable 28. Furthermore, the curve characteristic 40 can be a property of the extent to which the change of the first measured variable 28 in the monitoring period 34 differs from the change in the control phase 26 or in the time period outside the monitoring period 34.

[0102] In step S20 , it is then checked whether the curve characteristic 40 substantially corresponds to the predetermined characteristic 42 .

[0103] If the curve characteristic 40 does not substantially correspond to the predetermined characteristic 42 in the described sense, this is evaluated as an actuation phase 26 without purge 22 .

[0104] In addition, the method further comprises step S15:

[0105] - determining whether the second measured variable 32 has changed significantly before or during the monitoring period 34 .

[0106] Under criterion K1 (value 36 is greater than or equal to predetermined value 38), under criterion K2 (curve characteristic 40 substantially corresponds to predetermined characteristic 42) and under criterion K3 (second measured variable 32 does not change significantly before or during monitoring period 34), it is concluded that the opening and closing of anode purge valve 20 is causally related to the change of first measured variable 28. This is therefore evaluated as a control phase 26 with purge 22 (step S16).

[0107] In a not Figure 1 and Figure 2 In a step shown in the figure but preferred, the identification of the purge 22 can be averaged over a plurality of operating stages 26. This can improve the reliability of confirming that the anode purge valve 20 is correctly opened or closed. The averaging process can, for example, include calculating how many successful purge identifications there are and how many failed purge identifications there are, and giving a percentage (e.g., less than 75% successful) below which a fault signal or operating limit is triggered.

[0108] Figure 3 Shown according to Figure 2 An example of an evaluation of the method.

[0109] exist Figure 3 , a control signal 44 is also shown in the course of time. The control signal 44 comprises a first control phase 46, a second control phase 48 and a third control phase 50. The control phases 46, 48, 50 are highlighted from the background of the control signal 44 by an essentially rectangular signal.

[0110] The first measured variable 28 is Figure 3 The example shown in FIG. 5 corresponds to a pressure difference 52 formed by an anode inlet pressure 54 and an anode outlet pressure 56 .

[0111] also, Figure 3 Also shown is the changed value 36 of the first measured variable 28 , ie the changed value 36 of the pressure difference 52 (for standard K1 ).

[0112] also, Figure 3 Also shown in FIG. 3 is a curve characteristic 40 of the change in the first measured variable 28 , ie, a curve characteristic 40 of the change in the pressure difference 52 , within the monitoring period 34 (for standard K2 ).

[0113] also, Figure 3 Also shown is a second measured variable 32 (for standard K3 ).

[0114] The first control phase 46 and the third control phase 50 meet the three criteria K1, K2 and K3. Therefore, these control phases 46, 50 are evaluated as control phases 26 with purge 22. The second control phase 48 meets the criteria K1 and K2, but does not meet the criterion K3. Therefore, the second control phase 48 is evaluated as a control phase 26 without purge 22.

[0115] Figure 4 Shows that according to Figure 2 Another example of an evaluation of a method comprising four manipulation stages 46 , 48 , 50 , 58 .

[0116] Over the entire course of change, second measured variable 32 shows no significant change. Criterion K3 is therefore satisfied for these four control phases 46 , 48 , 50 , 58 .

[0117] The first control phase 46 satisfies neither the criterion K1 nor the criterion K2 . This control phase 46 is therefore evaluated as a control phase 26 without purge 22 .

[0118] The second control phase 48 , the third control phase 50 and the fourth control phase 58 all meet the criteria K1 and K2 . Therefore, these control phases 48 , 50 , 58 are evaluated as control phases 26 with purge 22 .

[0119] Figure 5 and Figure 6 An example of a measurement performed on a fuel cell 10 having an anode purge valve 20 of a vehicle 60 during dynamic driving operation is shown.

[0120] The control signal 44 comprises a plurality of periodically executed control phases 26. The first measured variable 28 is Figure 3 and Figure 4The same corresponds to the pressure difference 52 formed by the anode inlet pressure 54 and the anode outlet pressure 56. This method reliably distinguishes between control phases 26 with purge 22 (marked with a tick) and control phases 26 without purge 22 (marked with a cross).

[0121] The ellipses 70 represent current influences which do not rise in a step-like manner and can be distinguished from brief control phases 22 with purge processes 22 .

[0122] The method Figure 1 and Figure 2 The embodiments described in the aspects can be executed by a suitable data processing device 64, not shown in further detail. In addition, the embodiments of the method can also be implemented in a computer program 66, not shown in further detail, by means of program instructions. In addition, the computer program 66 can also be stored on a computer-readable storage medium 68, not shown in further detail.

[0123] An alternative to the method described is Figure 1 and Figure 2 In the embodiment described in the above aspect, a fuel sensor, such as a hydrogen sensor, can be used. However, other sensors are required for this purpose. Figure 7 A measurement example of this alternative method for detecting purge is shown in , in which the time profile of fuel measured variable 72 is shown.

[0124] The principles of the preferred embodiments of the present invention can therefore be summarized as follows:

[0125] In order to detect the purging, it is therefore preferred to continuously monitor whether the anode purge valve is being actuated. Once an actuation phase is detected, a first measured variable, for example the pressure difference between the anode inlet pressure and the anode outlet pressure, is evaluated. The actuation phase can be detected, for example, based on a switching signal that should control the anode purge valve.

[0126] During the actuation phases of the anode purge valve, the change in a first measured variable is monitored and analyzed for known influences caused by the anode purge valve. The first measured variable can be a pressure level, in particular the pressure difference between two pressure sensors in the anode. If the actuation of the anode purge valve is estimated to have a significant influence on the pressure, a clear and adequate actuation of the anode purge valve can be obtained from this and in particular preferably averaged over a plurality of actuation phases. In this way, effective or successful actuation can be confirmed, so that the functionality is monitored continuously and in particular even in dynamic driving operation.

[0127] The term "substantially" (e.g., "substantially vertical axis") in the context of the technology disclosed in the present invention includes an exact property or exact value (e.g., "vertical axis"), respectively, as well as deviations that are insignificant to the function of the property / value (e.g., "tolerable deviations or tolerances"), respectively.

[0128] The above description of the present invention is for illustrative purposes only and is not intended to limit the present invention. Various changes and modifications may be made within the scope of the present invention without exceeding the scope of the present invention and its equivalents.

[0129] List of reference numerals:

[0130] - 10 Fuel Cells

[0131] - 12 Anode

[0132] - 14 membranes

[0133] - 16 cathode

[0134] - 18 runners

[0135] - 20 Anode purge valve

[0136] - 22 Purge

[0137] - 24 Control Devices

[0138] - 26 Control Phase

[0139] - 28 First measurement variable

[0140] - 30 Pressure

[0141] - 32 Second measurement variable

[0142] - 34 Monitoring time period

[0143] - 35 Threshold

[0144] - 36 Value

[0145] - 38 Predetermined value

[0146] - 40 curve features

[0147] - 42 Predetermined features

[0148] - 44 Control Signal

[0149] - 46 First Control Stage

[0150] - 48 Second Control Stage

[0151] - 50 Third Control Stage

[0152] - 52 Pressure difference

[0153] - 54 Anode inlet pressure

[0154] - 56 Anode outlet pressure

[0155] - 58 Fourth Control Stage

[0156] - 60 vehicles

[0157] - 62 Current

[0158] - 64 Data processing devices

[0159] - 66 Computer Programs

[0160] - 68 Computer readable storage medium

[0161] - 70 ellipse

[0162] - 72 fuel measurement parameters

[0163] - K1, K2, K3 standards

Claims

1. A computer-implemented method for identifying a purge (22) initiated in a fuel cell (10) by means of an anode purge valve (20), the method comprising the following steps: - detecting the anode purge valve (20) for initiating a control phase (26) of the purge (22); - monitoring a first measured variable (28) which is related to a pressure (30) which changes when the purge (22) actually takes place; - monitoring a second measured variable (32), which has an influence on the pressure (30) independently of the purging (22) that occurs; - if the first measured variable (28) changes during the control phase (26) and / or within a monitoring period (34) after the control phase, determining whether the second measured variable (32) has changed by more than or equal to a predetermined value before or during the monitoring period (34); and - under the criterion (K3), the occurrence of purging (22) is detected, ie, the second measured variable (32) has undergone a change in value that is less than a predetermined threshold value before or during a monitoring period (34).

2. The method according to claim 1, further comprising the steps of: - if the first measured variable (28) changes within the monitoring period (34), determining the magnitude (36) of the change in the first measured variable (28); and - under the additional criterion (K1), the occurrence of purging (22) is detected, namely, the value (36) is greater than or equal to a predetermined value (38).

3. The method according to any one of the preceding claims, further comprising the steps of: - if the first measured variable (28) changes within the monitoring period (34), determining a curve characteristic (40) of the change of the first measured variable (28); and - under the additional criterion (K2), the occurrence of purging (22) is detected, namely, the curve characteristic (40) corresponds to a predetermined characteristic (42).

4. A method according to any one of the preceding claims, wherein: The first measured variable (28) is the pressure (30) in a flow channel (18) of the fuel cell (10), wherein the flow channel (18) is preferably arranged at the anode (12) of the fuel cell (10); and / or the first measured variable (28) is the pressure difference (52) between an inlet pressure or anode inlet pressure (54) on one side and an outlet pressure or anode outlet pressure (56) in the flow channel (18) on the other side.

5. A method according to any one of the preceding claims, wherein: The second measured variable (32) is the pressure at the anode of the fuel cell (10) and / or the current in the fuel cell (10) and / or the fuel consumption in the fuel cell (10).

6. The method according to any one of the preceding claims, wherein: The determination begins 10-1000 ms, preferably 50-250 ms, particularly preferably 90-110 ms before the monitoring time period (34); and / or the monitoring time period (34) begins with the start of the control phase (26); and / or the monitoring time period (34) ends with the end of the control phase or ends after the control phase (26) until 1000 ms, preferably until 500 ms, particularly preferably until 250 ms.

7. The method according to any one of the preceding claims, further comprising the steps of: - The detection of the occurrence of a purge (22) is averaged over a plurality of control phases (26).

8. Data processing apparatus (64) comprising means for executing the method as claimed in any one of the preceding claims.

9. Computer program (66) comprising program instructions which, when executed by a microprocessor, cause the microprocessor to perform the method of any one of claims 1 to 7.

10. A computer-readable storage medium (68) on which the computer program (66) of claim 9 is stored. 11 . A motor vehicle comprising an anode purge valve ( 20 ) in a fuel cell ( 10 ) and comprising a data processing device ( 64 ) as claimed in claim 8 .

Images