Diagnostic procedure for damage diagnosis during the operation of a fuel cell system
Patent Information
- Application Number
- DE102022120620
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2021-08-17
- Filing Date
- 2022-08-16
- Publication Date
- 2025-02-06
- Estimated Expiration
- 2042-08-16
Smart Images
Figure 00000000_0000_ABST
Abstract
Description
The present invention relates to a diagnostic method for a damage diagnosis during operation of a fuel cell system, to a diagnostic device for carrying out such a diagnostic method, and to a fuel cell system having such a diagnostic device.The present invention is based on the known diagnostic possibility for detecting damage to fuel cells in a fuel cell system. These are in particular damage to the membrane within fuel cells, which separates the cathode side from the anode side of a fuel cell. During continuous operation of fuel cells, it may happen that the membrane is thinned out and becomes so thin, in particular at local locations, that so-called pin holes are produced. This may result in leakage occurring between the cathode side and the anode side, which may result in the failure of this fuel cell. It is also known that leaks can occur during the clamping of the membrane, but also into the environment with continued operation, so that a diagnosis is desired in order to monitor this continued damage process.It is known to diagnose the damage described above with a so-called bleed down. For this purpose, the fuel cell system is transferred into a special operating mode. In this case, the gas supply to the cathode side is stopped and it is subsequently waited for how the voltage conditions develop at the individual fuel cell, a plurality of fuel cells or a fuel cell stack. If the voltage falls over a defined period of time, a diagnostic period of time can be determined, the length of which gives a statement about the current damage situation. If the diagnostic time period is very short, a damaged membrane or a leak in the fuel cell stack can be assumed. If the diagnosis period lies in a predefined longer time horizon, this can lead to the conclusion as to an intact membrane and a lack of a leak.It is disadvantageous in the known solutions that such diagnoses can only be carried out on test stands for fuel cells. This is based in particular on the fact that no current production of the fuel cell system is possible during the execution of such a diagnosis. Moreover, the time period for performing such diagnosis is relatively long, and is in the range of, for example, one or more minutes. This has the result that in known solutions fuel cell systems are checked for test stands before their later use and information about the damage processes is later used in characteristic maps in real operation, for example when using the fuel cell system in a vehicle. A real determination and a diagnosis during operation in a vehicle are therefore not possible with the known solutions. Inaccuracies in the characteristic maps, but also variations in the operating mode of the fuel cell system, can therefore not be taken into account, or can only be taken into account to a very limited extent, in the solutions to date.It is the object of the present invention to at least partially eliminate the disadvantages described above in a cost-effective and simple manner. In particular, it is the object of the present invention to integrate a diagnosis of the fuel cell system into the vehicle operation in a cost-effective and simple manner.The above object is achieved by a diagnostic method having the features of claim 1, a diagnostic device having the features of claim 11 and a fuel cell system having the features of claim 12. Features and details which are described in connection with the diagnostic method according to the invention naturally also apply in connection with the diagnostic device according to the invention and the fuel cell system according to the invention and vice versa, so that with regard to the disclosure reference is or can always be made to the individual aspects of the invention reciprocally.According to the invention, a diagnostic method is intended to enable damage diagnosis during operation of a fuel cell system having a fuel cell stack with an anode section and a cathode section. This diagnostic method is preferably used for operation in a vehicle. Such a diagnostic method is characterized by the following steps:receiving an operation stop signal for stopping the operation of the fuel cell system,stopping gas supply to the cathode portion of the fuel cell stack,presetting a discharge current for discharging the fuel cell stack,monitoring a cell voltage at the fuel cell stack,terminating the specification of the discharge current when a diagnosis starting voltage is reached at the fuel cell stack,detecting a diagnosis time period from the diagnosis start time when the diagnosis start voltage is reached to the diagnosis end time when a diagnosis end voltage is reached.The core idea of a diagnostic method according to the invention is that it can be integrated in the operation of the fuel cell system, for example for driving a vehicle. If a fuel cell system is used for driving a vehicle, this is effected, for example, for a hybrid drive functionality. A battery device is coupled to an electric motor, wherein the fuel cell system can make available electrical power for the electric motor in addition to the battery device. At the same time, charging of the battery device from the fuel cell system is also possible.In the context of the invention, the reception of an operation stop signal can preferably also be understood to mean a signal of a power demand of a predetermined limit being undershot.In the context of the invention, the final voltage is preferably understood to mean a single cell voltage, a cell group voltage and / or a fuel cell stack voltage. The diagnosis time voltage can preferably also be constant within the scope of the invention, wherein a voltage difference is then used as a measure.In addition to a maximum limitation for power generation using fuel cell systems, minimum power outputs are usually also predefined, which should or may not be undershot during efficient operation of the fuel cell system. A minimum power limit is usually in the range of approximately 10% of the maximum possible power of such a fuel cell system. If the power demand now falls below such an operating minimum value during operation of the vehicle, this leads to an operation stop of the fuel cell system in known fuel cell systems in vehicles in order to avoid inefficient operation. However, in the case of such a transition into an operation stop, there is the risk of degradation of the membrane of the individual fuel cells in the fuel cell stack. This is based in particular on the fact that after the gas supplies have been switched off, a high electrical potential forms on the membranes of the fuel cells, which can lead to membrane damage. In order to avoid this, a discharge current is specified in a first step according to the invention. This discharge current now leads to the result that in an operation stop situation after the gas supply to the cathode section is stopped, the discharge current results in a rapid lowering of the electrical potential and thus of the voltage at the individual fuel cells. This makes it possible for the discharge of the fuel cells in the fuel cell stack to take place as quickly as possible, in order to keep the period of time during which a high electrical potential advances a damage mechanism very short.In contrast to known solutions, in which the operation stop with a discharge current leads to the most complete and rapid possible discharge of the individual fuel cells, however, a diagnostic method according to the invention stops the specification of the discharge current as soon as a diagnostic starting voltage is reached. Such a diagnostic starting voltage is greater than 0 volt and is in particular predefined at 0 volt at a predefined distance from the zero point. It thus becomes possible, on the one hand, to very rapidly reduce an electrical potential at the individual fuel cells of the fuel cell stack in a first step by the specification of the discharge current. However, a residual potential is maintained which, as will be explained later, is preferably below a damage limit in order subsequently to carry out a diagnostic method.The diagnostic method proceeds in a similar manner to the known bleed-down method. Thus, a time period is measured which detects, as a diagnosis time period, the time between a diagnosis start time and a diagnosis end time. The diagnosis start time is the start of a corresponding timer when the cell voltage on the fuel cell stack has reached the diagnosis start voltage after the discharge current has been specified. It is now waited how, without imposing a discharge current or external potential reductions, the voltage develops at the fuel cells of the fuel cell stack. A drop over time leads to a diagnostic end voltage lying below the diagnostic start voltage being reached after a diagnostic time period. The longer this diagnosis time period between this diagnosis start time and the diagnosis end time, the higher the tightness inside the fuel cell stack. Thus, as will also be explained later, limit values can be predefined which can either positively diagnose damage to the membrane within the fuel cell stack and / or negatively diagnose integrity of the membrane in the fuel cell system.It should also be noted that the cell voltage can be monitored for individual fuel cells, for individual groups of fuel cells, but also across the stack for all fuel cells in a fuel cell stack. Preferably, the monitoring of the cell voltage takes place both after the reception of the operation stop signal until the diagnosis start voltage is reached and proceeding further until the diagnosis end voltage is reached.A diagnostic method according to the invention now brings various advantages. On the one hand, it becomes possible to additionally use stop situations in real operation in a vehicle for the fuel cell system for a diagnostic method. This allows, in addition to the use of known characteristic maps which are determined at the test stand for fuel cell systems, also a diagnostic to be integrated into the operation of the vehicle in order in this way to allow damage situations in the fuel cell system to be monitored even more accurately.In addition, the diagnosis is integrated into a stop situation of the fuel cell system, which essentially fulfills two core ideas. In the first step, the fastest possible lowering below a damage potential takes place, while in the second step, a relatively slow diagnosis takes place. The slow diagnosis has the additional advantage that simple and cost-effective sensors and measurement diagnostics can be used, so that the cost for monitoring and carrying out a diagnostic method according to the invention remains manageable or is even possible with sensors present in existing fuel cells.It can entail advantages if, in a diagnostic method according to the invention, the diagnostic starting voltage is below a damage voltage for the fuel cell stack. Depending on the material selection and the structural design of the respective fuel cell, a defined damage voltage can be predefined above which, without a power loss, i.e. in a stop situation for this fuel cell, an in particular irreversible damage to the membrane could occur. This damage voltage can also be understood as a limit value under which the discharge current must bring the potential of this fuel cell in a stop situation in order to avoid or at least reduce damage. The correlation of this embodiment, namely that the diagnosis starting voltage lies below the damage voltage, leads to the mentioned advantages, namely that a rapid lowering of the voltage at the respective fuel cell below the damage voltage for protection thereof becomes possible. Subsequently, below a risk of damage, the diagnosis is carried out with the aid of the diagnosis starting voltage up to the diagnosis end voltage. The damage voltage is therefore above the diagnosis start voltage, wherein the diagnosis start voltage is still high enough to carry out the explained bleed-down diagnosis method according to the present invention. Preferably, a safety margin of, for example, 10% to 20% is present between the diagnosis starting voltage and the damage voltage. It is thus possible for the diagnosis starting voltage to be predefined at approximately 90% or approximately 80% of the damage voltage.Advantages are furthermore obtained if, in a diagnostic method according to the invention, the cell voltage for a plurality of fuel cells and / or for a plurality of fuel cell groups of the fuel cell stack is monitored separately, wherein in particular the minimum cell voltage of all monitored cell voltages is used for the detection of the diagnostic time period and / or the maximum cell voltage is used for the termination of the specification. In this embodiment, it is again apparent that a plurality of different cell voltages can be detected for individual fuel cells or fuel cell groups. This also results in individual values for the cell voltage for each detected fuel cell and / or for each monitored fuel cell group. According to the invention, the minimum cell voltage of all monitored cell voltage or, for example, the standard deviation or other static evaluation methods is now used, for example, for the diagnostic time period to be detected. In other words, the diagnosis start time is started by reaching the diagnosis start voltage by the minimum cell voltage, and the diagnosis end time is also defined by reaching the minimum cell voltage at the diagnosis end voltage. This makes it possible to carry out the monitoring for the cell, which in particular first reaches the end point in time of the diagnosis and thus entails the greatest risk of damage to the membrane. The alternative or additional use of the maximum cell voltage of all monitored cell voltages for the specification has the result that, in particular in combination with a damage voltage according to the preceding paragraph, it is ensured that all individually monitored fuel cells or fuel cell groups are also reliably below this damage limit.Advantages can entail if, in a diagnostic method according to the invention, the operation stop signal is received if, in a comparison of a detected power demand with an operation minimum value for the operation power of the fuel cell system, the detected power demand is below the operation minimum value. This allows, in a diagnostic method according to the invention, a so-called stop mode for temporarily stopping the operation of the fuel cell system to be combined with the diagnostic method. As has already been explained at the beginning, the diagnostic method can thus be carried out if the power requirement is so low that a stop of the fuel cell system is to take place. This operation stop signal can be provided from the outside in such a case, but can also be generated within the scope of the diagnostic method according to the invention itself.It can be further advantageous if, in a diagnostic method according to the invention, a prediction and / or a simulation of an operation stop duration takes place, wherein the steps of ending the specification of the detection and the detection of the diagnostic time period are blocked in at least one of the following cases:operation stop duration is shorter than a damaged diagnostic time period,Operation stop duration is shorter than an intact diagnostic period.The above list is a non-exhaustive list. Blocking of the steps of ending the specification and of detecting the diagnostic time period accordingly leads to blocking of the diagnostic steps in a diagnostic method according to the invention. This is of course useful if the diagnosis could not be successfully carried out by the prognosis or simulation that was carried out. This is the case when the expected operation stop duration is shorter than a damaged diagnosis period. A damaged diagnostic time period is the diagnostic time period when the fuel cell is undershot, said diagnostic time period being identified as damaged. In contrast, an intact diagnostic period is the diagnostic period when the fuel cell is defined as intact when exceeded. The damaged diagnosis time period and the undershoot thereof can thus be understood as a positive test for the damage situation and an overshoot of the intact diagnosis time period can be understood as a positive test for an intactness of the fuel cell. Blocking when the operation stop duration is shorter than one of these two diagnosis periods leads to ensuring that a diagnosis is carried out only when at least one of these two tests can also actually be carried out with the receipt of a comparison statement with at least one of the two periods.Further advantages can be achieved if, in a diagnostic method according to the invention, the discharge current is predetermined as a function of the cell voltage, in particular directly proportional. As has been explained, such a discharge current serves to reduce the cell voltage at the fuel cell stack as quickly as possible. A high voltage at the respective fuel cells during the switch-off operation leads in this embodiment to a higher discharge current than a correspondingly lower cell voltage during the switch-off situation. This has the effect, in particular, that the speed at which the discharge current achieves a reduction in the cell voltage is adapted to the difference to be bridged between the cell voltage in the switch-off situation and the diagnosis starting voltage. Such adjustability leads to adjustability of the speed of the reduction of the cell voltage, so that the protection mechanisms can be adapted accordingly to the respective operating situation before switching off.Further advantages can be achieved if, in a diagnostic method according to the invention, the discharge current is kept constant or substantially constant until the diagnosis starting voltage is reached. Readjustment is therefore preferably not effected. This leads to a very simple controllability of the discharge current, since its value is specified and maintained as a setpoint value. Safety in performing these discharge steps for the fuel cells is also increased in this way.Further advantages can be achieved if, in a diagnostic method according to the invention, the discharge current is completely or substantially completely blocked for the diagnostic period. This can be provided, for example, by electrically isolating possible current drains at the fuel cell stack. Maximizing electrical resistances at the fuel cell stack can also bring about this effect. In particular, corruption of the diagnosis is avoided in this way, since parallel discharge processes on the fuel cell stack are excluded as completely as possible or at least reduced to a minimum by current consumption.It is furthermore advantageous if, in a diagnostic method according to the invention, the detected diagnostic time period is checked for a damaged diagnostic time period being undershot and / or for an intact diagnostic time period being exceeded. This diagnostic result as a result of this comparison can now be output as a diagnostic signal in a diagnostic method according to the invention. If the detected diagnostic time interval falls below a damaged diagnostic time interval, the cell voltage at the fuel cell stack has dropped so quickly that a damaged membrane or a leak can be assumed. However, if the diagnostic time period exceeds an intact diagnostic time period, it is highly likely that the membrane is intact and that any leakage from the fuel cell stack is not present. When the intact diagnostic time period is exceeded, the detection of the diagnostic time period can preferably be discontinued, even if the final diagnostic voltage has not yet been reached at this point in time.It is likewise advantageous if, in a diagnostic method according to the invention, at least one of the following parameters is designed to be adjustable:- Diagnosis starting voltage- Final diagnosis voltageThe above list is a non-exhaustive list. It goes without saying that evaluations or the already explained periods of time which are to be used for the comparison can also be adapted and varied accordingly.Another object of the present invention is a diagnostic apparatus for damage diagnosis in operation of a fuel cell system including a fuel cell stack having an anode portion and a cathode portion. Such a diagnostic device includes a receiving module for receiving an operation stop signal for stopping the operation of the fuel cell system. Further, a stop module for stopping the supply of gas to the cathode portion of the fuel cell stack is provided. With the aid of a specification module, a specification of a discharge current for discharging the fuel cell stack and for ending the specification of the discharge current when a diagnosis starting voltage is reached at the fuel cell stack is possible. A monitoring module is provided for monitoring the cell voltage at the fuel cell stack. Finally, a detection module is provided for detecting a diagnostic time period from the diagnostic start time when the diagnostic start voltage is reached to the diagnostic end time when a diagnostic end voltage is reached. In this case, the receiving module, the stop module, the specification module, the monitoring module and / or the detection module are designed in particular for carrying out a diagnostic method according to the invention. Thus, a diagnostic device according to the invention provides the same advantages as have been explained in detail with reference to a diagnostic method according to the invention.The present invention likewise relates to a fuel cell system having at least one fuel cell stack having an anode section and a cathode section. The anode section includes an anode supply section for supplying anode supply gas and an anode discharge section for discharging anode off-gas. The cathode portion is provided with a cathode supply portion for supplying cathode supply gas and a cathode discharge portion for discharging cathode off gas. Further, the fuel cell system includes a diagnostic apparatus according to the present invention. Thus, a fuel cell system according to the invention brings with it the same advantages as have been explained in detail with reference to a diagnostic device according to the invention and with reference to a diagnostic method according to the invention.Further advantages, features and details of the invention will become apparent from the following description, in which exemplary embodiments of the invention are described in detail with reference to the drawings. They show schematically: FIG. 1 shows an embodiment of a diagnostic device according to the invention, FIG. 2 shows an embodiment of a fuel cell system according to the invention, FIG. 3 shows a possible execution of a diagnostic method according to the invention, FIG. 4 shows a further possible execution of a diagnostic method according to the invention, and FIG. 5 shows a power curve on a fuel cell system.FIG. 1 schematically shows a diagnostic device 10 as can be used in fuel cell systems 100 when used in a vehicle. The diagnostic device 10 is capable of starting a diagnostic method according to the invention with the reception of an operation stop signal BSS. For this purpose, a receiving module 20 is provided, which receives this operation stop signal BSS or even generates itself, for example as a function of a power request LAF, as shown in FIG. 5. A relay to a stop module 30 leads to the gas supply to the cathode section 130 of the fuel cell stack 110 now being stopped in the control of the fuel cell system 100. In order subsequently to avoid damage to the membrane of the individual fuel cells in the fuel cell stack 110, the diagnostic device 10 has a specification module 40, which now specifies a discharge current IEto the fuel cell stack 110 and applies it thereto. The discharge current IE reduziert the cell voltage UZwhich is continuously monitored by the monitoring module 50.As soon as the monitoring module 50 establishes during the monitoring of the cell voltage UZ that a diagnosis starting voltage UDS is undershot, this essentially leads to two further actions. Firstly, the specification of the discharge current IEis stopped by the specification module 40 and secondly a timer is started by the detection module 60 in order to detect the diagnostic time period DZuntil the end diagnosis voltage UDEis reached when the cell voltage UZis continuously monitored.FIG. 2 schematically illustrates an application of such a diagnostic device 10 to a fuel cell system 100. Here, a part of the fuel cell system 100 is illustrated with a fuel cell stack 110 divided into an anode portion 120 and a cathode portion 130. Anode feed gas is supplied to the anode portion 120 via an anode feed portion 122, and cathode feed gas is supplied to the cathode portion 130 via a cathode feed portion 132. Anode off-gas is discharged via the anode discharge section 124, and cathode off-gas is discharged via the cathode discharge section 134. Not shown in detail are further functional components of such a fuel cell system 100, such as reformer devices, heat exchanger devices, recirculation devices or the like. The diagnostic device 10 is preferably configured according to the embodiment of FIG. 1 and is capable, illustrated here with dashed lines, of carrying out a signal exchange with the fuel cell stack 110, the anode supply section 122 and / or the cathode supply section 132 for carrying out a diagnostic method according to the invention.FIG. 3 schematically shows the performance and effect of one possibility of a diagnostic method according to the invention. The diagram illustrated in FIG. 3 plots the cell voltage UZon the Y axis over time along the X axis. Initially, the cell voltage UZ is at a high level during a regular operation for generating electric current until an operation stop signal BSS announces an pending operation stop of the fuel cell system 100. During operation of the fuel cell system 100, this operation stop signal BSS is followed by the stop of the supply of gas via the cathode section 130, so that this high cell voltage UZ would now bring about a damage potential for the damage of the respective fuel cells. In order to reduce this damage potential, a discharge current IEis predefined, which in FIG. 3 acts in such a way that the cell voltage UZis rapidly lowered over a short period of time. This lowering in an active manner by the specification of the discharge current IEis stopped as soon as the cell voltage UZhas reached a diagnosis start voltage UDS. At this time, each specification of a discharge current IEis preferably completely blocked, in particular the fuel cell stack 110 is even electrically insulated, so that the now beginning reduction of the cell voltage UZ erfolgt by processes within the fuel cell stack 110. Reaching the diagnosis start voltage UDS starts a timer, this start time being defined as the diagnosis start time DZS. During the continuous monitoring of the cell voltage UZ, it is detected when a lower-lying diagnosis end voltage UDE is reached. This reaching time is set as the diagnosis end time DZE, and the timer is stopped. The difference between the diagnosis start time DZS and the diagnosis end time DZE can thus be defined as the diagnosis time period DZ.As explained above, the lowering of the cell voltage UZ over the diagnostic time period DZ takes place exclusively or essentially exclusively on the basis of internal processes in the fuel cell stack 110. It can thus be concluded that, in the event of a rapid drop to the diagnostic end voltage UDE, there must accordingly be a leakage or damage to the membrane. If the diagnostic time period DZ is long, in particular longer than an intact diagnostic time period DZI, it can be assumed that the membrane is intact and that there is no local leakage.FIG. 4 also shows how a monitoring of more than one cell voltage UZ, here a minimum cell voltage UZ' and a maximum cell voltage UZ max, is carried out. In this case, the maximum cell voltage UZmax is used to ensure that this maximum cell voltage UZ at the beginning of the diagnostic method at the diagnostic start time DZS is also below a damage voltage US. Moreover, it is ensured that the diagnostic steps themselves are carried out at the minimum cell voltage UZ', in particular with regard to the diagnostic end time DZE.FIG. 5 shows the operating profile on the basis of the operating power BL of a fuel cell system 100 when a diagnostic method of the present invention is run through. Thus, a power demand LAF schwankt operation of the fuel cell system 100 depending on the operation of the vehicle. In a part-load or low-load operation, this power request LAF drops below an operation minimum value BM, which in this case triggers an operation stop signal BSS. At this point in time, it now becomes possible to carry out a diagnostic method according to the invention, i.e. in a first step to fall below a damage voltage US very quickly via the discharge current IE. In the embodiment of FIG. 5, it is additionally checked, before the performance of the diagnostic steps, with the aid of a prediction and / or a simulation, whether the entire operation stop duration BSD is long enough to perform a successful diagnosis. In FIG. 5, this is because both a damaged diagnosis period DZB and an intact diagnosis period DZI are shorter than the operation stop period BSD.The above explanations regarding the embodiments describe the present invention exclusively within the scope of examples.List of reference characters10 Diagnostic device 20 Receiving module 30 Stop module 40 Specification module 50 Monitoring module 60 Detection module 100 Fuel cell system 110 Fuel cell stack 120 Anode section 122 Anode supply section 124 Anode discharge section 130 Cathode section 132 Cathode supply section 134 Cathode discharge section BSS Operation stop signal BSD Operation stop duration LAF Power request BM Operation minimum value BL Operation power IE Discharge current UZ Cell voltage US Damage voltage UDS Diagnosis start voltage UDE Diagnosis end voltage DZ Diagnosis time period DZB Damaged diagnosis time period DZI Intact diagnosis time period DZS Diagnosis start time DZE Diagnosis end time
Claims
Diagnostic method for a damage diagnosis during operation of a fuel cell system (100) having a fuel cell stack (110) having an anode section (120) and a cathode section (130), characterized bythe following steps: - receiving an operation stop signal (BSS) for stopping operation of the fuel cell system (100), - stopping the gas supply to the cathode section (130) of the fuel cell stack (110), - presetting a discharge current (IE) for discharging the fuel cell stack (110), - monitoring a cell voltage (UZ) at the fuel cell stack (110), - ending presetting of the discharge current (IE) when a diagnosis start voltage (UDS) at the fuel cell stack (110) is reached, detecting a diagnostic time period (DZ) from the diagnosis start time (DZS) when the diagnosis start voltage (UDS) is reached to the diagnosis end time (DZE) when a diagnosis end voltage (UDE) is reached.Diagnostic method according to Claim 1, characterized in that the diagnostic starting voltage (UDS) is below a damage voltage (US) for the fuel cell stack (110).Diagnostic method according to one of the preceding claims, characterized in that the cell voltage (UZ) for a plurality of fuel cells and / or for a plurality of fuel cell groups of the fuel cell stack (110) is monitored separately, wherein in particular the minimum cell voltage (UZ min) of all monitored cell voltages (UZ) is used for the detection of the diagnostic time period (DZ) and / or the maximum cell voltage (UZ max) is used for the termination of the specification.Diagnostic method according to one of the preceding claims, characterized in that the operation stop signal (BSS) is received if, in the case of a comparison of a detected power demand (LAF) with an operation minimum value (BM) for the operation power (BL) of the fuel cell system (100), the detected power demand (LAF) is below the operation minimum value (BM).Diagnostic method according to one of the preceding claims, characterized in that a prediction or a simulation of an operation stop duration (BSD) takes place, wherein the steps of ending the specification and of detecting the diagnostic time period (DZ) are blocked in at least one of the following cases: - operation stop duration (BSD) is shorter than a damaged diagnostic time period (DZB) - operation stop duration (BSD) is shorter than an intact diagnostic time period (DZI)Diagnostic method according to one of the preceding claims, characterized in that the discharge current (IE) is predetermined, in particular directly proportional, as a function of the cell voltage (UZ).Diagnostic method according to one of the preceding claims, characterized in that the discharge current (IE) is kept constant or substantially constant until the diagnosis starting voltage (UDS) is reached.Diagnostic method according to one of the preceding claims, characterized in that the discharge current (IE) is completely or substantially completely blocked for the diagnostic time period (DZ).Diagnostic method according to one of the preceding claims, characterized in that the detected diagnostic time period (DZ) is checked for a damaged diagnostic time period (DZB) being undershot and / or for an intact diagnostic time period (DZI) being exceeded.Diagnostic method according to one of the preceding claims, characterized in that at least one of the following parameters is designed to be adjustable: - Diagnosis starting voltage (UDS) - Diagnosis ending voltage (UDE)Diagnostic device (10) for a damage diagnosis during operation of a fuel cell system (100) having a fuel cell stack (110) having an anode section (120) and a cathode section (130), having a receiving module (20) for receiving an operation stop signal (BSS) for stopping operation of the fuel cell system (100), a stop module (30) for stopping the gas supply to the cathode section (130) of the fuel cell stack (110), a specification module (40) for specifying a discharge current (IE) for discharging the fuel cell stack (110) and for terminating the specification of the discharge current (IE) when a diagnosis starting voltage (UDS) is reached at the fuel cell stack (110), a monitoring module (50) for monitoring the cell voltage (UZ) on the fuel cell stack (100) and a detection module (60) for detecting a diagnosis time period (DZ) from the diagnosis start time (DZS) when the diagnosis start voltage (UDS) is reached to the diagnosis end time (DZE) when a diagnosis end voltage (UDE) is reached, wherein the receiving module (20), the stop module (30), the specification module (40), the monitoring module (50) and / or the detection module (60) are designed in particular for carrying out a diagnosis method having the features of one of claims 1 to 10.Fuel cell system (100) having at least one fuel cell stack (110) having an anode section (120) and a cathode section (130), the anode section (120) having an anode feed section (122) for feeding anode feed gas and an anode discharge section (124) for discharging anode exhaust gas, the cathode section (130) having a cathode feed section (132) for feeding cathode feed gas and a cathode discharge section (134) for discharging cathode exhaust gas, wherein a diagnostic apparatus (10) having the features of claim 11 is further provided for carrying out a diagnostic method having the features of one of claims 1 to 10.