Methods for operating a hydrogen production plant and hydrogen production plant

Abstract

A method for operating a hydrogen production plant (10) with several electrolysis devices (11) configured to produce hydrogen from water using electric current, wherein the electrolysis devices (11) preferably form several cascades (12) connected in parallel, each consisting of electrolysis devices (11) connected in series, wherein the electrolysis devices (11) are supplied with water via a water circuit (13) for hydrogen production, and wherein the electrolysis devices (11) are supplied with electrical power from an electric current source or electric voltage source for hydrogen production. To check whether a hydrogen leak is present at an electrolysis device (11), the following steps of a safety routine are performed: The electrolysis device (11) to be checked is subjected to a defined electrical input quantity.At the electrolysis device (11) under test, an actual electrical output variable, dependent on the defined electrical input variable, is recorded. This actual output variable is then compared to a target output variable. Based on this comparison, a conclusion is drawn regarding the presence or probability of a hydrogen leak at the electrolysis device (11).

Description

[0001] The invention relates to a method for operating a hydrogen production plant and a hydrogen production plant.

[0002] DE 10 2017 108 413 A1 discloses an electrolysis device with a cell stack comprising several cell stack elements, wherein the cell stack elements of the cell stack form several electrolysis cells. Furthermore, the electrolysis device known from this prior art has a force application unit by which a force can be exerted on the cell stack to compress the cell stack elements of the cell stack in a fluid-tight manner. The force application unit has opposing end plates between which the cell stack is arranged and compressed. The force application unit also includes compression devices comprising spring elements and struts, wherein the spring force of the spring elements presses the end plates against each other, compressing the cell stack. Connections are provided on the end plates of the electrolysis device, namely water supply connections, water discharge connections, and hydrogen connections.Water is supplied to the electrolysis device via the water inlet connections, while water and oxygen are removed from the electrolysis device via the water outlet connections. The hydrogen connections are used to remove or pass through the hydrogen produced during electrolysis from the electrolysis device.

[0003] In a hydrogen production plant, several electrolysis devices are installed, wherein the electrolysis devices are expediently connected in series in the respective cascade, forming at least one cascade, preferably several parallel cascades.

[0004] The electrolysis units of a hydrogen production plant are connected to a water circuit within the plant, which supplies the electrolysis units with hydrogen for hydrogen production. For hydrogen production, the electrolysis units, which are connected to an electrical current or voltage source within the hydrogen production plant, are supplied with electrical power by this source.

[0005] During the operation of a hydrogen production plant, a hydrogen leak can occur at an electrolysis unit, particularly at an electrolysis cell. This can lead to hydrogen escaping, either together with oxygen, into the water discharged from the electrolysis unit via the water drain connections, or alternatively into the environment, specifically into a process chamber housing the electrolysis unit. When hydrogen enters the water along with oxygen, this is known as an H₂-in-O₂ leak.

[0006] Hydrogen is an explosive gas. It becomes ignitable when mixed with oxygen. Therefore, for safety reasons, it is essential to monitor whether a hydrogen leak from an electrolysis unit or electrolysis cell could allow hydrogen to enter a process chamber of the hydrogen production plant or, along with the oxygen produced during hydrogen production, reach the water side of the plant. Hydrogen sensors can be used for this purpose, measuring the hydrogen concentration in the process chamber or in the area of ​​an oxygen separator, which separates the oxygen from the water in the hydrogen production plant's water circuit.

[0007] If a hydrogen sensor detects an unacceptably high hydrogen concentration in the process room or in the area of ​​an oxygen separator, it can be concluded that there is a hydrogen leak in the hydrogen production plant.

[0008] However, it is not yet reliably possible to attribute a hydrogen leak to an electrolysis device of the hydrogen production plant or even to an electrolysis cell of an electrolysis device of the hydrogen production plant.

[0009] WO 2018 / 193 067 A1 discloses an electrolysis device with multiple electrolysis cells, wherein each electrolysis cell has at least one contact element that is distinct from a connection element for a power supply. The at least one contact element of the respective electrolysis cell serves to detect an individual cell parameter in order to identify a fault operating condition based on the individual cell parameter.

[0010] EP 3 336 947 B1 discloses a hydrogen compression device comprising a proton-conducting membrane, an anode, and a cathode. The hydrogen compression device includes a detector configured to detect hydrogen crossover through the proton-conducting membrane. This is achieved by detecting an electric current flowing between the anode and the cathode when an electrical voltage is applied to both and the flow paths of the hydrogen compression device are sealed at the anode and the cathode.

[0011] WO 03 / 096 459 A1 discloses a method for detecting a gas leak in a fuel cell.

[0012] There is a need, particularly when a hydrogen leak is detected in a hydrogen production plant, to trace it back to an electrolysis unit, preferably an electrolysis cell within the electrolysis unit. However, it should also be possible to reliably verify, even in other operating situations of a hydrogen production plant, whether a hydrogen leak exists at an electrolysis unit, especially at an electrolysis cell within the electrolysis unit.

[0013] The object of the invention is to provide a method for operating a hydrogen production plant that makes it possible to check whether a hydrogen leak exists in an electrolysis device of a hydrogen production plant. Furthermore, there is a need for a corresponding hydrogen production plant.

[0014] This problem is solved by a method for operating a hydrogen production plant according to claim 1 and a hydrogen production plant according to claim 18.

[0015] According to the invention, the following steps of a safety routine are performed to check whether a hydrogen leak is present in at least one electrolysis device: The electrolysis device to be checked is supplied with a defined electrical input. An actual electrical output variable, dependent on the defined electrical input variable, is measured at the electrolysis device. The actual output variable measured at the electrolysis device is compared with at least one target output variable. Depending on the comparison of the actual output variable with the at least one target output variable, a conclusion is drawn about a hydrogen leak or the probability of a hydrogen leak at the respective electrolysis device.

[0016] The invention makes it possible to check whether a hydrogen leak or a probability of a hydrogen leak exists in an electrolysis device, and, if necessary, to locate the site of the hydrogen leak or the probability more precisely.

[0017] The method according to the invention can be applied to a hydrogen production plant with a plurality of electrolysis devices. However, it can also be applied to hydrogen production plants with only one electrolysis device. The method according to the invention can be used to identify a hydrogen leak and / or to locate a hydrogen leak. Localization can mean identifying a specific electrolysis device or even identifying a specific electrolysis cell within an electrolysis device. The term "safety routine" is to be understood broadly and can also include purposes of predictive maintenance and maintenance recommendations—for example, when or which electrolysis device should be replaced.

[0018] The defined electrical input can be interpreted as an excitation signal or pulse. The actual electrical output can be interpreted as a response signal or system response.

[0019] The electrical input and output variables can be point-like or quasi-point-like quantities. They can also be time-related. For example, a constant or variable voltage pulse can be used as the electrical input. The output variable can conveniently be the time course over a predefined time interval. Alternatively or additionally, the start and end of the output variable can be limited by exceeding and / or falling below defined threshold values. The target output variable can be, for example, an empirically determined expected value of an output variable.

[0020] Actual and / or target output variables can, in particular, be voltage gradients (rate of decrease of the applied voltage).

[0021] Preferably, the safety routine is performed automatically, on a timer, and / or event-driven basis. In particular, the safety routine is performed on a timer after a defined operating time of the hydrogen production plant or the respective electrolysis device to be checked has been reached.Alternatively or preferably additionally, the safety routine is then carried out in an event-driven manner when the hydrogen production plant or the respective electrolysis device to be checked is switched from a production operation to a shutdown operation or is in shutdown operation, and / or when the hydrogen production plant or the respective electrolysis device to be checked is switched from a production operation to a standby operation or is in standby operation, and / or when a hydrogen sensor either in the area of ​​an oxygen separator of the water circuit of the hydrogen production plant or in a process room accommodating the electrolysis devices of the hydrogen production plant detects a hydrogen concentration that is greater than a limit value, and / or when a pressure sensor in a hydrogen discharge line of the hydrogen production plant detects a pressure drop that is greater than a limit value.The time-controlled and / or event-controlled execution of the safety routine is particularly preferred in order to check whether a hydrogen leak is present at a respective electrolysis device of the hydrogen production plant.

[0022] The safety routine can be triggered automatically. However, it can also be triggered manually, for example, if maintenance personnel are on site and the hydrogen production plant is in maintenance mode.

[0023] Standby operation refers to a state of the hydrogen production plant in which at least the hydrogen lines are under operating pressure and water continues to circulate through the water circuit, but essentially no hydrogen is being produced. The water typically has a temperature close to the target operating temperature, preferably above 50°C. The hydrogen production plant can be switched back from standby operation to production operation within a period of less than 5 minutes.

[0024] Shutdown operation refers to a state of the hydrogen production plant in which the hydrogen lines are depressurized, meaning they are no longer under operating pressure, and no water is circulating through the water circuit. Hydrogen is not produced in this state. The water typically reaches ambient temperature. Before hydrogen production can be resumed, a start-up routine must be performed, during which, among other things, the water is heated back to the target operating temperature before full hydrogen production capacity is available again.

[0025] Preferably, at least one target output variable for the respective electrolysis device under test depends on the age or operating hours of the device. This takes into account the fact that, for example, the expected target output variable shifts with increasing age. This prevents false detection of a hydrogen leak. Alternatively or additionally, the target output variable can be made dependent on pressure cycles, temperature cycles, or shutdown cycles (number of shutdowns).

[0026] Preferably, a defined electrical voltage signal is used as the defined electrical input variable applied to the electrolysis device under test, wherein the actual electrical output variable recorded at the electrolysis device under test is preferably a time-dependent measurement of an electrical voltage signal over a specific period of time, and wherein the target electrical output variable can be an empirically determined target output variable based on historical operating data or a variable dependent on the actual output variables of other electrolysis devices. This is particularly preferred for reliably verifying whether a hydrogen leak is present at an electrolysis device or whether the probability of a hydrogen leak exceeds a certain threshold.

[0027] Preferably, an individual actual electrical output variable is recorded at each electrolysis cell of the respective electrolysis device under test. This actual output variable is then compared with at least one target output variable to determine whether or not hydrogen leakage has occurred at that specific electrolysis cell, or the probability of such leakage. The target electrical output variable for each electrolysis cell of the respective electrolysis device under test is preferably an expected value determined empirically or theoretically beforehand.

[0028] Alternatively or additionally, the actual output values ​​of the other electrolysis cells, or a value dependent on the actual output values ​​of the other electrolysis cells in this electrolysis device under test, can be used. In other words, it is possible either to make a relative comparison between the actual values ​​of the individual electrolysis cells in an electrolysis device and, if the actual value of one electrolysis cell deviates from the actual value of the other electrolysis cells, to conclude that there is a hydrogen leak or a leakage probability. Alternatively, the comparison of the actual value of an electrolysis cell can also be made with the previously determined, empirically or theoretically calculated expected value as the target value in order to conclude that there is a hydrogen leak or a leakage probability. A combination of both approaches is possible.

[0029] The actual electrical output variable is preferably recorded over a time range of at least 30 seconds and up to 600 seconds.

[0030] This design allows for the determination of hydrogen leakage or the probability of such hydrogen leakage for individual electrolysis cells.

[0031] Preferably, the safety routine is performed for one of the cascades of the hydrogen production plant, while the other cascades are used for hydrogen production. This has the advantage that, while the safety routine is being executed, the hydrogen production plant can continue to be used for hydrogen production in those cascades where the safety routine is not performed.

[0032] In a further advantageous embodiment of the invention, the method can also include repeating one or more steps of the safety routine, for example, by deliberately performing these steps under different environmental conditions. This allows verification of whether a similar (or differing) behavior of an electrolysis device or cell also occurs under varying environmental conditions. This can improve the predictive accuracy of a fault or fault probability. Examples of environmental conditions that can be varied include different pressures, different temperatures, or different flow rates on the water side, i.e., of process water in the process water circuit, or different pressures on the hydrogen side, i.e., in the hydrogen lines.For example, the steps of the safety routine can be performed once at a first temperature and a second time at a second temperature lower than the first. The results of the first run are then compared with the results of the second run of the safety routine to determine or further improve the assessment of leakage or leakage probability. Preferably, an increased leakage probability is inferred precisely when the results of both runs yield the same or at least qualitatively comparable results. If multiple runs under different environmental conditions lead to different results, this may indicate that causes other than leakage are responsible for the deviant behavior of an electrolysis device (or an electrolysis cell).

[0033] Preferred embodiments of the invention are described in the dependent claims and the following description. Exemplary embodiments of the invention are explained in more detail with reference to the drawing, without being limited thereto. The drawing shows: Fig. 1 a schematic representation of a hydrogen production plant according to the invention with several electrolysis devices to illustrate the process according to the invention, Fig. 2 Actual output variables recorded at electrolysis cells of an electrolysis device, Fig. 3 target output variables that depend on the age or operating hours of the electrolysis device, Fig. 4 a diagram to illustrate a further development of the method according to the invention.

[0034] Fig. Figure 1 schematically shows a hydrogen production plant 10 with several electrolysis devices 11, which are set up to produce hydrogen H2 from water H2O using electric current. In the Fig. In the embodiment shown in Figure 1, the electrolysis devices 11 are arranged in three parallel cascades 12, each consisting of two electrolysis devices 11 connected in series. The number of parallel cascades 12 and the number of electrolysis devices 11 per cascade 12 are purely illustrative.

[0035] The electrolysis devices 11 are supplied with water via a water circuit 13 of the hydrogen production plant 10 for hydrogen production. Furthermore, the electrolysis devices 11 are connected to an electrical current source or voltage source (not shown) of the hydrogen production plant 10 to supply the electrolysis devices 11 with electrical power for hydrogen production.

[0036] The basic structure of an electrolysis device 11 is known. An electrolysis device 11 has several cell stack elements arranged to form a cell stack, the cell stack being arranged and pressed between end plates. The end plates of the electrolysis devices 11 are provided with at least one water supply connection, at least one water outlet connection, and at least one hydrogen connection. Water (H₂O) can be supplied to the respective electrolysis device 11 via the water supply connection, and water (H₂O) and oxygen (O₂) can be discharged from the respective electrolysis device 11 via the respective water outlet connection. As already described, the water is supplied via the water circuit 13, which includes a pump 14 for pumping the water and an oxygen separator 15 for separating the oxygen (O₂) from the water (H₂O).Hydrogen H2 produced during electrolysis by the respective electrolysis device 11 can be discharged via the at least one hydrogen connection, whereby the generated hydrogen H2 can be stored in a hydrogen tank 16 or discharged in another way.

[0037] As previously explained, each electrolysis device 11 has a cell stack consisting of cell stack elements, which are bipolar plates and membranes, in particular polymer exchange membranes. The cell stack elements of a cell stack of an electrolysis device 11 form several electrolysis cells. If a membrane of an electrolysis cell of an electrolysis device 11 fails, hydrogen can escape from the cathode-side hydrogen side of the respective electrolysis cell to the anode-side water side of the respective electrolysis cell. This hydrogen, along with the oxygen (O2) produced during electrolysis, can then be carried away from the respective electrolysis device 11 via the discharged water (H2O) and enter the oxygen separator 15. A mixture of hydrogen (H2) and oxygen (O2), which is explosive, can then accumulate in the oxygen separator 15.

[0038] Furthermore, as a result of a hydrogen leak at an electrolysis device 11, for example due to defective seals in the electrolysis device 11, the generated hydrogen H2 can enter a process room in which the electrolysis devices 11, in particular the entire hydrogen production plant 10, are located, whereby an explosive mixture of hydrogen H2 and oxygen O2 can then form in the process room.

[0039] To check whether there is a hydrogen leak at an electrolysis device 11 of a hydrogen production plant 10, the following steps of a safety routine are carried out for the respective electrolysis device 11 to be checked: The respective electrolysis device 11 to be tested is supplied with a defined electrical input quantity (excitation quantity).

[0040] The defined electrical input variable is preferably a defined electrical voltage signal applied to the electrolysis device 11 under test, specifically to its electrical current connections, via which the electrolysis device 11 under test is connected to an electrical voltage source or electrical current source of the hydrogen production plant 10. This electrical voltage signal can be a constant voltage with a defined voltage level, applied to the electrolysis device 11 under test for a defined period of time. However, the defined electrical voltage signal applied as an electrical input variable to the electrolysis device 11 under test can also change over time according to a predetermined pattern.The magnitude of the voltage signal applied as an electrical input to the electrolysis device 11 to be tested can correspond to the magnitude of an operating voltage or only to a defined percentage of the operating voltage.

[0041] The electrolysis device 11 to be tested is subjected to the defined electrical input quantity over a defined time interval of several minutes, in particular over a time interval of 5 to 60 minutes or 5 to 50 minutes or 5 to 40 minutes or 5 to 30 minutes or 5 to 20 minutes or 5 to 10 minutes or 10 to 60 minutes or 10 to 50 minutes or 10 to 40 minutes or 10 to 30 minutes or 10 to 20 minutes or 20 to 60 minutes or 20 to 50 minutes or 20 to 40 minutes or 20 to 30 minutes or 30 to 60 minutes or 30 to 50 minutes or 30 to 40 minutes or 40 to 60 minutes or 40 to 50 minutes or 50 to 60 minutes.

[0042] During the application of the defined electrical input quantity to the electrolysis device 11 under test and / or immediately after the defined electrical input quantity has been applied to the electrolysis device 11 under test and subsequently switched off, an actual electrical output quantity, dependent on the defined electrical input quantity, is recorded at the electrolysis device 11 under test.

[0043] This actual electrical output variable of the electrolysis device 11 to be checked can be recorded on the electrolysis device 11 as a whole or individually on the electrolysis cells of the respective electrolysis device 11 to be checked. Fig. Figure 1 shows that actual output variables recorded at the electrolysis devices 11 are fed via a signal line 17 to a control unit 18, which further processes the actual output variables.

[0044] The actual output value is preferably recorded over a time range of at least 30 seconds and up to 600 seconds.

[0045] Preferably, the electrical voltage signal passively generated at each electrolysis cell of the electrolysis device 11 to be tested is recorded as the actual electrical output variable for the respective electrolysis device 11 to be tested, which forms in response to the application of the defined electrical input variable to the electrolysis device 11.

[0046] The actual output variable or variables recorded at the respective electrolysis device 11 to be checked are compared with a respective target output variable, whereby this comparison of the respective actual output variable and the stored target output variable takes place in the control unit 18.

[0047] As already explained, a corresponding target output variable can be compared with a corresponding actual output variable for the electrolysis device 11 as a whole or for each of the electrolysis cells of the respective electrolysis device 11.

[0048] Depending on the comparison of the respective actual output variable with the respective target output variable, a conclusion is drawn about a hydrogen leakage or the probability of a hydrogen leakage at the respective electrolysis device 11.

[0049] The safety routine by which a hydrogen leakage or the probability of a hydrogen leakage at a respective electrolysis device 11 is carried out preferably automatically, namely in a time-controlled and / or event-controlled manner, via the control unit 18.

[0050] If the safety routine is performed on a timer, it is carried out after the hydrogen production plant 10 has reached a defined operating time or age, in particular after the respective electrolysis device 11 of the hydrogen production plant 10 has reached a defined operating time or age. This makes it possible to subject an electrolysis device 11 to a safety routine whenever it has reached a defined number of operating hours of hydrogen production.

[0051] If the safety routine is event-driven, it can be executed in this manner, in particular, when the hydrogen production plant 10 or the respective electrolysis device 11 to be checked is switched from hydrogen production operation to shutdown mode and / or when the hydrogen production plant 10 or the respective electrolysis device 11 is in shutdown mode. Alternatively or additionally, the safety routine is executed in an event-driven manner when the hydrogen production plant 10 or the respective electrolysis device 11 to be checked is switched from hydrogen production operation to standby mode and / or when the hydrogen production plant 10 or the respective electrolysis device 11 is in standby mode. Other events are possible.

[0052] Preferably, the safety routine at a respective electrolysis device 11 is then carried out in an event-controlled manner when an impermissibly high hydrogen concentration is detected based on a hydrogen sensor, for example in a process room accommodating the electrolysis devices 11 or in the area of ​​the oxygen separator 15. Thus, it shows Fig. 1. A hydrogen sensor 19 is installed in the area of ​​the oxygen separator 15 and detects the hydrogen concentration in the area of ​​the oxygen separator 15. If the detected hydrogen concentration is greater than a limit value, the control unit 18 can automatically trigger the execution of the safety routine.

[0053] Another event that triggers the automatic execution of the safety routine is the evaluation of a pressure sensor 20, which is located in a hydrogen discharge line 21 leading to the hydrogen tank 16 and measures the pressure in the hydrogen discharge line 21. If the control unit 18 determines, based on the measurement signal provided by the pressure sensor 20, that the pressure drop in the hydrogen discharge line 21 is greater than a limit value, the control unit 18 can again automatically trigger the execution of the safety routine.

[0054] As already explained above, to carry out the safety routine at a respective electrolysis device 11, a defined electrical input quantity, namely a defined electrical input voltage, is applied to it, wherein preferably for each electrolysis cell of the electrolysis device 11 a developing time course of an electrical voltage signal is measured as the actual output quantity.

[0055] The defined electrical input voltage is applied to the respective electrolysis device 11 via the electrical current source or electrical voltage source, which is connected to connection elements of the electrolysis device 11 for its electrical energy supply and from which the electrolysis device 11 is supplied with electrical power during hydrogen production operation.

[0056] The measurement of the actual output variable, namely the developing time course of the electrical voltage signal at preferably each of the electrolysis cells of the electrolysis device 11, is carried out via contact elements of the respective electrolysis cell, which are different from a connection element for the power supply.

[0057] Fig. Figure 2 shows the time course of actual output variables 22a, 22b, 22c and 22d measured at electrolysis cells of an electrolysis device 11, where these actual output variables are each the time course of an electrical voltage V measured at the respective electrolysis cell of the electrolysis device 11.

[0058] The actual output variables 22a, 22b, 22c and 22d of the Fig. 2 are formed in response to a constant electrical voltage applied to the electrolysis device 11 as an input variable, which is present at the respective electrolysis device 11 until time t1 and is switched off at time t1.

[0059] In Fig. 2. It is determined that for the electrolysis device 11 under test, the actual output variable 22c of an electrolysis cell of the electrolysis device 11 under test deviates by more than a limit value from the actual output variables 22a, 22b and 22d of the other electrolysis cells of the electrolysis device 11 under test.

[0060] According to this, in Fig. 2. Each of the actual output variables 22a, 22b, 22c and 22d of a respective electrolysis cell of the electrolysis device 11 to be tested is compared with the actual output variables 22a, 22b, 22c, 22d of the other electrolysis cells of the electrolysis device 11 to be tested, so that therefore in Fig. 2. The actual output values ​​of the other electrolysis cells of the electrolysis device 11 under test can be used as the target output value for one electrolysis cell, or alternatively, an average of the actual output values. In contrast, it is also possible to use a target output value based on historical operating data, i.e., one determined empirically, as the electrical target output value.

[0061] Preferably, the respective target output variable with which an actual output variable is compared is dependent on the age or the operating hours of the respective electrolysis device 11 or of the electrolysis cells thereof, if the target output variable has been empirically determined and is stored in the control unit 18.

[0062] Thus, in Fig. Figure 3 shows several different target output variables 23a, 23b, 23c and 23d, which depend on the age of the respective electrolysis device 11 or its electrolysis cells. Corresponding target output variables 23a, 23b, 23c and 23d can be stored in the control unit 18.

[0063] The individual electrolysis units 11 installed in the hydrogen production plant 10 may have different operating hours or different ages, particularly if, for example, one or more electrolysis units 11 have already been replaced due to a defect. In this case, the individual electrolysis units 11, and thus their electrolysis cells, have different ages and therefore different operating hours. The control unit 18 then stores target output variables 23a, 23b, 23c, and 23d that depend on the age or operating hours.

[0064] When the control unit 18 detects a hydrogen leak or a probability of a hydrogen leak exceeding a threshold value, it can automatically generate a signal that triggers maintenance or a replacement of the respective electrolysis unit 11. This signal can be stored in the control unit 18 and kept ready for retrieval. Alternatively or additionally, a shutdown event can be triggered, shutting down at least the affected electrolysis unit, or a cascade comprising this electrolysis unit, or a sub-branch comprising this electrolysis unit. This makes it possible to prevent a fault and (only) take the leakage-prone electrolysis unit out of service as a precaution, without shutting down the entire hydrogen production plant.In other words, continued partial operation is possible. This eliminates or at least reduces the risk of an emergency shutdown due to a malfunction.

[0065] That is, if the control unit 18 concludes that there is a hydrogen leak or a probability of a hydrogen leak that exceeds a limit value, the control unit 18 can also automatically shut down or shut down the hydrogen production plant 10, at least partially.

[0066] Alternatively or additionally, the control unit 18 can transmit a corresponding signal, for example via the internet, to a higher-level control unit. The recorded actual electrical output variables from the electrolysis device 11, together with their respective electrical input variables, can also be transmitted to a higher-level control unit to create a database and thus provide target output variables for the electrolysis devices 11 based on that database.

[0067] As already explained, the respective safety routine at the respective electrolysis device 11 is preferably executed in a time-controlled and / or event-controlled manner. Then, when the electrolysis devices 11, as described in Fig. As shown in Figure 1, the hydrogen production plant 10 is arranged in several cascades 12. The safety routine is preferably performed for one of the cascades 12 while the other cascades 12 are used for hydrogen production. This is advantageous in order to use the hydrogen production plant 10 continuously for hydrogen production and thus avoid restricting its availability through the safety routine.

[0068] Fig. Figure 4 illustrates for the hydrogen production plant 10 of the Fig. 1. Different operating scenarios occur over time t. During the time interval Δt1, all cascades 12 of the hydrogen production plant 10 are used for hydrogen production. At time t2, i.e., at the end of the time interval Δt1, the control unit 18 automatically triggers a hydrogen leakage test for the hydrogen production plant 10, for example, event-driven, by the control unit 18 determining, for example, at time t2, that the hydrogen concentration measured by the hydrogen sensor 19 is greater than a limit value. Then, in Fig. 4 during the time interval Δt2, initially for one of the cascades, for example for the one in Fig. 1 upper cascade shown, on the electrolysis devices 11 of this cascade 12 a respective safety routine is executed, whereby the other cascades, i.e. in Fig. 1. The middle cascade 12 and the lower cascade 12 continue to be used for hydrogen production. If it is determined that a hydrogen leak exists at one of the electrolysis devices 11 of the cascade 12 checked during the time interval Δt2, or that the probability of a hydrogen leak is greater than a limit value, then the cascade 12 can be shut down or taken out of service for hydrogen production. Conversely, if it is determined that no hydrogen leak exists at any of the electrolysis devices 11 of the cascade 12 checked, or that the probabilities are less than a limit value, then the respective cascade 12 can be put back into operation for hydrogen production after the time interval Δt2.

[0069] Following the time interval Δt2, the safety routines for the electrolysis devices 11 of another cascade are carried out during the time interval Δt3, for example in Fig.1 for the middle cascade 12, while the lower cascade 11 continues to operate, i.e., continues to be used for hydrogen production, and depending on whether a hydrogen leak was detected at an electrolysis device 11 of the upper cascade 12 or not, the upper cascade 12 is either restarted for hydrogen production or remains shut down. Similarly, during the time interval Δt4, the lower cascade 12 consisting of electrolysis devices 11 can be subjected to the safety routine, whereby during the time interval Δt5, i.e., after completion of the execution of the safety routines in all cascades 12, depending on the result of the safety routines, either all electrolysis devices 11 of all cascades 12 are used for hydrogen production or only a subset of the cascades 12.

[0070] Depending on the result of the safety routine, the hydrogen production plant 10 can be shut down completely or only partially. Furthermore, it is possible to initiate the replacement or maintenance of individual electrolysis units 11, and in particular preventively, before a fault occurs. In the event of a fault—especially an H2-in-O2 fault—it is also possible to identify the cascade with at least one faulty electrolysis unit 11 by temporarily partially shutting down individual cascades, and then to perform the safety routine only in the cascade with the faulty electrolysis unit 11.The cascade with the faulty electrolysis device 11 can be identified, in particular, by the fact that the hydrogen concentration in the O2 separator 15, measured via the H2 sensor 19, is higher when the cascade with the faulty electrolysis device 11 is operating, and lower when the cascade with the faulty electrolysis device 11 is temporarily out of service. Only after identifying the faulty cascade can the aforementioned safety routine then be performed for all electrolysis devices 11 in the faulty cascade to identify the faulty electrolysis device 11. Therefore, the safety routine does not need to be performed for all electrolysis devices 11 in all cascades.This means, particularly in large plants with several dozen, a shock (5 dozen), a gros (12 dozen) or more electrolysis devices, significant time savings in identifying the faulty electrolysis device.

[0071] The invention further relates to the hydrogen production plant 10 with the multiple electrolysis devices 11 and the control unit 18, which is configured to automatically execute the method according to the invention. For this purpose, the control unit 18 has data interfaces for exchanging data with the components involved in carrying out the method according to the invention, for example, with the electrolysis cells of the electrolysis devices 11, where the actual output variables are recorded, and with the hydrogen sensor 19. Furthermore, the control unit 18 has a processor for data processing and a memory for data storage. In addition, program modules are implemented in the control unit 18 that serve to carry out the method according to the invention. Reference symbol list 10 hydrogen production plants 11 Electrolysis device 12 Cascade 13 Water cycle 14 Pump 15 Water treatment plant 16 hydrogen storage 17 Signal line 18 Control unit 19 Hydrogen sensor 20 pressure sensor 21 Hydrogen discharge line 22a Actual output value 22b Actual output value 22c Actual output value 22d Actual output value 23a Target output value 23b Target output value 23c Target output value 23d Target output value QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] DE 10 2017 108 413 A1

[0002] EP 3 336 947 B1

[0010]

Claims

[1] Method for operating a hydrogen production plant (10) with several electrolysis devices (11) designed to produce hydrogen from water using electric current, wherein the electrolysis devices (11) preferably form at least one cascade (12) of electrolysis devices (11) connected in series, preferably several cascades (12) connected in parallel of electrolysis devices (11) connected in series, wherein, for hydrogen production, the electrolysis devices (11) are supplied with water via a water circuit (13) of the hydrogen production plant (10), wherein, for hydrogen production, the electrolysis devices (11) are supplied with electrical power from an electrical current source or electrical voltage source, characterized by , that To check whether there is a hydrogen leak at at least one electrolysis device (11), the following steps of a safety routine are performed: The respective electrolysis device (11) to be tested is supplied with a defined electrical input quantity, At the respective electrolysis device (11) to be checked, an actual electrical output variable is recorded which depends on the defined electrical input variable, The actual output variable recorded at the respective electrolysis device (11) to be checked is compared with at least one respective target output variable. Depending on the comparison of the actual output variable with the at least one target output variable, a conclusion is drawn about a hydrogen leakage or the probability of a hydrogen leakage at the respective electrolysis device (11). [2] Method according to claim 1, characterized bythat the security routine is performed automatically, on a time-controlled and / or event-controlled basis. [3] Method according to claim 2, characterized by , that the safety routine is carried out in a time-controlled manner after reaching a defined operating time of the hydrogen production plant (10) or the respective electrolysis device (11) to be checked. [4] Method according to claim 2 or 3, characterized by , that the safety routine is then carried out in an event-driven manner when the hydrogen production plant (10) or the respective electrolysis device to be checked (11) is switched from a production operation to a shutdown operation or is in a shutdown operation. [5] Method according to any one of claims 2 to 4, characterized by, that the safety routine is then carried out in an event-driven manner when the hydrogen production plant (10) or the respective electrolysis device to be checked (11) is switched from a production operation to a stand-by operation or is in stand-by operation. [6] Method according to any one of claims 2 to 5, characterized by , that the safety routine is then carried out in an event-driven manner when a hydrogen sensor (19) in the area of ​​an oxygen separator (15) of the water circuit (13) of the hydrogen production plant (10) and / or in a process room accommodating the electrolysis devices (11) detects a hydrogen concentration that is greater than a limit value. [7] Method according to any one of claims 2 to 6, characterized by, that the safety routine is then carried out in an event-driven manner when a pressure sensor (20) in a hydrogen discharge line (21) of the hydrogen production plant (10) detects a pressure drop that is greater than a limit value. [8] Method according to any one of claims 1 to 7, characterized by , that at least one target output variable for the respective electrolysis device (11) to be checked depends on the age or operating hours of the electrolysis device (11) to be checked. [9] Method according to any one of claims 1 to 8, characterized by , that the safety routine is carried out for one of the cascades (12) of the hydrogen production plant (10), while the other cascades (12) of the hydrogen production plant (10) are used for hydrogen production. [10] Method according to any one of claims 1 to 9, characterized by, that if a hydrogen leak or the probability of a hydrogen leak at the respective electrolysis device (11) is detected that exceeds a limit value, a signal is generated which triggers a shutdown event, maintenance and / or replacement of the respective electrolysis device (11). [11] Method according to any one of claims 1 to 10, characterized by , that a defined electrical voltage signal is used as the defined electrical input quantity with which the respective electrolysis device (11) to be tested is supplied. [12] Method according to any one of claims 1 to 11, characterized by , that the electrical actual output variable, which is recorded at the respective electrolysis device (11) to be checked, is a time course of an electrical voltage signal. [13] Method according to any one of claims 1 to 12, characterized by, that the electrical target output variable of the respective electrolysis device (11) to be tested is an empirically determined target output variable based on historical operating data. [14] Method according to any one of claims 1 to 12, characterized by , that the electrical target output variable of the respective electrolysis device (11) to be checked is a target output variable which depends on the actual output variables of the other electrolysis devices (11). [15] Method according to any one of claims 1 to 14, characterized by, that preferably at each electrolysis cell of the respective electrolysis device (11) to be checked an individual electrical actual output variable is recorded, which is compared with a target output variable in order to conclude, depending on the comparison of the respective actual output variable with the respective target output variable, about a hydrogen leakage or the probability of a hydrogen leakage at the respective electrolysis cell of the respective electrolysis device (11) to be checked of the hydrogen production plant (10). [16] Method according to claim 15, characterized by , that the actual output variables of the other electrolysis cells or a variable dependent on the actual output variables of the other electrolysis cell of this electrolysis device (11) to be tested is used as the electrical target output variable of an electrolysis cell of the respective electrolysis device (11). [17] Method according to any one of claims 1 to 16, characterized by, that the respective actual electrical output variable is recorded during the application of the defined electrical input variable to the electrolysis device (11) to be tested and / or immediately after the application of the defined electrical input variable to the electrolysis device (11) to be tested and the subsequent switching off of the same. [18] Hydrogen production plant (10) with several electrolysis devices (11) which are set up to produce hydrogen from water using electric current, wherein the electrolysis devices (11) preferably form at least one cascade (12) of electrolysis devices (11) connected in series, preferably several cascades connected in parallel of electrolysis devices (11) each connected in series, and with a control unit which is set up to carry out the method according to one of claims 1 to 17.

Images