Use of pressure sensors in fuel cell systems as a replacement for a hydrogen process sensor

Abstract

Method for determining at least one amount of a reactant in a first volume (V1) of a fuel cell system (10) with at least one fuel cell (21), wherein a material stream containing the amount of reactant and an inert amount of at least one inert gas flows from the first volume (V1) into the fuel cell (21) and then into a second volume (V2), wherein a recirculation element (16) conveys at least part of the material flow from the second volume (V2) into the first volume (V1), and wherein a first pressure is measured in the first volume (V1) and a second pressure is measured in the second volume (V2), characterized in that the following steps are carried out, a. Setting up a state system of at least a part of the fuel system (10), wherein in the state system: ▪ a change in the amount of reactant in the first volume (V1), ▪ a change in the amount of reactant in the second volume (V2), ▪ at least one change in the amount of inert material in the first volume (V1), ▪ at least one change in the amount of inert material in the second volume (V2) is taken into account, b. Estimation of the amount of substance and inert substance in the first volume (V1) and the second volume (V2) based on the state system, c. Calculation of an estimated first pressure and an estimated second pressure, whereby a relationship between the amount of substance and the amount of inert material in the first volume (V1) and the first pressure and a relationship between the amount of substance and the amount of inert material in the second volume (V2) and the second pressure is assumed according to a gas law, d. Feedback of the estimated first pressure and the estimated second pressure with the measured first pressure and with the measured second pressure, e. and thereby correcting and determining the amount of substance and inert substance in the state system in the first volume (V1) and in the second volume (V2).

Description

[0001] The invention relates to a method for determining at least one molar amount of a reactant in a first volume of a fuel cell system having at least one fuel cell according to the preamble of claim 1, wherein a material stream containing the molar amount of the reactant and an inert amount of at least one inert gas flows from the first volume into the fuel cell and then into a second volume, wherein a recirculation element conveys at least a portion of the material stream from the second volume into the first volume, and wherein a first pressure is measured in the first volume and a second pressure is measured in the second volume. Furthermore, the invention relates to a fuel cell system having at least one fuel cell for the electrochemical conversion of at least one reactant, according to the preamble of claim 14. State of the art

[0002] To operate a fuel cell system, reactants are fed to both the anode and the cathode of the fuel cell system. These are usually gas mixtures which, in addition to the reactants, also contain other gases which do not serve as starting materials for electrochemical conversion in the fuel cell system and are referred to below as inert gases. For example, air is fed to the cathode, which contains inert gases such as nitrogen as well as oxygen as a reactant. An anode stream containing a fuel, for example hydrogen, is fed to the anode. The anode stream can also contain, for example, carbon dioxide from a reforming process or water vapor from a humidifier. Even when using almost pure hydrogen, recirculation, i.e.When the anode stream is returned to the fuel cell after leaving the fuel cell, impurities in the form of inert gases are introduced. These impurities can be introduced into the anode stream through the diffusion of nitrogen or water through a fuel cell membrane that connects the cathode and anode, or through leaks. These introduced impurities, as well as the impurities in the hydrogen used, accumulate through recirculation, so that when the anode stream is fed to the fuel cell at a constant pressure, the partial pressure of the hydrogen drops. If the partial pressure of hydrogen is too low, the performance of the fuel cell system decreases. To maintain the operation of the fuel cell, an outlet valve is opened to release the anode stream, which has an insufficient hydrogen content, to the environment.

[0003] In order to avoid unnecessary release of hydrogen into the atmosphere due to premature opening of the outlet valve, the hydrogen content in a volume can be measured using a hydrogen process sensor. The hydrogen process sensor is preferably located in the first volume upstream of the fuel cell. However, hydrogen process sensors are only produced in small quantities and are therefore cost-intensive. They also take up a lot of space. A further disadvantage of using a hydrogen process sensor is the measurement inaccuracies of the hydrogen process sensor, which measures non-linearly and inaccurately, i.e., not repeatably. Another disadvantage is that a hydrogen process sensor measures substance-specifically and can therefore only measure the proportion of hydrogen, but not other reactants or inert gases.

[0004] DE 10 2004 063 533 A1 discloses a method for determining the composition of a volume flow in a fuel cell system, in which a pressure, a pressure difference, and a temperature in the fuel cell system are measured. The measured pressure difference is compared with a theoretical pressure difference, wherein the theoretical pressure difference is determined based on characteristic maps of a conveying device that conveys the volume flow and based on an assumed composition of the volume flow. The assumed composition of the volume flow is considered to be the actual composition in the fuel cell system if the difference between the measured and theoretical pressure difference falls below a limit value. A disadvantage of this method is that characteristic maps of the conveying device for different substances must be known, which must preferably be determined experimentally.Determining different inert gases is hardly possible due to similar characteristic curves.

[0005] JP 2007-012532 A describes a method for estimating impurities in gas mixtures. Disclosure of the invention

[0006] The object of the invention is to determine the amount of a reactant and the amount of inert material in the first volume in order to calculate the partial pressure or the molar fraction of a reactant in a material flow through the fuel cell, in particular the molar fraction of a fuel in an anode flow. The determination should be cost-effective, space-saving, and precise, avoiding the disadvantages of a hydrogen process sensor. Furthermore, a method should be found that is independent of the reactant and characteristic maps used and that enables the determination of additional proportions of inert gases.

[0007] To achieve this object, a method for determining at least one amount of a reactant in a first volume in a fuel cell system having at least one fuel cell with the features of claim 1, in particular the characterizing part, is proposed. Advantageous developments of the invention are specified in the dependent method claims. Furthermore, this object is achieved by a fuel cell system having at least one fuel cell with the features of claim 14, in particular the characterizing part. Features and details described in connection with the method according to the invention naturally also apply in connection with the fuel cell system according to the invention, and vice versa. The features mentioned in the claims and in the description can each be essential to the invention individually or in any combination.

[0008] The invention provides that the following steps are carried out in the method according to the invention: a. Establishing a state system of at least a part of the fuel system, wherein in the state system: ▪ a change in the amount of reactant in the first volume, ▪ a change in the amount of reactant in the second volume, ▪ at least one change in the amount of inert material in the first volume, ▪ at least one change in the amount of inert material in the second volume is taken into account, b. Estimation of the amount of substance and inert material in the first and second volumes based on the state system, c. Calculation of an estimated first pressure and an estimated second pressure, assuming a relationship between the amount of substance and the amount of inert material in the first volume and the first pressure and a relationship between the amount of substance and the amount of inert material in the second volume and the second pressure according to a gas law, d. Feedback of the estimated first pressure and the estimated second pressure with the measured first pressure and with the measured second pressure, e. and thereby correcting and determining the amount of substance and inert substance in the first and second volumes in the state system.

[0009] The advantage here is that only two pressures are required as measured variables, eliminating the need for a costly, bulky, and imprecise hydrogen process sensor. The process can also be used for any reactant and is not limited to hydrogen. Furthermore, no complex preparatory work, such as creating a characteristic map, which must also be created anew for each delivery device and each component, is required. The molar fraction of the reactant in the first volume corresponds to the molar fraction of the reactant flowing through the fuel cell and is thus responsible for the sufficient supply of the reactant to the fuel cell, since the first volume is located directly upstream of the fuel cell.The molar fraction of the reactant in the first volume is x1 / (x1+x2), where the molar fraction x1 of the reactant in the first volume and the molar fraction x2 of the inert substance in the first volume are determined in the process. The partial pressure of the reactant, p, can be determined from the molar fraction of the reactant in the first volume. v1, Reaktand in the first volume can be calculated using equation (1), since the first pressure p v1 is measured in the procedure. pV1,reactant=x1x1+x2pV1

[0010] In the simplest case, when determining the gas composition in the volumes under consideration, the method can only distinguish between the reactant and a quantity of inert substance, where the quantity of inert substance includes all inert gases, i.e., all gases except the reactant. For example, the method can determine the composition of the gases present in the first volume, distinguishing between hydrogen on the one hand and the total of nitrogen, water, and carbon dioxide on the other. For this purpose, in step a. of the method, the change in the quantity of inert substance of all inert gases is balanced as the change in the quantity of inert substance.

[0011] The method also makes it possible to differentiate between the individual inert gases. For this purpose, it can be provided that the amount of inert material comprises a first inert gas and that, in particular, in step a. at least one change in a second inert gas in the first volume and at least one change in a second inert gas in the second volume are additionally balanced. This can mean, for example, that in step a. only the amount of inert material of one inert gas, for example nitrogen, is balanced and the other inert gases are neglected. However, it can also be provided that the amounts of inert material of different inert gases are considered individually and that, for example, a distinction is made between the amount of inert material of nitrogen and the amount of inert material of water.In this case, in addition to the first to fourth equations for the change of hydrogen and nitrogen in the first and second volumes, a fifth equation for the change of the inert amount of water in the first volume and a sixth equation for the change of the inert amount of water in the second volume are added to the state system.

[0012] In order to obtain a state system that is easy to calculate in a further embodiment, it is planned that in step a. the balancing is carried out under simplified boundary conditions. For example, it can be assumed that ▪ that the outlet valve is closed, ▪ that a first and / or second temperature in the first and / or second volume is constant ▪ that the change in the amount of reactant and the change in the amount of inert substance due to diffusion through the fuel cell membrane is small and slow and therefore these changes cannot be balanced but treated as disturbances ▪ that due to a water separator in the recirculation element, the amount of inert water passing from the recirculation element into the first volume is negligible.

[0013] Advantageously, in step a., a state system can be established in which the changes in the respective amount of substance and / or inert material depend on at least one measured variable and at least one amount of substance and / or inert material. The measured variable can be the first or second pressure, the first or second temperature, but also the speed of a conveying device in the recirculation element that conveys the material flow from the second to the first volume, a control voltage of an inlet valve through which the reactant enters the first volume, or the current generated by the fuel cell, etc.

[0014] It has proven advantageous if the state system has been set up before the start of operation of the fuel cell, during which the amount of hydrogen is to be determined, so that steps b. to e. of the method are carried out repeatedly, in particular continuously, during operation of the fuel cell. This means that the amounts of substance and inert material in the first and second volumes are repeatedly estimated one after the other using the state system, and the estimated first and second pressures are calculated using the estimated amounts of substance and inert material. A gas law such as the ideal gas law or a real gas law such as the van der Waals gas law can be used for this. The estimated pressure can be compared with the respective measured pressure and conclusions can be drawn from this as to the extent to which the values ​​under b.Whether the estimate made was correct or needs to be corrected, whereby steps b. to e. can be performed multiple times for a measurement point. As a rule, subsequent measured values ​​are also taken into account for the next estimate.

[0015] To estimate the amounts of substance and the amounts of inert material according to step b, the system of states can be solved under static boundary conditions, i.e., the changes in the amounts of substance and the amounts of inert material are set to zero. When considering several inert gases individually, it is important to ensure that sufficient measured values ​​or boundary conditions are known to solve the system of equations. Another measurable quantity that can be considered, for example, is the change in the first and second pressures over time.

[0016] Alternatively, the amounts of substance and the amounts of inert material can be estimated according to step b. using a control-engineering observer, in particular a Luenberger observer, or a Kalman filter. In this case, the amounts of substance and the amounts of inert material are assumed at the beginning and inserted into the state system. For example, it is assumed that pure hydrogen is present in the first volume at the beginning of fuel cell operation, so that the molar fraction of hydrogen in the first volume can be set to one and the molar fraction of inert material to zero.

[0017] The procedures for performing steps d. and e. also differ when solving the state system under static conditions and when using a Luenberger observer or Kalman filter. For example, a Kalman filter can be used to filter out measurement noise.

[0018] The method according to the invention can be improved by additionally measuring a first temperature in the first volume and a second temperature in the second volume for the estimation according to step b. and / or the calculation according to step c. This makes it possible to generate estimated values ​​that exhibit a smaller error compared to the actual amounts of substance and inert material present in the fuel cell system.

[0019] It is also advantageous to determine additional measured variables as accurately as possible for the estimation according to step b. This includes the amount of reactant flowing through the inlet valve into the first volume to replace the reactant consumed by the electrochemical reaction. For this purpose, it is useful to additionally measure a third pressure in a third volume located upstream of the first volume by the inlet valve. This also includes measuring the speed of the conveyor or the moisture that, for example, a water separator removes from the fuel cell system.

[0020] According to the invention, it can further be provided that a plurality of operating points of the fuel cell system are included to implement the method, since a plurality of different measured values ​​improves the comparison and thus the correction options for the estimated values. Different operating points can involve both the operation of the fuel cell system at different temperatures and the operation of the conveyor system at different speeds.

[0021] The determination of the amount of reactant in the first volume can be used to open an outlet valve through which the material flow escapes from the fuel cell system when the amount of reactant in the first volume falls below a certain limit.

[0022] Measuring a second pressure in a second volume and determining the individual material flows can also be used to detect a leak through which the material flow is escaping from the fuel cell system. This can be identified by an unusually high pressure drop between the first and second volumes. Another indication is that the amount of material flowing through the inlet valve into the first volume is significantly greater than the amount of material consumed by the electrochemical reaction.

[0023] The object of the invention is further achieved by a fuel cell system having at least one fuel cell for the electrochemical conversion of at least one reactant, wherein a first pressure sensor is arranged in the first volume to measure a first pressure, wherein a second pressure sensor is arranged in the second volume to measure a second pressure, and wherein the fuel cell system has a monitoring unit to carry out the method according to the invention.

[0024] The fuel cell system can contain more than one fuel cell, in particular a fuel cell stack, and can be used with various fuel cell types as well as reactants or gas compositions. In particular, the process can be applied to reformer gas in addition to pure hydrogen. Examples of implementation

[0025] Further measures improving the invention will become apparent from the following description of an exemplary embodiment of the invention, which is schematically illustrated in the figure. All features and / or advantages arising from the claims, the description, or the drawing, including structural details, spatial arrangement, and method steps, may be essential to the invention both individually and in various combinations. It shows: Fig. 1 a fuel cell system according to the invention with a first and a second pressure sensor

[0026] In Fig.1 shows a fuel cell system 10 with a fuel cell stack 20 with two exemplary fuel cells 21, each fuel cell 21 having an anode compartment 22 with an anode and a cathode compartment 23 with a cathode. The anode compartment 22 and the cathode compartment 23 are separated from each other by a membrane 24. In the fuel cells 21, an electrochemical reaction takes place at the anode and the cathode, wherein, for example, oxygen and hydrogen react to form water. For this purpose, air containing, for example, oxygen and nitrogen is fed as a cathode current (not shown) into the cathode compartment 23, and a hydrogen-containing anode current ṅ FC into the anode chamber 22 of each fuel cell 21 according to the arrows 42. To determine the anode current ṅ FCTo produce hydrogen, hydrogen flows from a hydrogen tank according to arrow 40 through a shut-off valve 12, through a third volume V3 and through an inlet valve 13 into a first volume V1, which is located upstream of the fuel cells 21 in the flow direction. The hydrogen flowing from the inlet valve 13 into the first volume V1 is referred to below as the inlet flow ṅ ̇̇E In the first volume V1, the inlet flow ṅ E with the recirculation flow ṅ R which flows according to arrow 41 through a recirculation element 16 into the first volume V1. The recirculation flow ṅ R arises from the fact that the anode current ṅFC, which has passed through the fuel cells 21, is conveyed back to the fuel cells 21 by a conveying device 11 in order to make the hydrogen not reacted in the fuel cells 21 available to the fuel cells 21 again. The recirculation current ṅ Rcontains unused hydrogen and inert gases. The anode current ṅ FC leaves the first volume V, according to the arrow 42, whereby the anode current ṅ FC the gas composition of the first volume V1. This allows the change ẋ1 in the amount of substance x of the hydrogen in the first volume V1 and the change ẋ2 in the amount of substance x2 of all inert gases in the first volume V1 to be balanced according to equations (2) and (3), where x3 is the amount of hydrogen in the second volume V2 and x4 is the amount of substance of all inert gases in the second volume V2. Contrary to common convention, “x” will be used below to refer to an amount of substance itself, i.e. the number of molecules considered, and not to a molar fraction. In the context of this invention, “ṅ” refers to a material flow which can have both an amount of substance and an amount of inert substance. x˙1=−x1x1+x2n˙FC+x3x3+x4n˙R+n˙E x˙2=−x2x1+x2n˙FC+x4x3+x4n˙R

[0027] In the fuel cells 21, a quantity of hydrogen ṅ i consumed by the electrochemical reaction. Further changes in the composition of the anode current result from the fact that gases can diffuse through the membrane 24 between the anode compartment 22 and the cathode compartment 23 of each fuel cell 21. For example, nitrogen and / or reaction water can diffuse from the cathode compartment 23 into the anode compartment 22 according to arrows 43, 44. Hydrogen can pass from the anode compartment 22 into the cathode compartment 23 according to arrow 45. The anode current ṅ FC flows from the fuel cells 21 according to the arrow 46 into a second volume V2. Usually, not all of the hydrogen is consumed by the electrochemical reaction in the fuel cell, so that hydrogen with the anode current ṅ FCflows into the second volume V2. Should the amount of hydrogen in the second volume V2 fall below a limit, an outlet valve 14 can be opened briefly and a stream of hydrogen can escape into the environment 18. In order to ensure that as little hydrogen as possible goes unused and to increase the efficiency of the fuel cell system 10, the outlet valve 14 will remain closed for a large part of the operating time of the fuel cell system 10, and the gas mixture in the volume V2 will be recirculated with the aid of a recirculation element 16 according to arrow 47 and returned to the volume V1, from where the gases return to the fuel cell stack 20.

[0028] Thus, the change ẋ3 of the amount of substance x3 of hydrogen in the second volume V2 and the change ẋ4 of the amount of substance x4 of all inert gases in the second volume V2 can be balanced according to equations (4) and (5), whereby the outlet valve 14 is assumed to be closed and the changes in the amount of substance x3 and the amount of inert substance x4 due to diffusion through the membrane 24 are not taken into account: x˙3=−x1x1+x2n˙FC−x3x3+x4n˙R−n˙i x˙4=−x2x1+x2n˙FC−x4x3+x4n˙R

[0029] The mass flows ṅ E , ṅ R , ṅ FC and ṅ i all depend on quantities known or measurable in the system according to equations (6) to (9), where p v1 a first pressure, p, measured according to the invention by a first pressure sensor 30 in the first volume V1 v2 is a second pressure measured according to the invention by a second pressure sensor 31 in the second volume V2. T V1is either the estimated or measured first temperature in the first volume V1. T V2 is either the estimated or measured second temperature in the second volume V2. If the first temperature T V1 and the second temperature T V2 If the measured temperatures are to be measured, the first and second pressure sensors 30, 31 can be designed as combined pressure and temperature sensors. Furthermore, u Ein a control voltage applied to the inlet valve 13, K ein a valve, material, temperature and pre-pressure dependent constant of the inlet valve 13, n d a speed of the conveyor device 11, V h a displacement volume of the conveyor device 11, ε is a relative clearance of the conveyor device 11, κ is an adiabatic coefficient, a0, a1, f specific parameters of the fuel cell stack 20, R is the general gas constant, i is the current of the fuel cell stack 20, N is the number of fuel cells 21 in the fuel cell stack 20, F is the Faraday constant n˙E=KEin⋅uEin n˙R=nd⋅Vh60(1−ε((pV1pV2)1κ_−1))pV2TV2⋅R n˙FC=(pV1−pV2−a0a1+TV1⋅R⋅fpV1⋅i⋅N2⋅F)pV1TV1⋅R⋅x1x1+x2 n˙i=i⋅N2⋅F

[0030] Thus, the quantities to be determined in equations (2) to (5) are the amounts of substance x2, x3 and inert substance amounts x2, x4, so that a non-linear state system according to equation (10) with ẋ and x according to equations (11) and (12) can be set up. x˙=f(x) x˙=(x˙1,x˙2,x˙3,x˙4)T x˙=(x1,x2,x3,x4)T

[0031] In addition to this state system according to equation (10), output equations are created according to equation (13), in which the measured quantities y=(p V1 p V2)" with the amounts of substance x1, x3 and inert amounts x2, x4 are linked to each other via the ideal gas law according to equations (14),(15). y=g(x) pV1=TV1⋅RV2(x1+x2) pV2=TV2⋅RV2(x3+x4)

[0032] A control-engineered observer can be used to determine the amounts of substance x3, x3 and the amounts of inert substance x2, x4 based on the state system and the initial equations. To enable feedback, a Luenberger observer is used in this example.

[0033] Plausible initial boundary conditions are assumed, e.g., that at the beginning of operation of the fuel cell stack 20, pure hydrogen is present in the first volume V1, so that initially x1 / (x1+x2) =1 and x2 / (x1+x2) =0 can be set. With these and possibly further assumptions, the state system is calculated according to equation (10), so that with the assumptions made, estimated amounts of substance x̂3, x̂3 and inert material quantities x̂2, x̂4 are obtained. With the estimated amounts of substance x̂1, x̂3 and inert material quantities x̂2, x̂4, the first and second estimated pressures p̂ V1 , p V2 calculated according to equation (14) and equation (15). Based on the difference Δp V1 of the first measured pressure p V1 with the first estimated pressure p̂ V1 , or based on the difference Δp̂ V2 of the second measured pressure p̂ V2 with the second estimated pressure p̂ V2The system of states can now be corrected and the amounts of substance x1, x3 and inert substance x2, x4 can be determined, for which the differences Δp V1 , Δp V2 are reduced, so that the state system approximates reality. The characteristics of a water separator 17 can be taken into account as further boundary conditions. It is also possible to use a third pressure p V3 in the third volume V3 using a third pressure sensor 32 and thus to determine equation (6) more precisely.

Claims

[1] Method for determining at least one amount of a reactant in a first volume (V1) of a fuel cell system (10) with at least one fuel cell (21), wherein a material stream containing the amount of reactant and an inert amount of at least one inert gas flows from the first volume (V1) into the fuel cell (21) and then into a second volume (V2), wherein a recirculation element (16) conveys at least part of the material flow from the second volume (V2) into the first volume (V1), and wherein a first pressure is measured in the first volume (V1) and a second pressure is measured in the second volume (V2), characterized by that the following steps are carried out, a. Setting up a state system of at least a part of the fuel system (10), wherein in the state system: ▪ a change in the amount of reactant in the first volume (V1), ▪ a change in the amount of reactant in the second volume (V2), ▪ at least one change in the amount of inert material in the first volume (V1), ▪ at least one change in the amount of inert material in the second volume (V2) is taken into account, b. Estimation of the amount of substance and inert substance in the first volume (V1) and the second volume (V2) based on the state system, c. Calculation of an estimated first pressure and an estimated second pressure, whereby a relationship between the amount of substance and the amount of inert material in the first volume (V1) and the first pressure and a relationship between the amount of substance and the amount of inert material in the second volume (V2) and the second pressure is assumed according to a gas law, d. Feedback of the estimated first pressure and the estimated second pressure with the measured first pressure and with the measured second pressure, e. and thereby correcting and determining the amount of substance and inert substance in the state system in the first volume (V1) and in the second volume (V2). [2] Method according to claim 1, characterized by that the amount of inert material includes all inert gases. [3] Method according to claim 1, characterized by that the amount of inert substance comprises a first inert gas and that in particular additionally in step a. • at least one change of a second inert gas in the first volume and ▪ at least one change of a second inert gas in the second volume must be accounted for. [4] Method according to one of the preceding claims, characterized by that in step a. the accounting is carried out under simplified boundary conditions. [5] Method according to one of the preceding claims, characterized bythat in step a a state system is established in which the changes in the respective amount of substance and / or amount of inert substance depend on at least one measured variable and at least one amount of substance and / or amount of inert substance. [6] Method according to one of the preceding claims, characterized by that steps b. to e. are carried out continuously. [7] Method according to one of the preceding claims, characterized by that for the estimation according to step b. and / or calculation according to step c., a first temperature is additionally measured in the first volume and a second temperature is measured in the second volume. [8] Method according to one of the preceding claims, characterized by that for the estimation according to step b. a third pressure is additionally measured in a third volume (V3) which is located in front of the first volume (V1). [9] Method according to one of the preceding claims, characterized in that a plurality of operating points of the fuel cell system are included to carry out the method. [10] Method according to one of the preceding claims, characterized by that the amounts of substances and inert gas quantities are determined based on the state system under static conditions with subsequent result filtering. [11] Method according to one of the preceding claims, characterized by that the amounts of substance and inert gas are determined on the basis of the state system by a control observer, in particular a Luenberger observer, or a Kalman filter. [12] Method according to one of the preceding claims, characterized by that when the amount of reactant in the first volume (V1) falls below a limit, an outlet valve (14) is opened through which the material flow escapes from the fuel cell system (10). [13] Method according to one of the preceding claims, characterized by that a leak through which the material flow escapes from the fuel cell system (10) is detected by measuring the second pressure. [14] Fuel cell system (10) with at least one fuel cell (21) for the electrochemical conversion of at least one reactant, wherein in the fuel cell system (10) a first volume (V1) is arranged fluidically upstream of the fuel cell (21) and a second volume (V2) is arranged fluidically downstream of the fuel cell (21), wherein a material flow comprising a reactant and at least one inert gas can be guided through the first volume (V1), the fuel cell (21) and the second volume (V2), wherein a recirculation element (16) connects the first volume (V1) to the second volume (V2) in order to convey at least a part of the material flow from the second volume (V2) to the first volume (V1), wherein a first pressure sensor (30) is arranged in the first volume (V1) in order to measure a first pressure, wherein a second pressure sensor (31) is arranged in the second volume (V2) to measure a second pressure, characterized by , that the fuel cell system (10) has a monitoring unit (15) to carry out the method according to one of claims 1 to 13.

Images