Method, control unit and exhaust air installation having control unit for determining the hydrogen concentration

Abstract

The present invention relates to a method for determining an actual hydrogen concentration c of an exhaust air flow of a fuel cell, comprising the following steps: - detecting a first hydrogen concentration signal S1, corresponding to an upstream hydrogen concentration, of the exhaust air flow; - detecting a second hydrogen concentration signal S2, corresponding to a downstream hydrogen concentration, of the exhaust air flow; - determining a hydrogen concentration function c(x) representing the actual hydrogen concentration, where x defines a position in the direction of flow of the exhaust air flow; and - calculating the actual hydrogen concentration on the basis of the hydrogen concentration function c(x); characterised in that the detected hydrogen concentration signals S1, S2 are transferred as two separate input variables into the hydrogen concentration function c(x).

Description

[0001] Method, control unit and exhaust system with control unit for determining the hydrogen concentration

[0002] The present invention relates to a method for determining the actual hydrogen concentration of an exhaust air stream from a fuel cell, particularly for a fuel cell electric vehicle (FCEV). Furthermore, the present invention relates to a control device for carrying out such a method, as well as an exhaust air system for a fuel cell, comprising such a control unit for determining the actual hydrogen concentration.

[0003] The emission values ​​of fuel cell vehicles are subject to legal regulations regarding the maximum hydrogen concentration in the exhaust air stream. The maximum hydrogen concentration at the end of the exhaust system before release into the environment must not exceed 8% by volume and must not average over 3 seconds to more than 4% by volume.

[0004] During operation of the fuel cell, drain and purge pulses are generated by the removal of water produced in the cathode via the drain valve and by the purging of the hydrogen-carrying anode. These drain and purge pulses cause brief increases in the hydrogen concentration in the exhaust air stream.

[0005] To comply with prescribed emission limits, the hydrogen concentration in the exhaust air stream must be precisely determined and monitored within the exhaust system. This is typically achieved by installing two hydrogen concentration sensors, positioned one behind the other and spaced apart in the direction of airflow. These sensors measure the hydrogen concentration in the exhaust air stream separately and independently. However, since the sensors have a response time of approximately 3 seconds, while the drain and purge pulses are usually shorter, the sensors often measure a lower value than the actual hydrogen concentration in the exhaust air stream. The hydrogen concentration is not present at the sensors for a sufficient duration.

[0006] To achieve a shorter sensor response time and thus more accurately determine the actual hydrogen concentration, a measuring device has been developed according to CN 118 090 838 A. This device comprises two identical metal oxide hydrogen concentration sensors arranged at a defined distance from each other. The two sensors are electrical resistors whose resistance changes with the hydrogen concentration. They are connected in series in an electrical circuit. The output signal, a voltage, is measured between the two sensors. In principle, the electrical circuit represents a voltage divider with two variable resistances that depend on the prevailing hydrogen concentration. The output signal is therefore dependent on both sensors.Furthermore, a flow generator is provided, which is connected upstream of the hydrogen concentration sensors and establishes a defined, constant flow velocity of the gas stream to be measured, which flows towards the sensors. Due to the distance between the sensors in the direction of flow, the resistances of the sensors change with a time delay when the hydrogen concentration in the gas stream changes, and the output signal describes a voltage curve that can be attributed to the hydrogen concentration. The advantage of the integrated series connection of two hydrogen concentration sensors is an improved response time of the measuring device.

[0007] A disadvantage of the measuring device according to CN 118090838 A is, among other things, that the device requires a predefined flow velocity, set by the flow generator, to determine the hydrogen concentration, regardless of the exhaust air flow velocity. However, since the exhaust air flow velocity is not constant, and the set flow velocity in the measuring device can differ from that of the exhaust air flow, the device does not reliably and accurately measure the actual hydrogen concentration. Furthermore, the measuring device is very complex in its design and requires a large number of components, which increases costs and the failure rate.

[0008] The object of the present invention is to overcome the disadvantages of the known prior art, in particular to provide a method that is easier to implement and / or more precise for determining the actual hydrogen concentration in an exhaust air stream.

[0009] This task is solved by the characteristics of independent claims.

[0010] A method for determining, in particular the profile of, the actual hydrogen concentration of an exhaust air stream from a fuel cell is provided. In the method according to the invention, a first hydrogen concentration signal of the exhaust air stream, corresponding to an upstream hydrogen concentration, and a second hydrogen concentration signal of the exhaust air stream, corresponding to a downstream hydrogen concentration, are determined. Furthermore, a hydrogen concentration function c(x), representing the actual hydrogen concentration, is determined, where x defines a position in the flow direction of the exhaust air stream. The actual hydrogen concentration is calculated based on the hydrogen concentration function c(x). The determined hydrogen concentration signals are fed into the hydrogen concentration function c(x) as two separate input variables.In particular, the exhaust air stream flows away from the fuel cell, preferably in an exhaust air duct, and especially into the ambient air. The actual hydrogen concentration in the exhaust air stream can be understood as the hydrogen concentration in the air of the exhaust air stream, preferably with minimal, and preferably no, measurement error. The hydrogen concentration is preferably expressed as a percentage by volume. A minimal measurement error can, for example, be a deviation of no more than 0.5%, preferably no more than 0.1%, from the actual value. In particular, each of the hydrogen concentration signals is determined by means of a hydrogen concentration sensor, preferably an independent one. The hydrogen concentration signal is, for example, a voltage signal that corresponds to a hydrogen concentration, so that each voltage value can be assigned a hydrogen concentration.Each of the two hydrogen concentration signals can be assigned a sampling point, which can be defined by the position of the corresponding hydrogen concentration sensors in the direction of the exhaust air flow. Specifically, the sampling point for the first hydrogen concentration signal is located in the direction of flow between the fuel cell and the sampling point for the second hydrogen concentration sensor. The sampling points for the first and second hydrogen concentration signals can be arranged at a defined distance from each other in the direction of flow. In particular, the determination of the first and second hydrogen concentration signals is carried out continuously over time, preferably simultaneously. The determined hydrogen concentration signals may deviate from the actual hydrogen concentration present in the exhaust air flow and are subject to measurement errors.The hydrogen concentration function c(x) specifically represents the profile of the actual hydrogen concentration in the exhaust air stream, allowing a value of the actual hydrogen concentration to be evaluated, and in particular calculated, at any given time and at any position in the flow direction. The determined hydrogen concentration signals are fed into the hydrogen concentration function c(x) as two separate inputs. This means, in particular, that the signals are independent of each other and are individually incorporated into the hydrogen concentration function c(x). Specifically, the hydrogen concentration function c(x) establishes a mathematical relationship between the first and second hydrogen concentration signals, which, for example, corrects the measurement errors of the hydrogen concentration signals and preferably extrapolates the actual hydrogen concentration in the exhaust air stream from the hydrogen concentration signals.The hydrogen concentration can be calculated continuously over time. Preferably, the hydrogen concentration function is evaluated at the downstream end of the exhaust system. This method offers the advantage of more precise measurement of the hydrogen concentration in the exhaust stream, thus enabling more efficient control of the fuel cell operation without exceeding emission limits. Furthermore, the method can be applied to existing hydrogen concentration sensors already installed in exhaust systems of vehicles, eliminating the need to develop and install new components. It also compensates for the slow response time of conventional hydrogen concentration sensors.This allows for the detection of short-term changes in hydrogen concentration in the exhaust air stream, particularly those shorter than the response time of the hydrogen concentration sensors. Response time refers specifically to the tau 90, or preferably the tau 63, time of the hydrogen concentration sensors, i.e., the time until the sensor reaches 90% or 63% of its initial reading.

[0011] It measured 63% of the actual hydrogen concentration.

[0012] In one exemplary implementation, the flow velocity of the exhaust air stream is passed as an additional input variable to the hydrogen concentration function. The flow velocity can be determined using a suitable sensor, such as a flow sensor. Taking the flow velocity into account offers the advantage of a more precise calculation of the actual hydrogen concentration.

[0013] In one exemplary implementation, the flow velocity of the exhaust air stream is determined using hydrogen concentration signals and / or a compressor speed signal from the fuel cell and / or a control signal from a purge valve of the fuel cell and / or a control signal from a drain valve of the fuel cell. The signals can be used individually or in any combination to determine the flow velocity of the exhaust air stream. The flow velocity can be determined in an intermediate step and passed to the hydrogen concentration function, or it can be determined directly within the hydrogen concentration function. The accuracy of the flow velocity determination can be increased by using multiple signals.The compressor speed signal corresponds in particular to the speed of a compressor located in the cathode of the fuel cell, which regulates the air supply to the fuel cell. Specifically, the flow velocity of the exhaust air stream depends on the compressor speed and can be determined from it. The purge valve control signal corresponds in particular to information about the opening state of the purge valve located at the anode of the fuel cell. In particular, when the purge valve is open, gas, especially hydrogen, enters the exhaust air stream from the anode through the purge valve, which can lead to a change in the flow velocity. The drain valve control signal corresponds in particular to information about the opening state of the drain valve located at the cathode of the fuel cell.In particular, when the drain valve is open, water and gas, especially air, from the cathode enter the exhaust air stream, which can lead to a change in flow velocity. Using the hydrogen concentration signals and / or the compressor speed signal and / or the purge valve control signal and / or the drain valve control signal, individually or in combination, offers the advantage that no additional sensor is required to measure the flow velocity, as existing signals can be used. In particular, the flow velocity can be predicted using all of these signals.

[0014] According to one exemplary embodiment, the flow velocity is determined based on the time difference between a change in the first hydrogen concentration signal and the second hydrogen concentration signal, and / or based on a predetermined distance between the sampling points of the hydrogen concentration signals. In particular, the distance between the sampling points of the first and second hydrogen concentration signals in the direction of the exhaust air flow allows for the detection of changes in the hydrogen concentration of the exhaust air flow with a time delay. Specifically, the upstream, first hydrogen concentration signal changes first, followed by the downstream, second hydrogen concentration signal after a time delay.By determining the time difference between the change in the first hydrogen concentration signal and the corresponding change in the second hydrogen concentration signal, and taking into account the known distance between the sampling points, the flow velocity can be calculated. In particular, the hydrogen concentration signals do not need to be evaluated qualitatively; only the time of a change in both signals needs to be recorded. According to an exemplary embodiment, the hydrogen concentration function c(x) is determined based on the geometry of the exhaust system, in particular its diameter, and / or the position and / or distance of the sampling points to determine the hydrogen concentration signals.The hydrogen concentration function establishes a mathematical relationship between the first and second hydrogen concentration signals. Specifically, it extrapolates the actual hydrogen concentration in the exhaust air stream from the change in hydrogen concentration between the first and second signals. This mathematical relationship is influenced by the geometry of the exhaust system, particularly the exhaust duct, and can be determined through simulation and / or experimentation. Thus, a specific hydrogen concentration function can be determined for different exhaust systems, positions, and / or distances between the hydrogen concentration signal sampling points.

[0015] In an exemplary procedure, the gradient of the hydrogen concentration signals is determined, and the actual hydrogen concentration is determined by comparing the determined gradients. The gradient of the first and second hydrogen concentration signals can be determined by taking the first derivative of the respective signal with respect to time. In particular, the gradients of the hydrogen concentration signals depend on the flow velocity. Specifically, the actual hydrogen concentration can be determined by comparing the gradient of the first hydrogen concentration signal, caused by a change in the hydrogen concentration in the exhaust air stream, with the corresponding gradient of the second hydrogen concentration signal, also caused by a change in the hydrogen concentration in the exhaust air stream.

[0016] According to an exemplary embodiment, the determination of the first and / or second hydrogen concentration signal is based on the principle of thermal conductivity, in particular by means of a TCD sensor.

[0017] According to an exemplary embodiment, the method is designed to determine a hydrogen concentration distribution along position x in the direction of flow of the exhaust air stream. In particular, the actual hydrogen concentration can be determined for each position in the direction of flow of the exhaust air stream using the hydrogen concentration function. Specifically, this makes it possible to determine the actual hydrogen concentration at the downstream end of the exhaust air stream, regardless of the position of the hydrogen concentration signal sampling points.

[0018] According to a further aspect of the present invention, which can be combined with the preceding aspects and exemplary embodiments, a control unit is provided for determining, in particular the course, of an actual hydrogen concentration of an exhaust air stream from a fuel cell, in particular according to the method of the preceding aspect and exemplary embodiments.The control unit comprises a first signal input for receiving an upstream hydrogen concentration signal of the exhaust air stream, measured by an upstream hydrogen concentration sensor; a second signal input for receiving a downstream hydrogen concentration signal of the exhaust air stream, measured by a downstream hydrogen concentration sensor; and a processing unit containing a hydrogen concentration function c(x) that represents the actual hydrogen concentration, where x defines a position in the direction of the exhaust air stream flow. The processing unit considers the determined hydrogen concentration signals as two separate input variables in the hydrogen concentration function c(x).The control unit can be a standalone data processing device, preferably detachably integrated into the vehicle, such as a microchip or processor, or a permanently integrated processor unit of the vehicle. The processing unit can be a microchip or processor. The hydrogen concentration function c(x) stored on the processing unit can, for example, be stored in integrated memory or in a database that the processing unit can access. The first and second hydrogen concentration signals can be received by the control unit via a physical connection, such as a cable, or by wireless technology, such as Bluetooth or WLAN. The two signals can be received by the control unit simultaneously or with a time delay.Furthermore, the control unit can include additional signal inputs to receive, for example, the compressor speed signal and / or the control signal of the purge valve of the fuel cell and / or the control signal of the drain valve of the fuel cell.

[0019] According to a further aspect of the present invention, which can be combined with the preceding aspects and exemplary embodiments, an exhaust system for a fuel cell is provided. The exhaust system comprises an exhaust duct for conveying an exhaust air stream from the fuel cell, as well as an upstream hydrogen concentration sensor arranged in the exhaust duct. It further comprises a downstream hydrogen concentration sensor arranged in the exhaust duct downstream of the upstream hydrogen concentration sensor, and a control unit designed according to the invention for determining the actual hydrogen concentration of the exhaust air stream. The exhaust duct can be connected to the fuel cell directly or indirectly at an upstream end in the direction of the exhaust air stream. The downstream end in the direction of the exhaust air stream can discharge directly or indirectly into the ambient air of the vehicle.In particular, the exhaust air stream is directed from the fuel cell into the ambient air of the vehicle via the exhaust air duct. The direction of flow of the exhaust air stream can be defined as away from the fuel cell. Upstream and downstream refer to the direction of flow of the exhaust air stream. Specifically, the upstream hydrogen sensor is arranged between the fuel cell and the downstream hydrogen sensor in the exhaust air duct in the direction of flow of the exhaust air stream. The upstream and downstream hydrogen sensors can be spaced apart from each other in the direction of flow of the exhaust air stream. The two hydrogen sensors can protrude completely or at least partially into the exhaust air duct, in particular in such a way that they detect the hydrogen concentration in the exhaust air duct.

[0020] Preferred embodiments are given in the dependent claims.

[0021] Further properties, features and advantages of the invention will be clarified below by describing preferred embodiments of the invention with reference to the accompanying exemplary drawings, which show:

[0022] Figure 1 schematic representation of an exhaust air system;

[0023] Figure 2 schematic flowchart of the method according to the invention; and

[0024] Figure 3 schematic signal flow diagram of the method according to the invention.

[0025] In the following description of exemplary embodiments, the method according to the invention is generally designated by reference numeral 100 and the exhaust air system by reference numeral 1. Figure 1 shows the schematic structure of an exhaust air system 1. The exhaust air system 1 basically comprises an exhaust air duct 3 and a fuel cell 5 arranged at its upstream end in the direction of flow R. The fuel cell 5 comprises a compressor 21, a purge valve 23, and a drain valve 25. The exhaust air duct has a diameter D and a length L. Furthermore, the exhaust air system 1 comprises an upstream hydrogen concentration sensor 7 and a downstream hydrogen concentration sensor 11. A first sampling point 9 in the exhaust air duct 3 in the direction of flow R of the exhaust air stream is associated with the upstream hydrogen concentration sensor 7 at position xi.A second sampling point 13 in the exhaust air duct 3, in the direction R of the exhaust air stream, is assigned to the downstream hydrogen concentration sensor 11 at position X2. Position X describes the location, or its distance in the direction R of the exhaust air stream, in the exhaust air duct 3 of the exhaust air system 1 with respect to the upstream end of the exhaust air duct 3. The first and second sampling points 9 and 13 are arranged at a distance A from each other in the direction R of the exhaust air stream. Each of the hydrogen concentration sensors 7 and 11 has a signal output 31 and 33, respectively, at which the hydrogen concentration signal Si and S2, respectively, determined by the respective hydrogen concentration sensor 7 and 11, respectively, is present. The first hydrogen concentration signal S1 corresponds to an upstream hydrogen concentration CAUF in the exhaust air stream at position xi, determined by the hydrogen concentration sensor 7.The second hydrogen concentration signal S2 corresponds to a downstream hydrogen concentration CAB in the exhaust air stream at position X2, as determined by the hydrogen concentration sensor 11. A line 27, transmitting the first hydrogen concentration signal S1, runs from the upstream hydrogen concentration sensor 7, from signal output 31 to signal input 17 of a control unit 15. Similarly, a line 29, transmitting the second hydrogen concentration signal S2, runs from signal output 33 of the downstream hydrogen concentration sensor 11 to signal input 19 of the control unit 15. It is conceivable that lines 27 and 29 could be replaced by wireless data transmission such as Bluetooth or WLAN technology. The control unit 15 may include further signal inputs not shown. Furthermore, the control unit includes a processing unit 37, such as a processor or microchip.

[0026] The hydrogen concentration distribution 35 exemplifies the graph of a fictitious actual hydrogen concentration c over the entire length L of the exhaust air duct 3 at a time t. Each position xi, X2, X3 is assigned an actual hydrogen concentration Ci, C2, C3. Figure 2 shows a schematic flow diagram of the process of the inventive method 100. The method 100 basically comprises four steps. Preferably, the method 100 is carried out continuously over time during the operation of the fuel cell 5 in a fuel cell vehicle, so that a calculated actual hydrogen concentration c is available and can be monitored at any given time, ensuring that prescribed limits are not exceeded.

[0027] In step 110, the first hydrogen concentration signal Si is determined. This first hydrogen concentration signal Si corresponds to an upstream hydrogen concentration CAUF. The upstream hydrogen concentration CAUF is measured by the upstream hydrogen concentration sensor 7 at the first sampling point 9 in the exhaust air duct 3. The determination of the first hydrogen concentration signal Si is preferably continuous over time, but can alternatively be performed discretely. Preferably, the first hydrogen concentration signal Si contains a time component. The upstream hydrogen concentration sensor 7 can be a TCD sensor based on the principle of thermal conductivity.

[0028] In step 120, the second hydrogen concentration signal S2 is determined. This second process step 120 occurs simultaneously with the first process step 110, but can also be performed with a time delay. The second hydrogen concentration signal S2 corresponds to a downstream hydrogen concentration CAB. The downstream hydrogen concentration CAB is measured by a downstream hydrogen concentration sensor 11 at a second sampling point 13 in the exhaust air duct 3. The determination of the second hydrogen concentration signal S2 preferably occurs continuously over time, but can also be performed discretely. Preferably, the second hydrogen concentration signal S2 contains a time component. The upstream hydrogen concentration sensor 9 can be a TCD sensor based on the principle of thermal conductivity.

[0029] In the third step, 130, of procedure 100, the hydrogen concentration function c(x) is determined. Step 130 can also be performed before or simultaneously with steps 110 and 120 and, for example, be stored in a database and then retrieved from it. The hydrogen concentration function c(x) represents the actual hydrogen concentration c in the exhaust air stream. The hydrogen concentration function c(x) takes the first and second hydrogen concentration signals Si and S2 as input parameters and indicates the actual hydrogen concentration c in the exhaust air stream as a function of the position x. In step 130, the mathematical relationship between the first hydrogen concentration signal Si and the second hydrogen concentration signal S2 is established, particularly taking into account the flow velocity v of the exhaust air stream, and is provided as the hydrogen concentration function c(x).In other words, the hydrogen concentration function c(x) can extrapolate the actual hydrogen concentration c based on the determined hydrogen concentration signal S1, the determined hydrogen concentration signal S2, and, in particular, the flow velocity v. The geometry of the exhaust system 1, in particular, influences how the hydrogen concentration changes along the exhaust duct 3, as the hydrogen contained in the exhaust stream dilutes, for example, and thus the hydrogen concentration decreases along the flow direction R. The diameter D of the exhaust duct is especially crucial and is included in the determination of the hydrogen concentration function c(x). Another influencing factor that can be considered in step 130 is the respective position xi, X2 of the first and second sampling points 9, 13, and their distance A in the flow direction R.In particular, the hydrogen concentration function c(x) can be determined by means of simulation or experiments on the test bench for any geometries of exhaust air systems and hydrogen concentration sampling points.

[0030] An intermediate step can be performed before calculating the actual hydrogen concentration c in step 140. In this intermediate step, the flow velocity v of the exhaust air stream can be determined, which can then be included in step 140. For example, the flow velocity v can be determined using a flow velocity sensor. Alternatively, the hydrogen concentration signals S1 and S2 determined in steps 110 and 120 can be used to determine the flow velocity v of the exhaust air stream. In particular, a change in the hydrogen concentration of the exhaust air stream due to the position of the first and second sampling points 9, 13 first leads to a change in the first hydrogen concentration signal S1 and subsequently to a change in the second hydrogen concentration signal S2.Since the positions xi and X2 of sampling points 9 and 13, and thus their distance A, are known, the flow velocity v can be deduced from the time difference of the change in the hydrogen concentration signals S1 and S2. Another alternative implementation of the intermediate step can be based on a compressor speed signal S. n a compressor 21 arranged in the cathode of the fuel cell 5. In particular, the exhaust air flow from the exhaust air system 1 is generated by the compressor 21 in the cathode of the fuel cell 5. The flow velocity v of the exhaust air flow depends in particular on the compressor speed n. Thus, the compressor speed signal S nThe flow velocity v of the exhaust air stream can be determined. The flow velocity can also depend on the control of the purge valve 23 and drain valve 25 of the fuel cell 5. When the purge valve and / or the drain valve opens, the flow velocity can increase because additional gas from the fuel cell is added to the exhaust air stream. The flow velocity can be determined from the control signal SP of the purge valve and the control signal SD of the drain valve, each of which contains information about the valve positions. Alternatively, the flow velocity v of the exhaust air stream can also be determined from the combination of the hydrogen concentration signals Si and S2 and the compressor speed signal S. n and the control signals SP and SD of the purge valve 23 and the drain valve 25 to increase accuracy.

[0031] In step 140, the actual hydrogen concentration c is calculated. For example, this calculation takes place in processing unit 37 of control unit 15. The calculation is based on the previously determined hydrogen concentration signals S1 and S2, and in particular the flow velocity v determined in an intermediate step, using the hydrogen concentration function c(x). The hydrogen concentration signals S1 and S2 are passed separately to the hydrogen concentration function c(x); the signal flow is shown schematically in Figure 2. In an exemplary embodiment, the flow velocity v is also passed to the hydrogen concentration function c(x). Only in step 140, the calculation of the actual hydrogen concentration c, are the two hydrogen concentration signals S1 and S2 combined.For example, in the hydrogen concentration function c(x), the gradients of the first and second hydrogen concentration signals S1 and S2 are determined. This can be done by taking the first derivative with respect to time. The two gradients can then be compared, particularly considering the previously determined flow velocity v, and the actual hydrogen concentration c can be deduced from this. The result of the calculation of the actual hydrogen concentration c depends on the position x. Thus, for every time point at every position x along the direction of the exhaust air flow, a discrete actual hydrogen concentration value c can be determined. xThe calculated actual hydrogen concentration at the first and second sampling points 9 and 13 at positions xi and X2, as well as at the downstream end of the exhaust air duct 3 at position X3, is preferably evaluated. At each time point t, this corresponds to an actual hydrogen concentration Ci, C2, and C3, as shown in Figure 3. Figure 3 schematically shows the signal flow of the hydrogen concentration signals Si and S2 determined in steps 110 and 120. The signals are fed separately into the hydrogen concentration function c(x). In an exemplary embodiment not shown, the two signals can additionally be fed into a further flow velocity function, in which the flow velocity v of the exhaust air stream is determined, the result of which can in turn be fed into the hydrogen concentration function c(x) as a third independent input variable.In addition to the two hydrogen concentration signals S1 and S2, the compressor speed signal S can also be included in the flow velocity function. n The flow velocity function of compressor 21, as well as the control signal SP of purge valve 23 and the control signal SD of drain valve 25, are incorporated. Alternatively, the flow velocity function can be integrated into the hydrogen concentration function c(x). For example, in addition to the hydrogen concentration signals S1 and S2, the compressor speed signal S can also be included. n of the compressor 21, as well as the control signal SP of the purge valve 23 and the control signal SD of the drain valve 25 are directly transferred to the hydrogen concentration function c(x).

[0032] The features disclosed in the foregoing description, the figures, and the claims can be important for the realization of the invention in its various embodiments, both individually and in any combination. 1 Exhaust system

[0033] 3 Exhaust air duct

[0034] 5 Fuel cell

[0035] 7 upstream hydrogen concentration sensor 9 upstream sampling point

[0036] 11 downstream hydrogen concentration sensor 13 downstream sampling point

[0037] 15 Control unit

[0038] 17 first signal input

[0039] 19 second signal input

[0040] 21 Compressor

[0041] 23 Purge valve

[0042] 25 Drain valve

[0043] 27 First hydrogen concentration signal leading line 29 Second hydrogen concentration signal leading line 31 Signal output of the upstream hydrogen concentration sensor

[0044] 33 Signal output of the downstream hydrogen concentration sensor

[0045] 35 Hydrogen concentration distribution

[0046] 37 processing units

[0047] 100 procedures

[0048] 110 first procedural step

[0049] 120 second procedural step

[0050] 130 third procedural step

[0051] 140 fourth procedural step

[0052] R Flow direction of the exhaust air stream

[0053] Si first hydrogen concentration signal

[0054] S2 second hydrogen concentration signal

[0055] Sn Compressor speed signal

[0056] S P Purge valve control signal

[0057] S D Control signal of the drain valve

[0058] c(x) hydrogen concentration function

[0059] X Position

[0060] c actual hydrogen concentration c x actual hydrogen concentration at position x CAUF upstream hydrogen concentration

[0061] CAB downstream hydrogen concentration

Claims

Patent claims 1. Method (100) for determining an actual hydrogen concentration (c) of an exhaust air stream of a fuel cell (5), comprising the following steps: - Determining (110) a first hydrogen concentration signal (Si) of the exhaust air stream corresponding to an upstream hydrogen concentration (CAUF); - Determine (120) a second hydrogen concentration signal (S2) of the exhaust air stream corresponding to a downstream hydrogen concentration (CAB); - Determine (130) a hydrogen concentration function (c(x)) representing the actual hydrogen concentration (c), where (x) defines a position in the direction of flow (R) of the exhaust air stream; and - Calculate (140) the actual hydrogen concentration (c) using the hydrogen concentration function (c(x)) ; characterized in that the determined hydrogen concentration signals (Si), (S2) are passed as two separate input variables to the hydrogen concentration function (c(x)).

2. Method (100) according to claim 1 , characterized in that the flow velocity (v) of the exhaust air stream is passed as a further input variable to the hydrogen concentration function (c(x)).

3. Method (100) according to claim 2, characterized in that the flow velocity (v) of the exhaust air stream is determined on the basis of the hydrogen concentration signals (Si), (S2) and / or on the basis of a compressor speed signal (S n ) of the fuel cell (5) and / or based on a control signal (SP) of a purge valve (23) of the fuel cell (5) and / or based on a control signal (SD) of a drain valve (25) of the fuel cell (5).

4. Method (100) according to claim 2 or 3, characterized in that the flow velocity (v) is determined based on a time difference between a change in the hydrogen concentration signal (S1) and a change in the hydrogen concentration signal (S2) and / or based on a predetermined distance (A) from sampling points (7), (11) the hydrogen concentration signals (Si), (S2) are determined.

5. Method (100) according to one of the preceding claims, characterized in that the hydrogen concentration function (c(x)) is determined based on a geometry of the exhaust air system (1), in particular a diameter (D) of the exhaust air system (1), and / or a position (x) and / or the distance (A) of the sampling points (7), (11) for determining the hydrogen concentration signals (S1), (S2).

6. Method (100) according to one of the preceding claims, characterized in that the gradient of the hydrogen concentration signals (S1), (S2) is determined and the actual hydrogen concentration (c) is determined by comparing the determined gradients.

7. Method (100) according to one of the preceding claims, characterized in that the determination (110, 120) of the first and / or second hydrogen concentration signal (Si), (S2) is carried out on the principle of thermal conductivity, in particular by means of a TCD sensor.

8. Method (100) according to one of the preceding claims, characterized in that the method (100) is configured to determine a hydrogen concentration distribution (35) along the position (x) in the flow direction (R) of the exhaust air stream.

9. Control unit (15) for determining an actual hydrogen concentration (c) of an exhaust air stream of a fuel cell (5), in particular according to the method (100) according to one of the preceding claims, comprising: - a first signal input (17) for receiving an upstream hydrogen concentration signal (S1) of the exhaust air stream; - a second signal input (19) for receiving a downstream hydrogen concentration signal (S2) of the exhaust air stream; a processing unit (37) on which a hydrogen concentration function (c(x)) representing the actual hydrogen concentration (c) is stored, where (x) defines a position in the direction of flow (R) of the exhaust air stream; This is characterized by the fact that the processing unit (37) takes into account the determined hydrogen concentration signals (Si), (S2) as two separate input variables in the hydrogen concentration function (c(x)).

10. Exhaust system (1) for a fuel cell (5), comprising: - an exhaust air duct (3) for conveying an exhaust air stream from the fuel cell (5); an upstream hydrogen concentration sensor (7) arranged in the exhaust air duct (3); a downstream hydrogen concentration sensor (11) arranged in the exhaust air duct (3) downstream of the upstream hydrogen concentration sensor (7); and - a control unit (15) designed according to claim 9 for determining the actual hydrogen concentration (c) of the exhaust air stream.

Images