Diagnostic procedure for a coolant system of a fuel cell stack

Abstract

Diagnostic procedure for detecting gas bubbles in a coolant system of a fuel cell stack of a fuel cell system, comprising the following steps: - Determining a target pressure (Ps), - Determining an actual pressure (P(t)), - Determining an integral (I(t)) of a difference between the actual pressure (P(t)) and the target pressure (Ps) over a time window (F), - Comparing the integral (I(t)) over the time window (F) with a reference value (R), and - Outputting a fault signal (80) based on the comparison result, characterized by the fact that Additional external actual pressure influences (PE(t)) that affect the actual pressure (P(t)) are determined and taken into account for comparison, whereby Impacts to the fuel cell stack caused by driving on an uneven road surface are detected, and these impacts are taken into account as additional external actual pressure influences (PE(t)).

Description

[0001] The present invention relates to a diagnostic method for a coolant system of a fuel cell stack of a fuel cell system, a computer program product for carrying out such a diagnostic method and a diagnostic device.

[0002] It is known in the art that a fuel cell stack of a fuel cell system can be cooled with a coolant. If there is too little coolant in the cooling system, this is referred to as underfilling. In addition to the coolant, gas bubbles are then also present in the cooling system. The more gas bubbles are present in the cooling system, the worse the cooling effect. It is known to calculate an integral over the pressure fluctuations in the cooling system over time to obtain a measure of the number of gas bubbles in the cooling system. However, external influences that lead to additional pressure fluctuations in the cooling system are not taken into account. This can lead to inaccurate results regarding the presence of gas bubbles in the cooling system.For example, current technology does not take into account the additional pressure fluctuations in the coolant system when a vehicle with a fuel cell system drives on an uneven road surface. In such fuel cell systems, even small air bubbles can cause significant damage.

[0003] Against this background, one object of the present invention is to at least partially overcome the disadvantages described above. In particular, it is an object of the present invention to provide a diagnostic method for detecting gas bubbles in the coolant system of a fuel cell stack, in which external influences on the pressure in the coolant system can be taken into account.

[0004] The foregoing problem is solved by a diagnostic method with the features of claim 1, a computer program product with the features of claim 11, and a diagnostic device with the features of claim 12. Further features and details of the invention will become apparent from the dependent claims, the description, and the drawings. Features and details described in connection with the diagnostic method according to the invention naturally also apply in connection with the computer program product and the diagnostic device according to the invention, and vice versa, so that the disclosure of the individual aspects of the invention always makes, or can make, reciprocal references.

[0005] Accordingly, a diagnostic procedure for detecting gas bubbles in the coolant system of a fuel cell stack of a fuel cell system is described. The diagnostic procedure comprises the following steps: - Determining a target pressure, - Determining an actual pressure, - Determining the integral of the difference between the actual pressure and the target pressure over a time window, - Comparing the integral over the time window with a reference value, and - Outputting a fault signal based on the comparison result, where additional external actual pressure influences that affect the actual pressure are determined and taken into account for comparison.

[0006] By calculating the integral from the difference between the actual pressure and the target pressure, a measure can be determined that is characteristic of the gas bubbles in the coolant system. If gas bubbles are present in the coolant system, then the system is not sufficiently filled with coolant. The more gas bubbles there are in the coolant system, the greater the fluctuations in the actual pressure and thus the higher the value of the integral. A fault signal can be output if the value of the integral is greater than the reference value.

[0007] If additional external pressure influences occur, these can trigger further pressure fluctuations. These additional pressure fluctuations can distort the result, leading to an incorrect assessment of the gas present in the coolant system.

[0008] Accordingly, the additional external influences on the actual pressure are determined and taken into account for comparison with the reference value. This provides an accurate indication of the gas quantity in the coolant system or at least concrete evidence of underfilling.

[0009] The gas bubbles in the coolant system could be air. The described diagnostic procedure is specifically suitable for fuel cell stacks in a vehicle.

[0010] The actual pressure is the pressure that is actually present in the coolant system. The actual pressure fluctuates over time. The more gas bubbles are present in the coolant system, the more the actual pressure fluctuates.

[0011] The target pressure is the pressure without movement of the fuel cell stack or other external influences. Therefore, the target pressure does not fluctuate. The target pressure is the pressure one would expect, for example, in a fuel cell stack of a stationary vehicle under standardized ambient conditions, regarding temperature, air pressure, humidity, etc. The target pressure is typically regulated by a fuel cell control unit. The target pressure can also be the theoretical pressure that the cooling system should maintain.

[0012] The fuel cell stack contains multiple fuel cells. The fuel cell system, in turn, can contain multiple fuel cell stacks.

[0013] The time window over which integration takes place can range from 1 second to 100 minutes, and especially from 1 minute to 10 minutes.

[0014] The reference value is a value that is set to determine from which value of the integral over the time window there is too much gas in the coolant system.

[0015] According to one embodiment of the diagnostic procedure, the additional external actual pressure influences are combined with the actual pressure before the integral is determined. Advantageously, this takes the additional external actual pressure influences into account. These additional external actual pressure influences can be any influences that change the actual pressure. This means that "the additional external actual pressure influences are combined with the actual pressure," specifically, that the additional external actual pressure influences are subtracted from the actual pressure. The newly determined actual pressure is then used to determine the integral.

[0016] According to another embodiment of the diagnostic procedure, an influence integral is determined using the additional external actual pressure influences. This influence integral is then combined with the integral. Advantageously, this takes the additional external actual pressure influences into account. "Combining the influence integral with the integral" means, in particular, that the influence integral is subtracted from the integral. The newly determined integral is then compared with the reference value.

[0017] According to another embodiment of the diagnostic procedure, an influence integral is determined using the additional external actual pressure influences. The reference value is then shifted based on this influence integral. Advantageously, this takes the additional external actual pressure influences into account. The influence integral can also be integrated over the same time window as the integral. Therefore, a specific value is obtained for the influence integral integrated over this time window. The reference value is then shifted using this specific value.

[0018] In principle, the actual pressure, the integral, and the reference value can all be recalculated based on the additional external influences on the actual pressure. Alternatively, a combination of two or three of these methods can be used to account for the additional external influences on the actual pressure.

[0019] According to another embodiment of the diagnostic procedure, the difference between the actual pressure and the target pressure is integrated over the time window. Subtracting the target pressure from the actual pressure can leave fluctuations around the zero line. Advantageously, using the absolute value of the difference prevents positive and negative values ​​from partially canceling out during integration. If these values ​​were to cancel each other out, it would distort the result, leading to incorrect conclusions about the amount of gas in the coolant system after integration.

[0020] Alternatively, one can integrate the square of the difference between the actual pressure and the target pressure over the time window to ensure that individual areas do not cancel each other out. In principle, other mathematical methods are conceivable to ensure that individual areas do not cancel each other out during integration.

[0021] According to another embodiment of the diagnostic procedure, the time window over which integration is performed is shifted. The comparison with the reference value is then repeated. It is possible to shift the time window, perform a new integration within the shifted time window, and compare the result with the reference value. It is also possible to shift the time window multiple times, perform an integration over the shifted time window with each shift, calculate an average of the multiple integrations, and compare this average with the reference value.

[0022] The time window can also be used, for example, as a so-called rolling time window. For this purpose, the time window is repeatedly shifted by 1 second between, for example, 5 seconds and 100 seconds, between 5 seconds and 50 seconds, and especially between 5 seconds and 15 seconds.

[0023] It is also possible to change the length of the time window when shifting the time window.

[0024] According to another embodiment of the diagnostic procedure, the end time of one time window is the start time of a shifted time window. Advantageously, such a sequence of time windows covers the entire time range.

[0025] According to another embodiment of the diagnostic procedure, impacts to the fuel cell stack caused by driving on an uneven road surface are detected. These impacts are then taken into account as additional external pressure influences. Vibrations of the fuel cell stack and / or impacts to the fuel cell stack can affect the actual pressure in the coolant system. It is therefore advantageous to consider these additional external pressure influences when assessing whether there is too much gas in the coolant system.

[0026] According to another embodiment of the diagnostic procedure, specific events that can influence the actual pressure are detected and taken into account as additional external influences on the actual pressure. A specific event could be, for example, strong acceleration, such as that which occurs when a vehicle brakes. Furthermore, a vehicle driving over cobblestones can result in specific events. Each individual cobblestone crossing could trigger a specific event. However, one could also consider the entire journey over the cobblestones as a single specific event.

[0027] According to another embodiment of the diagnostic procedure, a parameter that can influence the actual pressure is considered as an additional external influence on the actual pressure. Advantageously, parameters such as temperature, air pressure, humidity, etc., can also be considered if they have an influence on the actual pressure.

[0028] Furthermore, an object of the present invention is a computer program product comprising instructions which, when executed by a computer, cause it to perform the steps of a diagnostic method according to the invention. Thus, a computer program product according to the invention also offers the same advantages as those explained in detail with reference to a diagnostic method according to the invention.

[0029] Furthermore, a diagnostic device for detecting gas bubbles in the coolant system of a fuel cell stack of a fuel cell system is provided. The diagnostic device comprises a setpoint determination module for determining a setpoint pressure, an actual pressure determination module for determining an actual pressure, an integral determination module for determining the integral of the difference between the actual pressure and the setpoint pressure over a time window, a comparison module for comparing the integral over the time window with a reference value, and an output module for outputting a fault signal based on the comparison result. The diagnostic device also includes an actual pressure influence consideration module for determining additional external actual pressure influences that affect the actual pressure and for taking these additional external actual pressure influences into account for comparison with the comparison module.Additional external pressure influences (PE(t)) are in particular impacts against the fuel cell stack caused by driving on an uneven road surface.

[0030] The target pressure determination module, the actual pressure determination module, the integral determination module, the comparison module, the output module and / or the actual pressure influence consideration module are specifically designed for carrying out a diagnostic procedure as described.

[0031] Further advantages, features, and details of the invention will become apparent from the following description, in which exemplary embodiments are described in detail with reference to the drawings. The drawings show: Fig. 1 a schematic view of a diagnostic device; Fig. 2. a graph to describe the influence of the actual pressure influences on the actual pressure; and Fig. 3. A graph to describe the influence of the influence integral on the integral.

[0032] Fig. Figure 1 shows a schematic view of a diagnostic device 10. The diagnostic device 10 is used to detect gas bubbles in a coolant system of a fuel cell stack of a fuel cell system. The diagnostic device 10 comprises a setpoint pressure determination module 20, an actual pressure determination module 30, an integral determination module 40, an actual pressure influence consideration module 50, a comparison module 60, and an output module 70. The diagnostic device 10 is specifically designed to perform a diagnostic method according to the invention.

[0033] Below the diagnostic device 10 in the Fig. 1. Graphs and comparisons are shown to explain the individual modules 20, 30, 40, 50, 60.

[0034] The target pressure determination module 20 is used to determine a target pressure Ps. For this purpose, the target pressure Ps is read, for example, by a sensor or received from a computer. This is in Fig. 1 is symbolized by the dashed arrow pointing towards the target pressure determination module 20. The two graphs below the target pressure determination module 20 show the target pressure Ps versus time t.

[0035] The actual pressure determination module 30 is used to determine an actual pressure P(t). For this purpose, the actual pressure P(t) is read, for example, from a sensor or received from a computer. This is in Fig. 1 is symbolized by the dashed arrow pointing to the actual pressure determination module 30. The two graphs below the actual pressure determination module 30 show the actual pressure P(t) versus time t. How to the Fig. As can be seen from Figure 1, the upper graph shows the case where the actual pressure P(t) fluctuates significantly with time t. This occurs when there is a large amount of gas in the coolant system. The lower graph, on the other hand, shows the case where there is very little gas in the coolant system. The fluctuations in the actual pressure P(t) are therefore minimal.

[0036] The integral modulus 40 is used to determine the integral I(t) of the difference between the actual pressure P(t) and the target pressure Ps over a time window F. The value of this integral is a measure of the amount of gas in the cooling system. The upper graph under the integral modulus 40 shows the case with large fluctuations in the actual pressure P(t), i.e., with a large amount of gas in the coolant system. The integral I(t) therefore increases sharply with time t. The lower graph under the integral modulus 40 shows the case with small fluctuations in the actual pressure P(t), i.e., with very little gas in the coolant system. The integral I(t) therefore increases only slowly with time t.

[0037] In particular, the magnitude of the difference between the actual pressure P(t) and the target pressure Ps can be integrated over the time window F. This ensures that time periods of the difference with a positive sign and time periods of the difference with a negative sign do not cancel each other out during integration over the time window F.

[0038] The comparison module 60 is used to compare the integral I(t) over the time window F with a reference value R. Integrating over the time window F yields a specific value that can serve as a measure of the amount of gas in the coolant system. The reference value is the value above which there is too much gas in the coolant system. Fig. The equation I(F) > R is given directly below the comparison module 60. This applies to the case where large fluctuations in the actual pressure P(t) occur, i.e., where there is a large amount of gas in the coolant system. Further down, the equation I(F) < R is given. This applies to the case where only very small fluctuations in the actual pressure P(t) occur, i.e., where there is little gas in the coolant system.

[0039] Output module 70 is used to output a fault signal 80 based on the comparison result of comparison module 60. If the condition I(F) > R is met, then a fault signal 80 is output. If the condition I(F) < R is met, then no fault signal 80 is output.

[0040] The actual pressure influence consideration module 50 is used to determine additional external actual pressure influences PE(t) that affect the actual pressure P(t). For this purpose, the additional external actual pressure influences PE(t) are read in, for example, by a sensor or received from a computer. This is described in Fig. 1 symbolized by the dashed arrow, which points to the actual pressure influence consideration module 50.

[0041] Furthermore, the actual pressure influence consideration module 50 serves to take into account the additional external actual pressure influences PE(t) for comparison with the comparison module 60. In principle, the actual pressure P(t), the integral I(t), and / or the reference value R can be recalculated due to the additional external actual pressure influences PE(t) in order to account for the additional external actual pressure influences PE(t). Alternatively or additionally, the calculation of the integral can also be paused as long as the influence is present.

[0042] Fig. Figure 2 shows a graph describing the influence of the additional external actual pressure influences PE(t) on the actual pressure P(t). Before determining the integral I(t), the additional external actual pressure influences PE(t) are combined with the actual pressure P(t). The additional external actual pressure influences PE(t) are in Fig. 2 is simplified and represented as a constant value. However, the additional external actual pressure influences PE(t) can also exhibit a more complex temporal profile. If one considers the additional external actual pressure influences PE(t), i.e., if one subtracts the additional external actual pressure influences PE(t) from the actual pressure P(t), then a newly determined actual pressure P'(t) is obtained. This newly determined actual pressure P'(t) is then further considered for the detection of gas bubbles in the coolant system.

[0043] Fig.Figure 3 shows a graph describing the influence of an influence integral IE(t) on the integral I(t). Using the additional external actual pressure influences PE(t), an influence integral IE(t) can be determined. This influence integral IE(t) can be combined with the integral I(t) to determine a newly calculated integral I'(t). In particular, the influence integral IE(t) can be subtracted from the integral I(t). The newly calculated integral I'(t) is then further considered for the detection of gas bubbles in the coolant system.

[0044] Another possibility is to redetermine the reference value R. For this, an influence integral IE(t) is determined using the additional external actual pressure influences PE(t). The reference value R is then shifted due to this influence integral IE(t). This newly determined reference value R is then further considered for the detection of gas bubbles in the coolant system.

[0045] A variety of events can be considered as additional external pressure influences PE(t). According to the invention, impacts to the fuel cell stack caused by driving on an uneven road surface are considered. Furthermore, specific events that influence the actual pressure P(t) can be considered as additional external pressure influences PE(t). Such specific events include, for example, strong acceleration, such as during braking. Additionally, one or more parameters that can influence the actual pressure P(t) can be considered as additional external pressure influences PE(t). These parameters include, in particular, temperature, air pressure, humidity, etc.

[0046] The time window F over which integration is performed can be shifted. After shifting, the comparison with the reference value R can then be carried out again. In particular, the end time of time window F can become the start time of a shifted time window F. In this way, it is possible to continuously monitor whether there is too much gas in the coolant system. It is also possible to average over several time windows F and then perform the comparison with the reference value R. Reference symbol list 10 Diagnostic device 20 Target pressure determination module 30 Actual pressure determination module 40 Integral Determination Module 50 Actual pressure influence consideration module 60 Comparison module 70 Output module 80 Interference signal PS Target pressure P(t) Actual pressure P'(t) newly determined actual pressure I(t) integral I'(t) newly determined integral F Time window R Reference value PE(t) Actual pressure influences IE(t) Influence integral

Claims

[1] Diagnostic procedure for detecting gas bubbles in a coolant system of a fuel cell stack of a fuel cell system, comprising the steps: - Determining a target pressure (Ps), - Determining an actual pressure (P(t)), - Determining an integral (I(t)) of a difference between the actual pressure (P(t)) and the target pressure (Ps) over a time window (F), - Comparing the integral (I(t)) over the time window (F) with a reference value (R), and - Outputting a fault signal (80) based on the comparison result, characterized by , that Additional external actual pressure influences (PE(t)) that affect the actual pressure (P(t)) are determined and taken into account for comparison, whereby Impacts to the fuel cell stack caused by driving on an uneven road surface are detected, and these impacts are taken into account as additional external actual pressure influences (PE(t)). [2] Diagnostic method according to claim 1, characterized by , that before determining the integral (I(t)) the additional external actual pressure influences (PE(t)) are combined with the actual pressure (P(t)). [3] Diagnostic method according to claim 1 or 2, characterized by , that an influence integral (IE(t)) is determined using the additional external actual pressure influences (PE(t)), and that the influence integral (IE(t)) is combined with the integral (I(t)). [4] Diagnostic method according to any one of the preceding claims, characterized by , that an influence integral (IE(t)) is determined using the additional external actual pressure influences (PE(t)), and that the reference value (R) is shifted due to the influence integral (IE(t)). [5] Diagnostic method according to any one of the preceding claims, characterized by , that an amount of the difference between the actual pressure (P(t)) and the target pressure (Ps) is integrated over the time window (F). [6] Diagnostic method according to any one of the preceding claims, characterized by , that the time window (F) over which integration is performed is shifted in time, and the comparison with the reference value (R) is then performed again, wherein in particular an end time of the time window (F) is a start time of a shifted time window (F). [7] Diagnostic method according to any one of the preceding claims, characterized by , that special events which can influence the actual pressure (P(t)) are detected and taken into account as additional external actual pressure influences (PE(t)). [8] Diagnostic method according to any one of the preceding claims, characterized by, that a parameter which can influence the actual pressure (P(t)) is taken into account as additional external actual pressure influences (PE(t)). [9] Computer program product comprising instructions which, when executed by a computer, cause it to perform the steps of a diagnostic procedure having features of any one of claims 1 to 8. [10] Diagnostic device (10) for detecting gas bubbles in a coolant system of a fuel cell stack of a fuel cell system, comprising a target pressure determination module (20) for determining a target pressure (Ps), a pressure determination module (30) for determining a pressure (P(t)), an integral determination module (40) for determining an integral (I(t)) of a difference between the actual pressure (P(t)) and the target pressure (Ps) over a time window (F), a comparison module (60) for comparing the integral (I(t)) over the time window (F) with a reference value (R), and an output module (70) for outputting a fault signal (80) based on the comparison result, characterized by an actual pressure influence consideration module (50) for determining additional external actual pressure influences (PE(t)) which affect the actual pressure (P(t)) and for taking into account the additional external actual pressure influences (PE(t)) for comparison with the comparison module (60), wherein impacts against the fuel cell stack caused by driving on an uneven road surface are considered as additional external actual pressure influences (PE(t)).

Images