Method for determining a calibration curve
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- DAIMLER TRUCK AG
- Filing Date
- 2025-12-12
- Publication Date
- 2026-07-30
Smart Images
Figure EP2025086929_30072026_PF_FP_ABST
Abstract
Description
[0001] Daimler Truck AG Oezvatan December 12, 2025
[0002] Method for recording a calibration curve
[0003] The invention relates to a method for recording a calibration curve for the fill level of a tank for a liquefied gas according to the type defined in more detail in the preamble of claim 1. The invention also relates to a fuel cell vehicle.
[0004] Tanks for storing liquefied gases are fundamentally known from the prior art. They are also referred to as cryogenic storage systems and contain the gas at very low temperatures, at which it is liquid. The present method deals specifically, but not exclusively, with a tank for storing liquid hydrogen, which serves as fuel for a fuel cell system. This fuel cell system can, for example, be used in a vehicle to supply electrical power for the vehicle's drive system.
[0005] In the context of storing liquefied gases in tanks, it is generally known that level sensors can be used to monitor the fill level. US 2006 / 0123902 A1, for example, describes a capacitive level sensor inside a liquid hydrogen tank. DE 202019103696 U1 also describes a level sensor in a tank for liquefied gas used as fuel in a vehicle.
[0006] In practice, it is typically the case that with all current setups and measurement methods, the accuracy of the fill level indicator is relatively imprecise, so that accuracies on the order of + / - 10% can only be achieved. This has the significant disadvantage that only a very unreliable range indication is possible when used in a vehicle. US 202410270065 A1 uses pressure measurements as an alternative to such fill level sensors, in this case in a system with two tanks. Here, too, the disadvantage arises that the accuracy depends on many external factors, so that it can only be determined whether one of the tanks is full or empty.
[0007] From US patent 2020 / 0248638 A1, it is also known to perform measurements on such a tank when it is empty and when it is full, using both a pressure signal and a level signal. A calibration line is then interpolated between these two points to improve the prediction of the actual tank contents based on the recorded measurements.
[0008] The object of the present invention is to further improve the accuracy of the method described in the aforementioned prior art.
[0009] According to the invention, this problem is solved by a method with the features in claim 1, and in particular in the characterizing part of claim 1. Further embodiments and developments of the method according to the invention are set forth in the dependent claims.
[0010] Furthermore, a fuel cell vehicle according to claim 8 solves the problem. Advantageous further developments also arise from the dependent subclaims.
[0011] The inventive method for acquiring a calibration curve for the fill level of a tank utilizes, similarly to prior art, measured values correlated with the fill level, which are also referred to as raw measurement data or ticks. A calibration line is assumed as a linear extrapolation between fill level points and the measured values when the tank is empty and full. Based on the acquired measurement data, the tank contents or fill level, e.g., in percent, can be determined. The measured values are thus converted into the fill level via the calibration line.
[0012] According to the invention, the fuel consumed in the fuel cell is measured and integrated based on a current value from the fuel cell to determine an expected fill level. Based on the deviation between the measured fill level over time and the expected fill level over time, a deviation between these values is determined for each point in time. The linear calibration curve, which is fundamentally similar to the prior art, is then adjusted or corrected to a calibration curve based on these deviations. The calibration curve then represents a relationship, individually determined for each tank, between the measured values and the corresponding fill level. It is typically no longer linear. It should be recorded separately for each individual tank, for example, during its initial commissioning.For this purpose, preferably starting from a fully fueled state to capture the starting point of the calibration curve, the stored fuel can be consumed, preferably completely, to capture the endpoint of the calibration curve. During fuel consumption in the fuel cell, the current value of the fuel cell is simultaneously integrated to determine the amount of fuel consumed very precisely. The linear calibration curve can then be extrapolated between the measured values of the full and empty states, and subsequently adjusted using the measured deviations between the fill level measured by level sensors and the expected fill level calculated from the fuel cell current.This results in a typically non-linear calibration curve that very accurately depicts the relationship between the recorded measurements and the actual fill level for each individually calibrated tank. With such a calibration curve stored individually for each tank, it is then possible to increase the accuracy of the fill level readings to + / - 1 to 2%, i.e., by a factor of 5 to 10 compared to the state of the art.
[0013] So-called boil-off gases, i.e., vaporized gas that is released from the tank at certain pressure limits, are responsible in practice for at least some of the inaccuracies in predicting the fill level. In the method according to the invention, a particularly advantageous embodiment therefore provides that boil-off from the tank is prevented during the acquisition of the calibration curve. This avoids any inaccuracies that could arise during calibration due to escaping boil-off gas. In principle, the optimization of the calibration curve could be carried out in individual ranges of the tank's fill level by comparing the measured fill level values or raw data from the level sensor with the expected fill level based on the integrated current of the fuel cell. This would be possible, for example, when emptying the tank from an 80% fill level to a 40% fill level.However, it is particularly efficient if a fully filled tank is completely emptied during the calibration curve recording process, with the fuel cell consuming all the fuel. Such a complete emptying, starting from a 100% fill level, allows the tank to be individually calibrated in a single, automated process, for example, during its first use.
[0014] If, for example, a vehicle system has multiple fuel tanks, the calibration according to the described procedure is performed on exactly one of these tanks at a time, while the others are shut off. Then, with multiple tanks, each tank is subjected to the procedure sequentially to achieve an ideal and reliable calibration for each individual tank. This ultimately results in a very well-calibrated overall system, which, based on measured values and using the previously recorded and stored calibration curve, allows for a very accurate determination of the actual fill level of each individual tank.
[0015] As already mentioned, according to an advantageous embodiment, calibration can be performed after the initial filling of the fuel cell or in a system with multiple fuel cells. Preferably, the calibration is carried out automatically, in particular by using the fuel from the corresponding fuel cell during its first use, starting from a 100% fill level in that fuel cell, until the fuel cell is completely empty.
[0016] As mentioned above, such a method can be used for any type of tank, and various liquefied gases and, in principle, different consumers are conceivable. However, the method is particularly advantageous for use in a fuel cell vehicle, and especially in a heavy commercial vehicle, which has at least one tank for liquefied hydrogen and is equipped with sensors for recording measurements correlated with the fill level and at least one processing unit configured to carry out the method.Although the vehicle can in principle be any type of vehicle on land, water or in the air, fuel cells currently play a role primarily in non-rail-bound land vehicles, and especially in commercial vehicles, particularly in so-called heavy commercial vehicles, which according to common definitions are vehicles with a permissible total weight of more than 3.51 tons.
[0017] Further advantageous embodiments of the method according to the invention and of the fuel cell vehicle according to the invention can also be seen from the exemplary embodiment, which is described in more detail below with reference to the figures.
[0018] This shows:
[0019] Fig. 1 shows a schematic representation of a fuel cell vehicle;
[0020] Fig. 2 shows a tank system and a fuel cell to illustrate the method according to the invention;
[0021] Fig. 3 shows a calibration line between the raw measurement data of a sensor system and the fill level of the respective tank container;
[0022] Fig. 4 shows a diagram of the fuel level over time during a journey of the fuel cell vehicle; and
[0023] Fig. 5 shows a corrected calibration curve of the respective tank container compared to the calibration line according to Fig. 3.
[0024] Figure 1 shows an example of a heavy commercial vehicle 1 in the form of a tractor unit with a semi-trailer. This heavy commercial vehicle 1 is to be electrically powered, with a fuel cell system 2, designated as , supplying the majority of the required electrical drive power. This fuel cell system 2 is supplied with oxygen from the air and hydrogen from a tank system 3, which is shown here, purely by way of example, in the lower part of the semi-trailer of the heavy commercial vehicle 1. This tank system 3 is a so-called cryogenic tank system 3, which comprises one or more tanks 4 in which the hydrogen is stored in liquid form. The tank system 3 is shown again in a further schematic representation in Figure 2.In the embodiment shown here, it consists of two individual tank containers 4, each of which has a level sensor 5. The level is determined by means of these sensors in a known manner, for example, by capacitive level sensors, by detecting the liquid surface using an ultrasonic sensor, or the like. These raw level measurement data, also referred to as "ticks," are then transmitted to a processing unit 6.
[0025] Similar to the prior art, a characteristic curve 7 can now be determined by linear extrapolation from a completely empty state of the respective tank 4 and a full state of the respective tank 4. This curve is shown in Figure 3. The y-axis represents the fill level L (Level) in percent, and the x-axis represents the measured values or raw measurement data or ticks x. The two recorded points 1 and 11 are labeled x0 on the x-axis at a fill level of 0% and xi at a correspondingly higher fill level. Extrapolation can be performed through these points I and II up to a completely full tank 4, i.e., a fill level L of 100%.
[0026] When the measured values or ticks are transmitted to the processing unit 6, the fill level L of the respective tank 4 can, in principle, be determined via the calibration line 7. However, this is associated with relatively high inaccuracies, as explained in the description at the beginning. For this reason, a fuel cell 9 of the fuel cell system 2, connected to the tank 4 via a pipe 8, is used to improve the calibration of the tank 4 by consuming liquid hydrogen.
[0027] Starting from a fully filled tank 4, an operating mode is initiated via valves present in the piping 8 (not shown here) in which hydrogen is only drawn from one of the two tanks 4 in the example shown in Figure 2. The hydrogen was consumed in the fuel cell 9, for example, during the first drive of the commercial vehicle 1. Points I and II at ticks x0 and xi in the representation of Figure 3, and thus ultimately the characteristic curve 7, can also be recorded. The level sensor 5 of the corresponding tank 4 detects the behavior of the fill level L over time t shown in Figure 4 during the operation of the fuel cell 9, i.e., while the liquid hydrogen from the tank 4 is completely consumed. The solid line A shows the measured values.The line A, labeled A, plots the fill level L over time t based on the number of ticks, for example, during a journey in which the liquid hydrogen in one of the two tanks 4 is completely consumed. The dotted sections of this line A represent time periods without hydrogen consumption, during which the fill level L does not change. In addition to this curve A, which shows the measured fill level L over time t, the amount of hydrogen consumed by the fuel cell 9 can now be determined very precisely using a current sensor 10 in the fuel cell 9. It is sufficient to integrate the current generated by the fuel cell 9 of the fuel cell system 2.
[0028] Starting from an initial quantity, particularly a maximum quantity, of liquid hydrogen that has been filled into tank 4, the remaining quantity and thus the fill level L can be calculated at any given time using the processing unit 6. In Figure 4, this time course of the calculated fill level is labeled B and shown with a dashed line. At each individual time t, there is a more or less significant deviation D between this curve B of the expected fill level L and the measured fill level A. In Figure 4, this deviation D is shown for a time period labeled t as a purely exemplary example. Based on this deviation D at each individual time t, a corresponding adjustment of the calibration curve 7 can be calculated.From the calibration line 7 mentioned at the outset, which is shown in Figure 3, a calibration curve 11 is derived, taking into account these adjustments, which consider external influences such as tolerances in the shape of the tank container and other non-linearities. This calibration curve 11 is shown again together with the calibration line 7 in Figure 5. By using such a calibration curve 11, which can be determined and stored for each individual tank container 4 successively by completely or almost completely emptying the tank container 4 through the consumption of the filled liquid hydrogen via the fuel cell system 2, the prediction of the fill level L for precisely this tank container 4 can be improved after calibration.
[0029] If a tank system 3 now includes several tank containers 4, as indicated in the illustration of Figure 2, then the calibration is carried out first for one of the tank containers 4 and then for the other.
Claims
Daimler Truck AG Oezvatan December 12, 2025 Patent claims 1. Method for recording a calibration curve (11) of the fill level (L) of a tank container (4) for a liquefied gas as fuel for a fuel cell (9), wherein measured values (x) correlating with the fill level (L) are recorded, wherein a calibration line (7) is assumed as a linear extrapolation between two points (I, II) at the measured values (x0, xi) of an empty state and a filled state of the tank container (4) and the associated fill level (L), characterized in that the fuel consumed in the fuel cell (9) is recorded on the basis of an integrated current value of the fuel cell (9), from which an expected fill level (L) is derived, and the calibration line (7) is adapted to a calibration curve (11) on the basis of the deviation (D) between the time course of the measured fill level (A) and the time course of the expected fill level (B).
2. Method according to claim 1, characterized by the fact that During the acquisition of the calibration curve (11), boil-off from the tank container (4) is prevented.
3. Method according to claim 1 or 2, characterized by the fact that The tank container (4) is completely emptied during the acquisition of the calibration curve (11) due to the consumption of fuel in the fuel cell (9).
4. Method according to claim 1, 2 or 3, characterized by the fact that the fuel tank (4) is fully filled by the consumption of fuel in the fuel cell (9) before the calibration curve (11) is recorded.
5. Method according to any one of claims 1 to 4, characterized by the fact that In the case of several tank containers (4), the calibration curve (11) is recorded for exactly one of the tank containers (4), whereby the recording of the calibration curve (11) is carried out for all tank containers (4) one after the other.
6. Method according to any one of claims 1 to 5, characterized by the fact that the calibration curve (11) is recorded after the initial refueling of the (respective) tank (4).
7. Method according to any one of claims 1 to 6, characterized by the fact that the acquisition of the calibration curve (11) is automated.
8. Fuel cell vehicle (1), comprising at least one tank (4) for liquefied hydrogen, with sensors (5, 10) for recording measured values (x) correlating with the fill level (L), and with at least one computing unit (6) which is equipped to carry out the method according to one of claims 1 to 7.
9. Fuel cell vehicle (1) according to claim 8, characterized by its training as a commercial vehicle.
10. Fuel cell vehicle (1) according to claim 8 or 9, characterized by its training as a heavy commercial vehicle.