Method and system for operating a plurality of fuel cell systems

Abstract

The invention presented relates to a method (100) for operating a plurality of fuel cell systems (201, 203), the method (100) comprising: - measuring (101) physical properties of a respective fuel cell system (201, 203) by means of a sensor system of the respective fuel cell system (201, 203), - transmitting (103) fuel cell system data to a central server (205), the fuel cell system data including values measured by the sensor system of the respective fuel cell system (201, 203), - determining (105) settings for the respective fuel cell system (201, 203) by means of the server (205) on the basis of the fuel cell system data and at least one mathematical model executed on the server (205) specifically for the respective fuel cell system (201, 203), and - transmitting (107) the settings to at least one controller (207, 209, 215) for controlling the respective fuel cell system (201, 203).

Description

[0001] R.413844

[0002] - 1 -

[0003] Description

[0004] title

[0005] Method and system for operating a variety of fuel cell systems

[0006] The presented invention relates to a method and a system for operating a plurality of fuel cell systems according to the attached claims.

[0007] State of the art

[0008] Fuel cell systems often use controllers that employ internal models with static parameterization. These are typically validated for a beginning-of-life and end-of-life state.

[0009] Furthermore, methods for predicting system states are known in which model errors are determined as an additive term using a so-called "Learning-Based Predictive Control".

[0010] Numerous methods exist for characterizing the aging states of a fuel cell system.

[0011] Disclosure of the invention

[0012] Within the scope of the presented invention, a method and a system for operating a plurality of fuel cell systems are introduced. Further features and details of the invention will become apparent from the respective dependent claims, the description, and the drawings. Features and details described in connection with the method according to the invention naturally also apply in connection with the invention R.413844.

[0013] - 2 -

[0014] system and vice versa, so that with regard to the disclosure of the individual aspects of the invention, mutual reference is always made or can be made.

[0015] The invention presented here serves in particular to operate a large number of fuel cell systems in an optimized manner throughout their life cycle according to one or more predetermined control objectives.

[0016] Thus, according to a first aspect of the presented invention, a method for operating a large number of fuel cell systems is presented.

[0017] The presented method comprises measuring the physical properties of a respective fuel cell system using sensors of the respective fuel cell system, transmitting fuel cell system data to a central server, wherein the fuel cell system data includes values ​​measured by the sensors of the respective fuel cell system, determining settings for the respective fuel cell system by the server, based on the fuel cell system data and at least one mathematical model specifically executed on the server for the respective fuel cell system, and transmitting the settings to at least one controller for regulating the respective fuel cell system.

[0018] In the context of the presented invention, "settings" refers to control commands or parameters, as well as means for determining control commands or parameters, such as mathematical models or sub-models.

[0019] The presented invention is based on the provision of computing power by a central server for adapting the settings of a fuel cell system to its current state throughout its lifetime. Accordingly, computing units used locally in a given fuel cell system can be designed to be particularly compact and energy-efficient. R.413844

[0020] - 3 -

[0021] The server according to the invention determines settings for a given fuel cell system based on values ​​measured by sensors in the fuel cell system and transmits the settings to a number or at least one controller for regulating the fuel cell system. The determined settings, for example, override the data of a given internal model of a given model predictive controller.

[0022] Such controllers can be, for example, power controllers, temperature controllers, air pressure controllers and / or mass flow controllers.

[0023] To determine the settings, at least one individual mathematical model, adapted to the current state of a respective fuel cell system, is executed on the server, which can be used, for example, to determine a voltage applied to a cell stack of the fuel cell system and / or a total system power of the fuel cell system.

[0024] The fuel cell system data can be preprocessed, for example by filtering, in order to minimize the amount of data required for communication between the server and the fuel cell system.

[0025] For example, the presented method enables specific trajectory planning of setpoints depending on the current state of a respective fuel cell system, in particular control unit functions that plan setpoint profiles as part of an operating strategy. Such setpoint profiles can be, for example, fuel cell power, fuel cell stack temperature, and / or a permissible air pressure or air mass flow in the cathode path.

[0026] In the case of multiple model-based control functions in a given fuel cell system, the underlying models of these control functions can either be set in parallel or as required in individualized increments, or transferred as whole models or sub-models to a respective controller.

[0027] In hierarchical, cascaded, or distributed controller structures with different models at different levels, the models can be updated accordingly. R.413844

[0028] - 4 - It must be ensured that they remain robust and consistent across all levels. For example, models executed at different levels can operate on the same data basis. In particular, models executed at different levels can be derived as sub-models from a central model, so that a change to the central model leads to a change in all sub-models.

[0029] It may be intended that the fuel cell system data include diagnostic data determined based on values ​​measured by the sensors.

[0030] Diagnostic data, such as error messages or parameters determined by diagnostic functions, provide information about the condition of a given fuel cell system. Accordingly, specific settings optimized for the current condition, such as an aging or degradation state of a particular fuel cell system, can be determined based on this diagnostic data.

[0031] It may also be provided that at least one characteristic parameter is determined by means of the model executed on the server, which describes a state of the respective fuel cell system, and that at least one characteristic parameter is transferred as a setting to the at least one controller and is used there to adapt a mathematical model of the at least one controller to the state of the respective fuel cell system.

[0032] Key parameters, such as characteristic values ​​or function parameters, can be directly transferred to a respective controller and used there to control a fuel cell system.

[0033] It may also be provided that the settings include parameters for a mathematical model and / or a mathematical model itself or a sub-model of a mathematical model of a respective model predictive controller.

[0034] By determining parameters or mathematical models or sub-models using the central server, the computing units executing the controllers can be significantly relieved of their workload. R.413844

[0035] - 5 -

[0036] It may also be stipulated that at least one controller is selected from the following list of controllers: power controller, temperature controller, air pressure controller and / or mass flow controller.

[0037] It may also be provided that at least one controller is executed on the server, on a computing unit of a system encompassing the respective fuel cell system and / or a computing unit of the respective fuel cell system.

[0038] In particular, a given controller can be hierarchically organized across different computing units in order to, for example, adjust controlled systems or controlled variables with different time constants at different speeds.

[0039] For example, a power and temperature controller that regulates with a high time constant can be implemented at a higher level than a mass flow and pressure controller that regulates with a small time constant.

[0040] It may also be provided that a large number of controllers are executed on different computing units, with each controller of the large number of controllers being updated in parallel by the settings determined by the server.

[0041] Parallel updates of different controllers on different computing units ensure consistent, i.e., coordinated control of a respective fuel cell system by the different controllers.

[0042] It may also be possible for specific mathematical models to be created by the respective controllers based on the settings for the respective fuel cell system.

[0043] Mathematical models specific to each fuel cell system accurately represent the state of that system, enabling precise adjustments and optimal energy, power, and longevity of operation. R.413844

[0044] - 6 -

[0045] In particular, a power controller set by the presented method or adapted to a current state of a respective fuel cell system leads to a particularly energy-efficient operation of the respective fuel cell system, while in particular a temperature controller set by the presented method or adapted to a current state of a respective fuel cell system enables a particularly component-friendly operation of the respective fuel cell system and, as a result, ensures a long service life of the respective fuel cell system.

[0046] As an alternative to parameters for physical or empirical prediction models, Gaussian processes, neural networks, etc., can also be trained on the server and transmitted as settings to a respective fuel cell system. The behavior of the controlled system can be trained as an absolute model or a total model, or as a difference model in comparison to another, e.g., physical, model, resulting in a hybrid model overall.

[0047] It may also be provided that the respective fuel cell system comprises several fuel cell stacks and that the fuel cell system data and settings are determined specifically for each fuel cell stack.

[0048] By determining fuel cell system data specifically for each fuel cell stack, each fuel cell stack can be operated or adjusted specifically and accordingly, optimized for its respective condition.

[0049] It may also be provided that, in the event that the server determines a state of the respective fuel cell system that corresponds to a predefined error state when determining the settings, a corresponding error message is transmitted to the respective fuel cell system.

[0050] By transmitting an error message from the server to a respective fuel cell system, the respective fuel cell system is particularly affected by R.413844.

[0051] - 7 - quickly, possibly bypassing further process steps, directly informed about its faulty condition and can activate appropriate countermeasures, such as emergency operation.

[0052] It may also be provided that the at least one mathematical model specifically executed on the server for the respective fuel cell system is converted into a simplified metamodel and the settings are determined based on the metamodel.

[0053] A central model is particularly suitable for setting up, and especially for updating, respective models or sub-models of hierarchical model predictive controllers, from which sub-models are derived that only process signals relevant to the respective level.

[0054] It may also be possible to statistically evaluate fuel cell system data from all fuel cell systems of the multitude of fuel cell systems on the server and to output an overview of the state of the multitude of fuel cell systems on an output unit.

[0055] A statistical analysis of all fuel cell system data from all fuel cell systems across a large number of fuel cell systems allows for the identification of outliers and their investigation into potential problems. Furthermore, the large number of fuel cell systems can be centrally monitored.

[0056] According to a second aspect, the presented invention relates to a system for operating a number of fuel cell systems, wherein the system comprises a central server, the central server being configured to determine settings for each fuel cell system based on fuel cell system data of a respective fuel cell system, which includes physical properties of the respective fuel cell system measured by means of sensors of the respective fuel cell system, using at least one mathematical model specifically executed on the server for the respective fuel cell system and to transmit these settings to at least one controller for controlling at least one fuel cell system.

[0057] - 8 - material cell systems of the number of fuel cell systems to be transferred. In particular, the system can be operated by a method according to the first aspect of the invention.

[0058] It may be provided that the system further comprises a number of computing units, wherein each computing unit of the number of computing units is assigned to a respective fuel cell system of the number of fuel cell systems, wherein the respective computing units are configured to execute a number of controllers to control the correspondingly assigned fuel cell system, and wherein the server is configured to transmit settings to the number of controllers in order to update them in parallel.

[0059] Through a parallel, i.e., temporally parallel or synchronous, transmission of settings to different controllers, a coordinated or mutually aligned control of a respective fuel cell system is achieved by the controllers.

[0060] It may also be provided that the number of computing units are part of a respective fuel cell system or part of a system comprising a respective fuel cell system.

[0061] For example, individual controllers can be executed on a vehicle computing unit of a vehicle that includes a respective fuel cell system and interact with other controllers executed on a computing unit of the respective fuel cell system itself in order to regulate the fuel cell system based on settings provided by the central server.

[0062] Advantages described in detail in relation to a method for operating a plurality of fuel cell systems according to the first aspect of the invention apply equally to the system for operating a plurality of fuel cell systems according to the second aspect of the invention and vice versa.

[0063] Further advantages, features and details of the invention will become apparent from the following description, which refers to the drawings R.413844

[0064] - 9 -

[0065] Exemplary embodiments of the invention are described in detail. The features mentioned in the claims and in the description can each be essential to the invention individually or in any combination.

[0066] They each show schematically:

[0067] Figure 1 shows a possible embodiment of the presented method and

[0068] Figure 2 shows a possible embodiment of the presented system.

[0069] Figure 1 shows a method 100 for operating a number of fuel cell systems.

[0070] The method 100 comprises a measurement step 101, in which physical properties of a respective fuel cell system are measured by means of sensors of a respective fuel cell system, and a first transmission step 103, in which fuel cell system data are transmitted to a central server, wherein the fuel cell system data include values ​​measured by the sensors of the respective fuel cell system.

[0071] Furthermore, the procedure 100 includes a determination step 105, in which settings for the respective fuel cell system are determined by the server on the basis of the fuel cell system data and at least one mathematical model specifically executed on the server for the respective fuel cell system, and a second transmission step 107, in which the settings are transferred to at least one controller for regulating the fuel cell system.

[0072] Fig. 2 shows a system 200 for operating a number, here by way of example a large number of fuel cell systems 201 , 203.

[0073] A central server 205 of system 200 is configured to use fuel cell system data from a respective fuel cell system 201, 203 to determine the physical properties of the respective fuel cell system 201, 203 measured by means of sensors of the respective fuel cell system R.413844

[0074] - 10 - include determining settings for the respective fuel cell system 201, 203 using at least one mathematical model 211, 213 executed specifically for the respective fuel cell system 201, 203 on the server 205 using the fuel cell system data and transferring the determined settings to respective controllers 207, 209, 215 to control at least one of the fuel cell systems 201, 203.

[0075] In this process, the server 205 executes a detailed data model 211 for the first fuel cell system 201, e.g., based on a stack voltage, an air pressure and / or a stack temperature, on the basis of which an air pressure and mass flow controller 207 of the first fuel cell system 201 is set.

[0076] The air pressure and air mass flow controller 207 of the first fuel cell system 201 is implemented, for example, on a vehicle control unit of a vehicle comprising the fuel cell system 201 or in a computing unit of the fuel cell system 201 itself.

[0077] For the second fuel cell system 203, the server 205 executes a further detailed data model 213, e.g., based on a stack voltage, an air pressure and / or a stack temperature, on the basis of which an air pressure and air mass flow controller 209 of the second fuel cell system 203 is set.

[0078] The air pressure and air mass flow controller 209 of the second fuel cell system 203 is implemented, for example, on a vehicle control unit of a vehicle comprising the fuel cell system 203 or in a computing unit of the fuel cell system 203 itself.

[0079] In contrast, a power and temperature controller 215 is centrally and jointly implemented on the vehicle control unit for both systems and receives operating parameters directly from the server 205. Accordingly, the power and temperature controller 215, the air pressure and mass airflow controller 209, and R.413844 form

[0080] - 11 - The air pressure and air mass flow controller 207 has a hierarchical controller architecture, which is implemented as an example on three different computing units and is configured by the central server 205. For this purpose, the server 205 executes models 211 and 213, one of which is the

[0081] One sub-model represents the stack voltage U(i, T) as a function of current and temperature, and another represents the net electrical power of the fuel cell system P(i, T) as a function of current and temperature. Using these sub-models, the efficiency and waste heat generation of the fuel cell system 201 can be assessed at an operating point, which is why they are used in particular to adjust the power and temperature controller 215 and the air pressure and mass flow controller 209 or the air pressure and mass flow controller 207.

Claims

R.413844 - 12 - Claims 1. Method (100) for operating a number of fuel cell systems (201 , 203), wherein the method (100) comprises: - Measuring (101) physical properties of a respective fuel cell system (201 , 203) using sensors of the respective fuel cell system (201 , 203), - Transmitting (103) fuel cell system data to a central server (205), wherein the fuel cell system data includes values ​​measured by the sensors of the respective fuel cell system (201, 203), - Determining (105) settings for the respective fuel cell system (201, 203) by the server (205) based on the fuel cell system data and at least one mathematical model specifically executed on the server (205) for the respective fuel cell system (201, 203) and - Transferring (107) the settings to at least one controller (207, 209, 215) to control at least one fuel cell system of the number fuel cell systems (201, 203).

2. Method (100) according to claim 1, characterized in that the fuel cell system data comprise diagnostic data determined on the basis of values ​​measured by the sensors.

3. Method (100) according to claim 1 or 2, characterized in that at least one characteristic parameter is determined by means of the model executed on the server (205) which describes a state of the respective fuel cell system (201 , 203) and which the at least one characteristic parameter is transmitted as a setting to the at least one controller (207, 209, 215). R.413844 - 13 - is used there to adapt a mathematical model of the at least one controller (207, 209, 215) to the state of the respective fuel cell system (201 , 203).

4. Method (100) according to one of the preceding claims, characterized in that the settings comprise parameters for a mathematical model and / or a mathematical model itself or a sub-model of a mathematical model of a respective model predictive controller (207, 209, 215).

5. Method (100) according to claim 4, characterized in that the at least one controller (207, 209, 215) is selected from the following list of controllers: power controller, temperature controller, air pressure controller and / or mass flow controller.

6. Method (100) according to one of the preceding claims, characterized in that the at least one controller (207, 209, 215) is implemented on the server (205), on a computing unit of a system comprising the respective fuel cell system (201, 203) and / or a computing unit of the respective fuel cell system (201, 203).

7. Method (100) according to one of the preceding claims, characterized in that a plurality of controllers (207, 209, 215) is executed on different computing units, wherein each controller (207, 209, 215) of the plurality of controllers (207, 209, 215) is updated in parallel by the settings determined by means of the server (205).

8. Method (100) according to claim 7, characterized in that, R.413844 - 14 - that specific mathematical models are created for each controller (207, 209, 215) based on the settings for the respective fuel cell system (201, 203).

9. Method (100) according to one of the preceding claims, characterized in that the respective fuel cell system (201 , 203) comprises several fuel cell stacks and the fuel cell system data and settings are determined specifically for each fuel cell stack.

10. Method (100) according to one of the preceding claims, characterized in that, in the event that, when determining the settings by the server (205), a state of the respective fuel cell system (201 , 203) is determined which corresponds to a predetermined error state, a corresponding error message is transmitted to the respective fuel cell system (201 , 203).

11. Method (100) according to one of the preceding claims, characterized in that the at least one mathematical model specifically executed on the server (205) for the respective fuel cell system (201, 203) is converted into a simplified metamodel and the settings are determined on the basis of the metamodel or comprise a metamodel.

12. Method (100) according to one of the preceding claims, characterized in that fuel cell system data from all fuel cell systems (201 , 203) of the plurality of fuel cell systems (201 , 203) are statistically evaluated on the server (205) and an overview of the state of the plurality of fuel cell systems (201 , 203) is output on an output unit. R.413844 - 15 - 13. System (200) for operating a number of fuel cell systems (201, 203), wherein the system (200) comprises: a central server (205), wherein the central server (205) is configured to determine settings for each fuel cell system (201, 203) based on fuel cell system data, which includes physical properties of each fuel cell system (201, 203) measured by means of sensors of the respective fuel cell system (201, 203), using at least one mathematical model specifically executed on the server (205) for each fuel cell system (201, 203), and to transmit the determined settings to at least one controller (207, 209, 215) for controlling at least one fuel cell system of the number of fuel cell systems (201, 203).

14. System (200) according to claim 13, characterized in that the system (200) further comprises a number of computing units, wherein each computing unit of the number of computing units is assigned to a respective fuel cell system (201, 203) of the number of fuel cell systems (201, 203), and wherein the respective computing units are configured to execute a number of controllers (207, 209, 215) for controlling the correspondingly assigned fuel cell system (201, 203), and wherein the server (205) is configured to transmit settings to the number of controllers (207, 209, 215) in order to update them in parallel. R.413844 - 16 - 15. System (200) according to claim 14, characterized in that the number of computing units are part of a respective fuel cell system (201 , 203) or part of a system comprising a respective fuel cell system (201 , 203).

Images