DE102022204020A1 - Method for operating at least one fuel cell system using a combined operating strategy - Google Patents

Abstract

The invention relates to a method for operating at least one fuel cell system (100), in particular for a vehicle (101), using a combined, preferably time scale-based, operating strategy (MPC), comprising: - Providing at least two operating strategies (MPC1, MPC2, .. ., MPCN) for operating the fuel cell system (100), the operating strategies having different dynamics (D1, D2, ..., DN), - evaluating operating strategies (MPC1, MPC2, ..., MPCN) depending on corresponding dynamics (D1, D2, ..., DN), - Operating the at least one fuel cell system (100) according to an operating strategy (MPCi) with a, in particular low, dynamic (Di), - Providing framework conditions (Ri), in particular for manipulated variables, State variables, result variables, etc., according to the operating strategy (MPCi) to a further operating strategy (MPCi+1) with a further, in particular faster, dynamic (Di+1), - operating the at least one fuel cell system (100) according to the further operating strategy (MPCi +1) with a further, in particular faster, dynamic (Di+1) taking into account general conditions (Ri) according to the operating strategy (MPCi).

Description

The invention relates to a method for operating at least one fuel cell system, in particular for a vehicle, using a combined, preferably time scale-based, operating strategy. The invention further relates to a corresponding computer program product. The invention further relates to a corresponding control unit for carrying out a corresponding method. The invention also relates to a corresponding fuel cell system with a corresponding control unit.

State of the art

• - flexible Bereitstellung von elektrischer Leistung (eng. „performance“), insbesondere unter dynamischen Bedingungen, bspw. in einem mobilen Betrieb eines Fahrzeuges,
• - Minimierung von Verbrauch an Treibstoffen, insbesondere von Brennstoffen, wie z. B. Wasserstoff,
• - Sicherstellung von Funktionalitäten unter variablen (weltweiten) Betriebsbedingungen, wie z. B: Gefrierstart, Heißland-Fahrt, Bergfahrt, Fahrt mit Anhänger, etc.,
• - Sicherstellung von Funktionalitäten über eine Betriebszeit, d. h. im Neuzustand wie im gealterten Zustand,
• - Minimierung von Alterung/Degradation-Effekten in einem Brennstoffzellenstack,
• - Minimierung von Alterung/Degradation-Effekten in Batterien, insbesondere in einer HV-Batterie,
• - Minimierung von Alterung/Degradation-Effekten in weiteren Komponenten des Brennstoffzellensystems, wie z.B. in einem Luftverdichter eines Luftsystems des Brennstoffzellensystems,
• - Komfortabler Betrieb in einem Fahrzeug,
• - Vermeidung von unangenehmen Geräuschen für die Insassen und auch für das Umfeld eines Fahrzeuges, etc.

For vehicles (fuel cell vehicle or FCV) with drive systems with one or more fuel cell systems (fuel cell system or FCS), selecting an operating strategy is important. Well-known operating strategies are often map-based, based on only one time scale, only one operating target, e.g. consumption. Different operating strategies have different operating cells:

• - flexible provision of electrical power (“performance”), especially under dynamic conditions, for example in mobile vehicle operation,
• - Minimizing fuel consumption, especially fuels such as: B. hydrogen,
• - Ensuring functionalities under variable (global) operating conditions, such as: B: Freeze start, hot country driving, mountain driving, driving with trailer, etc.,
• - Ensuring functionalities over an operating period, i.e. in new condition as well as in aged condition,
• - Minimization of aging/degradation effects in a fuel cell stack,
• - Minimization of aging/degradation effects in batteries, especially in an HV battery,
• - Minimization of aging/degradation effects in other components of the fuel cell system, such as in an air compressor of an air system of the fuel cell system,
• - Convenient operation in a vehicle,
• - Avoidance of unpleasant noises for the occupants and also for the surroundings of a vehicle, etc.

With different operating strategies and/or operating goals, it is challenging to take the different system dynamics into account in a complex system, starting from sensor detection, processing, etc., up to the control/regulation of different actuators with different dynamics, dead times, etc.

Disclosure of the invention

According to the first aspect, the present invention provides: a method for operating at least one fuel cell system, in particular for a vehicle, using combined operating strategies. According to the second aspect, the present invention provides: a computer program product with the features of the independent product claim. According to the third aspect, the present invention provides: a control unit with the features of the independent device claim. According to the fourth aspect, the present invention provides: a fuel cell system with the features of the independent device claim. Features and details that are described in connection with different embodiments and/or aspects of the invention naturally also apply in connection with the other embodiments and/or aspects and vice versa, so that with regard to the disclosure of the individual embodiments and/or aspects always is or can be mutually referenced.

• - Bereitstellen von mindestens zwei Betriebsstrategien zum Betreiben des Brennstoffzellensystems, wobei die Betriebsstrategien unterschiedliche Dynamiken aufweisen,
• - Bewerten (bspw. Anordnen, Sortieren, Systematisieren, Optimieren, Zusammenstellen und/oder Gruppieren) von Betriebsstrategien in Abhängigkeit von korrespondierenden Dynamiken,
• - Betreiben des mindestens einen Brennstoffzellensystems gemäß einer Betriebsstrategie mit einer, insbesondere niedrigen, Dynamik,
• - Bereitstellen von Rahmenbedingungen (Constraints/Limitierungen), insbesondere für Stellgrößen, Zustandsgrößen, Ergebnisgrößen usw., gemäß der ersten Betriebsstrategie an eine weitere Betriebsstrategie mit einer weiteren, insbesondere schnelleren, Dynamik,
• - Betreiben des mindestens einen Brennstoffzellensystems gemäß der weiteren Betriebsstrategie mit einer zweiten, insbesondere schnelleren, Dynamik unter Berücksichtigung von Rahmenbedingungen gemäß der, insbesondere langsameren, Betriebsstrategie.

The invention provides: Method for operating at least one or more fuel cell systems, in particular for a mobile application, for example in a vehicle, using a combined, preferably time scale-based, operating strategy (an exemplary control technology method, which has a combined, preferably time scale-based , operational strategy can be, for example: MPC process (eng. model predictive control); MHE process (eng. moving horizon estimation) or similar), having the following actions/process steps:

• - Providing at least two operating strategies for operating the fuel cell system, the operating strategies having different dynamics,
• - Evaluating (e.g. arranging, sorting, systematizing, optimizing, compiling and/or grouping) operating strategies depending on corresponding dynamics,
• - Operating the at least one fuel cell system according to an operating strategy with, in particular, low dynamics,
• - Providing framework conditions (constraints/limitations), especially for manipulated variables, state variables and result variables etc., according to the first operating strategy to a further operating strategy with a further, in particular faster, dynamic,
• - Operating the at least one fuel cell system according to the further operating strategy with a second, in particular faster, dynamic, taking into account general conditions according to the, in particular slower, operating strategy.

The last three or the last four steps of the method can, for example, be carried out together, simultaneously, time-synchronized, and/or iteratively from i=1, 2, ..., N - 1.

An operating strategy within the scope of the present disclosure is an optimization-based technical control method and/or regulation method, which can have different actions/procedural steps, starting from sensor detection, processing, etc., to control/regulation of various actuators in the fuel cell system.

An operating strategy may relate to a fuel cell system as a whole or to one or more subsystems of a fuel cell system, such as: B. an air system, an anode system, a cooling system and / or an electrical system. An operating strategy can also apply across multiple fuel cell systems.

• - modellprädiktive Regelung von elektrischen Größen (Spannung, Strom, elektrische Leistung),
• - modellprädiktive Regelung der gasförmigen und/oder fluidischen Strömungen/Medien, bspw. in einem Luftsystem (zur Sauerstoffversorgung eines Brennstoffzellenstacks) und/oder einem Wasserstoffsystem (zur Wasserstoffversorgung des Brennstoffzellenstacks),
• - modellprädiktive Regelung der flüssigen und/oder fluidischen Strömungen/Medien, bspw. in einem Kühlsystem und ggf. für die Wärmeübergänge im System (Wärmeübertrager, Pumpen, etc.),
• - modellprädiktive Regelung eines Wassermangementsystems (zur Regelung der Membranfeuchte bzw. des Membranwassergehalts),
• - modellprädiktive Regelung eines kurz-, mittel- und/oder langfristigen Stack-Betriebs- und/oder Alterungsverhaltens (zur Regelung reversibler Effekte und/oder zur Durchführung von Regenerationsmaßnahmen, Anpassungen der Betriebsstrategie, Lastprofile),
• - modellprädiktive Regelung langfristiges Stack-Betriebs- und Alterungsverhalten.

The operating strategies within the scope of the present disclosure may include the following control methods and/or regulation methods:

• - model-predictive control of electrical variables (voltage, current, electrical power),
• - model-predictive control of the gaseous and/or fluidic flows/media, for example in an air system (for supplying oxygen to a fuel cell stack) and/or a hydrogen system (for supplying hydrogen to the fuel cell stack),
• - model-predictive control of the liquid and/or fluid flows/media, for example in a cooling system and, if necessary, for the heat transfers in the system (heat exchangers, pumps, etc.),
• - model predictive control of a water management system (to control membrane moisture or membrane water content),
• - model-predictive control of short-, medium- and/or long-term stack operating and/or aging behavior (to control reversible effects and/or to carry out regeneration measures, adjustments to the operating strategy, load profiles),
• - Model predictive control of long-term stack operating and aging behavior.

• - Zeitschrittweite (Tstep), bspw. zur Berechnung der Modelle, zur Optimierung, zur Aktor-Ansteuerung, zur Sensorerfassung, und/oder
• - Zeithorizonte (Tpred), bspw. Zeitspanne der Prädiktion, die für die Betriebsführung incl. Optimierung in der jeweiligen zeitskalierten Betriebsstrategie verwendet wird.

In the context of the present disclosure, a dynamic for the respective operating strategy can be understood as a time scale of a corresponding operating strategy, which can have the following time parameters:

• - Time step width (Tstep), for example for calculating the models, for optimization, for actuator control, for sensor detection, and/or
• - Time horizons (Tpred), e.g. period of prediction, which is used for operational management including optimization in the respective time-scaled operating strategy.

Under a dynamic for the respective operating strategy, a quality function or a cost function, which can be constant and/or variable, and/or one or more operating goals of a corresponding operating strategy, also depending on operating modes, can also be taken into account.

The dynamics that can be taken into account in the overall control for one or more fuel cell systems and also in a fuel cell vehicle range from milliseconds to years.

With the help of the invention, different operating strategies can be flexibly optimized for the different dynamic classes in a fuel cell system or in several fuel cell systems, with one or more energy storage devices, and also in the entire mobile vehicle application including vehicle movement control/regulation.

The method provides multiple timescale-based operational strategies that are evaluated, sorted, systematized, optimized, compiled and/or grouped according to their dynamics. Operating strategies for processes with similar dynamic behavior and, if necessary, with corresponding framework conditions (so-called constraints and/or limitations) are assigned to the same dynamic class. When carrying out the method for operating multiple fuel cell systems, different operating strategies can be sorted, systematized, optimized, compiled and/or grouped across systems.

The idea of the invention is that an operating strategy with a slower dynamic behavior meets its specifications Operating strategy communicated with faster dynamic behavior. In this way, the specifications of the operating strategy with a slower dynamic behavior can be taken into account in the operating strategy with a faster dynamic behavior.

• - die Dynamiken der unterschiedlichen Subsysteme aufeinander optimal abgestimmt und deren Potential ausgenutzt werden,
• - eine optimierte oder sogar optimale bzw. pareto-optimale Betriebsführung zur Erreichung von unterschiedlichen Betriebszielen bereitgestellt werden, und
• - eine adaptive Betriebsführung (incl. Anpassung von Dynamiken) erreicht werden.

Using the procedure you can:

• - the dynamics of the different subsystems are optimally coordinated and their potential is exploited,
• - an optimized or even optimal or Pareto-optimal operational management is provided to achieve different operational goals, and
• - Adaptive operational management (including adaptation of dynamics) can be achieved.

The method makes it possible to arrive at an (approximately global) optimal operating strategy for one or more fuel cell systems in a mobile application, including adaptation over lifetime, system and environmental conditions and various use cases.

• - Optimierung von Rechenzeit,
• - Durchführung in Echtzeit, hierzu können die jeweiligen Betriebsstrategien bspw. in einem jeweils ausreichenden Zeitraster bewertet werden,
• - Reduktion von Komplexität bei der Systemregelung,
• - Einfachere Übertragbarkeit auf Topologie-Varianten,
• - Adaptionsmöglichkeiten individuell je nach Dynamik,
• - Adaption über Lebenszeit, Anwendungsfall, System- und Umgebungsbedingungen, Infrastruktur, gesetzliche Vorgaben, etc.,
• - Berücksichtigung konstanter und variabler Rahmenbedingungen/Constraints/Limitierungen,
• - Dynamikübergreifende Berücksichtigung von Rahmenbedingungen über mehrere Ebenen,
• - Optimierte oder sogar optimale kombinierte Betriebsstrategie zur bestmöglichen Erreichung von unterschiedlichen Betriebszielen auf allen relevanten Ebenen,
• - Berücksichtigung konstanter und variabler Kostenfunktionen,
• - Betriebsmodi abhängige Kostenfunktionen bzw. Optimierungen,
• - Möglichkeit der Aufteilung auf unterschiedliche Steuergeräte. Ferner kann das Verfahren mindestens einen weiteren Schritt aufweisen:
• - Bereitstellen von Informationen, insbesondere Zustandsgrößen, Ergebnisgrößen usw., gemäß der weiteren Betriebsstrategie an die, insbesondere langsamere, Betriebsstrategie.

Several advantages can be achieved with the help of the invention:

• - optimization of computing time,
• - Implementation in real time, for this purpose the respective operating strategies can be evaluated, for example, in a sufficient time frame,
• - Reduction of complexity in system control,
• - Easier transferability to topology variants,
• - Adaptation options individually depending on the dynamics,
• - Adaptation over lifetime, use case, system and environmental conditions, infrastructure, legal requirements, etc.,
• - Consideration of constant and variable framework conditions/constraints/limitations,
• - Cross-dynamic consideration of framework conditions across multiple levels,
• - Optimized or even optimal combined operating strategy to best achieve different operating goals at all relevant levels,
• - Consideration of constant and variable cost functions,
• - Operating mode-dependent cost functions or optimizations,
• - Possibility of splitting between different control devices. Furthermore, the method can have at least one further step:
• - Providing information, in particular state variables, result variables, etc., in accordance with the further operating strategy to the, in particular slower, operating strategy.

In this way, an information flow from the second operating strategy with a further, in particular faster, dynamic to the, in particular slower, operating strategy with a, in particular low, dynamic can be made possible. In this way, the operating strategy, particularly the slower one, can take into account the information from the further, particularly faster, operating strategy and adapt its own strategy if necessary.

Furthermore, it can be provided that when evaluating the operating strategies depending on corresponding dynamics, different time step sizes and/or time horizons of the operating strategies are taken into account. In this way, the dynamics of operational strategies can be evaluated, sorted, systematized, optimized, compiled and/or grouped in an advantageous manner.

Furthermore, it can be provided that when evaluating the operating strategies depending on corresponding dynamics, different, in particular constant and/or variable, quality functions and/or operating objectives of the operating strategies are taken into account. In this way, different operating goals can be taken into account at all relevant dynamic levels and depending on the operating mode.

The method can advantageously be carried out for several fuel cell systems.

It is also conceivable that framework conditions and/or information are exchanged between two adjacent operating strategies. In other words, it is conceivable that framework conditions and/or information are exchanged between operating strategies in a cascade manner. In this way, a hierarchical consideration of framework conditions and/or information from operating strategy to operating strategy can be made possible, depending on the dynamics of the respective operating strategies.

In addition, it is conceivable that framework conditions and/or information between several, preferably all, operating strategies arranged one after the other can be exchanged. It is even further conceivable that framework conditions and/or information are exchanged broadcast and/or peer-to-peer between operating strategies. This allows the availability of framework conditions and/or information across multiple dynamic levels.

According to the second aspect, the present invention provides: a computer program product comprising instructions which, when the computer program product is executed by a computer, cause the computer to carry out a method which can run as described above. With the help of the computer program product according to the invention, the same advantages are achieved that were described above in connection with the method according to the invention. These advantages are fully referenced here.

According to the third aspect, the present invention provides: a control unit, comprising a computing unit and a storage unit in which a code is stored, which, when at least partially executed by the computing unit, carries out a method which can run as described above. With the help of the control unit according to the invention, the same advantages are achieved that were described above in connection with the method according to the invention. These advantages are fully referenced here.

The control unit can act as a central control unit of a vehicle, a common control unit for at least one or more fuel cell systems, as a local control unit for at least one or more subsystems in one or more fuel cell systems and/or as a distributed control unit.

Advantageously, it can be provided in the control unit that the computing power for carrying out the method is at least partially outsourced to an external computing unit, in particular in the case of operating strategies with low dynamics.

The present invention, according to the fourth aspect, provides: a fuel cell system comprising a control unit which can be implemented as described above. With the help of the fuel cell system according to the invention, the same advantages are achieved that were described above in connection with the method according to the invention. These advantages are fully referenced here.

Preferred embodiments:

• 1 eine beispielhafte Struktur mit multiplen zeitskalenbasierten Betriebsstrategien mit verschiedenen Dynamiken,
• 2 eine weitere beispielhafte Struktur mit multiplen zeitskalenbasierten Betriebsstrategien mit verschiedenen Dynamiken, und
• 3 eine beispielhafte Struktur mit multiplen zeitskalenbasierten Betriebsstrategien mit verschiedenen Dynamiken bei Anwendungen mit mehreren Brennstoffzellensystemen bzw. mehreren Bren nstoffzel lenstacks.

The invention and its further developments as well as their advantages are explained in more detail below with reference to drawings. They show schematically:

• 1 an exemplary structure with multiple timescale-based operating strategies with different dynamics,
• 2 another exemplary structure with multiple timescale-based operating strategies with different dynamics, and
• 3 an exemplary structure with multiple timescale-based operating strategies with different dynamics in applications with multiple fuel cell systems or multiple fuel cell stacks.

In the different figures, the same parts of the invention are always provided with the same reference numerals, which is why they are generally. d. R. can only be described once.

The 1 until 3 serve to explain a method which was developed for operating at least one or more fuel cell systems 100, in particular for a mobile application, for example in vehicles 101. The method is based on a combined, preferably time scale-based, operating strategy MPC, which is carried out, for example, by means of an MPC method (model predictive control), MHE method (moving horizon estimation) or similar. can be implemented. The combined, preferably time scale-based, operating strategy MPC can be printed out as a function of several operating strategies MPC1, MPC2, ..., MPCN.

• - Bereitstellen von mindestens zwei Betriebsstrategien MPC1, MPC2, ..., MPCN zum Betreiben des Brennstoffzellensystems 100, wobei die Betriebsstrategien unterschiedliche Dynamiken D1, D2, ..., DN aufweisen,
• - Bewerten, umfassend bspw. Anordnen, Sortieren, Optimieren, Systematisieren, Zusammenstellen und/oder Gruppieren, von Betriebsstrategien MPC1, MPC2, ..., MPCN in Abhängigkeit von korrespondierenden Dynamiken D1, D2, ..., DN,
• - Betreiben des mindestens einen Brennstoffzellensystems 100 gemäß einer Betriebsstrategie MPCi mit einer, insbesondere niedrigen, Dynamik Di,
• - Bereitstellen von Rahmenbedingungen Ri (Constraints/Limitierungen), insbesondere für Stellgrößen, Zustandsgrößen, Ergebnisgrößen usw., gemäß der Betriebsstrategie MPCi an eine weitere Betriebsstrategie MPCi+1 mit einer weiteren, insbesondere schnelleren, Dynamik D i+1,
• - Betreiben des mindestens einen Brennstoffzellensystems 100 gemäß der weiteren Betriebsstrategie MPC i+1 mit einer weiteren, insbesondere schnelleren, Dynamik D i+1 unter Berücksichtigung von Rahmenbedingungen Ri gemäß der, insbesondere langsameren, Betriebsstrategie MPCi.

The process has the following actions/process steps:

• - Providing at least two operating strategies MPC1, MPC2, ..., MPCN for operating the fuel cell system 100, the operating strategies having different dynamics D1, D2, ..., DN,
• - Evaluate, including, for example, arranging, sorting, optimizing, systematizing, compiling and/or grouping, operating strategies MPC1, MPC2, ..., MPCN depending on corresponding dynamics D1, D2, ..., DN,
• - Operating the at least one fuel cell system 100 according to an operating strategy MPCi with a, in particular low, dynamic Di,
• - Providing framework conditions Ri (constraints/limitations), especially for Control variables, state variables, result variables, etc., according to the operating strategy MPCi to a further operating strategy MPCi+1 with a further, in particular faster, dynamic D i+1,
• - Operating the at least one fuel cell system 100 according to the further operating strategy MPC i+1 with a further, in particular faster, dynamic D i+1, taking into account general conditions Ri according to the, in particular slower, operating strategy MPCi.

The last three or the last four steps of the method can, for example, be carried out together, simultaneously, time-synchronized, and/or iteratively from i=1, 2, ..., N - 1.

An operating strategy MPC1, MPC2, ..., MPCN in the context of the present disclosure can be understood to mean an optimization-based technical control method and / or regulation method, which can have different actions / method steps, including, for example, sensor detection, processing, etc., up to and including control /Control of various actuators in the fuel cell system.

An operating strategy MPC1, MPC2, ..., MPCN may relate to a fuel cell system 100 as a whole or to one or more subsystems of a fuel cell system 100, such as. B. an air system, an anode system, a cooling system and / or an electrical system. However, an operating strategy can also apply across multiple fuel cell systems 100.

• MPC1 modellprädiktive Regelung von elektrischen Größen (Spannung, Strom, elektrische Leistung),
• MPC2 modellprädiktive Regelung der gasförmigen und/oder fluidischen Strömungen/Medien, bspw. in einem Luftsystem (zur Sauerstoffversorgung eines Brennstoffzellenstacks) und/oder einem Wasserstoffsystem (zur Wasserstoffversorgung des Brennstoffzellenstacks),
• MPC3 modellprädiktive Regelung der flüssigen und/oder fluidischen Strömungen/Medien, bspw. in einem Kühlsystem,
• MPC4 modellprädiktive Regelung eines Wassermangementsystems (zur Regelung der Membranfeuchte bzw. des Membranwassergehalts),
• MPC5 modellprädiktive Regelung eines kurz-, mittel- und/oder langfristigen Stack-Betriebs- und/oder Alterungsverhaltens (zur Regelung reversibler Effekte und/oder zur Durchführung von Regenerationsmaßnahmen, Anpassungen der Betriebsstrategie, Lastprofile),
• MPC6 modellprädiktive Regelung langfristiges Stack-Betriebs- und Alteru ngsverhalten

The operating strategies MPC1, MPC2, ..., MPCN within the scope of the present disclosure may include the following control methods and/or regulation methods:

• MPC1 model predictive control of electrical variables (voltage, current, electrical power),
• MPC2 model predictive control of the gaseous and/or fluidic flows/media, for example in an air system (to supply oxygen to a fuel cell stack) and/or a hydrogen system (to supply hydrogen to the fuel cell stack),
• MPC3 model predictive control of liquid and/or fluidic flows/media, for example in a cooling system,
• MPC4 model predictive control of a water management system (for controlling membrane moisture or membrane water content),
• MPC5 model predictive control of short-, medium- and/or long-term stack operating and/or aging behavior (to control reversible effects and/or to carry out regeneration measures, adjustments to the operating strategy, load profiles),
• MPC6 model predictive control long-term stack operating and aging behavior

• - Zeitschrittweite Tstep, bspw. zur Berechnung der Modelle, zur Optimierung, zur Aktor-Ansteuerung, zur Sensorerfassung, und/oder
• - Zeithorizonte Tpred, bspw. Zeitspanne der Prädiktion, die für die Betriebsführung incl. Optimierung in der jeweiligen zeitskalierten Betriebsstrategie verwendet wird.

A dynamic D1, D2, ..., DN for the respective operating strategy MPC1, MPC2, ..., MPCN can be understood as a time scale of a corresponding operating strategy MPC1, MPC2, ..., MPCN, comprising, for example, the following time parameters:

• - Time step width Tstep, for example for calculating the models, for optimization, for actuator control, for sensor detection, and/or
• - Time horizons Tpred, e.g. period of prediction that is used for operational management including optimization in the respective time-scaled operating strategy.

Under a dynamic D1, D2, ..., DN for the respective operating strategy MPC1, MPC2, ..., MPCN can also be a quality function or a cost function, which can be constant and / or variable, and / or one or more operating goals a corresponding operating strategy, also depending on operating modes, must be taken into account.

The dynamics D1, D2, ..., DN, which can be taken into account in the overall control for one or more fuel cell systems 100 and also in a fuel cell vehicle, range from milliseconds to years.

Advantageously, using the method, different operating strategies MPC1, MPC2, ..., MPCN can be flexibly implemented for the different dynamic classes in a fuel cell system 100 or in several fuel cell systems 100, with one or more energy storage devices, and also in the entire mobile vehicle application including vehicle movement control/regulation be combined in an optimized manner.

The method provides multiple timescale-based operating strategies MPC1, MPC2, ..., MPCN, which are evaluated, sorted, optimized, systematized, compiled and/or grouped according to their dynamics. Operating strategies MPC1, MPC2, ..., MPCN for processes with similar dynamic behavior and, if necessary, with corresponding framework conditions (so-called constraint specifications and/or limitations) can be assigned to the same dynamic class. The implementation of the method for operating multiple fuel cell systems 100 can involve different operating strategies gies and/or limitations can be sorted, systematized, optimized, compiled and/or grouped across systems.

Thereafter, an operating strategy MPCi (i = 1, 2, ..., N - 1) with a slower dynamic behavior can transmit its specifications to an operating strategy MPCi+1 with a faster dynamic behavior, as in the 1 until 3 indicated by arrows. In this way, the specifications of the MPCi operating strategy with a slower dynamic behavior can be taken into account in the MPCi+1 operating strategy with a faster dynamic behavior.

Using the process, the dynamics D1, D2, ..., DN of different subsystems can be optimally coordinated with one another. With the help of the method, optimized or even optimal or Pareto-optimal operational management can be provided to achieve different operational goals. The process can be used to achieve adaptive operational management (including adjustment of dynamics). The process also enables inclusive adaptation or adaptation over lifetime, system and environmental conditions and various use cases.

• - unterschiedliche Systemdynamiken D1, D2, ..., DN in unterschiedlichen Klassen 1, 2, ..., N aufzuteilen, die in unterschiedlichen bzw. multiplen Zeitskalen:
  • - die Aktionen, wie z. B. Berechnung, Optimierung, Messung, Abtastung, Trajektorienplanung, Vorsteuerung, Ansteuerung, Regelung, Diagnosen, Beobachterfunktionen, Plausibilisierungen, etc. durchführen (subsummiert unter Zeitschrittweite Tstep) und
  • - auch einen jeweils eigenen Horizont prädizieren (subsummiert unter Zeithorizonte Tpred) und in den jeweiligen Optimierungen berücksichtigen.

In other words, the concept according to the invention suggests:

• - divide different system dynamics D1, D2, ..., DN into different classes 1, 2, ..., N, which are in different or multiple time scales:
  • - the actions, such as B. carry out calculation, optimization, measurement, scanning, trajectory planning, pre-control, control, regulation, diagnoses, observer functions, plausibility checks, etc. (subsumed under time step width Tstep) and
  • - also predict their own horizon (subsumed under time horizons Tpred) and take them into account in the respective optimizations.

The time step width Tstep (also called Tsample) and the respective predictive horizon Tpred (for prediction) can be defined separately for each operating strategy MPC1, MPC2, ..., MPCN or each system dynamic.

This means that only as much computing power as necessary is invested specifically for the relevant system dynamics D1, D2, ..., DN.

Real-time implementation in control devices 102 in the vehicle is possible. Slower operating strategies MPC1, MPC2, ..., MPCN or slower dynamics D1, D2, ..., DN can optionally be calculated on an external server, for example in the cloud.

• Die durchgezogenen Pfeile bedeuten die Vorgaben der „langsamer“ getakten Betriebsstrategien MPC1, MPC2, ..., MPCN bzw. der Betriebsstrategien MPC1, MPC2, ..., MPCN mit geringerer Dynamik. Diese Daten enthalten insbesondere Rahmenbedingungen Ri (Constraints/Limitierungen), insbesondere für Stellgrößen, Zustandsgrößen, Ergebnisgrößen usw., die von den Vorgängen mit geringerer Dynamik bestimmt werden.

The multiple timescale-based operating strategies MPC1, MPC2, ..., MPCN are coupled as in the 1 until 3 shown as an example and described below:

• The solid arrows mean the specifications of the “slower” clocked operating strategies MPC1, MPC2, ..., MPCN or the operating strategies MPC1, MPC2, ..., MPCN with lower dynamics. This data contains in particular framework conditions Ri (constraints/limitations), in particular for manipulated variables, state variables, result variables, etc., which are determined by the processes with lower dynamics.

The faster operating strategy MPC1, MPC2, ..., MPCN with faster dynamics D1, D2, ..., DN cannot significantly influence these framework conditions Ri (constraints/limitations) in their dynamics D1, D2, ..., DN, so that these can be taken into account as framework conditions Ri.

In addition to the general conditions Ri, further data can be specified.

Framework conditions Ri can also be passed on from a slower/higher level via “passing through” data to a faster/lower level.

The arrows shown in dashed lines mean the feedback of the “faster” clocked operating strategies MPC1, MPC2, ..., MPCN or the operating strategies MPC1, MPC2, ..., MPCN with higher dynamics. This data contains, for example, the current performance, also the predictive performance for a certain horizon, the accruing costs or input data for the cost functions of the connected operating strategies MPC1, MPC2, ..., MPCN, states of components/subsystems/systems may be relevant for the MPC with slower dynamics than information.

Data can also be passed on in this direction across multiple levels.

Advantageously, cost functions/quality functions and/or operating goals can now also be applied individually for the respective dynamics D1, D2, ..., DN. Thus, a respective optimized operational management can be implemented for the respective system dynamics D1, D2, ..., DN, although goals for different dynamics D1, D2, ..., DN can be defined differently.

• - Zeitschrittweite Tstep,
• - Zeithorizonte Prädiktion Tpred,
• - Gütefunktion bzw. Betriebsziele (bspw.: jeweilige Alterung der beteiligten Komponenten, elektrischer Leistungsverbrauch der jeweiligen Aktoren, Abweichung zwischen Soll- und Istwerten, NVH, etc.), wobei diese auch betriebsmodiabhängig definiert werden können,
• - Rahmenbedingungen R1 (Limitierungen/Constraints, bspw. Max- und Min-Limitierung von Aktoren, max. zulässige Gradienten, etc.),
• - Stellgrößen (bspw. für Aktoren, Ventilstellungen, Drehmomente von Antrieben, Heizer, etc.),
• - Zustandsgrößen (bspw. Drücke, Drehzahlen, Temperaturen, Temperaturdifferenzen, Aktivität, Konzentrationen, etc.)
• - Ergebnisgrößen (bspw. elektrische Leistung, Abwärmen, Gesamtspannung, elektrischer Verbrauch, etc.).

The following parameters can be defined for the respective operating strategies MPC1, MPC2, ..., MPCN:

• - time step width Tstep,
• - time horizon prediction Tpred,
• - Quality function or operating goals (e.g.: respective aging of the components involved, electrical power consumption of the respective actuators, deviation between setpoint and actual values, NVH, etc.), whereby these can also be defined depending on the operating mode,
• - Framework conditions R1 (limits/constraints, e.g. max and min limits of actuators, max. permissible gradients, etc.),
• - Control variables (e.g. for actuators, valve positions, torques of drives, heaters, etc.),
• - State variables (e.g. pressures, speeds, temperatures, temperature differences, activity, concentrations, etc.)
• - Result variables (e.g. electrical power, waste heat, total voltage, electrical consumption, etc.).

Below, exemplary dynamics D1, D2, ..., DN and different example values are given, which can show the different magnitudes of the dynamics D1, D2, ..., DN. The dynamics D1, D2, ..., DN depend on whether the respective systems to be operated are operated dynamically or rather slowly. The time step width Tstep and the time horizons Tpred of the prediction can also be variable, for example depending on a specific application and/or an operating mode.

A possible operating strategy MPC1 can e.g. B, be intended for electrical variables or Powernet.

• - Zeitschrittweite Tstep z.B. 400 ns ..... 5 ms.
• - Zeithorizonte Tpred für die Prädiktion z.B. 4ms ....400ms.

The dynamics D1 of the electrical system is very fast. The calculation/time step size/sampling must be carried out at high frequency. The predictive time horizon is short.

• - Time step width Tstep e.g. 400 ns ..... 5 ms.
• - Time horizons Tpred for prediction e.g. 4ms ....400ms.

A possible operating strategy MPC2 can be provided for the gaseous and/or fluid flows/media, for example in an air system (for supplying oxygen to a fuel cell stack) and/or a hydrogen system (for supplying hydrogen to the fuel cell stack).

The dynamics D2 for the gaseous and/or fluidic flows/media is dominated by different systems in different directions. When pressure builds up, the oxygen supply (air system and cathode path) is slower than the hydrogen supply. When pressure is reduced, the dynamics are essentially limited by the anode circuit (due to hydrogen consumption and/or purge/drain) and the cathode path is usually faster.

• - Zeitschrittweite Tstep z.B. 10ms .... 100ms.
• - Zeithorizonte Tpred für die Prädiktion z.B. 200ms .... 1500ms.

In addition, it is extremely important to maintain the maximum permissible pressure difference across the membrane (ie between the cathode and anode sides). The points mentioned mean that the two gas systems are therefore preferably calculated with the same dynamics D2 and thus with the same time scale and can therefore also be combined in one operating strategy MPC2 (two different operating strategies MPC2 with the same time scale would of course also be conceivable).

• - Time step width Tstep e.g. 10ms .... 100ms.
• - Time horizons Tpred for prediction e.g. 200ms .... 1500ms.

A possible operating strategy MPC3 can be provided for the liquid and/or fluid flows/media, for example in a cooling system and possibly for the heat transfers in the system (heat exchangers, pumps, etc.).

• - Zeitschrittweite Tstep z.B. 100 ms .... 500ms.
• - Zeithorizonte Tpred für die Prädiktion z.B. 1s ....20s.

The dynamics D3 are slower here than for the gaseous and/or fluid flows/media.

• - Time step width Tstep e.g. 100 ms .... 500ms.
• - Time horizons Tpred for the prediction e.g. 1s ....20s.

A possible operating strategy MPC4 can be intended for water management or membrane moisture/water content.

• - Zeitschrittweite Tstep z.B. 500ms ....5s.
• - Zeithorizonte Tpred für die Prädiktion z.B. 3s ....200s.

The water content or moisture content of the membrane or water management is crucial for the good functionality of the stack and for avoiding excessive aging. Since the fuel cell stack or the individual fuel cells also have internal water storage options, this results in somewhat slower dynamics D4 or inertia effects, which require a previous operating period to be taken into account in order to be able to make a prediction.

• - Time step width Tstep e.g. 500ms ....5s.
• - Time horizons Tpred for the prediction e.g. 3s...200s.

The MPC4 operating strategy can, for example, regulate the exit moisture of the stack at the cathode output.

Due to the strong connection between water management and the stack temperature, the MPC3 operating strategy can also be combined with the MPC4 operating strategy (cf. 2 and 3 ).

A possible operating strategy MPC5 can be intended for medium-term stack operating and aging behavior.

This operating strategy MPC5 is responsible for assessing and adapting the medium-term stack operating and aging behavior. The stack or stacks (in multi-stack systems) have both reversible and irreversible aging effects. The reversible effects that lead to loss of performance or deterioration in operation can be reversed. This can be done through appropriate operating points, load profiles (which occur when driving anyway) or through targeted regeneration functions (which are specifically triggered and can lead to a temporary change in overall operation).

• - Zeitschrittweite Tstep z.B. Minute(n)...stunde(n).
• - Zeithorizonte Tpred für die Prädiktion z.B. Stunden ....Tage.

This operating strategy MPC5 can, so to speak, pass on variable framework conditions R5 due to reversible effects to the faster operating strategies MPC1 to MPC4, whereby this can mean both a tightening of the framework conditions R5 and a relaxation of the framework conditions R5 (e.g. after regeneration).

• - Time step width Tstep e.g. minute(s)...hour(s).
• - Time horizons Tpred for prediction e.g. hours...days.

A possible operating strategy MPC6 can be intended for long-term stack operating and aging behavior.

This operating strategy MPC6 is responsible for assessing and adapting the long-term stack operating and aging behavior. The stack or stacks (in multi-stack systems) also age/degrade irreversibly.

• - Zeitschrittweite Tstep z.B. Stunden ... Tage.
• - Zeithorizonte Tpred für die Prädiktion z.B. Woche(n).... Monate.

These effects can no longer be reversed. The operating behavior of the fuel cell systems but also of the higher-level systems, such as power splitting between the battery and fuel cell systems, must be adapted.

• - Time step width Tstep e.g. hours ... days.
• - Time horizons Tpred for the prediction e.g. week(s)...months.

In the 3 is the structure of the combined, preferably timescale-based, operating strategies MPC1, MPC2, ..., MPCN
for different dynamics D1, D2, ..., DN in the system with several fuel cell systems 100 shown. The structure from the 2 (or from the 1 ) can be expanded in such a way that instead of one operating strategy MPCi for a dynamic class Di, several operating strategies for fuel cell systems 100 are indicated.

However, if several fuel cell systems 100 are supplied by a subsystem, for example by an air system, then an operating strategy MPCi can be provided for the corresponding dynamic class Di.

How it is 3 in particular indicates, for example, a common electrical system can be provided for several fuel cell systems 100.

The above description of the figures describes the present invention exclusively in the context of examples. Of course, individual features of the embodiments can, if it makes technical sense, be freely combined with one another without departing from the scope of the invention.

Claims (11)

Method for operating at least one fuel cell system (100), in particular for a vehicle (101), using a combined, preferably time scale-based, operating strategy (MPC), comprising: - Providing at least two operating strategies (MPC1, MPC2, ..., MPCN) for operating the fuel cell system (100), the operating strategies having different dynamics (D1, D2, ..., DN), - evaluating operating strategies (MPC1, MPC2, ..., MPCN) depending on corresponding dynamics (D1, D2 , ..., DN), - Operating the at least one fuel cell system (100) according to an operating strategy (MPCi) with a, in particular low, dynamic (Di), - Providing framework conditions (Ri), in particular for manipulated variables, state variables, result variables, etc ., according to the operating strategy (MPCi) to a further operating strategy (MPCi+1) with a further, in particular faster, dynamic (Di+1), - operating the at least one fuel cell system (100) according to the further operating strategy (MPCi+1) with a further, in particular faster, dynamic (Di+1) taking into account general conditions (Ri) according to the operating strategy (MPCi). Procedure according to Claim 1 , characterized in that the method has at least one further step: - Providing information (I), in particular state variables, result variables, etc., according to the further operating strategy (MPCi+1) to the operating strategy (MPCi). Procedure according to Claim 1 or 2 , characterized in that when evaluating the operating strategies (MPC1, MPC2, ..., MPCN) depending on corresponding dynamics (D1, D2, ..., DN) different time step widths (Tstep) and / or time horizons (Tpred) of the operating strategies (MPC1, MPC2, ..., MPCN) are taken into account. Method according to one of the preceding claims, characterized in that when evaluating the operating strategies (MPC1, MPC2, ..., MPCN) depending on corresponding dynamics (D1, D2, ..., DN) different, in particular constant and / or variable , quality functions and/or operating objectives of the operating strategies (MPC1, MPC2, ..., MPCN) are taken into account. Method according to one of the preceding claims, characterized in that the method is carried out for several fuel cell systems (100). Method according to one of the preceding claims, characterized in that framework conditions (R) and/or information (I) are exchanged between two adjacent operating strategies (MPCi, MPCi+1), and/or that framework conditions (R) and/or information ( I) are exchanged in a cascade manner between operating strategies (MPC1, MPC2, ..., MPCN), and/or that framework conditions (R) and/or information (I) are exchanged between several, preferably all, successively arranged operating strategies (MPCi, MPCi+1, MPCi+2) are exchanged, and/or that framework conditions (R) and/or information (I) are exchanged broadcast and/or peer-to-peer between operating strategies (MPC1, MPC2, ..., MPCN). Method according to one of the preceding claims, characterized in that the combined, preferably time-scale-based, operating strategy (MPC) is provided for different operating strategies with different dynamics and/or time scales: - time-scale-based operating strategy (MPC1) for control and/or regulation of electrical variables, in particular electrical voltage, electrical current, and/or electrical power, - time-scale-based operating strategy (MPC2) for control and/or regulation of fluid media supplies, in particular an anode media supply and a cathode media supply, - time-scale-based operating strategy (MPC3) for control and/or or control of liquid media supplies, in particular a media supply for coolant flows and a media supply for heat transfers in the fuel cell system (100), - time scale-based operating strategy (MPC4) for controlling and/or regulating water management and/or a moisture/water content of a PEM membrane, - time-scale-based operating strategy (MPC5) for adapting the operation of the fuel cell system (100) due to medium-term aging effects, in particular due to reversible stack degradation, preferably for triggering recovery functions, - time-scale-based operating strategy (MPC6) for adapting the operation of the fuel cell system (100) due to of long-term aging effects, especially due to irreversible stack degradation. Computer program product, comprising instructions which, when the computer program product is executed by a computer, cause the computer to carry out a method according to one of the preceding claims. Control unit (102), comprising a computing unit and a storage unit in which a code is stored which, when at least partially executed by the computing unit, carries out a method according to one of the preceding claims. Control unit (102) according to the preceding claim, characterized in that the computing power for carrying out the method can be at least partially outsourced to an external computing unit (200), in particular in the case of operating strategies (MPC1, MPC2, ..., MPCN) with low dynamics ( D1, D2, ..., DN). Fuel cell system (100), comprising a control unit (102) according to one of the preceding Claims 9 and 10 .

Images