Hydrogen fueling method and controller optimized using predictive time period control with artificial neural network model

Abstract

The method for refueling a mobile vehicle using hydrogen as fuel according to the present disclosure includes: a step of performing a first hydrogen refueling process controlled based on a first refueling protocol; a step of acquiring first refueling data obtained from the first hydrogen refueling process; and a step of providing control information for a second hydrogen refueling process based on the first refueling data.

Description

Technical Field

[0001] This disclosure relates to control technologies for hydrogen refueling of hydrogen fuel vehicles, and more particularly to hydrogen refueling methods and control devices optimized by predictive time-cycle control using artificial neural network models, and to a platform for the hydrogen refueling process and for improving the efficiency, convenience, speed and real-time performance of the refueling process. Background Technology

[0002] The content described in this section is merely background information about this exemplary embodiment and does not constitute prior art.

[0003] Hydrogen vehicles, or hydrogen electric vehicles, are zero-emission vehicles that operate using electricity generated through the reaction of high-pressure hydrogen stored in the vehicle with atmospheric air. Hydrogen electric vehicles are also known as fuel cell electric vehicles (FCEVs). Most hydrogen electric vehicles use hydrogen as an energy source to generate electricity through a fuel cell system. In addition to emitting pure water (H2O) during electricity generation, hydrogen electric vehicles also have the ability to remove ultrafine dust from the atmosphere during driving, making them a promising environmentally friendly mobility solution for the future. Given that hydrogen fuel is virtually unlimited on Earth and the energy generation process is environmentally friendly, this technology has attracted widespread attention due to its potential for broad industrial applications.

[0004] Hydrogen fuel cell vehicles are defined as mobile vehicles that use hydrogen as an energy source or generate electricity using hydrogen as fuel and use the generated electricity to drive an electric motor. Hydrogen fuel cell vehicles may include not only the aforementioned hydrogen-electric vehicles, but also air vehicles, industrial trucks, trains, ships, and aircraft, as well as devices that use hydrogen as fuel to generate electricity and use the generated electricity for propulsion.

[0005] In most hydrogen-electric vehicles, high-pressure hydrogen, safely stored in a hydrogen fuel tank, and oxygen, introduced through an air supply system, are delivered to a fuel cell stack, where an electrochemical reaction between the hydrogen and oxygen generates electricity. This electricity is then converted into kinetic energy to drive an electric motor, propelling the hydrogen-electric vehicle. Furthermore, hydrogen-electric vehicles emit only pure water through an exhaust outlet while in operation.

[0006] Meanwhile, unlike hydrogen electric vehicles, hydrogen fuel cell vehicles also use hydrogen as fuel. In hydrogen fuel cell vehicles, hydrogen is directly burned in the internal combustion engine (ICE), and the heat generated is used to drive the electric motor. The method of refueling hydrogen fuel cell vehicles is not significantly different from that of hydrogen electric vehicles.

[0007] In control technologies used to supply hydrogen to vehicles that use hydrogen as fuel, the goal is to ultimately control the temperature and pressure of the compressed hydrogen storage system (CHSS) on the fuel cell side, so that the CHSS operates within the extreme temperature and pressure conditions required for safe hydrogen refueling.

[0008] When the wired and wireless communication technologies used for control, as well as computing technologies, were not yet mature, the conventional hydrogen refueling process, control technologies, and protocols used for hydrogen electric vehicles were limited and therefore failed to adequately reflect the latest advancements in information and communication technologies (ICT). Consequently, conventional hydrogen refueling technologies for hydrogen electric vehicles are inefficient, slow, and unsuitable for large-scale hydrogen refueling. Summary of the Invention

[0009] [Technical Issues]

[0010] The purpose of this disclosure is to propose a hydrogen refueling process for hydrogen fuel vehicles and a communication protocol and refueling protocol for the process, which overcomes the limitations and defects of conventional one-way communication and improves the safety, compatibility, efficiency and reliability of hydrogen refueling.

[0011] Another objective of this disclosure is to improve the speed and real-time performance of the hydrogen refueling process and to enhance the accuracy of control and prediction of the hydrogen refueling process by utilizing additional means such as artificial neural network models and / or model predictive control.

[0012] Another objective of this disclosure is to provide a method for applying optimized communication and refueling protocols, wherein hydrogen fuel vehicles and dispensers utilize conventional or advanced communication media to efficiently achieve the goal of hydrogen refueling.

[0013] Another objective of this disclosure is to propose a refueling method that optimizes hydrogen refueling based on the surrounding environment and conditions.

[0014] [Technical Solution]

[0015] According to an exemplary embodiment of the present disclosure, a method for refueling a mobile vehicle using hydrogen as fuel may include: performing a first hydrogen refueling process controlled based on a first refueling protocol; acquiring first refueling data during the first hydrogen refueling process; and providing control information for a second hydrogen refueling process based on the first refueling data and according to a first prediction period.

[0016] In a method for refueling a mobile vehicle using hydrogen as fuel according to an exemplary embodiment of the present disclosure, the first prediction cycle may be determined based on user input.

[0017] In a method for refueling hydrogen fuel into a hydrogen fuel vehicle according to an exemplary embodiment of the present disclosure, the first prediction period may be determined based on one or more of the following: a) the reliability of the model that generates control information, b) the reliability of refueling data obtained through communication between the vehicle and a dispenser that refuels the vehicle, c) the reliability of the vehicle or the dispenser, or d) the stability of climatic conditions.

[0018] In a method for refueling hydrogen fuel into a hydrogen fuel vehicle according to an exemplary embodiment of the present disclosure, control information can be generated through model predictive control (MPC) operations.

[0019] In a method for refueling a hydrogen fuel vehicle according to an exemplary embodiment of the present disclosure, control information may be generated by an artificial neural network (ANN) model.

[0020] A method for refueling a hydrogen fuel vehicle according to an exemplary embodiment of the present disclosure may further include determining a second prediction cycle for predicting a third hydrogen fuel refueling process following a second hydrogen fuel refueling process.

[0021] The method for refueling hydrogen fuel into a hydrogen fuel vehicle according to an exemplary embodiment of the present disclosure may further include obtaining second fuel refueling data during a second hydrogen refueling process.

[0022] Determining the second prediction cycle includes: determining the second prediction cycle based on the second refueling data and the prediction state information predicted in the process of generating control information for the second hydrogen refueling process.

[0023] A method for refueling hydrogen fuel into a hydrogen fuel vehicle according to an exemplary embodiment of the present disclosure may further include a process of determining whether a first refueling protocol supports the generation of control information via model predictive control (MPC).

[0024] A method for refueling a hydrogen fuel vehicle according to an exemplary embodiment of the present disclosure may further include determining whether a communication protocol associated with a first refueling protocol supports bidirectional communication between the vehicle and a dispenser that refuels the vehicle.

[0025] A method for refueling a hydrogen fuel vehicle according to an exemplary embodiment of the present disclosure may further include determining whether the vehicle and a dispenser for refueling the vehicle support bidirectional communication between the vehicle and the dispenser.

[0026] According to an exemplary embodiment of the present disclosure, a control device for a process of refueling hydrogen fuel into a mobile body using hydrogen as fuel may include a memory storing at least one instruction and a processor executing the at least one instruction. The processor is configured via at least one instruction to execute a first hydrogen refueling process controlled based on a first refueling protocol, acquire first refueling data during the first hydrogen refueling process, and provide control information for a second hydrogen refueling process based on the first refueling data and according to a first prediction period.

[0027] In a control device for controlling the process of refueling hydrogen fuel into a mobile body using hydrogen as fuel according to an exemplary embodiment of the present disclosure, a first prediction cycle can be determined based on user input.

[0028] In a control device for controlling the process of refueling hydrogen fuel into a mobile body using hydrogen as fuel according to an exemplary embodiment of the present disclosure, the first prediction period may be determined based on one or more of the following: a) the reliability of the model that generates the control information, b) the reliability of the fuel refueling data obtained through communication between the mobile body and the dispenser that refuels the mobile body, c) the reliability of the mobile body or the dispenser, or d) the stability of climatic environmental conditions.

[0029] In a control device for controlling the process of refueling hydrogen fuel into a mobile body using hydrogen as fuel according to an exemplary embodiment of the present disclosure, control information can be generated through model predictive control (MPC) operation.

[0030] In a control device for controlling the process of refueling hydrogen fuel into a mobile body using hydrogen as fuel according to an exemplary embodiment of the present disclosure, control information can be generated by an artificial neural network (ANN) model.

[0031] In a control device for controlling the process of refueling hydrogen fuel into a mobile body using hydrogen as fuel, according to an exemplary embodiment of the present disclosure, a processor is configured to determine a second prediction period for predicting a third hydrogen fuel refueling process following a second hydrogen fuel refueling process.

[0032] The processor can be configured to acquire second fuel refueling data during the second hydrogen refueling process.

[0033] The processor is configured to determine the second prediction cycle based on the second refueling data and the predicted state information predicted during the generation of control information for the second hydrogen refueling process.

[0034] In a control device for controlling the process of refueling hydrogen fuel into a mobile body using hydrogen as fuel, according to an exemplary embodiment of the present disclosure, a processor may be configured to determine whether a first refueling protocol supports a process for generating control information via model predictive control (MPC).

[0035] In a control device for controlling the process of refueling hydrogen fuel into a mobile body using hydrogen as fuel, according to an exemplary embodiment of the present disclosure, a processor may be configured to determine whether a communication protocol associated with a first refueling protocol supports bidirectional communication between the mobile body and a dispenser that refuels the mobile body with hydrogen fuel.

[0036] In a control device for controlling the process of refueling hydrogen fuel into a mobile body using hydrogen as fuel according to an exemplary embodiment of the present disclosure, the processor may be configured to determine whether the mobile body and the dispenser for refueling hydrogen fuel into the mobile body support bidirectional communication between the mobile body and the dispenser.

[0037] [Beneficial Effects]

[0038] According to exemplary embodiments of this disclosure, the limitations and defects of conventional one-way communication in the hydrogen refueling process for hydrogen fuel vehicles, as well as in the communication protocols and refueling protocols used in the hydrogen refueling process, can be overcome, and the safety, compatibility, efficiency, and reliability of hydrogen refueling can be improved.

[0039] According to exemplary embodiments of this disclosure, by using additional means such as artificial neural network models and / or model predictive control, the speed and real-time performance of the hydrogen refueling process can be improved, and the accuracy of the control and prediction processes of the hydrogen refueling process can be improved.

[0040] According to exemplary embodiments of this disclosure, an optimized refueling process can be implemented based on the surrounding environment and conditions during hydrogen refueling. Attached Figure Description

[0041] Figure 1 This is a conceptual diagram illustrating a hydrogen refueling and control process for a hydrogen fuel vehicle applying an exemplary embodiment of the present disclosure.

[0042] Figure 2 This is a conceptual diagram illustrating an example of state changes that occur during the hydrogen refueling process for a hydrogen vehicle / hydrogen fuel mobile unit, applying an exemplary embodiment of the present disclosure.

[0043] Figure 3 This is a conceptual diagram illustrating a platform or test platform for applying exemplary embodiments of the present disclosure to a hydrogen fuel refueling process.

[0044] Figure 4This is a conceptual diagram illustrating a model predictive control (MPC) process according to an exemplary embodiment of the present disclosure.

[0045] Figure 5 This is a conceptual diagram illustrating a simulation or test platform used in the hydrogen refueling process according to an exemplary embodiment of the present disclosure.

[0046] Figure 6 This is a conceptual diagram illustrating the concept of using an artificial neural network (ANN) to model, predict, or control the hydrogen refueling process according to an exemplary embodiment of the present disclosure.

[0047] Figure 7 This is a conceptual diagram illustrating a concept of using model predictive control to control the hydrogen refueling process to achieve a target output, according to an exemplary embodiment of the present disclosure.

[0048] Figure 8 This is an operational flowchart illustrating the training and inference process of an artificial neural network when it is used for model predictive control during hydrogen refueling, according to an exemplary embodiment of the present disclosure.

[0049] Figure 9 This is a framework of functional blocks for performing a series of hydrogen refueling procedures according to an exemplary embodiment of the present disclosure, the procedures being capable of employing a bidirectional hydrogen refueling communication process.

[0050] Figure 10 This is an operational flowchart illustrating an exemplary embodiment of a hydrogen fuel refueling method according to the present disclosure.

[0051] Figure 11 It is shown on the timeline for implementation Figure 10 An illustration of an example of a prediction time period in an implementation method.

[0052] Figure 12 This shows the implementation Figure 10 A conceptual diagram illustrating an example of a platform for implementing the hydrogen refueling process.

[0053] Figure 13 It shows that it is executable. Figures 1 to 12 A conceptual diagram of an example of a general hydrogen refueling control device, hydrogen refueling control system, hydrogen refueling simulation device, hydrogen refueling simulation system, hydrogen refueling test platform, hydrogen refueling test system, or computing system for at least a portion of the process. Detailed Implementation

[0054] To better understand the features and advantages of this disclosure, exemplary embodiments of this disclosure will be described in detail with reference to the accompanying drawings.

[0055] However, it should be understood that this disclosure is not limited to the specific embodiments disclosed herein, but includes all modifications, equivalents, and substitutions falling within the spirit and scope of this disclosure. In the accompanying drawings, similar or corresponding components may be indicated by the same or similar reference numerals.

[0056] The terms including ordinal numbers (such as “first” and “second”) specified in this specification for interpreting various components are used to distinguish components from other components, but are not intended to limit to any particular component. For example, a second component may be referred to as a first component without departing from the scope of this disclosure, and similarly, a first component may be referred to as a second component. As used herein, the term “and / or” can include the presence of one or more associated listed items and any and all combinations of the listed items.

[0057] In the description of exemplary embodiments of this disclosure, "at least one of A or B" may mean "at least one of A and B" or "at least one combination of one or more of A and B". Furthermore, in the description of exemplary embodiments of this disclosure, "one or more of A and B" may refer to "one or more of A or B" or "one or more combinations of one or more of A and B".

[0058] When a component is referred to as "connected" or "coupled" to another component, that component may be directly logically or physically connected or coupled to the other component, or indirectly connected or coupled through an object in between. Conversely, when a component is referred to as "directly connected" or "directly coupled" to another component, it should be understood that there is no intermediate object between the components. Other terms used to describe relationships between components should be interpreted in a similar manner.

[0059] These terms are used herein for the purpose of describing specific exemplary embodiments only and are not intended to limit this disclosure. Unless the context clearly specifies otherwise, the singular form also includes the plural indicator. Moreover, the expressions “comprising,” “including,” “constructed,” and “configured” are used to mean the presence of a combination of the stated features, quantities, processing steps, operations, elements, or components, but are not intended to exclude the presence or addition of another feature, quantity, processing step, operation, element, or component.

[0060] Unless otherwise defined, all terms used herein (including technical or scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having the same meaning as they have in the context of the relevant literature and should not be interpreted as having an ideal or overly formal meaning unless expressly defined in this application.

[0061] The terms used in this disclosure are defined as follows.

[0062] "Hydrogen fuel cell vehicle" and "Hydrogen fuel cell vehicle: pollution-free vehicle or vehicle, including 1) vehicles or vehicles powered by electrical energy generated by the reaction of high-pressure hydrogen stored in the vehicle / vehicle with atmospheric air, such as "hydrogen electric vehicle" or "fuel cell electric vehicle (FCEV)", and / or 2) vehicles or vehicles powered by a propulsion system fueled by hydrogen or fluid, such as "internal combustion engine (ICE)".

[0063] In addition, hydrogen fuel cell vehicles refer to vehicles that use hydrogen as fuel, and can include conventional automobiles, automobiles that utilize both human power and fuel, and hybrid vehicles.

[0064] In the following exemplary embodiments, a refueling protocol and / or communication protocol for hydrogen refueling can be applied to hydrogen fuel vehicles.

[0065] Hydrogen fluid fuels can include gaseous hydrogen fuels or liquid hydrogen fuels.

[0066] "Compressed Hydrogen Storage System (CHSS)": A device used as part of a vehicle's fuel cell to compress and store hydrogen.

[0067] "Pressure relief device (PRD)": A device located in the CHSS that isolates the stored hydrogen from the refueling system and the rest of the environment and releases the hydrogen to the outside.

[0068] Hydrogen refueling essentially refers to the process of supplying high-pressure hydrogen from a hydrogen station distributor to compress the hydrogen and store it in the vehicle's tank. Hydrogen refueling can be used interchangeably with "refueling" in a simplified sense, as the distributor supplies hydrogen fuel to the hydrogen-electric vehicle. That is, in this specification, "refueling" can be used to refer to fuel supply, hydrogen refueling, or hydrogen filling, and hydrogen filling can refer to the filling of hydrogen fuel. For example, a refueling agreement can be referred to as a hydrogen filling agreement, a refueling period can be referred to as a hydrogen filling period, and a refueling method can be referred to as a hydrogen refueling method or a refueling method.

[0069] "Pressure ramp rate (PRR)": The rate at which CHSS pressure increases and is measured in megapascals per minute (MPa / min).

[0070] "Average Pressure Ramp Rate (APRR)": The average rate of pressure increase from the start to the end of hydrogen refueling.

[0071] "Pre-cooling": The process of cooling hydrogen at a hydrogen refueling station before refueling.

[0072] "Distributor": A component that supplies pre-cooled hydrogen to the CHSS.

[0073] "Nozzle": A device that connects to the hydrogen distribution system at a hydrogen refueling station and can be coupled to a receiver in a hydrogen electric vehicle to supply hydrogen fuel to the vehicle.

[0074] The term "refueling period" can be used to refer to the communication period that is carried out throughout the entire use of hydrogen refueling.

[0075] "Interoperability": The state in which components of a system interact with corresponding components of the system to perform operations that the system is intended to achieve. Additionally, information interoperability can refer to the ability of two or more networks, systems, devices, applications, or components to effectively share and easily use information without causing inconvenience to users.

[0076] Related or associated processes may include the process of establishing a relationship between two peer communicating entities.

[0077] "Command and control communications": communications used to exchange information required to start, control, and terminate wireless power transmission processes between the electric vehicle power supply equipment and the electric vehicle.

[0078] In the following detailed description, exemplary embodiments relating to hydrogen electric vehicles or fuel cell electric vehicles (FCEVs) are shown for ease of illustration. However, it will be apparent to those skilled in the art that the spirit of this disclosure can be applied to various types of hydrogen fuel cell vehicles. A hydrogen fuel cell vehicle refers to a mobile device that uses hydrogen as an energy source or uses hydrogen as fuel to generate electrical energy and uses the generated electrical energy to drive an electric motor. Hydrogen fuel cell vehicles may include not only hydrogen electric vehicles, but also airborne vehicles such as aircraft, industrial trucks, trains, ships, airplanes, etc., that use hydrogen as fuel to generate electrical energy and use it for propulsion.

[0079] Furthermore, the two-way communication process for hydrogen fuel supply according to this disclosure can be applied not only in part to hydrogen fuel vehicles, but also in part to buildings or facilities that use hydrogen as an energy source.

[0080] In the following description, hydrogen fuel may include at least one of gaseous hydrogen and liquid hydrogen, and may substantially refer to compressed hydrogen, but is not limited thereto.

[0081] Furthermore, although for ease of explanation, vehicles that use a two-way communication process for hydrogen fuel supply are described primarily in relation to hydrogen electric vehicles (FCEVs), the vehicles are not limited to this and may include hybrid electric vehicles (EVs) that use hydrogen as fuel, internal combustion engine (ICE) vehicles, etc.

[0082] In the following description, some or all of the processes of the communication method, communication protocol negotiation method, hydrogen refueling (fuel supply) protocol negotiation method, and hydrogen refueling (fuel supply) parameter negotiation method performed in the fuel vehicle may be executed by the electronic control unit (ECU), communication device, or communication control device in the hydrogen fuel vehicle.

[0083] In the following description, some or all of the processes in the communication method, communication protocol negotiation method, hydrogen refueling (fuel supply) protocol negotiation method, hydrogen refueling (fuel supply) parameter negotiation method, hydrogen refueling (fuel supply) method, and hydrogen refueling (fuel supply) control method performed by the distributor may be executed by the distributor's controller, electronic control unit, communication device, or communication control device. Furthermore, some processes of these methods may also be executed by the controller, electronic control unit, communication device, or communication control device of the filling station associated with the distributor.

[0084] Furthermore, if necessary, technologies disclosed prior to the filing date of this application may be included as part of the configuration disclosed in this application, and such technologies will be described in this specification to the extent that they do not diminish the spirit of this disclosure. However, in describing the configuration disclosed in this application, detailed descriptions of matters that are well-known technologies and can be readily understood by those skilled in the art may be omitted, as such detailed descriptions may obscure the spirit of this disclosure. For example, technologies using thermodynamic models for hydrogen refueling control, technologies applying model predictive control (MPC) techniques for generalized dynamic control, and technologies for configuring and controlling artificial neural networks for training and inference may utilize technologies well-known prior to the filing date of this application, and at least some of these well-known technologies may be applied as essential component technologies for implementing this disclosure.

[0085] However, the spirit of this disclosure is not intended to claim any rights to such known technology, and the content of the known technology may be included as part of this disclosure, provided that it does not depart from the spirit of this disclosure.

[0086] In the following text, it will be explained through Figures 1 to 13 The exemplary embodiments shown herein are used to describe detailed aspects of this disclosure.

[0087] Figure 1 This is a conceptual diagram illustrating a hydrogen refueling and control process for a hydrogen fuel vehicle applying an exemplary embodiment of the present disclosure.

[0088] like Figure 1As shown, pre-cooled hydrogen is supplied from hydrogen station 200 to hydrogen fuel cell vehicle (hydrogen fuel vehicle) 300 via distributor 100. In this case, the hydrogen refueling process can be described by parameters including the average pressure ramp rate (APRR).

[0089] Typically, hydrogen storage systems installed in vehicles can be mainly divided into high-pressure hydrogen tanks, pressure-controlled high-pressure pipelines, and an external frame. High-pressure hydrogen tanks have been developed and commercialized, with capacities ranging from tens to hundreds of liters. For vehicle applications, multiple small and lightweight storage tanks are connected in parallel to achieve large capacities.

[0090] High-pressure hydrogen tanks are widely referred to as compressed hydrogen storage systems (CHSS) 310, and in this specification, for convenience, the term "tank" refers to CHSS 310, and for ease of explanation, the terms "tank" and CHSS 310 may be used interchangeably.

[0091] In a typical hydrogen storage system, a boss unit is used to control hydrogen storage, through which hydrogen gas flows into and out of storage tank 310. Due to the characteristic that hydrogen injection and use cannot occur simultaneously, valves, pressure reducing mechanisms, and various sensors for measurement are assembled at a single boss unit to control hydrogen storage.

[0092] The interface between the hydrogen station 200 and the hydrogen fuel vehicle 300 is handled by the distributor 100, which controls the target pressure and injection rate based on vehicle tank information and fuel filling information from the station 200 and the tank. An example of the control logic currently in use is control logic conforming to the SAE J2601 (2020-05) standard.

[0093] In conventional technology, information can be delivered from the hydrogen fuel vehicle 300 to the distributor 100 using both communication and non-communication methods. Even when communication is used, in conventional technology, only the temperature and pressure values ​​of the vehicle tank 310 are transmitted unidirectionally to the distributor 100, and the distributor 100 does not actively utilize such information, but only uses it for safety criteria, such as emergency shutdown under extreme temperatures or pressures.

[0094] The control logic for safe and rapid refueling is entirely managed by the distributor 100, and the vehicle tank 310 has only minimal safety management devices that automatically release hydrogen via a pressure relief device (PRD) 320 under conditions such as overheating, without requiring an active safety management scheme.

[0095] exist Figure 2In response to the phenomenon of hydrogen temperature rise during hydrogen refueling, which will be described later, station 200 includes a high-pressure hydrogen storage unit 220 and a precooler 210. The precooler 210 lowers the temperature of the hydrogen by precooling and supplies the hydrogen to the hydrogen electric vehicle 300 via distributor 100.

[0096] In an exemplary embodiment of this disclosure, although a basic configuration similar to conventional technology is used, the hydrogen refueling controller 110 within the dispenser 100 actively controls the hydrogen refueling process by utilizing status information (such as temperature, pressure) and filling status information (such as the filling status (SOC) of CHSS 310), which is received from the hydrogen fuel vehicle 300 and station 200.

[0097] According to an exemplary embodiment of this disclosure, real-time temperature data of the vehicle tank 310 is used to control the fuel filling rate in real time, and the system is designed to operate at the highest feasible fuel filling rate when conditions below safety limits are met, thereby enabling a reduction in fuel filling time within the available range.

[0098] In conventional refueling protocols, safety boundary conditions are overly constrained, resulting in overcooling during refueling, causing the temperature of most tanks 310 to be measured at approximately 40°C-50°C at the end of refueling.

[0099] According to an exemplary embodiment of this disclosure, the amount of precooling required and provided is actively adjusted to optimize the cooling load of the hydrogen station 200 and improve the operating efficiency of the hydrogen station 200.

[0100] The standard protocol, which is mainly designed for light-duty hydrogen electric vehicles, raises the following issues: when a new type of vehicle is added, all variables must be redefined and reflected in the standard.

[0101] According to exemplary embodiments of this disclosure, control techniques using ANN-based learnable refueling logic allow for broad application across different mobile domains through a learning and training process to update the logic itself when new devices are applied.

[0102] In conventional technology, the only countermeasure to prevent overheating of the storage tank in a hydrogen electric vehicle is to release gas via a pressure relief device (PRD) 320 when overheating occurs above a predetermined temperature.

[0103] According to an exemplary embodiment of this disclosure, by installing the cooling system 330, which will be described later, in the storage tank 310 itself, the fuel refueling rate can be increased simultaneously and the overheating of the storage tank 310 can be actively responded to, thereby improving the safety of the hydrogen fuel vehicle 300.

[0104] According to exemplary embodiments of this disclosure, hydrogen fuel can be safely charged / supplied while improving the efficiency of the hydrogen charging / supply process and enhancing its speed and real-time performance.

[0105] According to exemplary embodiments of the present disclosure, a hydrogen charging control technique based on model predictive control (MPC) with guaranteed real-time performance can be provided.

[0106] According to exemplary embodiments of this disclosure, a control technique can be provided to improve the prediction accuracy of hydrogen refueling results based on an artificial neural network (ANN) model. The accuracy of the prediction results can be improved by reflecting real-time measurements in the ANN model, which is trained using actual refueling data and theoretical simulation results.

[0107] According to an exemplary embodiment of this disclosure, the efficiency of hydrogen charging control can be improved by using an intelligent metasystem (IMS) to manage actual measurement data and predicted state information obtained from the model as a whole.

[0108] As described above, during conventional hydrogen refueling, the distributor 100 is responsible for controlling the interaction between the hydrogen fuel vehicle 300 and the hydrogen station 200. The distributor 100 is equipped with a protocol that is a method for injecting hydrogen into the hydrogen fuel vehicle 300 according to predetermined standards, and the distributor 100 supervises the overall control.

[0109] Examples of protocols installed in the distributor 100 include those based on the international standard SAE J2601 (2020-05), which are also applicable to these exemplary embodiments of the present disclosure to the extent that they are consistent with the purposes of the present disclosure.

[0110] For minimum safety requirements, various situations are simulated through thermodynamic modeling, and parameters derived from this simulation are utilized in methods such as table-based single-injection methods and partially real-time correction methods based on MC-formulas.

[0111] Minimum safety requirements include upper limits for temperature and pressure conditions in CHSS 310, as well as guidelines for state of charge (SOC).

[0112] Simulations can be performed using thermodynamic modeling that includes boundary conditions from best to worst.

[0113] To the extent that it is consistent with the purposes of this disclosure, this configuration may be applied to the configuration of exemplary embodiments of this disclosure.

[0114] Even in following Figure 1During configuration, the following problems were found in conventional techniques where the distributor 100 does not actively control the state values. These problems are also evident in conventional techniques that rely solely on simulations using simple thermodynamic models.

[0115] In conventional techniques, worst-case expected boundary conditions (excessively large boundary conditions) are assumed to predetermine the injection rate, resulting in unnecessary precooling and a reduction in overall fuel loading speed. In such cases, the injection rate is simply determined by the average pressure ramp rate (APRR), which can hinder active response depending on changing conditions. Unnecessary precooling can lead to excessive energy consumption and operating costs.

[0116] Because it relies on simulation-based results, there are limitations to the applicable capacity or configuration of tank 310. Furthermore, for new systems, development and application require additional resources, leading to limitations on applicable targets.

[0117] Thermodynamic models require a considerable amount of time to derive calculation results from mathematical formulas, and are therefore limited in application when pre-calculated parameters are not derived from the model. This method also lacks flexibility in practical control.

[0118] In conventional table-based methods, parameters such as the temperature of pre-cooled hydrogen provided by station 200 or the temperature of storage tank 310 measured in hydrogen fuel carrier 300 are not utilized, resulting in very low efficiency and difficulty in responding flexibly to changes in ambient conditions.

[0119] Conventional techniques based on the MC formula are used to correct the precooling temperature in real time. However, due to limitations in applicable targets, the calculation and application of the MC formula-based method are complex and difficult to scale.

[0120] These protocols were developed with the primary goal of ensuring safe refueling and lack alternatives for proactively controlling unforeseen circumstances such as excessive precooling or overheating of tank 310. Consequently, problems may arise such as increased operating costs due to overcooling or refueling delays due to overheating.

[0121] The features of this disclosure are derived to address these problems in conventional techniques, and are characterized by reducing reliance on simulation and attempting active control of state variables by reflecting real-time measurement data.

[0122] Figure 2 This is a conceptual diagram illustrating an example of state changes that occur during hydrogen refueling of a hydrogen fuel vehicle 300, applying an exemplary embodiment of the present disclosure.

[0123] like Figure 2As shown, when hydrogen is injected into the hydrogen storage tank 310, the internal temperature rises due to the heat of compression, causing the temperature of the hydrogen inside the hydrogen storage tank 310 to rise.

[0124] During the hydrogen refueling process, which is used to control the temperature of the storage tank 310, pre-cooled hydrogen is supplied so that the internal temperature of the storage tank 310 remains below 85°C at the final refueling completion point.

[0125] During operation, the storage tank 310 is configured to minimize heat transfer by forming the tank dome and body with carbon fiber, which surrounds the tank and has low thermal conductivity, in order to block heat exchange between ambient air and the hydrogen stored inside.

[0126] When the temperature of the hydrogen in the storage tank 310 rises during refueling, the surface temperature of the storage tank 310 rises only slightly until refueling is complete due to the low heat transfer characteristics of the storage tank 310.

[0127] These characteristics impede almost all heat exchange with ambient air, which mitigates the rapid temperature rise that occurs within the hydrogen storage tank 310 during refueling and therefore necessitates a separate temperature management program.

[0128] However, in conventional technology, there is no cooling device other than receiving pre-cooled hydrogen supplied by station 200.

[0129] The hydrogen refueling time is managed by controlling the pre-cooling and hydrogen injection rate at the hydrogen station 200 in order to keep the temperature of the vehicle's hydrogen storage tank below the upper limit of 85°C, but there is no separate temperature management plan for the storage tank 310 of the hydrogen fuel vehicle 300.

[0130] Therefore, especially in summer when the ambient air temperature is high, it becomes difficult to control the temperature of the vehicle storage tank 310 at station 200, leading to problems such as fuel refueling delays.

[0131] In an exemplary embodiment of this disclosure, Figure 2 The characteristic curves in the model are used as the basis. However, unlike conventional techniques, this disclosure optimizes the operating load of station 200 during phase I (pre-cooling phase) and the fuel loading rate (pressure ramp rate PRR) during phases II to IV, taking into account data of actual environmental variables (ambient temperature, atmospheric pressure, weather conditions, etc.), thereby deriving the optimal control conditions for the actual environment in this disclosure.

[0132] Figure 3 This is a conceptual diagram illustrating a platform or test platform that applies a hydrogen refueling process according to an exemplary embodiment of this disclosure.

[0133] like Figure 3As shown, an exemplary embodiment is illustrated, wherein an artificial neural network model 120 is assembled inside a dispenser 100 having a hydrogen fuel refueling controller 110.

[0134] Hydrogen refueling control based on the ANN-MPC method can be implemented when the artificial neural network model 120 performs a model predictive control (MPC) process to predict the next state value. Figure 3 In the case of using a hydrogen refueling protocol that allows for ANN-MPC-based hydrogen refueling control, the ANN-MPC-based hydrogen refueling protocol is typically implemented in the dispenser 100 located between station 200 and mobile unit 300. In an alternative embodiment of this disclosure, the ANN-MPC-based hydrogen refueling protocol may be implemented within a hydrogen refueling controller 110 or an artificial neural network model 120. Furthermore, in another alternative embodiment of this disclosure, even if not physically located inside the dispenser 100, the ANN-MPC-based hydrogen refueling protocol, the hydrogen refueling controller 110, or the artificial neural network model 120 may be implemented as part of a controller capable of electronically communicating with the dispenser 100 and influencing the hydrogen refueling process within the dispenser 100.

[0135] Refer again Figure 3 The artificial neural network model 120 can receive data from the station 200 and / or the mobile body 300 using real-time communication, and can deliver information for controlling hydrogen refueling to optimize hydrogen refueling based on the data to the hydrogen refueling controller 110.

[0136] Real-time communication data can include both static and dynamic data. Static data refers to information that does not change over time, such as the capacity, configuration, or equipment type of a storage tank, while dynamic data refers to information that can change during the refueling process, such as temperature or pressure.

[0137] Static data may include, for example, the volume, pressure rating, maximum permissible pressure, tank type, number of tanks, tank size, tank serial number, tank manufacturer, tank usage data, and permissible temperature range of the hydrogen fluid in the tank for the mobile body 300 or station 200.

[0138] Dynamic data may include, for example, fuel filling commands (e.g., start, stop, pause, abort, increase flow rate, decrease flow rate, tank change commands, etc.), control parameters for fuel filling (e.g., average pressure ramp rate (APRR), pressure ramp rate, etc.), real-time measurements of pressure and / or temperature fluctuations in the tank (which can indicate leaks and / or impending failures), the state of charge (SOC) of the tank on the station or mobile body side, the ambient temperature outside the tank, and real-time measurements of the flow rate of hydrogen fluid entering the tank.

[0139] Refer again Figure 3 The hydrogen refueling control method based on the ANN-MPC refueling protocol can set the basic conditions required for hydrogen refueling based on static data before hydrogen refueling begins, and perform real-time control based on dynamic data during the hydrogen refueling process.

[0140] like Figure 3 As shown, exception and problem handling modules 400 can be added. The ANN-MPC model is basically based on real-time bidirectional communication and can achieve optimal performance under conditions that allow real-time bidirectional communication.

[0141] However, as described below, various situations may occur during the verification process prior to hydrogen refueling or when the real-time communication status is deviated from during actual hydrogen refueling.

[0142] (1) No communication: Communication is not feasible because the mobile body 300 or station 200 is not equipped with communication function, the communication function is malfunctioning, or the communication function is malfunctioning during refueling.

[0143] (2) Communication error: Communication is possible using the mobile body 300 or station 200, but abnormal operation of components, sensors or equipment results in the loss or error of some data.

[0144] (3) Unverified equipment, components, or vehicles: Communication with mobile body 300 or station 200 is feasible and data is collected normally, but the reliability of the data provided by mobile body 300 or equipment is low and requires verification or limited use.

[0145] (4) Unspecified communication protocol: It is feasible to communicate with the mobile body 300 or the station 200, but the communication method is not defined by the standard or certified in the relevant country, and therefore security, reliability and stability cannot be guaranteed.

[0146] ANN-MPC's real-time communication-based control can provide optimal performance in terms of functionality, but it requires a strategy that can effectively, progressively, and differentially respond to the various communication-related problems that may occur in order to be applicable to real-world situations. The implementations described later present means to address these issues.

[0147] Figure 4 This is a conceptual diagram illustrating a model predictive control (MPC) process according to an exemplary embodiment of the present disclosure.

[0148] In control methods using MPC, the control accuracy increases as the accuracy of the prediction model improves. Figure 4The control method shown is ANN-MPC, in which artificial neural network model 120 is used as a prediction model, and a feedback control loop including hydrogen refueling controller 110 is used as a control scheme.

[0149] like Figure 4 As shown, the temperature, pressure, and ambient temperature of the mobile body 300 and station 200 are used as inputs, and the temperature and pressure of the storage tank 310 of the mobile body 300 are set as outputs.

[0150] although Figure 4 The present invention illustrates an embodiment that predicts the output temperature and pressure of the hydrogen refueling mobile unit 300, but in an alternative embodiment of this disclosure, an embodiment that predicts the output temperature and pressure of the storage tank of the station 200 may be implemented.

[0151] When new inputs are provided to the artificial neural network model 120, the output can be predicted. The artificial neural network model 120 has learned the correlation between inputs and outputs using simulation data from theoretical models and field measurements.

[0152] When a specified SOC value (target value) is provided to controller 110, the pressure ramp rate (PRR) as the fuel injection rate required to reach the target value can be predicted. When PRR_pd is given as a control parameter, artificial neural network model 120 can predict SOC_pd. SOC_pd can replace the initial target SOC_sp and be input to controller 110. PRR_c is applied, and the resulting SOC_m with respect to the moving body 300 can be fed back to artificial neural network model 120 to improve the prediction accuracy of SOC_pd by using it as part of the objective function or loss function used in training.

[0153] like Figure 3 and Figure 4 As shown, the real-time temperature, pressure, refueling rate (which may correspond to PRR), and fault signals of station 200 can be delivered to artificial neural network model 120. In an alternative embodiment of this disclosure, these same signals can also be delivered to hydrogen refueling controller 110.

[0154] Real-time temperature, pressure, and fault signals of the moving body 300 can also be transmitted to the artificial neural network model 120. In an alternative embodiment of this disclosure, real-time temperature, pressure, and fault signals can also be transmitted to the hydrogen refueling controller 110.

[0155] The hydrogen refueling controller 110 generates actual control commands, based on which the dispenser 100 performs hydrogen refueling. Some control commands, such as a temperature drop command, may also be delivered to the mobile body 300.

[0156] Commands or information related to hydrogen refueling control may include the pressure ramp rate (PRR) for hydrogen refueling in the tank 310 of the mobile body 300, control commands for the pressure ramp rate, and / or state information generated by the dispenser 100 after the execution of the control commands. Furthermore, information related to hydrogen refueling control may include variables affecting the state changes of hydrogen in the vehicle tank 310 of the mobile body 300, such as the real-time PRR or the mass flow rate of compressed hydrogen [m_dot, kg / s] derived during the feedback control process, and / or result information from controls performed using such variables. This control-related data may also affect the weights or parameters of the hidden layers of the artificial neural network model 120.

[0157] A hydrogen refueling control request based on a hydrogen refueling protocol may include a control command for the pressure ramp rate (PRR) for refueling hydrogen into vehicle tank 310.

[0158] Hydrogen refueling control related information may include one or more of the following: hydrogen refueling control requests, or status information generated by the execution of hydrogen refueling control requests. Control requests for changes in the state of hydrogen may include one or more of the following: control requests for the temperature or pressure of hydrogen in vehicle tank 310.

[0159] The state of hydrogen in vehicle tank 310 may include one or more of the following: temperature, pressure, or state of charge (SOC).

[0160] As a theoretical simulation, a thermodynamic model such as H2FillS (hydrogen filling simulation) can be used. This thermodynamic model may include a model and / or software designed to track and report one or more of the instantaneous changes in hydrogen temperature, pressure, and mass flow rate, and / or the instantaneous changes in the state of hydrogen within the vehicle tank during hydrogen refueling into a hydrogen-fueled vehicle or mobile body. Of course, the concepts of this disclosure are not limited to implementations using a specific thermodynamic model.

[0161] Hydrogen refueling agreements may include, for example, those defined in SAE J2601.

[0162] When provided with the same input data as an artificial neural network, this thermodynamic model can generate output data based on modeling and simulation. The variables of the thermodynamic model can be adjusted according to hydrogen refueling protocols, and for the same input data, different output data can be derived for different hydrogen refueling protocols.

[0163] In an exemplary embodiment of this disclosure, field data collected from a test platform can be used as input and output data, instead of input / output data from a thermodynamic model, to train the artificial neural network. That is, some field data collected from the test platform can be provided as input data to the artificial neural network, while another portion can be provided as ground-based real data corresponding to the output data of the artificial neural network.

[0164] The intrinsic parameters of an artificial neural network can be trained without initialization, or after initialization with predetermined values ​​based on field data. For example, the intrinsic parameters of an artificial neural network can be initialized based on input and output data derived from a thermodynamic model and used in the initial training.

[0165] Training artificial neural networks does not necessarily require deep learning and can use shallow learning.

[0166] The test platform according to exemplary embodiments of this disclosure can rely on dynamic field data to optimize the hydrogen refueling process.

[0167] Exemplary embodiments of this disclosure may use model predictive control (MPC) techniques based on artificial neural networks to predict the next state. The artificial neural network model 120 may be trained using theoretical results, field data, or both.

[0168] The artificial neural network model 120 may be a trained model that receives information related to hydrogen refueling control, the result of hydrogen refueling control request, and the state information of hydrogen in vehicle tank 310, and predicts changes in the state of hydrogen in vehicle tank 310.

[0169] The artificial neural network model 120 may be a trained model that receives information related to hydrogen refueling control, the current state of hydrogen in the vehicle tank 310, the state of hydrogen supplied from the dispenser 100 to the mobile body 300, and the ambient temperature, and predicts future changes in the state of hydrogen in the vehicle tank 310.

[0170] The artificial neural network model 120 may be a trained model that predicts the state changes of hydrogen in the vehicle tank 310 based on each target state of hydrogen in the vehicle tank 310 specified by hydrogen refueling control information and according to each hydrogen refueling protocol.

[0171] The artificial neural network model 120 can be trained to update its parameters based on hydrogen refueling control information and hydrogen state information in vehicle tank 310, using field data of state changes in vehicle tank 310 as ground real data to predict changes in the state of hydrogen in vehicle tank 310.

[0172] Artificial neural network model 120 can be a trained model that uses model predictive control (MPC) technology to predict changes in the state of hydrogen in vehicle tank 310.

[0173] In the event that the pre-cooling function of station 200 or the cooling system 310 of mobile body 300 is unavailable, hydrogen refueling control can be performed by taking these factors into account.

[0174] It can collect big data corresponding to each type and individual ID of dispenser 100, station 200, and mobile unit 300, control target status (temperature, pressure, SOC), initial status (temperature, pressure), and hydrogen refueling protocol for each type. Furthermore, it can derive standardized items that enable optimization and precise specification of the hydrogen refueling process by training the test platform with dynamic changes in field data corresponding to each situation.

[0175] According to exemplary embodiments of this disclosure, the reliability of the hydrogen refueling process can be improved by collecting and processing on-site hydrogen refueling data.

[0176] Field data regarding the state changes of hydrogen in vehicle tank 310 can be obtained first from mobile body 300. That is, mobile body 300 can obtain field data on the state changes of hydrogen in vehicle tank 310 regardless of whether a control request is transmitted to the hydrogen vehicle. Alternatively, in another exemplary embodiment of this disclosure, the control request may include a request for field data, and mobile body 300 may obtain the field data in response to the control request / field data request.

[0177] The field data may include data obtained from a test environment configured with test setups or response (with feedback capability) for the hydrogen vehicle side, test setups or response (with feedback capability) for the hydrogen storage tank and distributor side of the hydrogen station, and / or test setups or execution devices that can be installed on module A of the distribution control system side.

[0178] Field data refers to data obtained from station 200 and / or mobile vehicle 300 during an actual hydrogen refueling / supply process. In this context, field data may include data obtained not only when using actual stations and actual vehicles, but also when part or all of the environment is a test environment.

[0179] The field data may include data obtained from a test environment consisting of: a hydrogen vehicle equivalent test environment or a feedback-enabled device, a station-side hydrogen storage tank and distributor equivalent test environment or a feedback-enabled device, and / or a test environment or actuator with module A corresponding to the distribution control system installed.

[0180] In this case, field data can be obtained from a refueling site or test environment consisting of all three of the above-mentioned test environments or devices, or from a refueling site or test environment that includes any one of them.

[0181] For example, field data may include data from configurations such as... Figure 5 The data are obtained from the test environment or refueling site of the simulation model associated with modules B150 and C160 shown.

[0182] The device that enables feedback can refer to a device that includes a database built based on real-world data and can respond to requests from module A.

[0183] Field data may include static and dynamic data, and may be categorized into static data (e.g., the type of vehicle tank 310, the volume of vehicle tank 310, the number of modules / bottles of vehicle tank 310) and dynamic data (e.g., the temperature and pressure of hydrogen in vehicle tank 310).

[0184] In station 200, multiple hydrogen storage cylinders can be arranged and operated as a storage cylinder system. Real-time field information fed back to distributor 100 in response to requests to station 200 may include temperature and pressure information for each storage cylinder.

[0185] Based on requests from distributor 100 and / or selections from station 200, multiple storage cylinders can be switched and connected to distributor 100. In this case, temperature and pressure information for each storage cylinder included in real-time field information fed back from station 200 in response to requests from distributor 100 can influence the selection and / or switching of these storage cylinders.

[0186] Based on the temperature and pressure information of each storage bottle included in the real-time field information fed back from the station 200 based on the request from the distributor 100, the state of at least one storage bottle in the storage bottle system can be adjusted.

[0187] For example, as a preparatory step, based on the current status information of these storage cylinders within station 200, at least one of these cylinders can be adjusted to have a temperature and pressure suitable for fuel refueling. This adjustment can be performed in response to a request from dispenser 100 or by the control logic of station 200.

[0188] Figure 5 This is a conceptual diagram illustrating a simulation platform or test platform used in the hydrogen refueling process according to an exemplary embodiment of the present disclosure.

[0189] like Figure 5As shown, a mobile body portion 300a for simulating an actual mobile body 300, a station portion 200a for simulating an actual station 200, and a distributor portion 100a for simulating an actual distributor 100 are shown.

[0190] exist Figure 5 In this process, the mobile body section 300a, the station section 200a, and the distributor section 100a can be achieved through merging, combining, or competing between simulation models and data-based models that use real-time field dynamic data.

[0191] In order to analyze the process state changes between distributor 100 and station 200, a learning model based on field data between distributor 100 and mobile body 300 can be used as a reference for the interaction process between distributor part 100a and mobile body part 300a.

[0192] Conversely, in order to analyze the process state changes between distributor 100 and mobile body 300, a learning model based on field data between distributor 100 and station 200 can be used as a reference for the interaction process between distributor part 100a and station part 200a.

[0193] In alternative embodiments of this disclosure, control Figure 5 The simulated integrated management system can be implemented in the form of a cloud system and / or a remote server.

[0194] While exemplary embodiments focusing on artificial neural network model 120 are shown in this specification, other embodiments of this disclosure are not necessarily limited to artificial neural network model 120 or model predictive control techniques. In another embodiment of this disclosure, real-time field data may be fed back as a response after hydrogen refueling control-related information and / or feedback control based on the hydrogen refueling protocol are transmitted. Based on this real-time field data, hydrogen refueling control-related information and / or feedback control may be updated, or the next hydrogen refueling control-related information and / or feedback control may be generated.

[0195] When the station section 200a is used as a simulation model, the moving body section 300a can be used for temperature modulation and optimization of CHSS 310.

[0196] When the moving part 300a is used as a simulation model, the station part 200a can be used for the stabilization and control of the pre-cooling temperature.

[0197] control Figure 5 The simulation-integrated management system can obtain simulation results of hydrogen state changes in the vehicle tank corresponding to information related to hydrogen refueling control, and the difference between these results and field data on state changes in the vehicle tank.

[0198] control Figure 5The simulation-integrated management system can update the model 120 for the hydrogen refueling process of hydrogen vehicles based on the differences between the simulation results and the field data corresponding to information related to hydrogen refueling control.

[0199] control Figure 5 The simulation-integrated management system can obtain simulation results related to changes in the state of hydrogen in the vehicle tank in response to a control request by using a thermodynamic model that tracks one or more instantaneous changes between the temperature, pressure, or mass flow of hydrogen in the vehicle tank.

[0200] control Figure 5 The simulation-integrated management system can obtain simulation results related to changes in the state of hydrogen in the vehicle tank corresponding to the hydrogen refueling control request by inputting future predicted hydrogen refueling control requests into the model used for the hydrogen refueling process and applying model predictive control (MPC) technology.

[0201] After updating model 120, control Figure 5 The simulation integrated management system can replace the hydrogen fuel vehicle 300 with model 120, and in response to a second control request between the distributor 100 supplying hydrogen to the hydrogen fuel vehicle 300 and the station 200 supplying hydrogen to the distributor 100, obtain one or more of the simulation data or field data relating to the state changes of the hydrogen supplied from the station 200 to the distributor 100.

[0202] After updating model 120, control Figure 5 The simulation-integrated management system can replace station 200 with model 120, and in response to a third control request between hydrogen fuel vehicle 300 receiving hydrogen from distributor 100 and distributor 100, obtain one or more of the simulation data or field data relating to the state changes of hydrogen in the vehicle tank of hydrogen fuel vehicle 300.

[0203] Conventional hydrogen refueling processes or control technologies face challenges in achieving specified target SOC, nozzle end temperature / pressure, and CHSS 310 temperature / pressure. Furthermore, theoretically simulated hydrogen refueling methods do not match actual field data due to pressure variations, unstable flow rates, and high environmental variability.

[0204] The discrepancy between simulation results and actual field data stems from device-specific characteristics and environmental variability that are difficult to fully account for in theoretical simulations. Even under the same hydrogen refueling protocol, the final field data can vary depending on the initial or final target values. Conversely, when using different hydrogen refueling protocols, the final field data may differ even under the same target values ​​or initial conditions.

[0205] To address these conventional problems, exemplary embodiments of this disclosure may employ real-time field data utilization, bidirectional communication between devices, predictive control techniques, integrated system-wide control including both stations and vehicles, and the use and standardization of user-demand-based hydrogen refueling data.

[0206] Furthermore, exemplary embodiments of this disclosure can enhance existing hydrogen refueling protocols, standardize on-site refueling data formats, and compile and diagnose dynamic on-site data.

[0207] Exemplary embodiments of this disclosure may include the following enhancements to existing hydrogen refueling protocols.

[0208] Existing hydrogen refueling protocols can be selected as test targets.

[0209] Artificial neural network-based model predictive control (ANN-MPC) can be performed under the same refueling conditions as existing hydrogen refueling protocols, and the results of refueling control using ANN-MPC can be compared with the results of control using existing hydrogen refueling protocols to enhance or improve conventional hydrogen refueling protocols.

[0210] The monitoring system 130 can embed the test target protocol 131 and can request fuel filling control based on the embedded protocol, and compare the control value with the “predictive output” input from the ANN-MPC.

[0211] The monitoring system 130 or controller can execute existing hydrogen refueling protocols and transmit hydrogen refueling control-related information to the hydrogen refueling vehicle 300 via communication interfaces 140 and 160, and can obtain on-site data on changes in the state of hydrogen in the vehicle tank via communication interfaces 140 and 160.

[0212] The monitoring system 130 or controller can obtain the hydrogen refueling control sequence of the hydrogen refueling process generated by the model 120 (in the case of using ANN-MPC technology, future time series control sequence inputs can be predicted), and based on the hydrogen refueling control sequence generated by the model 120, it can enhance the existing hydrogen refueling protocol.

[0213] The monitoring system 130 or controller can obtain result data from the execution of a hydrogen refueling control sequence generated by a model for the hydrogen refueling process, and can use the hydrogen refueling control sequence to enhance a part of the existing hydrogen refueling protocol based on a comparison between the result data from the execution of the hydrogen refueling control sequence and field data from the existing hydrogen refueling protocol.

[0214] The refueling control results using the artificial neural network model 120 can be obtained by operating the model 120 or from a pre-built database. The basic specifications used in the test target refueling protocol 131 can remain unchanged, and detailed aspects such as the refueling table or logic can be enhanced or improved by referencing the refueling control values ​​of the artificial neural network model 120. In this case, the detailed elements of the test target refueling protocol 131 can be partially improved when a comparison between the refueling control results using the artificial neural network model 120 and the field data obtained from executing the test target refueling protocol 131 demonstrates superior performance from the refueling control results of the artificial neural network model 120. This comparison between the refueling control results and the execution results of the test target refueling protocol 131 can be performed on all or part of the hydrogen refueling process.

[0215] In an exemplary embodiment of this disclosure, a centralized thermodynamic model for artificial neural networks in distributor-vehicle interactions can be adopted by considering the following characteristics.

[0216] - Mass and energy balance in 0-dimensional unstable states

[0217] - 1D heat transfer through the vehicle's walls

[0218] CoolProp for evaluating hydrogen properties

[0219] Exemplary embodiments of this disclosure can perform comparative analysis between theoretical simulation results and actual field data.

[0220] In addition to the conditions set forth in the hydrogen refueling agreement, exemplary embodiments of this disclosure may also perform predictive analytics under specific conditions.

[0221] Figure 6 This is a conceptual diagram illustrating the concept of using an artificial neural network (ANN) to model, predict, or control the hydrogen refueling process according to an exemplary embodiment of the present disclosure.

[0222] like Figure 6 As shown, the current state measurement value is input into the input layer.

[0223] Ambient temperature Tamb, precooling gas temperature Tpre, and precooling gas pressure Ppre can be measured at the nozzle of distributor 100 or station 200.

[0224] Hydrogen temperature T CHSS and hydrogen pressure P CHSS The values ​​are measured on CHSS 310 of the hydrogen fuel cell vehicle 300, and the actual measured values ​​can be input into the input layer.

[0225] In the training process of an artificial neural network, the current measurement is provided to the input layer, and the next measurement is provided to the output layer, serving as the ground truth data for the network's learning process. This learning process can be described as the artificial neural network learning to predict the next measurement at the output layer based on a combination of input measurements. By learning the correlation between the input and output data, predictions can be made using both real-world dynamic refueling data and theoretical results.

[0226] During the inference or output process using an artificial neural network, actual field measurements are provided to the input layer, and the prediction of the next measurement is obtained as the output through the operation of the artificial neural network.

[0227] The learning process of the artificial neural network used in the exemplary embodiments of this disclosure can be shallow learning or deep learning, and among known neural networks, the network type suitable for the purposes of this disclosure can be adopted.

[0228] The values ​​input through the input layer are passed to the output layer after undergoing weight-based operations based on the hidden layers.

[0229] The state values ​​(predictions of the next state) output by the output layer can be used to calculate the charge state variables, such as the charge state (SOC), by using at least a part of the thermodynamic model.

[0230] In exemplary embodiments of this disclosure, hybrid control combining theoretical simulation models and artificial neural networks is also feasible. Thus, even with limited data, desired results can be achieved through learning, and suitable performance can be achieved using lightweight neural networks.

[0231] The real-time pressure ramp rate (PRR) or the mass flow rate of compressed hydrogen (kg / s) [m_dot] derived during the feedback control process can affect the weights or parameters of the hidden layers of an artificial neural network.

[0232] The hydrogen refueling technology based on artificial neural networks disclosed herein can use this model to enhance the accuracy of predicted refueling results. Since actual refueling data is used in conjunction with theoretical simulation results, real-time measurements can be reflected, further improving prediction accuracy.

[0233] While conventional control protocols compute and predict outcomes by performing simulations tailored to each individual case, the exemplary implementations of this disclosure differ in that they improve accuracy through repeated training across different cases.

[0234] Due to this difference, in the exemplary embodiments of this disclosure, accuracy is gradually improved through updates by adding various theoretical values ​​and empirical results. Even when introducing new refueling processes using different configurations or flow rate variations of tank 310, the functionality of the new refueling process can be updated in the model by adding real data via training, allowing for broad applicability across various mobile domains.

[0235] As an exemplary embodiment of the hydrogen refueling control technology for testing hydrogen refueling of a hydrogen fuel vehicle 300 according to this disclosure, model predictive control (MPC) may also be used.

[0236] Figure 7 This is a conceptual diagram illustrating a concept of using model predictive control (MPC) to control the hydrogen refueling process to achieve a target output according to an exemplary embodiment of the present disclosure.

[0237] like Figure 7 As shown, Model Predictive Control (MPC) is a control method that uses a process model to predict future outputs based on inputs, optimizes those predictions, and uses the resulting control inputs. In the process of controlling a system, the process response must be kept within a defined range to satisfy different boundary conditions, and compared to other control methods, MPC can effectively ensure that the process response remains within the specified range.

[0238] MPC obtains a process response close to the target output by predicting the time-domain time-planning input, using a predictive model to obtain the expected response, and only the control input is used as the control signal. After obtaining the planned output, the same process is repeated based on that output.

[0239] In an exemplary embodiment of this disclosure, when the hydrogen refueling model has been sufficiently validated, future refueling results can be predicted from the hydrogen refueling model and current measurements. Based on the predictions and refueling values, the pressure ramp rate (PRR) can be controlled in real time, allowing specific variables in CHSS 310 (e.g., hydrogen temperature T) to be adjusted. CHSS or hydrogen pressure P CHSS To achieve optimal refueling targets without violating any constraints.

[0240] like Figure 7 As shown in the exemplary embodiments of this disclosure, MPC-based control can be used to calculate future output values ​​based on current measurements and model predictions. Operating parameters can be adjusted so that the predicted future response moves optimally toward the setpoint (target).

[0241] For example, at the current time i, n model-based predictions can be generated. These n predictions form the prediction time domain.

[0242] Each model-based prediction or prediction time-domain point corresponds to the control time domain. That is, in order to achieve n model-based predictions, n control actions or control commands form the control time domain.

[0243] In practice, only the first control action at time i+1 out of the n model predictions and control actions generated at time i is passed to the system. As time progresses to time i+1, a new set of n model predictions and control actions is generated, forming new prediction time domains and control time domains respectively.

[0244] This method of extending or shifting the time domain while controlling the system is called MPC. In an exemplary embodiment of this disclosure, MPC-based control can be performed using both measured and predicted values ​​of state information (state values) of hydrogen temperature and pressure from CHSS 310.

[0245] Figure 8 This is an operational flowchart illustrating the training and inference process of an artificial neural network when using model predictive control during hydrogen refueling according to an exemplary embodiment of the present disclosure.

[0246] like Figure 8 The diagram illustrates the training process of an artificial neural network for artificial neural network model predictive control (ANN-MPC) technology according to an exemplary embodiment of the present disclosure.

[0247] exist Figure 8 In this context, it is assumed that the artificial neural network has been trained to obtain prediction time domain and control time domain based on MPC, and specifically, it has been trained to obtain n future predictions and corresponding control commands in a manner that optimizes the process of reaching the set point through MPC in order to achieve the future response.

[0248] like Figures 6 to 8 As shown, in an exemplary embodiment of this disclosure, the control system can be configured based on an artificial neural network (ANN) model 120, and a test platform system for real-time control based on model predictive control (MPC) can be constructed by ensuring the accuracy of the ANN model 120.

[0249] The test platform system for real-time control predicts future filling results and compares these results with actual measurements to control the filling speed, pressure ramp rate, or pressure increase rate. Constraints, control time intervals, sensitivity, and other parameters can be individually set to execute control within the optimal range of the system logic.

[0250] Primarily, optimal control can be performed based on real-time data from the refueling station 200 and the hydrogen fuel vehicle 300. However, under specific circumstances during system operation, control authority can be granted to directly control the pre-cooling temperature of the precooler 210 and the cooling system of the hydrogen fuel vehicle 300, thereby increasing the overall efficiency of the hydrogen refueling process.

[0251] like Figure 8 As shown, the control process begins by receiving a specified SOCsp from the customer (t=0 in step S710). For example, the current SOC could be 50%, and the SOCsp could be 85%.

[0252] SOC(t) as T CHSS (t) and P CHSS The function of (t) is given, and the process can be performed based on a general dynamic model.

[0253] If the current SOC(t) is equal to or greater than SOCsp (in step S720), hydrogen charging can be stopped. If the current SOC(t) is less than SOCsp (in step S720), i is set to t, and a moving-time domain prediction involving an artificial neural network is performed (in step S730).

[0254] Step S730 can be performed by generating MPC predictions using an artificial neural network 120, etc. In step S740, it can be determined whether the obtained n predictions have been optimized and meet the expected goals.

[0255] If the obtained n predictions are optimized, the control command PRR(t) can be obtained based on the n predictions and control commands, and PRR(t) can be applied to the distributor 100 and the vehicle tank 310 (in step S750).

[0256] Then, time t is incremented, and a new measurement value T is obtained. CHSS (t) and P CHSS (t), and input it into step S720.

[0257] If the n predictions obtained in step S730 are not optimized, step S730 can be executed again to obtain new n predictions and control commands.

[0258] exist Figure 8 In step S730, state prediction values ​​(T, P) that satisfy temperature and pressure limits can be generated for all arbitrary i and k.

[0259] Based on the current time i (= t), n state prediction values ​​and corresponding control commands can be derived.

[0260] Figure 8Step S740 can be understood as the process of searching for a set of n predictions that minimize the cost function representing whether the final control objective SOCsp has been achieved.

[0261] The output of the hydrogen fuel cell 300 (which includes state measurements of CHSS 310, such as temperature and pressure) can be fed as feedback into the artificial neural network model 120.

[0262] The output of the hydrogen charging station 200 (which includes state measurements such as the temperature and pressure of the pre-cooled hydrogen) can be fed as feedback into the artificial neural network model 120.

[0263] The artificial neural network model 120 transmits the predicted output to the hydrogen refueling controller 110, and the supervisory system 130 can input future inputs obtained through simulation or model-based predictions into the artificial neural network model 120 via module A140.

[0264] The ANN-MPC-based control process is a control technique that utilizes both simulation and actual measurement data, and uses an artificial neural network model 120 to perform at least a portion of the simulation and utilize the prediction results during the control process.

[0265] Exemplary embodiments of this disclosure are intended to configure an integrated hydrogen charging control protocol based on real-time data and to implement the system using various basic technologies.

[0266] The protocol installed on the distributor 100 can control the charging rate, pressure ramp rate, or pressure increase rate (PRR or [m_dot]) as outputs by using data on precooled hydrogen provided from the charging station 200 and data on the storage tank 310 provided from the hydrogen fuel vehicle 300 as real-time input values.

[0267] When events such as changes in the external environment occur, the pre-cooling temperature of the charging station 200 and the cooling system of the hydrogen fuel vehicle 300 can be directly controlled to comprehensively control the charging speed, pressure ramp rate, pressure increase rate (PRR or [m_dot]), processing efficiency, etc.

[0268] To complement the control protocol, the precooling system or precooler 210 of the hydrogen charging station 200 may be independently equipped with a self-cooling stabilization system.

[0269] Regarding temperature stability, the cooling stability system of the precooler 210 can be controlled independently, and the overall control target value can be adjusted within the protocol of the distributor 100.

[0270] To improve the economic efficiency of the charging station 200 and to supplement the functionality of the integrated control protocol, additional functions related to temperature stability can be granted to the precooler 210.

[0271] The precooling temperature may vary due to the initial temperature and flow rate of the hydrogen supplied to the precooler 210. To compensate for this, a new precooler structure for temperature stabilization is proposed as an exemplary embodiment of this disclosure.

[0272] The precooler 210 according to an exemplary embodiment of the present disclosure may include a temperature control mechanism and control logic for linkage with a protocol.

[0273] A forced cooling system may be installed in the storage tank 310 of the hydrogen fuel vehicle 300 and may cool some of the heat of compression generated during hydrogen charging to improve the charging rate. The protocol may also include the operation / control of the forced cooling system of the storage tank 310.

[0274] In an exemplary embodiment of this disclosure, temperature management functionality may be granted to the storage tank 310 of the hydrogen fuel vehicle 300 to increase the hydrogen charging rate and complement the functionality of the integrated control protocol.

[0275] In an exemplary embodiment of this disclosure, the vehicle tank 310 may be equipped with a self-cooling system to increase the total filling rate and improve the safety of the hydrogen fuel vehicle 300, and such a system may include control logic for autonomous operation and for linkage with protocols.

[0276] According to an exemplary embodiment of this disclosure, such integrated control can improve current filling efficiency and smoothly support preparation for the next filling.

[0277] When the precooling temperature of the precooler 210 is set to -40°C (T40), if the target value of the precooling temperature is reached, but the ambient temperature is higher than the set value, and the temperature rise on the storage tank 310 is greater than the expected value, a control signal or current status information can be transmitted to the hydrogen fuel mobile body 300 / storage tank 310, so that the self-cooling system of the storage tank 310 can be operated.

[0278] Conversely, when the precooling temperature of the precooler 210 is set to -40°C, but given the external environment and actual data that overcooling is occurring, the target value of the precooling temperature can be adjusted (e.g., adjusted to -35°C).

[0279] When additional control of the precooling target temperature and the temperature on the storage tank 310 is necessary, control information or control commands can be transmitted from the distributor 100 to both the hydrogen fuel vehicle 300 and the filling station 200.

[0280] According to an exemplary embodiment of this disclosure, the self-cooling system of the hydrogen fuel vehicle 300 and the charging station 200 can be controlled independently or by transmitting signals from the distributor 100.

[0281] An integrated control method for hydrogen charging according to an exemplary embodiment of this disclosure may further include a step of evaluating whether a measurement of the current state satisfies a constraint.

[0282] The constraint can be that the temperature and pressure of the hydrogen storage system of the hydrogen vehicle do not exceed their respective temperature and pressure limits.

[0283] According to exemplary embodiments of this disclosure, hydrogen fuel can be safely charged / supplied while improving the efficiency of the hydrogen charging / supplied process and enhancing its speed and real-time responsiveness.

[0284] According to exemplary embodiments of this disclosure, a testing method and testing platform capable of accurately modeling the hydrogen charging / supply process based on real-time field dynamic data can be implemented.

[0285] According to exemplary embodiments of this disclosure, a testing method and testing platform can be implemented that provides a model capable of precisely controlling the hydrogen charging / supply process by taking into account the differences between modeling results and simulation results based on theoretical models and real-time field data, or by taking into account modeling results, simulation results, and real-time field data.

[0286] According to exemplary embodiments of this disclosure, a test technique for hydrogen charging control with real-time performance based on model predictive control (MPC) can be implemented.

[0287] According to exemplary embodiments of this disclosure, a testing technique can be implemented to improve the control accuracy of hydrogen charging results based on an artificial neural network (ANN) model.

[0288] When receiving a user-specified charge amount (e.g., SOC), by using, for example Figures 6 to 8 The control logic for real-time communication and computation of the sequence shown can charge / supply hydrogen until the target amount is reached.

[0289] The charge volume can be set using various variables such as SOC, target pressure, time, and temperature, and the charge rate can be controlled to improve efficiency.

[0290] Figure 9 This is a framework of functional blocks (hereinafter referred to as the "hydrogen refueling framework") for performing a series of hydrogen refueling processes according to an exemplary embodiment of the present disclosure, which are capable of using bidirectional hydrogen refueling communication processes.

[0291] like Figure 9As shown, the hydrogen refueling framework may include a discovery and pairing function block (hereinafter referred to as "UC1" or "UC-1") as a function block for each use case (UC), a communication security function block (UC2 or UC-2), a communication protocol negotiation function block (UC3 or UC-3), a refueling protocol negotiation function block (UC4 or UC-4), a refueling parameter negotiation function block (UC5 or UC-5), a security check-in function block (UC6 or UC-6), a monitoring and control function block (UC7 or UC-7), a security check-out function block (UC8 or UC-8), a terminal function block (UC9 or UC-9), an error handling function block (UC10 or UC-10), and an emergency handling function block (UC11 or UC-11).

[0292] like Figure 9 As shown, each of the functional blocks UC1 to UC11 corresponds to time series steps S401 to S411. In this case, Figure 9 It can also be understood as an operation flowchart including time series steps S401 to S411.

[0293] UC10 and UC11 can be connected to UC3 through UC8 separately and can be configured to perform error handling and / or emergency handling in each use case.

[0294] The above use cases provide a consistent and unified functional block representation of the entire hydrogen refueling process for a hydrogen refueling system, ensuring safe and secure refueling communication. Vehicles and dispensers can execute these use cases sequentially in a specific order to achieve the hydrogen refueling objective.

[0295] Furthermore, after the dispenser nozzle is connected to the vehicle receiver, the corresponding vehicle and dispenser can be connected via a button. Figure 9 The sequence shown is used to perform fuel refueling communication according to the corresponding usage scenarios. However, specific usage scenarios can be omitted from the vehicle and dispenser when required, depending on predetermined requirements.

[0296] Each of the above usage scenarios can be implemented through communication between the distributor control system of the distributor and the hydrogen fuel vehicle, the distributor supplying hydrogen as fuel to the hydrogen fuel vehicle according to the refueling protocol for hydrogen fuel vehicles.

[0297] Simultaneously, the hydrogen fuel cell vehicle (hereinafter referred to as the "vehicle") and distributor implementing the above-described usage can perform data exchange for vehicle identification in UC-1. For this purpose, the vehicle may include sensors, an electronic control unit (ECU), a transmitter, and a receiver. When performing bidirectional communication, the receiver may be integrally coupled to the transmitter.

[0298] Furthermore, the dispenser can be configured to receive specific data from the vehicle. The dispenser can store data for data logging or can store specified data in the station's programmable logic controller (PLC) for use in the refueling protocol. Data logging can refer to the process of collecting data over a period of time to analyze specific operational states of the hydrogen refueling system or recording data-based events / operations of the system or network environment, or it can refer to data collected by such a process. In the case of bidirectional communication, the station can include sensors specified by the refueling protocol, and the station PLC or electronic control unit can obtain measurements from these sensors and transmit these measurements to the vehicle. The aforementioned vehicle or station can communicate using existing communication protocol standards, such as infrared communication, Wi-Fi, Bluetooth, etc.

[0299] Furthermore, a communication channel can be established between the physically connected vehicle and the distributor at the vehicle-distributor interface. The pairing process used to establish the communication channel can be performed using wired, optical, or wireless technologies.

[0300] The discovery and pairing procedure or pairing process may have the prerequisite that the dispenser nozzle is inserted into the vehicle fuel filler receiver and securely connected. The vehicle fuel filler receiver may be simply referred to as the vehicle receiver or receiver.

[0301] Furthermore, the vehicle and dispenser are essentially aware of which communication protocol to follow. Therefore, communication conforming to UC-1 may depend solely on a communication protocol agreed upon under current usage conditions as a condition for the discovery and pairing process or a subsequent condition of the pairing process. If the vehicle or dispenser selects a communication protocol outside the agreed scope, the selected communication will not be executed. That is, even if the pairing process is successfully completed, authorization for fuel or authorization for refueling may not be authorized.

[0302] All methods used to pair the vehicle with the distributor are configured not to increase the risk of ignition or explosion to an acceptable level. For example, all wired pairing methods are configured to mitigate or prevent the risk of sparks caused by electrostatic discharge.

[0303] Regarding the effectiveness of physical pairing, all methods for pairing a vehicle and a dispenser can be integrated into the vehicle-dispenser interface or can be mounted close to the fuel filling container of the vehicle and the nozzle and hose assembly of the dispenser. Here, the interface can refer to being physically integrated into the nozzle and container interface. Furthermore, proximity can be defined by the hardware associated with the pairing method. For example, the physical geometry for infrared communication, including the permissible distance between the transmitter and receiver, can be specified. Additionally, the physical shape of the hydrogen refueling hardware can be predefined. This proximity excludes relatively long-range wireless communication technologies, such as Bluetooth, which pose a risk of pairing physically unconnected vehicles with the dispenser. Infrared communication can be referred to as Infrared Data Association (IrDA) communication and can include bi-directional infrared (bi-IrDA) communication.

[0304] In steps S401, S403, S404 and / or S405, information regarding the interoperability and compatibility between the mobile body and the distributor can be shared and mutually checked.

[0305] In steps S401, S403, S404 and / or S405, it can be identified whether the mobile body and the distributor support bidirectional communication respectively.

[0306] In steps S401, S403, S404, and / or S405, a communication protocol associated with the refueling protocol can be negotiated, and refueling parameters according to the refueling protocol can be negotiated. In step S401, information related to the interoperability and compatibility of the communication protocol and / or the refueling protocol can be shared in advance, and information related to preferences or priorities on the mobile body side or the dispenser side can be shared.

[0307] Based on the information shared in step S401 regarding interoperability, compatibility, preferences, or priorities, the communication protocol, refueling protocol, and refueling parameters can be negotiated in steps S403, S404, and / or S405.

[0308] When there are differences between the information shared in step S401 and the information identified in steps S403, S404 and / or S405 (e.g., changes in the level of bidirectional communication support), the communication protocol, fuel refueling protocol and fuel refueling parameters may be negotiated or renegotiated based on the interoperability, compatibility, preference or priority updated in each of steps S403, S404 and / or S405.

[0309] Whether refueling is permitted can be determined through the fuel refueling agreement. Figures 1 to 8 The artificial neural network model described herein controls or supports / assists the hydrogen refueling process, and can... Figure 9In one or more of steps S401, S403, S404 and / or S405, it is determined whether the hydrogen refueling process is allowed to be controlled or controlled by an artificial neural network model.

[0310] The refueling agreement can be used to determine the passage. Figures 1 to 8 Whether the control or control support / auxiliary of the hydrogen refueling process using model predictive control (MPC) as described herein is permitted, and whether the control or control support / auxiliary of the hydrogen refueling process using model predictive control is permitted, can be determined by... Figure 9 The steps are determined in one or more of S401, S403, S404 and / or S405.

[0311] The communication registrations supported between the mobile entity and the distributor can be represented by Tables 1 and 2 below.

[0312] [Table 1]

[0313] Table 1 provides examples of communication levels defined from 0 to 3, differentiated based on reliability levels, to account for various scenarios such as communication failures, malfunctions, and insufficient data security / reliability. The communication levels in Table 1 may refer to those included in the currently developing ISO 19885-1 standard.

[0314] [Table 2]

[0315] Table 2 shows the control levels, where static and dynamic variables are specified according to the communication reliability level, and the permissible range of unreliable data and substitute variables is specified. At lower communication levels, some functions are restricted, but the operation of the refueling protocol is feasible at each communication level.

[0316] The hydrogen refueling protocol based on ANN-MPC proposed in the exemplary embodiments of this disclosure corresponds to Level 3 as defined in Tables 1 and 2, and it can be assumed that the static and dynamic data transmitted to and received from the mobile body 300 and station 200 are reliable and in a ready-to-use state.

[0317] Figure 10 This is an operational flowchart illustrating an exemplary embodiment of a hydrogen fuel refueling method according to the present disclosure.

[0318] According to an exemplary embodiment of the present disclosure, a method for refueling a mobile vehicle with hydrogen as fuel includes: performing a first hydrogen refueling process controlled based on a first refueling protocol (S1100); acquiring first refueling data during the first hydrogen refueling process (S1200); and providing control information for a second hydrogen refueling process based on the first refueling data and according to a first prediction period (S1300).

[0319] The method for refueling hydrogen fuel into a hydrogen fuel vehicle according to an exemplary embodiment of the present disclosure may further include performing a second hydrogen fuel refueling process (S1400) based on the control information provided in step S1300 and the first fuel refueling protocol.

[0320] In a method for refueling hydrogen fuel into a hydrogen fuel vehicle according to an exemplary embodiment of the present disclosure, the first prediction cycle may be determined by user input.

[0321] In a method for refueling hydrogen fuel into a hydrogen fuel vehicle according to an exemplary embodiment of the present disclosure, the first prediction period may be determined based on one or more of the following: a) the reliability of the model that generates control information, b) the reliability of refueling data obtained through communication between the vehicle and a dispenser that refuels the vehicle, c) the reliability of the vehicle or the dispenser, or d) the stability of climatic conditions.

[0322] In a method for refueling hydrogen fuel into a hydrogen fuel vehicle according to an exemplary embodiment of the present disclosure, control information can be generated through model predictive control (MPC) operations.

[0323] In a method for refueling a mobile vehicle with hydrogen fuel according to an exemplary embodiment of the present disclosure, control information can be generated by an artificial neural network (ANN) model.

[0324] A method for refueling hydrogen fuel into a hydrogen fuel vehicle according to an exemplary embodiment of this disclosure may further include: determining a second prediction period for predicting a third hydrogen fuel refueling process following a second hydrogen fuel refueling process. That is, in an exemplary embodiment of this disclosure, the prediction period may be dynamically changed.

[0325] In this case, the method for refueling hydrogen fuel into a hydrogen fuel vehicle according to an exemplary embodiment of the present disclosure may further include obtaining second fuel refueling data during a second hydrogen refueling process.

[0326] In this context, determining the second prediction period includes: determining the second prediction period based on the second refueling data and the state information predicted during the processing of control information for the second hydrogen refueling process. That is, in an exemplary embodiment of this disclosure, when the prediction period changes dynamically, the prediction time period (prediction period) can be dynamically changed based on the difference between the predicted value and the actual measured value.

[0327] A method for refueling hydrogen fuel into a hydrogen fuel vehicle according to an exemplary embodiment of the present disclosure may further include a process (S401, S403, S404, S405) for determining whether a first refueling protocol supports generating control information via model predictive control (MPC).

[0328] A method for refueling a hydrogen fuel vehicle according to an exemplary embodiment of the present disclosure may further include determining whether a communication protocol associated with a first refueling protocol supports bidirectional communication between the vehicle and a dispenser that refuels the vehicle (S401, S403, S404, S405).

[0329] A method for refueling a hydrogen fuel vehicle according to an exemplary embodiment of the present disclosure may further include determining whether the vehicle and a dispenser for refueling the vehicle support bidirectional communication between the vehicle and the dispenser (S401, S403, S404, S405).

[0330] The first hydrogen refueling process can refer to any nth sequence or control / prediction time domain, as described later. Figure 11 and Figure 12 As shown.

[0331] The second hydrogen refueling process can refer to any (n+1)th sequence or control / prediction time domain, as described later. Figure 11 and Figure 12 As shown.

[0332] For ease of description, Figure 11 and Figure 12 This assumes the use of a refueling protocol that supports the ANN-MPC operation used in the exemplary embodiments of this disclosure.

[0333] The predictive model is the most important component in control using MPC, and control accuracy can improve as the accuracy of the predictive model increases. ANN-MPC can refer to the case where an artificial neural network model 120 is used as the predictive model and MPC is used as the control method.

[0334] like Figure 3As shown, the basic platform constituting ANN-MPC uses the temperature and pressure of the mobile body 300 and station 200, as well as the outside air temperature, as inputs, and sets the temperature and pressure of the storage tank of the mobile body 300 as outputs. Furthermore, the accuracy of the correlation can be improved by learning the relationship between the inputs and outputs through the use of multiple theoretical and practical data.

[0335] like Figure 4 As shown, the operation of the ANN-MPC protocol can be performed as follows: the station is treated as a type of controlled object, the fuel filling rate control is set as the objective, an ANN is used as a predictive model, appropriate process inputs are calculated by performing ANN predictions and optimizing the objective function through an optimization process, and the pressure ramp rate (PRR) is controlled as the fuel filling rate through the appropriate process inputs.

[0336] Artificial neural network models can be trained using simulated values ​​and actual measurements from the mobile body, distributor, and station. These actual measurements are correlated with the temperature and pressure of the mobile body's tanks based on factors such as the temperature and pressure of the hydrogen fluid fuel injected from the station into the mobile body.

[0337] The reliability of the artificial neural network model 120 can be improved by continuously adding data obtained under various conditions and repeatedly performing training. The ANN-MPC model can be trained to optimize the hydrogen refueling time of a mobile vehicle by using the artificial neural network model 120 as the prediction model and configuring MPC as the controller.

[0338] To address known issues in applying the ANN-MPC model to hydrogen refueling, the following may be necessary.

[0339] Because the specifications of station facilities and the specifications of tanks in mobile bodies vary greatly depending on the type of equipment, specific artificial neural network models 120 that reflect the characteristics of mobile body models and stations may be required separately.

[0340] In cases where MPC is applied to a specific artificial neural network model 120 implemented based on the type of individual mobile body / station equipment used for control, individual and selective fuel filling rate control may be required to suit the individual case.

[0341] Since differences in facility specifications or constraints can occur in all individual cases when the same refueling rules are applied, individual settings control functions are required at the end-user stage to predict the range of refueling results based on the reliability of the artificial neural network model 120.

[0342] Figure 11 It is shown on the timeline for implementation Figure 10An illustration of an example of a prediction time period in an implementation method.

[0343] like Figure 11 As shown, the ANN-MPC model is used to illustrate the operations performed in the time domain during the hydrogen refueling control process.

[0344] The ANN-MPC model can predict future results at any time based on data from stations / distributors and mobile vehicles at any time, and can determine the refueling rate (PRR) based on the predicted values.

[0345] The prediction time period of the ANN-MPC model can be fixed based on the amount and type of data obtained, external conditions, etc. For example, the prediction time period can have a value of, for example, 10 seconds or 15 seconds. The ANN-MPC model can operate relative to the actual measured fuel refueling data in a direction that reduces the deviation between the predicted and actual values.

[0346] For example, when a 10-second prediction period is set, the ANN-MPC model can predict the situation 10 seconds later based on the actual measured data input at t=0 seconds, determine the refueling rate based on the prediction results, input the actual measured data again after 10 seconds to compare the predicted value and the actual value and reduce the deviation, and predict the result value again 10 seconds later based on the actual measured value.

[0347] In an exemplary embodiment of this disclosure, the prediction period can be set and operated as a fixed value in the ANN-MPC model through evaluation during the development phase.

[0348] In another exemplary embodiment of this disclosure, when a specific artificial neural network model 120 is implemented individually according to the type of mobile body and station equipment, additional control functions for predicting the cycle may be required.

[0349] exist Figure 11 In this process, the first cycle can correspond to the first hydrogen refueling process, while the second cycle can correspond to the second hydrogen refueling process.

[0350] In the first cycle, the first hydrogen refueling process (S1100) is performed. First refueling data (T) can be obtained during the first cycle, or immediately following the first cycle at t=10. CHSS，Real P CHSS，Real (S1200).

[0351] exist Figure 11 For ease of description, we assume the first prediction period is 10 seconds. Based on the first prediction period, the ANN-MPC model can obtain the prediction data (T) at t=20 seconds after the first prediction period through MPC operations. CHSS，Predicted PCHSS，Predicted The controller of the dispenser using the ANN-MPC model can generate a refueling rate (PRR) as a control command for the second hydrogen refueling process corresponding to the second cycle. The refueling rate is generated based on the predicted data at t=20 corresponding to the end time of the second cycle and the first refueling data at t=10 corresponding to the end time of the first cycle.

[0352] The ANN-MPC model can provide the distributor's controller with control information for the next cycle based on the predicted data for the next cycle and the actual measurement data for the current time (S1300).

[0353] The distributor's controller can execute the second hydrogen refueling process (S1400) based on the control command (fueling rate (PRR)) that reflects the control information.

[0354] In the same manner, based on the actual measured data after the second cycle and the predicted data after the third cycle, the ANN-MPC model and / or controller can generate control information and / or control commands for the third cycle and can execute the third hydrogen refueling process.

[0355] Figure 12 This shows the implementation. Figure 10 A conceptual diagram illustrating an example of a platform for implementing the hydrogen refueling process.

[0356] like Figure 12 The illustration shows a hydrogen refueling control process using an ANN-MPC model according to an exemplary embodiment of the present disclosure, and more specifically, shows a method for optimizing the operation of a hydrogen refueling protocol by controlling the predicted time period (Y) of an artificial neural network model 120.

[0357] The artificial neural network model 120 can predict the result value after the prediction period (X+Y(j+1)) based on the real-time data obtained in step S1200 and the current time point (X+Yj). The predicted value 122 can be used to provide control information for the fuel refueling speed (S1300). (j=0, 1, 2, etc.)

[0358] The controller 110 can generate control commands for the fuel filling speed (PRR or m_dot) based on control information, and can execute the first fuel filling process and the second fuel filling process (S1100, S1400) by executing the control commands.

[0359] During the development phase of the artificial neural network model 120, the prediction time period (Y) can be set to a short period to ensure initial reliability. Subsequently, as the artificial neural network model 120 develops, matures, and its prediction accuracy improves, the prediction time period (Y) can be gradually set to a longer period.

[0360] The prediction time period (Y) can also be directly related to the control period, which is changed by controlling the actual rate of pressure increase. Since the temperature and pressure of station 200 and moving body 300 do not respond immediately to control within a few seconds, the pressure increase rate control change period can typically be controlled at intervals of at least 5 seconds.

[0361] To improve refueling performance (e.g., reduce full-fill time), frequent controlled driving with short prediction cycles may be advantageous. On the other hand, in terms of operational efficiency, frequent controlled driving can increase system load and raise maintenance costs and failure rates. Therefore, in terms of operational efficiency, minimizing flow control with long prediction cycles is advantageous, while in terms of improving refueling performance, short prediction cycles are advantageous. Thus, it may be necessary to effectively set the prediction cycle through a trade-off between refueling performance and operational efficiency (stability).

[0362] The following can be considered representative factors affecting the determination of the prediction period (Y) of the artificial neural network model 120.

[0363] a) Reliability of Artificial Neural Network Model 120

[0364] The artificial neural network model 120 may have an initial model that is installed and operated in the station 200, and the reliability of the artificial neural network model 120 may be gradually improved by using actual fuel refueling results as data to perform learning and relearning.

[0365] The reliability of each artificial neural network model 120 can vary depending on factors such as the start time of operation of the artificial neural network model 120, the refueling frequency, and the characteristics of station 200. As reliability decreases, it may be necessary to correct the artificial neural network model 120 by using short prediction cycles to input predicted and actual measurements. The reliability of the artificial neural network model 120 can be improved by protecting the data as the number of refueling operations increases and performing retraining based on the protected data.

[0366] The initial model of the artificial neural network model 120 can be trained solely using theoretical simulations based on conservative boundary conditions to reflect various conditions, and therefore may have some differences from actual field data. Assuming the temperature and pressure of the tank in the current mobile body 300 are 20 degrees and 100 atm, the prediction period of the initialized artificial neural network model 120 is set to 60 seconds, and the temperature and pressure are predicted to be 30 degrees and 200 atm after 60 seconds, the fuel refueling rate is controlled at a PRR of 100 atm / 1 minute.

[0367] When the actual measurement after 60 seconds differs from the predicted value, the refueling rate is controlled to change, and the difference between the actual measurement and the predicted value can increase as the training level of the artificial neural network model 120 decreases. As the training level, maturity, and / or reliability of the artificial neural network model 120 decreases and the difference between the actual measurement and the predicted value increases, the variation in the refueling rate control command increases, and the variation in the control command can again cause an increase in error and further increase the control burden of station 200.

[0368] Therefore, in the case of the initial model, when the prediction period is set to a relatively short prediction period (e.g., 10 seconds or 20 seconds), the difference between the predicted value and the actual measurement value can be reduced, the data obtained by performing refueling can be relearned to improve reliability, and the prediction period can be gradually increased as the reliability of the artificial neural network model 120 (i.e., the ANN-MPC model) increases.

[0369] b) Reliability of real-time two-way data reception; As disclosed in Tables 1 and 2 above, the data received from the mobile body 300 and the station 200 is determined according to the communication process and protocol of the application and may have different levels of reliability.

[0370] When the reliability of transmitted data is low, keeping the prediction period as short as possible may be necessary to respond to inaccurate data acquisition.

[0371] The ANN-MPC model can perform speed optimization through control based on real-time communication data. When the reliability of the received communication data is guaranteed, the prediction cycle period can be set to be relatively long. However, when the received data may periodically or intermittently include noise or errors, and long-cycle predictions are performed based on erroneous real-time data, safety issues may arise due to erroneous control, and when normal data is received in the next cycle, the difference between actual measurements may cause rapid changes in the refueling speed control command, potentially increasing the control burden on station 200 and / or dispenser 100.

[0372] c) Reliability of facilities

[0373] When facility stability is insufficient in MPC control via the artificial neural network model 120, a shorter prediction cycle may be necessary. For example, when the response to ANN-MPC control deteriorates, or when a response deviates from normal operation due to design errors, facility deficiencies, or construction problems, control correction may be required through frequent prediction cycle control.

[0374] In other words, when the facilities of the mobile unit 300, station 200 and / or distributor 100 have low reliability, frequent predictive cycle control may be required to ensure safety.

[0375] The artificial neural network model 120 can predict the result value in advance after the prediction period based on the prediction period, and the predicted value 122 can be used to provide control information for the fuel refueling speed (S1300).

[0376] The values ​​learned by the artificial neural network model 120 represent information about the relationship between the output value and the input value, and it is expected that the relationship between the input and output values ​​follows a general thermodynamic relationship. Due to factors such as the response performance of the flow control valve of station 200, thermal insulation problems in the pipeline, equipment capacity, and design defects (such as friction), special responses and result values ​​that can affect the control process may occur. In particular, when a new design structure is applied, construction problems exist, or the facilities or components used are below standard, even if the control burden of station 200 is increased due to short control time cycles, it may be necessary to apply short time cycles in the initial stage.

[0377] d) Stability of climatic and environmental conditions

[0378] In environments where the assumed heat exchange conditions with the outside air are typically violated, the prediction cycle may need to be controlled during the initial model operation, depending on the circumstances. When the effects of high temperatures, low temperatures, local temperature differences, strong winds, gusts, typhoons, high humidity, etc., are significant, control corrections to improve reliability may be necessary through prediction cycle control.

[0379] The artificial neural network model 120 can start from a base model generated based on simulation results, based on various climatic environmental factors set as boundary conditions. As fuel refueling data is generated using the fuel refueling process of the base model and relearned to extend reliability, the climatic environmental factors set in the initial simulation phase can be diluted as learning progresses. When the artificial neural network model 120, trained in a cold region, operates new in a hot region with very different climatic environmental conditions, it may be necessary to re-ensure reliability through short-term evaluations.

[0380] Refer again Figure 12 The present invention illustrates a conceptual structure of a system for performing operations for optimizing a hydrogen refueling protocol by predicting time periodic control of an artificial neural network model 120 using an ANN-MPC model, according to an exemplary embodiment of the present disclosure.

[0381] In another exemplary embodiment of this disclosure, reference is made to Figure 12 An exemplary implementation of user-defined and inputted prediction period Y may be disclosed.

[0382] The refueling process between the hydrogen station 200 and the hydrogen vehicle is handled by the distributor 100, and users can perform variable settings and control operations related to refueling through the distributor 100.

[0383] The operator of station 200 can input a predicted cycle time setting suitable for the operating environment of station 200 via the control panel of dispenser 100, based on instructions presented for each situation. The user input value can be transmitted to artificial neural network model 120 to predict the result value after the input time and set the refueling rate, and controller 110 can use control information based on the predicted value 122 obtained through MPC to control the hydrogen refueling process. As a result of the control, the refueling rate can be determined, and hydrogen can be injected from hydrogen station 200 into hydrogen vehicle 300 via dispenser 100.

[0384] The actual measurement value after the input cycle time is transmitted from the hydrogen station 200 and the hydrogen mobile unit 300 to the distributor 100. Based on the actual measurement value, the future value from the current time (X+Yj) to the input time Y (X+Y(j+1)) is predicted, and the controller 110 can change and control the fuel filling speed (j=0, 1, 2, etc.).

[0385] As the reliability of the artificial neural network model 120 increases, it can predict the result value after a longer period of time, and the difference between the predicted value and the actual measurement value for each corresponding prediction time period can be reduced by high reliability, thereby enabling rapid control of the refueling rate.

[0386] At the start of refueling, the MPC operation and control process is executed based on the user-input prediction period Y, and can then be repeated according to the given prediction period Y.

[0387] During the training phase of the artificial neural network model 120, the reliability of the model can be improved by learning theoretical and actual measurement data that can reflect various situations, predicting and controlling the refueling rate after a predetermined time, and correcting the deviation between the predicted value 122 and the actual measurement value by comparing it with the measurement value after the actual running time.

[0388] The reliability of the artificial neural network model 120 can be improved when the absolute amount of data to be learned and the quality of the data that can reflect various situations are improved. The initial artificial neural network model 120 has a difference between the predicted value 122 and the actual measured value, and theoretically, the difference between the predicted value 122 and the actual measured value can increase as the prediction period Y increases. Therefore, in the initial model with a small amount of learning data, the prediction period Y can be set to a shorter value, and as learning progresses and reliability improves, the prediction period Y can be set to a longer value.

[0389] While exemplary implementations have been shown of setting the prediction period Y differently during the ANN-MPC operation learning process of the artificial neural network model 120, in another exemplary implementation of this disclosure, an exemplary implementation of dynamically changing the prediction period Y during hydrogen refueling can also be disclosed.

[0390] When the difference between the predicted value 122 and the measured value in the prediction period Y set in the initial stage of fuel refueling is stable within a predetermined threshold, the prediction period Y can be gradually changed to a longer value.

[0391] The hydrogen fuel delivered from distributor 100 to mobile unit 300 can be gaseous or liquid, and can typically be controlled at a single or variable rate. Average pressure ramp rate (APRR) is a method of controlling the rate of pressure increase to achieve the same rate of pressure increase from the start to the end of refueling, while pressure ramp rate (PRR) is a method of controlling the rate of pressure increase by varying the rate of pressure increase as needed during the refueling process. The rate of pressure increase can also be expressed as m_dot. To control the pressure in the tank of mobile unit 300 to increase at a constant or desired rate, distributor 100 controls the rate of pressure increase by adjusting the opening and closing degree of a flow control valve installed in the pipeline through which the hydrogen flows, and the mass flow rate can be measured at this time. The refueling rate m_dot, recorded as the mass flow rate, is closely related to APRR and PRR, and as the mass flow rate increases, the rate of pressure increase increases, but not necessarily in a direct proportional manner.

[0392] The artificial neural network model 120 according to the above exemplary embodiments of the present disclosure can use simulated data and actual field data to learn functions that control the hydrogen refueling process, or can learn a portion of the functions that control the hydrogen refueling process.

[0393] The actual field data used for learning can vary depending on the specifications of the tanks or facilities on the side of station 200, and when fuel refueling is actually performed for a specific mobile body 300, the reliability of the artificial neural network model 120 may be reduced when the data learned in actual application time is different from the hardware specifications.

[0394] To improve and optimize the reliability and efficiency of the artificial neural network model 120, individual models with specific types and detailed specifications for each type of station 200 and mobile body 300 can be trained.

[0395] In the case of station 200, the conditions used to classify low-level individual models may include maximum supply pressure, precooling temperature, maximum flow rate, etc.

[0396] In the case of the mobile body 300, the conditions may include vehicle model name, total CHSS volume, maximum container volume, maximum allowable flow rate, etc., and the vehicle model name may include serial number classified by the manufacturer, etc.

[0397] For safety reasons, the CHSS of the mobile body 300 can be implemented by including storage tanks of various sizes, and in the mobile body 300, not only the total volume of the CHSS but also the detailed specifications of the storage tanks can be taken into account.

[0398] The separately specialized and trained artificial neural network model 120 can be arranged and operated in the distributor 100, and according to an exemplary embodiment, can be stored in the mobile body 300 or a third-party location, and can be operated by communicating via a network.

[0399] A basic artificial neural network model can be provided to the distributor 100, and low-level artificial neural network models trained and differentiated according to individual specific conditions can be arranged and managed in the distributor 100. The basic artificial neural network model can be initialized by the manufacturer of the mobile body 300.

[0400] According to an exemplary implementation, even when the basic artificial neural network model is stored in the mobile body 300 or a third-party location, a separate model for hydrogen refueling can be stored and operated on the dispenser 100 side during actual field operations.

[0401] In yet another exemplary embodiment of this disclosure, the manufacturer of the mobile body 300 may use data based on the characteristics of the storage tank to generate and distribute an initialized basic artificial neural network model. The basic artificial neural network model may be stored in the memory of the mobile body 300 and may be transmitted to the distributor 100 after the mobile body 300 enters the station 200.

[0402] The differentiation of specialized artificial neural network models and the learning of low-level individual models can be primarily performed by the dispenser 100. Low-level individual models can be learned or relearned based on actual fuel dispensing data obtained by the dispenser 100.

[0403] To ensure the reliability of low-level individual models, the artificial neural network model can be relearned after sufficient real-world data has been accumulated. This process can be performed periodically or non-periodically in the allocator 100.

[0404] To ensure the reliability of the low-level individual model, the manufacturer of mobile unit 300 can collect fuel refueling data received from station 200 and can relearn the basic artificial neural network model. The updated basic artificial neural network model (which can be interpreted as a model specific to each mobile unit 300) can be transmitted to dispenser 100 again and updated. When mobile unit 300 enters station 200, the basic artificial neural network model can be transmitted to dispenser 100.

[0405] It can operate third parties that can ensure the reliability of low-level individual models and provide integrated management.

[0406] Third parties can be government agencies, associations, professional management companies, etc.

[0407] This association can be technology-related or standards-related.

[0408] In real-world applications, fuel refueling data can be collected via cloud servers and relearned using artificial neural network models. Third parties can represent Mobile300 to generate and distribute basic artificial neural network models.

[0409] Furthermore, low-level individual models can be collected periodically or non-periodically and upgraded as a whole, and the upgraded individual models can be provided again to individual stations 200. This process can be implemented as a joint learning type.

[0410] Figure 13 It shows that it is executable. Figures 1 to 12 A conceptual diagram of an example of a general hydrogen refueling control device, hydrogen refueling control system, hydrogen refueling simulation device, hydrogen refueling simulation system, hydrogen refueling test platform, hydrogen refueling test system, or computing system for at least a portion of the process.

[0411] like Figure 13 As shown, a computing system 3000 according to an exemplary embodiment of the present disclosure may include a processor 3100, a memory 3200, a communication interface 3300, a storage device 3400, an input interface 3500, an output interface 3600, and a bus 3700.

[0412] A computing system 3000 according to an exemplary embodiment of the present disclosure may include at least one processor 3100 and a memory 3200 storing instructions for instructing the at least one processor 3100 to perform at least one step. At least some steps of the method according to an exemplary embodiment of the present disclosure may be performed by the at least one processor 3100 executing instructions loaded from the memory 3200.

[0413] Processor 3100 may refer to a central processing unit (CPU), graphics processing unit (GPU), or dedicated processor on which methods according to exemplary embodiments of the present disclosure are performed.

[0414] Each of the memory 3200 and the storage device 3400 may consist of at least one of volatile and non-volatile storage media. For example, the memory 3200 may include at least one of read-only memory (ROM) and random access memory (RAM).

[0415] In addition, the computing system 3000 may include a communication interface 3300 for performing communication via a wireless network.

[0416] The computing system 3000 may also include a storage device 3400, an input interface 3500, an output interface 3600, etc.

[0417] The corresponding components included in the computing system 3000 can be connected via bus 3700 and can communicate with each other.

[0418] Examples of the computing system 3000 disclosed herein may include communicable desktop computers, laptop computers, notebooks, smartphones, tablet PCs, mobile phones, smartwatches, smart glass, e-book readers, portable multimedia players (PMPs), handheld gaming devices, navigation devices, digital cameras, digital multimedia broadcast (DMB) players, digital audio recorders, digital audio players, digital video recorders, digital video players, personal digital assistants (PDAs), etc.

[0419] An apparatus installed on a hydrogen-fueled mobile body 300 and controlling the hydrogen refueling process according to an exemplary embodiment of the present disclosure may include: a memory 3200 storing at least one instruction; and a processor 3100 executing at least one instruction.

[0420] In this case, the processor 3100 can execute a first hydrogen refueling process controlled by a first refueling protocol, obtain first refueling data obtained during the first hydrogen refueling process, and provide control information for a second hydrogen refueling process based on the first refueling data and according to a first prediction cycle.

[0421] In a control device for refueling a mobile body using hydrogen as fuel according to an exemplary embodiment of the present disclosure, a first prediction cycle can be determined by user input.

[0422] In a control device for refueling hydrogen fuel into a mobile vehicle using hydrogen as fuel according to an exemplary embodiment of the present disclosure, the first prediction period may be determined based on the reliability of the model that generates control information, the reliability of fuel refueling data obtained through communication between the mobile vehicle and a dispenser that refuels the mobile vehicle, the reliability of the mobile vehicle or the dispenser, or the stability of climatic environmental conditions.

[0423] In a control device for refueling a mobile vehicle using hydrogen as fuel according to an exemplary embodiment of the present disclosure, control information can be generated through model predictive control (MPC) operation.

[0424] In a control device for refueling a mobile body using hydrogen as fuel according to an exemplary embodiment of the present disclosure, control information can be generated by an artificial neural network (ANN) model.

[0425] In a control device for refueling a mobile body using hydrogen as fuel according to an exemplary embodiment of the present disclosure, a processor may determine a second prediction cycle for predicting a third hydrogen refueling process following a second hydrogen refueling process.

[0426] In this scenario, the processor 3100 can obtain second refueling data during the second hydrogen refueling process.

[0427] In this case, when determining the second prediction cycle, the processor 3100 can determine the second prediction cycle based on the second refueling data and the state information predicted during the generation of control information for the second hydrogen refueling process.

[0428] In a control device for refueling a mobile body using hydrogen as fuel according to an exemplary embodiment of the present disclosure, processor 3100 may determine whether a first refueling protocol supports a process for generating control information via model predictive control (MPC).

[0429] In a control device for a process of refueling hydrogen fuel into a mobile body using hydrogen as fuel according to an exemplary embodiment of the present disclosure, processor 3100 may determine whether a communication protocol associated with a first refueling protocol supports bidirectional communication between the mobile body and a dispenser that refuels the mobile body with hydrogen fuel.

[0430] In a control device for refueling a mobile body using hydrogen as fuel according to an exemplary embodiment of the present disclosure, processor 3100 may determine whether the mobile body and the dispenser for refueling the mobile body support bidirectional communication between the mobile body and the dispenser.

[0431] A control device for a hydrogen refueling process according to an exemplary embodiment of the present disclosure may be installed in a dispenser that performs hydrogen refueling to a mobile body, or may be implemented as a controller that is not installed in a dispenser but can communicate electronically with the dispenser and can affect the refueling of the dispenser.

[0432] The operation of the method according to embodiments of this disclosure can be implemented as a program or code stored in a computer-readable recording medium. The computer-readable recording medium includes any type of recording device in which information can be stored and read by a computer system. The computer-readable recording medium can also be distributed across a computer system connected via a network, and the program or code can be stored and executed in a distributed manner.

[0433] Computer-readable recording media may include hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Program instructions may include machine code generated by a compiler and high-level language code that can be executed by a computer using an interpreter.

[0434] Some aspects of this disclosure have been described in the context of apparatus; however, it should be understood that a corresponding method-based description is also applicable, wherein a module or apparatus corresponds to a method step or feature of a method step. Similarly, aspects described in the context of a method may also be represented as corresponding blocks or features of an apparatus. Some or all of the method steps may be performed by (or using) hardware apparatus (e.g., a microprocessor, a programmable computer, or electronic circuitry). In a predetermined embodiment, at least one key method step may be performed by such apparatus.

[0435] In some implementations, a programmable logic device (e.g., a field-programmable gate array (FPGA)) may be used to perform some or all of the functions of the methods described herein. In such implementations, the FPGA may be operated in conjunction with a microprocessor to perform one or more of the methods described herein. Typically, preferably, the method is performed via a hardware device.

[0436] Although this disclosure has been described with reference to preferred embodiments, those skilled in the art will understand that various modifications and changes may be made without departing from the spirit and scope of this disclosure as defined by the appended claims.

Claims

1. A method for fueling hydrogen fuel into a mobile vehicle using hydrogen as fuel, the method comprising: Execute the first hydrogen refueling process controlled by the first refueling protocol; First fuel refueling data is obtained during the first hydrogen refueling process; as well as Based on the first fuel refueling data, control information for the second hydrogen fuel refueling process is provided according to the first prediction cycle.

2. The method according to claim 1, wherein, The first prediction period is determined based on user input.

3. The method of claim 1, wherein, The first prediction period is determined based on one or more of the following: the reliability of the model that generates the control information; the reliability of the fuel refueling data obtained through communication between the mobile vehicle and the dispenser that refuels the mobile vehicle; the reliability of the mobile vehicle or the dispenser; and the stability of the climate conditions.

4. The method of claim 1, wherein, The control information is generated by Model Predictive Control (MPC) operations.

5. The method of claim 1, wherein, The control information is generated by an Artificial Neural Network (ANN) model.

6. The method according to claim 1, further comprising: A second prediction cycle is determined for predicting the third hydrogen refueling process following the second hydrogen refueling process.

7. The method according to claim 6, further comprising: Second fuel refueling data was obtained during the second hydrogen refueling process. The determination of the second prediction period includes: determining the second prediction period based on the second fuel refueling data and the prediction state information predicted during the generation of the control information for the second hydrogen fuel refueling process.

8. The method according to claim 1, further comprising: The process of determining whether the first fuel refueling protocol supports generating the control information via Model Prediction Control (MPC).

9. The method according to claim 1, further comprising: Determine whether the communication protocol associated with the first refueling protocol supports bidirectional communication between the mobile vehicle and the dispenser that refuels the mobile vehicle with hydrogen fuel.

10. The method according to claim 1, further comprising: Determine whether the mobile vehicle and the dispenser that dispenses hydrogen fuel to the mobile vehicle support bidirectional communication between the mobile vehicle and the dispenser.

11. A control device for a process of fueling hydrogen fuel into a mobile vehicle using hydrogen as fuel, the control device comprising: Memory, storing at least one instruction; as well as The processor executes at least one instruction. The processor is configured to execute the at least one instruction as follows: Perform the first hydrogen refueling process controlled by the first refueling agreement. First fuel refueling data was obtained during the first hydrogen refueling process, and Based on the first fuel refueling data, control information for the second hydrogen fuel refueling process is provided according to the first prediction cycle.

12. The control device according to claim 11, wherein The first prediction period is determined based on user input.

13. The control device of claim 11, wherein, The first prediction period is determined based on one or more of the following: the reliability of the model that generates the control information; the reliability of the fuel refueling data obtained through communication between the mobile vehicle and the dispenser that refuels the mobile vehicle; the reliability of the mobile vehicle or the dispenser; and the stability of the climate conditions.

14. The control device of claim 11, wherein, The control information is generated by Model Predictive Control (MPC) operations.

15. The control device of claim 11, wherein, The control information is generated by an Artificial Neural Network (ANN) model.

16. The control device of claim 11, wherein, The processor is configured to determine a second prediction period for predicting a third hydrogen refueling process following the second hydrogen refueling process.

17. The control device of claim 16, wherein, The processor is configured to: Second fuel refueling data was obtained during the second hydrogen refueling process, and The second prediction period is determined based on the second fuel refueling data and the predicted state information predicted during the generation of control information for the second hydrogen fuel refueling process.

18. The control device of claim 11, wherein, The processor is configured to determine whether the first fuel refueling protocol supports the process of generating the control information via Model Prediction Control (MPC).

19. The control device of claim 11, wherein, The processor is configured to determine whether a communication protocol associated with the first refueling protocol supports bidirectional communication between the mobile vehicle and the dispenser that refuels the mobile vehicle with hydrogen fuel.

20. The control device of claim 11, wherein, The processor is configured to determine whether the mobile body and the dispenser that dispenses hydrogen fuel to the mobile body support bidirectional communication between the mobile body and the dispenser.

Images