Real-time method for monitoring and simulating a supply chain for a green fuel

Abstract

A real-time method for monitoring and simulating a supply chain 1 for a green fuel 2,3. Historical data on an infrastructure 5 for producing a green fuel is held in a datastore. The infrastructure 5 i

Description

The invention relates to a real-time method for monitoring, simulating, and optimizing a selection of a supply chain for a green fuel. One of the most pressing problems currently facing humanity is energy generation and energy conversion with reduced release of climate-damaging gases, especially carbon dioxide. One promising approach is to obtain or synthesise what is referred to as green fuels (also referred to as eFuel or other hydrogen derivatives) using renewable energy and preferably from raw materials that can be extracted in a climate-neutral way. For example, the following approaches are currently being pursued in projects: The so-called power-to-liquid [PtL] designates a method in which electrical energy is used to produce a liquid fuel from renewable sources (such as wind turbines or wind farms). These liquid fuels can be used in existing internal combustion engines. In one project, exhaust gases from a steel mill are converted into liquid fuels using the resulting waste heat. In Iceland, methanol is produced by geothermal energy. The Fischer-Tropsch method is used to synthesise a high-quality diesel fuel that can be obtained from a variety of carbon-containing raw materials, such as biomass, methanol (produced using PtL), or carbon dioxide from the air (using carbon capture). Various projects are focused on diesel production using the Fischer-Tropsch method. In this context, the solid material carbon is converted into a liquid fuel, for example with the help of renewable energy. Methanol is a versatile fuel and chemical raw material for producing high quality fuels that is easier to store and transport than gaseous fuels. There are projects to convert carbon dioxide from air into methanol as well as to extract methanol from carbon. Dimethylether [DME] may be used as a replacement for diesel or as an alternative to liquid gas. This product bums with almost no soot and with low emissions of nitrogen oxides [NOx], Among other things, there are projects for producing DME from forestry waste, which is to be used in heavy trucks. In addition, research is being conducted on the production of DME from natural gas sources. The extraction of hydrogen [H2] from seawater appears obvious but requires a significant amount of electrical energy. Particularly wind-rich locations around the world are often thinly populated and are therefore ideal for the construction of wind farms. However, these remote locations render the direct supply of generated electrical energy as power uneconomical. Where there are long distances between hydrogen production and the end user, it may make sense not to liquefy the hydrogen by compression, but to convert it into methanol or (using the Fischer Tropsch method) into diesel fuel using carbon dioxide. The carbon dioxide required for this purpose can be obtained from the air, for example. Manufacturing and supplying green fuels requires the cooperation of numerous highly technological and / or highly specialised players. A smooth process from the source to the consumer of the green fuel is necessary to make it affordable enough to encourage the widespread use of green fuels instead of conventional fuels. In general terms, term “green fuel” encompasses fuels generated using waste products (e.g., carbon dioxide, heat) from an industrial process, fuels derived from biomass, and fuels generated using renewable energy. A green fuel also includes blends of green fuels with conventional fuels. Broadly, there are three main types of green-fuel: biobased fuels (a fuel derived from biomass), green hydrogen (hydrogen produced, e.g., by electrolysis, using renewable electricity), and e-fuels (a fuel generated from renewable electricity and carbon dioxide). Other specific examples of green fuels include fuels comprising: methanol, Dimethylether, hydrogen, and / or the like. Based on this, the underlying object of the present invention is to monitor a plurality of different supply chains for generating or otherwise obtaining a green fuel, and optionally selecting an optimal supply chain from the plurality and / or at least partially overcome the disadvantages known from the prior art. The features according to the invention emerge from the independent claims, for which advantageous configurations are shown in the dependent claims. The features of the claims can be combined in any technically meaningful manner, for which purpose it is also possible to consult the explanations from the following description and features from the figures, which comprise additional configurations of the invention. The invention relates to a real-time method for monitoring and simulating a supply chain for a green fuel, comprising the following constituents: in a data store, holding historical data on an infrastructure for producing a green fuel, the infrastructure being complex due to a plurality of subscribers; via a plurality of interfaces, obtaining real-time data relating to the infrastructure; and in a computer in a simulation environment based on the historical data held and the real-time data obtained, forecasting at least one optimal use of the infrastructure to form an efficient supply chain and / or to efficiently exploit a supply chain, wherein dedicated specifications are output in real-time for at least a plurality of the subscribers of the complex infrastructure. In some examples, the infrastructure includes physical equipment (e.g., machinery) and / or physical resources (e.g., energy sources, raw materials, etc.) from which a green fuel, or a derivative thereof, can be derived, otherwise obtained, stored, and / or distributed. Example energy sources include any one or more of: wind, nuclear, fossil fuels, geothermal, and solar power. The energy sources may be associated with corresponding physical equipment and / or a physical locality. For example, wind power may be associated with one or more wind farms; nuclear power may be associated with one or more nuclear (fusion or fission) plants; fossil fuels (e.g., natural gas, coal, crude oil) may be associated with one or more fossil fuel processing plants and / or one or more physical reserves of fossil fuels (e.g., a subsurface reservoir); geothermal power may be associated with one or more geothermal processing plants and / or one or more physical reserves of geothermal power (e.g., a volcano); and / or solar power may be associated with one or more solar farms. Example raw materials include any one or more of: green fuel; water (fresh or seawater); biomass; nuclear fuel; hydrogen; fossil fuels, such as natural gas, coal, or crude oil; and / or DME. In some examples, the physical equipment includes machinery for producing green fuel, or a derivative thereof, e.g., using raw materials and energy. In some examples, the physical equipment includes machinery for generating energy. For example, one or more wind turbines, one or more solar panels, one or more power stations (associated with nuclear or geothermal power or fossil fuels). In some examples, the physical equipment includes machinery for obtaining raw materials. For example, an oil rig for obtaining crude oil from a subsurface reservoir; a system for obtaining water from seawater (to be used to generate hydrogen via electrolysis); one or more carbon captures devices for capturing carbon dioxide; and / or agricultural equipment for obtaining biomass. In some examples, the physical equipment includes equipment associated with auxiliary processes, such as the storing of raw materials, storing energy, and / or transporting such raw materials and / or energy. For example, one or more containers for containing raw materials, such as green fuel, biofuel (e.g., methanol), hydrogen, biomass, water, or seawater; or one or more energy storage systems (e.g., batteries, flywheels, etc.) for storing electrical power. The physical equipment may further include transport equipment (e.g., a fleet of vehicles) for transporting the raw materials, or an electrical distribution system (e.g., the grid) for distributing electrical power. The physical equipment and / or physical resources may be associated with a respective locality. That is, the physical equipment and / or physical resources may not be co-located. The plurality of subscribers may be associated with different processes or parts of the supply chain, e.g., associated with different physical equipment recited above. The interests of the subscribers may, therefore, be conflicting. The infrastructure is associated with a plurality of possible supply chains, each of which may be associated with a plurality of subscribers, for generating and supplying a green fuel (e.g., hydrogen or methanol as the fuel). This makes a determination as to the optimal supply chain complex. The method involves obtaining real-time data relating to the infrastructure. The realtime data may be associated with: - For a given raw material, the production rate of the raw material, the quantity of the stored raw material, the spatial distribution of the stored material, and / or the availability and accessibility of containers for storing further raw material; - For machinery for producing green fuel, the production rate and quality of the green fuel being produced, and / or a status of the machinery (e.g., operational, or whether any error warnings are showing); - For equipment for generating energy, the power generation rate, the status of associated machinery (e.g., operational or whether any error warnings are showing). - For transportation equipment (of raw materials), the availability and spatial distribution of the fleet of transportation vehicles; and / or - For transportation equipment (of electrical power), the availability of electrical power to be distributed. The real-time data may also be associated with a physical measurement (e.g., the weather measurement, such as wind speed, light intensity, etc.) of the environment in which the infrastructure resides. For example, a wind speed measurement relating to a wind farm; a light intensity measurement relating to a solar farm; a humidity measurement; a temperature measurement and the like. The method then involves forecasting, in a simulation environment (e.g., using a digital twin), how the infrastructure may be used more efficiently to generate or supply green fuel or a derivate thereof. That is, which supply chain out of the plurality of possible supply chains is the optimal one (in real time) for generating or supplying green fuel. This forecasting makes use of at least a portion of the real-time data obtained (as described above) and historical data (associated with the previously acquired data from or relating to the infrastructure) held in a data store. In some implementations, the method described herein, and in particular the step of forecasting, in a simulation environment, as described above, may be performed by or using a machine learning model (e.g., a neural network). In some implementations, the model may be a deep neural network that includes an output layer and one or more hidden layers that each apply a non-linear transformation to an input in order to generate an output at the output layer. Machine learning models are models that receive an input and generate an output based on the received input. The mapping of an input to an output may correspond to a particular task (e.g., determining selecting a particular supply chain as “optimal”), and the machine learning model can be trained using an objective function to carry out that and / or other tasks. In some examples, the machine learning model comprises a set of model parameters, and the values of these parameters can be learnt (or updated) by minimizing the objective function. The objective function provides a numerical metric by which to assess the performance of the machine learning model. Example objective functions include a least-squares objective function, a cross-entropy objective function, a regression loss function, a classification objective function etc. The objective function may be parameterized by one or more factors (each of which contribute to the “cost” of the objective function). Each factor of the objective function may be associated with a respective weight, which can be the same or different. Example factors include: carbon footprint (e.g., the amount of greenhouse gases associated with producing the green fuel from a particular supply chain), time (e.g., the time required to create the green fuel and / or deliver it to an intended destination), and cost (e.g., the cost required to create the green fuel and / or deliver it to an intended destination). The terms “efficient” supply chain and “efficiently exploit a supply chain” are to be interpreted accordingly. That is, in terms of these optimizing these factors (carbon footprint, cost, and / or time). The machine learning model may be trained on a batch of training data. The batch of training data may include one or more training inputs (e.g., each corresponding to a particular supply chain) and their corresponding target outputs (e.g., the carbon footprint, cost, time associated with the corresponding supply chain). The machine learning model can process, using its model parameters, the training input to generate an output. The output is then compared with the target output according to the objective function (e.g., least-squares), and the model parameters for the machine learning model may then be updated (i.e., adjusted in value) to minimize the objective function (e.g., minimize a difference between the output and the target output). In some implementations, this involves computing a gradient of the objective function (with respect to one or more of the aforementioned factors, e.g., carbon footprint) and backpropagating the computed gradients through the neural network to update the values for the model parameters. In response to the forecasting, the method may include a step of outputting dedicated specifications (e.g., instructions) to at least one of the plurality of subscribers. The dedicated specifications may include instructions as to how the subscriber should operate or control its associated physical equipment so as to form at least part of the supply chain that is determined by the forecasting to be optimal. In some examples, the method includes a step of sending a respective set of dedicated specifications to a respective subscriber of the plurality. Also proposed here is a computer-readable storage medium storing instructions that, when executed using a processor, cause the processor to carry out the method steps recited above. Unless explicitly stated otherwise, ordinal numbers are used in the preceding and the following description only for the purposes of clear distinction and do not reflect any order or ranking of the designated components. An ordinal number greater than one does not imply that another such component has to necessarily be present. The method proposed here for monitoring (preferably as so-called event-based monitoring) and simulating a supply chain for a green fuel is carried out as a so-called digital twin, which is therefore executed like a model bound in software that represents reality, for example on a central server. In a preferred embodiment, the measured values used in the (modeled) real application or the (already processed) data used in its control software or monitoring software (as input data, for example) can be used without the need for conversion or processing. Rather, this input data in the model enables the same results to be achieved as in the real application. However, this allows us to simulate scenarios more quickly and respond to the resulting changes in a timely manner based on the system’s justified recommendations for action. In many areas, however, a change is not highly time critical or an event can be predicted with sufficient accuracy in advance, for example a weather change or an arrival of a transport freighter in a designated port. In many cases, alternatively or additionally, a response time is not critical, for example if there is a delay in delivery due to a traffic jam or if a supply store has reached or will soon reach a critical (e.g., upper) fill level. At the same time, however, the overall situation of such a supply chain, for example starting from the source (for example, sea water and wind energy for the production of green hydrogen) via logistics, possibly processing (for example, production of a liquid eFuel) to the delivery points (for example, gas stations), is so complex that time criticality arises and requires real-time processing. Such real-time processing takes place in the range of one minute to one second. It should be noted that the data situation is so complex - due to the direct use of real-time data from the subscribers -that this cannot be done by humans (at least not with reasonable effort). Historical data is at least advantageous, usually necessary, to draw reliable conclusions from (current) real-time data. The historical data is in one embodiment identical to the (current) real-time data, preferably wherein the newly obtained realtime data becomes historical data, for example supplementing or replacing it. Preferably, historical data is already available before the first application, for example as training data and / or knowledge data, in order to be able to generate appropriate recommendations for action. The data store is a single physical memory or is composed of a plurality of memories, wherein the individual memories are not necessarily configured to communicate with each other. For example, the data store or at least one of the plurality of stores is stored on a central server. A computer processor is capable of accessing the data held in the data store (directly and / or indirectly, as well as in processed and / or unprocessed form). In a simple embodiment, the number of interfaces corresponds to the number of subscribers in the supply chain. Preferably, the number of interfaces is determined by the measurement points of a respective subscriber, i.e., its measurement data can be recorded directly as real-time data. Alternatively or additionally, the real-time data has been processed before being forwarded via the associated interface and / or output data generated explicitly for the method proposed here (for example to protect know-how and / or personal rights). Preferably, a subscriber is free to decide in what form the respective real-time data is provided and / or via which and how many (proprietary) interfaces it is made available for the real-time process to be executed. For example, real-time data (direct and unprocessed) at an interface is a temperature measurement at a processing unit, for example in an electrolyser during water splitting to extract molecular hydrogen from sea water, for example). For example, real-time data (indirect and processed) at an interface is a status output of an energy store (for example, a salt reservoir for stored heat energy), such as the currently available or usable electrical energy and / or free storage capacity. As described above, the simulation environment is a model of (e.g., physical) reality, held as a digital twin. In one embodiment, insufficient real-time data is supplemented by fictitious scenarios or the system reacts accordingly to scenarios that have already occurred. For example, real-time data may be supplemented by observations such as the monitoring of freight ships at sea or in port if information provided by the shipping company is insufficient. Similarly, weather data based on historical observations can be used to derive the expected maximum wind yield (i.e., the electricity generation capacity) for a wind farm, provided that the wind farm operator’s information is limited to the current generation (i.e., available energy quantity), for example, and no forecast is output. The model provides clear instructions for action, i.e., dedicated specifications that allow the supply chain to be adapted. For example, due to the imminent absence of wind at a wind farm, a switch is made to a geothermal source for the production of molecular hydrogen in order to continue to meet current demand (cost-effectively and / or climate-neutrally). Conversely, when demand is low, for example, the currently available capacities are not fully utilised when storage facilities are full (including shiploads, for example), with available energy not being used or being used differently. For example, this is carried out in a market price-controlled manner by means of interfaces to (i.e., monitoring of) the relevant commodity exchanges and / or other subscribers, such as traders. In an advantageous embodiment, the output specifications are specified in such a way that they are returned to the respective subscriber as (direct or indirect) control variables via an associated interface (for example, the same interface from which the real-time data originates). These input variables resulting from the specifications then correspond to a (small-scale) order, for example for a hydrogen producer for a desired amount of hydrogen. It is further proposed, in an advantageous embodiment of the real-time method, that a plurality of sets of dedicated specifications be output. A set of specifications is compiled for one participant each, for example, or alternatively for a project controller. For example, a set always contains all specifications, i.e., also those that do not currently require a change. This serves as assurance and / or the possibility for human plausibility checks that the specifications are consistent and / or for documenting the processes. For example, if the specifications are communicated purely electronically via the interfaces, an error message is output if a set is incomplete. In one embodiment, human intervention (necessarily) only takes place in the event of such an error message (or an error message caused in another way). It is further proposed in an advantageous embodiment of the real-time method that the complex infrastructure comprises alternatively acting subscribers. In one embodiment, as already indicated above, several subscribers in a supply chain or in parallel supply chains of an (overall) infrastructure are in competition with one another, for example even though they have different tasks and produce different products. In terms of power generation, these include wind farms, geothermal sources, and solar plants. In fuel production, for example, these are the producer of hydrogen and the producer of methanol, although the producer of methanol is often downstream of the producer of hydrogen in a supply chain. It is further proposed in an advantageous embodiment of the real-time method that alternative green fuels may be produced from a common source and / or common raw material by means of the infrastructure. The alternative fuels are preferably produced based on a set of rules determined in advance and / or dynamically adapted. The decisions that are supported by real-time recording are derived from the set of rules. This infrastructure makes it possible to produce green fuels from hydrogen in the form of liquefied hydrogen or, alternatively, methanol (liquid under normal conditions) by adding carbon dioxide. Methanol may in turn be provided as a raw material for the production of diesel or gasoline. The same applies to the sources of energy supply, i.e., from wind, geothermal sources, or solar energy. In many cases, the raw material and the energy source are linked via the respective plant or its location, so that, for example, wind energy is used in Scotland, geothermal sources in Iceland, waste heat from a steelworks in Germany, and solar thermal energy or photovoltaics for water splitting and / or methanol production in Morocco. However, the location then again has a strong impact on the downstream logistics for transporting the green fuel to the location of demand from the customers or the sub-processors. It is further proposed in an advantageous embodiment of the real-time method that at least one of the following external influences and / or associated forecasts be considered in real-time: weather; prices for producible green fuels, raw materials, and / or services; demand; capacities of processing, storage, and / or logistics; unplanned cost changes; geopolitical influences; and deviations from upstream and / or downstream processes of subscribers in a potentially efficient supply chain. In one embodiment of the present invention, the real-time method for simulating a supply chain for a green fuel takes into account multiple external influences and / or associated forecasts, which are captured by a plurality of sensors, loT [Internet of Things] devices, databases, and / or interfaces to information services and processed in a computer. The external influences considered include, for example, the weather, which is monitored in real time (directly) by means of weather sensors and weather satellites or an external service provider, and / or which forecasts are obtained from meteorological data models. In one embodiment, the prices for producible green fuels, raw materials, and / or services are recorded in real time, whereby this information can, for example, come from commodity exchanges or price services. Demand for green fuels and capacity in processing, storage, and / or logistics are factors considered in one embodiment, which are taken into account in the computer’s simulation environment to ensure efficient utilisation of the existing infrastructure. In one embodiment, unplanned cost changes (for example, additional costs or rebates) that may arise from maintenance, breakdowns, wage increases, or similar factors are incorporated into the simulation. Geopolitical influences, which are preferably provided by news agencies and political analysis institutes, are also preferably incorporated into the simulation in order to be able to identify possible impacts on the supply chain early and react accordingly. For example, an upcoming change of government due to political sensibilities and / or specific electoral promises could prompt a forecast that changes are to be expected with regard to customs duties, taxes, and / or legal certainty in general. In one embodiment, deviations from upstream and / or downstream processes are considered, for example because they are slower or faster than expected. For example, these can occur due to strikes, breakdowns, or other unexpected events and directly impact the efficiency of the supply chain. By taking such factors into account, it is possible to optimise the supply chain for green fuels at all times and thus ensure the availability of these essential energy sources. It is further suggested, in an advantageous embodiment of the real-time method, that artificial intelligence is employed to identify possible decisions. The uncertainties that arise when transporting a green fuel, especially hydrogen, can be absorbed and minimised by artificial intelligence [Al], For example, the containers may be monitored to determine whether they could not be loaded onto the ship as planned. In this case, the system can automatically suggest possible solutions, such as alternative shipping routes, so that the delivery obligation to the customer can still be fulfilled. Al-powered analysis of historical data and real-time data on the status of all parts of the digital twin makes it possible to identify anomalies. This can be used to derive and optimise maintenance requirements, for example, in order to avoid production downtimes and to increase the efficiency of the entire production process. Taking into account data on future environmental conditions, such as the weather forecast, Al can be used to model various scenarios that provide information on the expected energy supply. Through a continuous learning process of the algorithms, the scenarios become more and more precise over time and thus enable accurate forecasts. Approaches using so-called neural networks have established themselves as particularly suitable and easy to handle. For large amounts of data, an approach with a so-called generative pre-trained transformer is particularly promising. A Generative Pre-trained Transformer (GPT) is an advanced algorithm based on the principle of machine learning. A GPT is able to process and analyse large amounts of data (for example, also known as big data) in order to independently derive forecasts and decisions. Preferably, such a GPT is trained in such a way that it is able to improve its prediction accuracy independently by continuously processing new input data. Preferably, such a GPT is set up to take into account so-called invisible or hidden boundary conditions, for example correlations that are generally not recognisable or observed by a person or events that are classified as unimportant. In one embodiment, the artificial intelligence used is a pre-trained generative transformer, a component of the computer that utilises pre-trained algorithms to identify specific targets in real time. This transformer is responsible for analysing and predicting optimal utilisation of the infrastructure by generating forecasts based on historical data and real-time data received via the simulation environment. The use of this transformer enables the efficient utilisation and efficient shaping of a supply chain, ensuring adaptation to constantly changing conditions and requirements. Preferably, the transformer is self-learning and / or self-enhancing, so that with each analysis it increases its accuracy in finding effective defaults for the participants in the complex infrastructure. It should be noted that (self-)learning represents a permanent change to the algorithm for future processing. Self-enhancing is executable within a task solution (i.e., approximation or optimisation) and does not have to lead to a permanent change in the algorithm. Preferably, however, self-enhancement is used as a meaningful form of self-learning. It should be noted that the algorithm only uses confidential data sources and / or treats the results confidentially. It should be noted that the simulation environment is a human-generated and comprehensible model or (if possibly additionally) a machine-learned model, wherein the internal links in the model are not comprehensible to a human (or at least not without further processing). In one embodiment, the transformer used is the model itself or a user of the model. In one embodiment, a mixture of both and / or a plurality of independent transformers are used. It is further proposed in an advantageous embodiment of the real-time method that the computer automatically outputs at least a portion of the specific specifications and / or reasons for identifying these specifications in the form of reports that can be read by both humans and computers. Creating reports is important for quality assurance and exchanging information within a network of participants. These reports are automatically generated but designed for human readability, for example as a computer file or on paper. In one embodiment, a (possibly graduated) certification is created which indicates how great the climate impact of the respective green fuel is. For example, a non-climate neutral energy source is used temporarily (e.g., as a supplement) to meet demand and / or to keep a process flow stable (e.g., to maintain a constant temperature and / or pressure). For example, a fuel batch is then assigned a reduced quality, e.g., designated as 80% climate-neutral, which then has consequences for the callable purchase price or required climate taxes, for example. Another important aspect here is also to identify future potential for improvement (for example, regarding ship propulsion systems and alternative energy sources) and to promote innovation in these areas. Specifically, it is preferably a boundary condition of the real-time method to achieve a particularly high quality. The above-described invention is discussed in detail in the following in the context of the relevant technical background with reference to the accompanying drawings which show preferred embodiments. The invention is not limited in any way by the purely schematic drawings, whereby it should be noted that the drawings are not true to scale and are not suitable for defining dimensional relationships. The figures show: Fig. 1: schematically an infrastructure and its integration into a controller using a digital twin; and Fig. 2: a decision tree for determining a supply chain within an infrastructure. Fig. 1 schematically shows an infrastructure 5 for producing and / or supplying a green fuel and its integration into a controller using a digital twin. Three levels are shown on the left in the illustration, namely, a network of participants 6 in a supply chain 1, at the top, wherein here, for example, a symbol for a factory 13 is shown for the source 9, a network for the logistics 11 (here pars-pro-toto the symbol for a freighter 14, a truck 15, a container 16, and a transshipment port 17) and a filling station 18 (as end customer 19 for green fuel 2,3). In the middle level, the currently active subscribers 6 are shown and below is the subscriber 6 (only optionally here in the truck 15), which is given a specification here. A computer 7 with a data store 4, in which the simulation environment 8 corresponding to the infrastructure 5 shown on the left and historical data are stored, and a screen (e.g., for reporting 12) are shown at the bottom right. The computer 7 (e.g., a central server) is used to generate specifications for the operation of infrastructure 5. That is, how the physical equipment of the infrastructure can be controlled to generate or supply a green fuel in a more optimal manner (e.g., more efficiently). At the top right, a satellite is shown as a symbol for data communication 20 of the resulting real-time data via the interfaces of the individual participants 6 with the computer 7, possibly over long distances. Examples of tasks to accomplish include: - monitoring the environment in which the physical equipment of the infrastructure resides (e.g., based on a weather forecast, humidity, wind speed, light intensity, etc.), - monitoring and controlling hydrogen production (e.g., in terms of production capacity, production quantity, production quality) associated with an electrolysis system, - monitoring and controlling transport of raw materials (e.g., biomass); - monitoring and optimising storage capacity of raw material; - monitoring and optimizing storage capacity of electrical power; - monitoring the current and expected status of physical equipment (e.g., with regards to electrical production capacity, material supply, and / or location); Other tasks to accomplish may include: managing digital networks, for example risk management through scenario analysis and / or network control and network monitoring; controlling transport or logistics 11, for example delivery management for transport to customers and material transport between the process steps; monitoring and optimising storage capacity and warehouse planning; and managing the goods handling, and for example monitoring the container availability, as well as its status and predicting required containers 16. Fig. 2 shows a decision tree for determining a supply chain 1 within an infrastructure 5. At the top of the diagram, the processing computer 7 is a central server with a symbolic representation of an input and output of a plurality of data via a plurality of interfaces in data communication 20 with a real infrastructure 5 (shown at the top left here with reference to Fig. 1). A sequence of stations or subscribers 6 is shown below, which are in competition with each other or represent alternative branches in the course shown in parallel. On the left, the process begins with source 9 for energy and raw materials 10 for a green fuel 2, 3. The energy source is shown at the top with a symbol for a wind turbine 21, whereby, for example, the electrical energy currently generated or made available and the usability of wind energy is monitored. In the middle is a symbol for available carbon dioxide 22, where, for example, quantity and quality are monitored, and to the right is a carbon dioxide store 23, where storage capacity and access to the store are monitored. Below is a symbol for water as a raw material source for green hydrogen 2, where, for example, the quantity and quality are monitored. A factory plant 13 for producing green hydrogen 2 is then shown, whereby, for example, the production capacity, production quantity and production quality, the energy source, and any external influences are monitored. A symbol for a hydrogen store is shown below, whereby, for example, the storage capacity and the required containers 16 are monitored. The process then splits into alternative green fuels 2, 3. In the top row, liquefied hydrogen 2 is produced and transported as a green fuel. The first symbol represents the liquefaction, whereby, for example, the production capacity, the production quantity and production quality, the storage capacity, the required containers 16, and the energy source used are monitored. The second symbol of this upper sequence section shows a symbol of a truck 15 for the transport of the liquefied (green) hydrogen 2, whereby, for example, real-time geolocation monitoring is performed, geopolitical and / or environmental conditions are monitored, and various scenarios of this usually complex logistics 11 are played out. This is followed by a symbol for the regasification 24 of the hydrogen 2, whereby, for example, the quality and the input amount are monitored. Finally, the distribution to the final recipients 19 (not shown here) is shown in the upper flow section with the symbol of a truck 15, whereby, for example, real-time geolocation monitoring and fill level monitoring are again carried out, and various scenarios of these usually complex distribution logistics are run through. In the bottom line, methanol 3 is produced and transported as a green fuel. The first symbol represents a tank canister for the production of methanol 3 by adding carbon dioxide 22, whereby, for example, the production capacity, the production quantity and production quality, the storage capacity, the required containers 16, and the energy source used are monitored. The second symbol of this lower sequence section shows a symbol of a truck 15 for the transport of the (green) methanol 3, whereby, for example, real-time geolocation monitoring is performed, geopolitical and / or environmental conditions are monitored, and various scenarios of this usually complex logistics 11 are played out. A symbol for a transshipment port 17 is shown next to it, whereby, for example, the availability of ports and the storage capacity are monitored. Finally, the distribution to the final recipients 19 (not shown here) is shown in the upper flow section with the symbol of a truck 15, whereby, for example, real-time geolocation monitoring and fill level monitoring are again carried out, and various scenarios of these usually complex distribution logistics are run through. The IT infrastructure required for communicating data includes the various interfaces along the supply chain 1, which are consolidated on a central server, and that all relevant data is displayed in a user-friendly way, with the option of accessing additional detailed views, such as the integrated digital part twins. For optimisation, simulations are performed using historical and real-time data to analyse different scenarios and identify the optimal action alternative. That is, identify the optimal supply chain for generating or supplying green fuel. Based on this identification, the system may involve controlling (or instructing subscribers) physical equipment of the infrastructure accordingly. For example, artificial intelligence algorithms (preferably a neural network) are employed to independently derive specific and ideal control strategies for follow-up actions. The following two application examples with certain (simplified and thereby comprehensible) scenarios or framework conditions are shown: Example 1: It is expected that wind capacity (for example, strength and duration) will decrease in the next week, meaning that production targets will not be met. However, the current wind capacity is so high this week that more hydrogen 2 could be produced than is needed for the current demand. Relevant and therefore particularly noteworthy parameters [KPI: Key Point of Interest] are, for example: wind availability, power source, production capacity, storage capacity. Other parameters [KPI] include: customer demand, and risk assessment. Strategies to optimize the combination of these parameters include, for example: all relevant parameters and real-time data are used to develop scenarios centrally, to meet set production targets, this week we will be using the wind capacity available to replenish the reservoirs, and more hydrogen 2 will be produced in the coming week to compensate for reduced production capacity. Example 2: A delivery to the end-users 19 is delayed due to problems with a subscriber 6 in the area of logistics 11. As a result, end-users 19 will receive their ordered amount of green fuel 2, 3 later than requested, and the containers 16 needed for storage will not arrive at the production site in a timely manner. Relevant and therefore particularly noteworthy parameters [KPI: Key Point of Interest] are, for example: geopolitical and environmental (weather) conditions, demand from end-users 19, storage data, real-time container tracking, number and / or type of container 16 required, production capacity, distribution scenarios, and risk assessment Strategies to achieve these parameters include, for example: centralised development and analysis of supply chain scenarios, an adjustment of the production quantity is possible because the customers have only limited storage capacities, and containers 16 must be procured for storage by other means in order to continue production. With the real-time method proposed herein for simulating a supply chain, green fuel can be produced cost-effectively and on demand.

Claims

1. A real-time method for monitoring and simulating a supply chain for a green fuel, comprising the following constituents:in a data store, holding historical data on an infrastructure for producing a green fuel, the infrastructure being complex due to a plurality of subscribers, wherein alternative green fuels are producible from a common source and / or common raw material by means of the infrastructure;via a plurality of interfaces, obtaining real-time data relating to the infrastructure, wherein the real-time data includes a weather measurement of an environment within which the infrastructure resides; andin a computer in a simulation environment based on the historical data held and the real-time data obtained, forecasting an optimal use of the infrastructure to form an efficient supply chain and / or to efficiently exploit a supply chain for producing the alternative green fuels,in response to said forecasting, outputting dedicated specifications to at least one of the plurality of the subscribers of the complex infrastructure to cause said at least one subscriber to operate respective physical equipment of the infrastructure for producing the alternative green fuels, in accordance with said optimal use of the infrastructure.

2. The real-time method according to claim 1, whereina plurality of sets of dedicated specifications are output.

3. The real-time method according to claim 1 or claim 2, wherein, the complex infrastructure comprises alternatively acting subscribers.

4. The real-time method according to any one of the preceding claims, wherein,at least one of the following external influences and / or associated forecasts are taken into account in real time:prices for producible green fuels, raw materials, and / or services;demand;capacities of processing, storage, and / or logistics;unplanned cost changes;geopolitical influences; anddeviations from upstream and / or downstream processes of subscribers in a potentially efficient supply chain.

5. The real-time method according to any one of the preceding claims, wherein artificial intelligence is used for identifying possible decisions.

6. The real-time method according to any one of the preceding claims, wherein the computer automatically outputs at least a portion of the specific specifications and / or reasons for identifying these specifications in the form of reports that can be read by both humans and computers.29 01 26A

Images