Methods for selecting energy storage systems for recycling in a circular economy based on digital twins

Abstract

The invention relates to a method for selecting energy storage devices for a recycling process, comprising the following steps: - Providing (S1) operational size profiles of a large number of energy storage devices in their respective applications; - Evaluating (S2) the operating parameter profiles of each energy storage system using an energy storage model to obtain, as energy storage data, a current aging state and one or more internal system states; - Providing (S3) energy storage information for each of the energy storage devices, wherein the energy storage information specifies at least one energy storage type of the respective energy storage device; - Provide (S4) at least one recycling request for at least one recycling process, wherein the recycling request specifies at least one material type and a corresponding quantity of material; - Selecting (S5, S6) energy storage devices depending on the at least one recycling request and the energy storage type, and depending on a suitability value that indicates the suitability of the respective energy storage device for immediate recycling, so that only energy storage devices with the worst suitability for the respective applications are selected that match the at least one recycling request; - Feeding (S7) the selected energy storage device to at least one recycling process.

Description

Technical field

[0001] The invention relates to methods for the needs-based selection of aged energy storage devices for a recycling process. Technical background

[0002] Energy storage devices, such as device batteries, especially vehicle batteries, may be designed as systems containing large quantities of valuable raw materials.

[0003] At the end of its service life, the materials used in energy storage devices are generally not depleted. Therefore, energy storage devices can usually be fully recycled to recover raw metals such as lithium, zinc, and the like. These can then be processed and used to manufacture new energy storage devices.

[0004] Furthermore, energy storage devices are often used in technical equipment beyond their planned lifespan, either in their initial application or in subsequent uses. Subsequent use may be intended for stationary systems, meaning that regular replacement is not always possible.

[0005] The performance of an energy storage system typically degrades over time and depending on its usage, so that, regardless of its calendar age, different aging states and internal system states can be reached due to cyclical aging. These internal system states can also be evaluated to identify potential anomalies in the energy storage system. These anomalies may necessitate immediate decommissioning or simply indicate accelerated future aging or degradation and an impending premature failure.

[0006] Energy storage devices can generally be operated beyond their planned end of life in an application if a reduced performance is acceptable. The end of life is typically defined by a drop in performance, particularly storage capacity, below a predetermined threshold. For example, vehicle batteries whose end of life is defined as 80% of their capacity-related state of health (SOH-C) can also be operated at aging states below this value. If necessary, the vehicle battery can be used, for instance, in a stationary battery operation mode.

[0007] The introduction of old energy storage devices into a recycling loop typically occurs on a request basis through a recycler or in the form of a futures contract, e.g., via a smart contract or similar mechanism. The aim is to remove from active use those device batteries that are particularly unsuitable for reuse. Disclosure of the invention

[0008] According to the invention, a method for selecting energy storage devices for a recycling process according to claim 1 and a corresponding device according to the dependent claim are provided.

[0009] Further details are specified in the dependent claims.

[0010] According to a first aspect, a procedure for selecting energy storage for a recycling process is provided, with the following steps: - Providing operational size profiles of a large number of energy storage systems in their respective current and / or planned applications; - Evaluating the operating parameter profiles of each energy storage device in order to obtain, as energy storage data, a current aging state and one or more internal system states; - Providing energy storage information for each of the energy storage devices, wherein the energy storage information specifies at least one energy storage type of the respective energy storage device; - Providing a recycling request for a recycling process, wherein the recycling request specifies at least one material type and a corresponding quantity of material; - Selecting energy storage devices based on the recycling request and the respective energy storage type, and based on a suitability score that indicates the suitability of the respective energy storage device for immediate recycling, so that only energy storage devices matching the recycling request with the worst suitability for the respective applications are selected; - Feeding the selected energy storage devices into the recycling process

[0011] It may be envisaged that the selection of energy storage devices is based on a query of a system containing a knowledge graph, with the recycling request and the respective energy storage type of the energy storage devices in question.

[0012] Furthermore, the suitability value of a given energy storage device can be determined depending on a suitability function, whereby the suitability function takes into account one or more of the following: - a quantity of CO2 released by feeding the specific energy storage into the recycling process, in particular by transporting it from the location of the energy storage to a recycling point in the recycling process, - the current state of aging, - a remaining lifetime determined using a predicted aging state trajectory, - the presence of an anomaly, - an exceedance or fall below a predefined threshold value by one or more internal system states.

[0013] The above method is based on the availability of historical operating parameter data for each of the energy storage devices under consideration. In the case of a device battery, the operating parameters include at least battery current and temperature, as well as battery voltage and state of charge at the pack, module, and cell levels. A corresponding usage pattern of the energy storage device can be determined from the historical operating parameter profiles. This makes it possible to model an aging state.

[0014] Furthermore, in the case of a device battery, an electrochemical (P2D) battery model can be parameterized as an energy storage model, allowing for the computational determination of internal battery states (system states). The P2D (pseudo-two-dimensional) battery model is a mathematical model for describing the electrochemical processes within a battery. The P2D battery model is capable of predicting the performance and behavior of batteries under various operating conditions and, in particular, of modeling the terminal voltage of the device battery.

[0015] The P2D model considers various physical and chemical processes in a battery, such as lithium-ion transport (the transport of lithium ions between and within the anode and cathode materials), electron transport (the transport of electrons through the anode, electrolyte, and cathode), electrochemical reactions (reactions that occur during charging and discharging the battery), and heat generation and dissipation during battery operation. The P2D model uses partial differential equations to describe these processes and predict how the battery will behave under different operating conditions. For example, the P2D battery model can be used to determine the internal battery states of a device battery, from which the battery's aging state can be derived.The parameterization of the model parameters of a P2D battery model is based on time series of operating parameters of the device battery using numerical optimization methods.

[0016] Thus, for example, each of the energy storage systems under consideration can be provided as a model in the form of a digital twin. Here, each physical unit of a vehicle or energy storage fleet can be viewed and modeled as its own instance of a digital twin in order to describe serial variation effects and the individual characteristics of multiple energy storage systems within a fleet.

[0017] Furthermore, the internal system states resulting from the energy storage model parameterized for a specific energy storage device can be evaluated, particularly on an energy storage device-specific basis in a digital energy storage instance twin, in order to detect, for example, a possible anomaly in the energy storage device in question.

[0018] Energy storage data can now include a current aging state, a predicted aging state trajectory, internal system states, and any anomalies.

[0019] Furthermore, the energy storage data for each energy storage device includes information on the energy storage type, nominal capacity, remaining minimum operating time, and the like.

[0020] It may be provided that the selection of energy storage devices is based on energy storage information about a remaining minimum operating time in the respective application.

[0021] Initiated by a recycling request, which can be submitted by a recycling company or via a forward contract, for example, a suitability score can now be determined based on the energy storage data and the specifications of the recycling request. The recycling request can include, for example, one or more of the following details: one or more materials required; a quantity of material; a recycling location; and the like.

[0022] This can be done, for example, using a knowledge graph.

[0023] A knowledge graph is a network of entities (things, concepts, people, places, etc.) and the relationships between them. These entities and connections are often derived from various sources. Traceability refers to the ability to trace the origin of a particular entity or relationship within a knowledge graph. It involves knowing and documenting the source or origin of the information. Traceability in a knowledge graph can be achieved through additional metadata associated with each entity or relationship, containing information about its provenance.

[0024] Based on the specifications of the recycling request and the energy storage data, a search can be performed in the knowledge graph, which uses semantic modeling to describe, for example, which active materials the device battery of a specific energy storage type contains and in what quantities. The search criteria used to query the knowledge graph can include whether the materials defined in the recycling request are present, the availability of the energy storage devices in terms of their remaining minimum service life, their location and accessibility, and their current value for the current application. This value can be determined, for example, based on the current state of aging, the predicted aging state trajectory, the internal system states, the presence of anomalies, and other energy storage data.The suitability rating is determined by the extent to which the energy storage device is suitable for remaining in use in the application assigned to the respective energy storage device.

[0025] The suitability value can be derived from a suitability function that evaluates the energy storage data and, in particular, considers or calculates the CO2 footprint in order to feed the energy storage device in question into the recycling process. The suitability value can be determined, for example, based on the distance and transport options to the recycling location, the current state of aging and remaining service life in the application, the presence of anomalies, and similar factors.

[0026] Based on the suitability values, the energy storage systems can now be sorted and prioritized in order to provide the requested amount of recycled material by feeding it into the recycling process.

[0027] Furthermore, the suitability value can be determined using a semantic traceability link, especially depending on the amount of CO2 released.

[0028] Furthermore, automated storage in a database with semantic linking in the knowledge graph is possible, so that a current effective amount of CO2 released is calculated and provided based on usage patterns, maintenance or material exchange.

[0029] The goal is to identify, based on a ranking of suitability values, those energy storage devices that are least suitable for their current application. These will be selected for removal from the applications so that they can be replaced by newer energy storage devices.

[0030] Once the appropriate energy storage systems for feeding into the recycling process have been identified, the recycling request can be answered automatically, if necessary, in particular by specifying a pricing model that relates to the suitability values.

[0031] The above method enables the automated evaluation and selection of energy storage devices used in various applications with regard to their suitability for recycling. This allows for the needs-based support of a recycling process in response to a recycling request.

[0032] According to another aspect, a device for carrying out the above procedure is provided. Brief description of the drawings

[0033] The embodiments are explained in more detail below with reference to the accompanying drawings. These show: Fig. 1. A system with a large number of device batteries in appropriate applications, monitored by means of a remote central unit; and Fig. 2. A flowchart illustrating a procedure for selecting device batteries for recycling; and Fig. 3. An exemplary visualization of a knowledge graph. Description of embodiments

[0034] Fig. Figure 1 schematically shows a system 1 with a variety of device batteries 3 as possible energy storage devices used in various applications 2. These applications can include mobile applications, such as a vehicle battery, as well as stationary applications, such as a home battery, solar battery, or the like. The energy storage devices can also include other electrochemical energy storage devices, such as a fuel cell system.

[0035] Using a respective battery management system 4 assigned to the device batteries, operating parameters of the device batteries, in particular battery current, battery voltage (pack, module and battery cells), state of charge and battery temperature, can be recorded as time series and transmitted to a central unit 5.

[0036] In the central processing unit 5, energy storage data can be determined based on the temporal operating parameter profiles, providing the best possible representation of the current battery state. This energy storage data can include a current aging state based on the temporal operating parameter profiles, internal battery states based on the parameterization of an electrochemical battery model, such as a P2D battery model, a predicted aging state trajectory, and an anomaly detected from the aforementioned energy storage data.

[0037] The central unit 5 is designed to monitor the cell aging states of each individual battery cell 3 by maintaining an associated electrochemical battery model and / or an aging state model. These models are updated or evaluated as soon as operating parameter profiles are available for the respective battery cell / device battery, enabling an update of the internal battery state and / or aging state associated with that battery cell. This results in a representation of the device battery as a "digital twin" in the central unit, where an internal battery state and an aging state can be specified for each battery cell of the monitored device batteries.

[0038] However, evaluating the battery state using the temporal operating parameter profiles, e.g., using the electrochemical battery model and / or the electrochemical aging model, requires providing the operating parameter profiles with high temporal resolution, i.e., with a high sampling frequency, e.g., between 1 Hz and 100 Hz. Transmitting operating parameter profiles for the battery cells and / or the device battery 3 with high temporal resolution results in large amounts of data that must be transmitted from the numerous applications to the central processing unit 5.

[0039] A knowledge graph, described via semantic modeling, can be implemented in database 6 of the central processing unit 5. The knowledge graph contains, for example, information about the materials used to construct the battery types of the applications and the proportions or quantities of each material. Fig. Figure 3 shows such a knowledge graph. The nodes represent entities, and the edges between the nodes indicate a relationship between the entities.

[0040] The knowledge graph can contain information about, among other things: - Digital product twins, such as: ◯ Master data and type data; ◯ Quality information; ◯ Service data; ◯ Logistics data; ◯ Requirements information and specifications for: ▪ Internal processes, such as manufacturing within the company; ▪ External processes, such as manufacturing at suppliers or customers; ◯ Design, layout and engineering information ▪ Changes to all specifications over time, such as engineering change requests and change orders; ◯ General causal chain modeling, such as: ▪ Behavioral models, for example for calculating a stress response; ▪ Aging models, for example for calculating and predicting at least one battery state, such as SEI thickness or available lithium or aging state; ◯ Validation data; ◯ Release data; ◯ Meta-information about the context; ◯ For information on the digital product passport, see also this source here: - Digital instance twins, such as instances for all the aforementioned product information. These include, in particular: ◯ Data for the unambiguous identification of the physical twin; ◯ Manufacturing data from a digital twin instance. Such as, at least; ▪ An end-of-line measurement; ▪ A process parameter; ◯ Quality instance data, such as ▪ Workshop data and history; ▪ Claim data and history; ▪ Time series data, specifically sensor data, that was recorded in the vehicle and sent to the cloud; ▪ Diagnostic data and history such as diagnostic trouble codes; ◯ Operational data, such as: ▪ at least one individually learned parameter in an observer model; ▪ at least one parameter of an operating strategy ▪ Data entry for at least one model, e.g. to calculate a CO2 footprint; ◯ Meta-information about the context ▪ For example, serial variation of the instance into which the product was integrated, e.g. the Hydrogen Gas Injector (HGI) is integrated into the Fuel Cell Power Model (FCPM), which may also exhibit serial variation.

[0041] Fig. Figure 2 shows, using a flowchart, the process for selecting and feeding device batteries into a recycling process. The process can be implemented as software and / or hardware in a processing unit of the central processing unit 5.

[0042] In step S1, operating parameter data from all device batteries of all applications 2 are continuously transmitted to the central unit 5. The central unit 5 stores the operating parameter data as operating parameter profiles for each of the device batteries 3 separately.

[0043] In step S2, one or more battery models are trained or parameterized for the individual device batteries. For example, an aging state model can be used, which is created with operating parameter data from a large number of device batteries of the same battery type. By evaluating the aging model based on differential equations or on a hybrid aging model, the current aging state of the respective device battery can be determined.

[0044] Alternatively, an electrochemical battery model can be created for each of the device batteries. This battery model can be a P2D battery model, where the model parameters are fitted using a least-squares or maximum likelihood approach to determine internal battery states.

[0045] The electrochemical battery model comprises a system of differential equations that, based on differential equations parameterized via model parameters, models internal battery states, in particular equilibrium states and, if applicable, kinetic states, using a time integration method and provides a relationship between the operating parameters of the device battery cells, namely battery current, battery voltage, battery temperature, and the state of charge of the device battery. Such electrochemical battery models are known, for example, from US 2016 / 023,566, US 2016 / 023,567, and US 2020 / 150,185.

[0046] This provides internal battery states and an aging state for each individual device battery, which are provided as battery data or energy storage data.

[0047] In addition, battery information is available for each of the device batteries in the form of battery data, which specifies the battery type, the nominal capacity, the remaining minimum operating time in the application in question, its calendar age, and the like.

[0048] In step S3, further evaluations from the internal battery states and the operating size trends can be added to the battery data.

[0049] For example, based on stress factors affecting the respective device battery, a prediction of its aging state can be made to obtain an indication of its future degradation. Such an evaluation can be performed regularly, e.g., once a week, and provides an aging state trajectory for predicted aging states. The aging state prediction is based on a load profile assigned to the respective application. If the application is a vehicle battery, the load profile can be driver-specific and derived from historical driving behavior.

[0050] Furthermore, anomaly detection can be performed based on the internal battery states to determine whether the behavior of the respective device battery corresponds to the normal behavior or whether the behavior deviates from it and an imminent failure of the corresponding device battery is to be expected.

[0051] Overall, the battery states, the current aging state, the predicted aging state trajectory, battery information and / or information about possible anomalies are provided as battery data.

[0052] The battery data can be stored in the central unit 5 in the form of a digital twin. The battery data of the digital twin can optionally be made available for retrieval with a signed usage certificate.

[0053] The digital twin in the device battery can be connected to a knowledge graph based on semantic technologies. This knowledge graph models ontologies in a data model, such as the material properties of all monitored device batteries and the battery types.

[0054] In step S4, the process checks whether a recycling request exists. This request can be based on a specific inquiry from a recycling company or on a request related to a futures contract (smart contract). The recycling request can specify, for example, one or more material types, a specific quantity of each material type, a timeframe for the availability of the recycled material, and similar parameters. An automated search can then be performed in the knowledge graph based on the recycling request's parameters.

[0055] The knowledge graph describes, for example, via semantic modeling, the cathode material used in the various battery types in the applications and the quantities of each. The search criteria used to query the knowledge graph can include the specific materials of the recycling request as well as information on the recyclability of the respective battery type. The individual device batteries identified as potential recycling candidates based on the knowledge graph search are then provided as a selection set. Furthermore, only those device batteries whose availability is sufficient to meet the minimum remaining service life requirements of the respective application can be included in the selection set.

[0056] In a subsequent step S5, a ranking of the device batteries in the selection set is created using a suitability value.

[0057] The suitability value for a specific device battery can, for example, take into account, using a predefined suitability function, the CO2 production resulting from feeding the specific device battery into the recycling process (e.g., through transport from the device battery's location to the recycling site and transport options), the current state of aging and the remaining service life, which can be determined using the predicted aging state trajectory, whether an anomaly exists, whether one or more internal battery states exceed or fall below a predefined threshold, and the like.

[0058] It may be provided that an automated estimate of the current CO2 footprint takes place, with automated storage in a database with semantic linking in the knowledge graph, so that a current effective CO2 footprint is calculated based on usage patterns or maintenance or material exchange and made available in real time or regularly, e.g. once a week.

[0059] The resulting suitability value can indicate the general suitability of a device battery in the application in question.

[0060] For the recycling process, step S6 involves selecting the weakest device batteries, i.e., those devices least suitable for the respective application. The number of devices selected from the sample set corresponds to the quantity specified in the recycling request, ensuring that the requested quantity of recycled material can be provided.

[0061] These device batteries can now be fed into the recycling process in step S7.

[0062] If necessary, the recycling request, provided it is based on a forward contract, can be answered automatically by assigning a bid price to the suitability value, which is then offered in response to the forward contract. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] US 2016 / 023,566

[0045] US 2016 / 023,567

[0045] US 2020 / 150,185

[0045]

Claims

[1] Procedure for selecting energy storage devices for a recycling process, comprising the following steps: - Providing (S1) operating variables of a plurality of energy storage devices in their respective applications; - Evaluating (S2) the operating variables of each energy storage device using an energy storage model in order to obtain a current ageing state and one or more internal system states as energy storage data; - Providing (S3) energy storage information for each of the energy storage devices, wherein the energy storage information indicates at least one energy storage type of the respective energy storage device; - Providing (S4) at least one recycling request for at least one recycling process, wherein the recycling request specifies at least one material type and a corresponding material quantity; - selecting (S5, S6) energy storage devices depending on the at least one recycling request and the energy storage type and depending on a suitability value that indicates a suitability of the respective energy storage device for immediate recycling, so that only energy storage devices matching the at least one recycling request with the worst suitability for the respective applications are selected; - feeding (S7) the selected energy storage device to the at least one recycling process. [2] Method according to claim 1, wherein the selection of the energy storage devices is based on the result of a query of a system comprising at least one knowledge graph with the recycling request and the respective energy storage type of the energy storage device in question. [3] The method according to claim 2, wherein the knowledge graph contains at least one of the following information: - Information on a Digital Product Twin, in particular at least one of master data and type data, quality information, service data, logistics data, requirement information and specifications for internal processes and external processes, design, layout and engineering information, general impact chain modelling, validation data, release data, meta-information on the context, and information on the digital product passport, and / or - Information about a digital instance twin, in particular instances of all the above-mentioned product information, manufacturing data of a digital instance twin, quality instance data, operational data, meta-information about the context. [4] Method according to one of claims 1 to 3, wherein the selection of the energy storage devices is based on energy storage information about a remaining minimum operating time in the respective application. [5] Method according to one of claims 1 to 4, wherein the suitability value of a respective energy storage device is determined as a function of a suitability function, wherein the suitability function takes into account one or more of the following: - an amount of CO2 released resulting from the supply of the specific energy storage device to the recycling process, in particular from the transport from the location of the energy storage device to a recycling location of the recycling process, - the current state of aging; - a remaining lifetime determined using a predicted ageing trajectory, - exceeding or falling below a specified threshold value by one or more internal system states. [6] Method according to one of claims 1 to 5, wherein an ontology is used in the knowledge graph, materializing and / or virtually. [7] Method according to one of claims 1 to 6, wherein a number of energy storage devices is selected depending on a quantity specified in the at least one recycling request depending on a suitability value. [8] Method according to one of claims 1 to 7, wherein the suitability value is determined using a semantic traceability link, in particular depending on an amount of CO2 released. [9] Method according to one of claims 1 to 8, wherein an automated storage in a database with semantic linking in the knowledge graph takes place, so that on the basis of usage patterns or maintenance or material exchange a current effective amount of released CO2 is calculated and provided. [10] Apparatus for carrying out one of the methods according to one of claims 1 to 9. [11] Computer program product comprising instructions which, when the program is executed by at least one data processing device, cause the device to carry out the steps of the method according to one of claims 1 to 9. [12] Machine-readable storage medium comprising instructions which, when executed by at least one data processing device, cause it to carry out the steps of the method according to one of claims 1 to 9.

Images