Method for categorizing used li-ion batteries

Abstract

Systems and methods disclosed herein comprise providing operational history and an electrolyte of a used Li-ion battery to a machine-learning model; receiving, from the machine-learning model, an estimate of a state of health (SoH) of the used Li-ion battery; reading parameters of the used Li-ion battery; providing the parameters and the estimate of the SoH of the used Li-ion battery to a machine-learning model trained to output a rate of degradation of the SoH of the used Li-ion battery in response to receiving parameters and a SoH; receiving, from the machine-learning model, a rate of degradation of the SoH of the used Li-ion battery; generating, based on the estimate of the SoH and the rate of degradation, a recommendation for an application of the used Li-ion battery, the application being a second-life application, recycling, or end-of-life; and providing the used Li-ion battery and a recommendation to a facility.

Description

Attorney Docket No.: BGW-00225METHOD FOR CATEGORIZING USED LI-ION BATTERIESRELATED APPLICATION

[0001] This application claims the benefit of U.S. Provisional Application. No. 63 / 701,850 filed October 1, 2024, which is hereby incorporated by reference in its entirety.TECHNICAL FIELD

[0002] Embodiments of the present disclosure are related to a system, method, and computer program product for battery lifecycle management. For example, some embodiments relate to a system, method, and computer program product for categorizing used lithium-ion (Li-ion) batteries into second-life applications or ready-to-recycle categories.BACKGROUND OF THE DISCLOSURE

[0003] Lithium-ion (Li-ion) batteries are rechargeable batteries that can be used in many applications. As Li-ion batteries age, their health can degrade. For example, a Li-ion battery may hold less charge, drain its charge faster, or charge slower as its health degrades.SUMMARY

[0004] In some embodiments, a process using minimal testing and characterization determines a State of Health (SoH) and Remaining Useful Life (RUL) of used Li-ion batteries.

[0005] The presently disclosed subject matter is directed to a method for categorizing a used lithium (Li)-ion battery. In some embodiments, the method comprises providing an at least partial set of an operational history and a representation of an electrolyte of a used Li- ion battery to a trained machine-learning model, the trained machine-learning model trained to output a state of health in response to receiving an operational history and a representation of an electrolyte of a Li-ion battery. The method includes receiving, from the trained machine-learning model, an estimate of a state of health of the used Li-ion battery. The method includes reading parameters of the used Li-ion battery, the parameters comprising- 1 -FH13093063.1Attomey Docket No.: BGW-00225 one or more of a diffusion coefficient, reaction kinetics, and thermal behavior. The method includes providing the parameters and the estimate of the state of health of the used Li-ion battery to a physics-based machine-learning model, the physics-based machine-learning model trained to output a rate of degradation of the state of health of the used Li-ion battery in response to receiving parameters and a state of health. The method includes receiving, from the physics-based machine-learning model, a rate of degradation of the state of health of the used Li-ion battery. The method includes generating, based on the estimate of the state of health and the rate of degradation, a recommendation for an application of the used Li-ion battery, the application being one or more of a second-life application, recycling, and end-of- life. The method includes providing the used Li-ion battery and an associated recommendation to a battery facility.

[0006] In some embodiments, a battery facility is a facility that processes, classifies, and employs for second-life usages used Li-ion batteries.

[0007] In some embodiments, the method further comprises reading a serial number associated with the used Li-ion battery and loading, from a memory and based on the serial number, the at least partial set of the operational history of the used Li-ion battery and a representation of an electrolyte of the used Li-ion battery.

[0008] In some embodiments, a serial number is any identifier that is marked on a battery that can be used to load operational history, data, or other metadata about a battery.Examples of a serial number can be a numeric code, alphanumeric code, a string of letters, a QR code, a bar code, or any other information that can identify a unique battery.

[0009] In some embodiments, data of the at least partial set of the operational history of the Li-ion battery is derived from one or more of: open circuit voltage testing, impedance testing, and capacity testing.

[0010] In some embodiments, the method further comprises measuring the at least partial set of the operational history of the used Li-ion battery using one or more open circuit voltage testing, impedance testing, and capacity testing.

[0011] In some embodiments, the method further comprises determining a level of completeness of the operational history and measuring the at least partial set of operational history of the used Li-ion battery when the level of completeness of the operational history is below a threshold.

[0012] In some embodiments, the representation of the electrolyte is selected from one or more of: lithium cobalt oxide, nickel manganese cobalt, and lithium iron phosphate.- 2 -FH13093063.1Attomey Docket No.: BGW-00225

[0013] In some embodiments, the operational history includes data fields for one or more of date or timestamp, voltage, current, capacity, and temperature associated with the used Li-ion battery.

[0014] In some embodiments, the operational history comprises usage data of the used Li- ion battery.

[0015] In some embodiments, the usage data includes one or more of charge cycles, discharge cycles, voltage, current, and temperature.

[0016] In some embodiments, the method further comprises determining a level of completeness of the usage data, wherein the level of completeness is one of complete, partially complete, and not available.

[0017] In some embodiments, estimating the state of health comprises using one or more of calendar aging and cycling aging based on the operational history.

[0018] In some embodiments, the operational history comprises a most recent operational data segment and one or more partial charge and / or discharge curves.

[0019] In some embodiments, generating the recommendation is based on a threshold of the state of health.

[0020] In some embodiments, the state of health is above the threshold and the used Li-ion battery is categorized as suitable for a second life application.

[0021] In some embodiments, the state of health is below the threshold and the used Li-ion battery is categorized as end of life.

[0022] In some embodiments, the method further comprises providing the used Li-ion battery to a battery recycling facility with a recommendation to recycle the used Li-ion battery.

[0023] In some embodiments, generating the recommendation for the application of the used Li-ion battery comprises running a remaining useful life (RUL) simulation on the used Li-ion battery.

[0024] In some embodiments, the application is a second-life application.

[0025] In some embodiments, the RUL simulation predicts one or more second-life operational scenarios based on the state of health.

[0026] In some embodiments, the method furthr comprises reading a serial number associated with the used Li-ion battery, loading, from a memory and based on the serial number, metadata of the battery associated with the serial number, the metadata comprising one or more of the electrolyte of the used Li-ion battery, a use case of the Li-ion battery, an age of the Li-ion battery, a capacity of the Li-ion battery, or a physical size of the Li-ion- 3 -FH13093063.1Attomey Docket No.: BGW-00225 battery, and determining the at least partial set of the operational history based on the metadata.

[0027] In some embodiments, the method further comprises determining an amount of testing and / or characterization required to efficiently categorize the used Li-ion battery into second-life application or ready to recycle.

[0028] In some embodiments, the method further comprises measuring an internal impedance of the used Li-ion battery and providing the measured impedance to the trained machine-learning model, wherein the trained machine-learning model is further trained to output the state of health in response to receiving the operational history, the representation of the electrolyte, and an impedance measurement, wherein receiving the estimate of the state of health is further based on the measured impedance provided to the trained machinelearning model.

[0029] In some embodiments, the internal impedance is measured by applying a predefined alternating-current excitation to the battery and recording a resulting voltage response.

[0030] In some embodiments, the method further comprises grouping the used Li-ion battery with one or more additional batteries having substantially similar internal impedance values as the internal impedance of the used Li-ion battery for integration into a battery pack.

[0031] In some embodiments, the trained machine-learning model is a physics-based simulation model and the operational history provided to the trained machine-learning model is free of battery capacity data.

[0032] In some embodiments, the method further comprises conducting a partial-capacity test by charging or discharging the used Li-ion battery for a predetermined duration and incorporating resulting test data into the state of health estimation.

[0033] The presently disclosed subject matter is directed to a system for categorizing a used lithium (Li)-ion battery. In some embodiments, the system comprises a computing node comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor of the computing node to cause the processor to perform the method of any of the preceding method claims.

[0034] The presently disclosed subject matter is directed to a computer program product for categorizing a used lithium (Li)-ion battery. In some embodiments, the computer program product comprises a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform the method of any of the preceding method claims.- 4 -FH13093063.1Attomey Docket No.: BGW-00225

[0035] In some embodiments, a system comprises a scanner configured to read a serial number of a used Li-ion battery, a test station configured to perform at least one of an opencircuit-voltage test, an impedance measurement, or a partial-capacity test on the used Li-ion battery, and a computing unit including at least one processor and a memory storing instructions that, when executed, cause the computing unit to perform any of the above methods.

[0036] The presently disclosed subject matter is directed to a method of training a machinelearning model to determine a state of health of a battery. In some embodiments, the method comprises reading, for each of a plurality of used Li-ion batteries: an operational history, the operational history comprising operational data of that battery, a representation of an electrolyte of that used Li-ion battery, and a state-of-health of that used Li-ion battery, the state-of-health representing a measure of electrical charge capacity of that Li-ion battery. The method includes generating a training dataset comprising: the operational history of each of the used Li-ion batteries, the representation of the electrolyte of each of the used Li-ion batteries, and the state of health of each used Li-ion battery, the generated training dataset comprising representations of at least two different electrolytes. The method includes providing the training dataset to a machine-learning model. The method includes training the machine-learning model on the training dataset, resulting in a trained machine-learning model trained to output a state of health in response to receiving an operational history and a representation of an electrolyte of a Li-ion battery.

[0037] In some embodiments, the operational history comprises usage data of the battery.

[0038] In some embodiments, the usage data comprises data fields of one or more of date, timestamp, voltage, current, capacity, and temperature.

[0039] In some embodiments, the method further comprises determining a level of completeness of the usage data, wherein the level of completeness is one of complete, partially complete, and not available.

[0040] In an embodiment, a method comprises reading a serial number associated with the used Li-ion battery; loading, based on the serial number, an operational history of the used Li-ion battery; determining a level of completeness of the operational history; estimating a state of health for the used Li-ion battery, based on the operational history of the used Li-ion battery; generating, based on the state of health, a recommendation for an application of the used Li-ion battery, the application being one or more of a second-life application, recycling, and end-of-life; and providing the used Li-ion battery and an associated category to a battery facility.- 5 -FH13093063.1Attomey Docket No.: BGW-00225

[0041] In some embodiments, loading the operational history comprises reading the operational history from a database.

[0042] In some embodiments, the database includes one or more date or timestamp, voltage, current, capacity, and / or temperature associated with the used Li-ion battery.

[0043] In some embodiments, the operational history comprises usage data of the battery.

[0044] In some embodiments, the usage data includes one or more of charge cycles, discharge cycles, voltage, current, and temperature.

[0045] In some embodiments, determining the level of completeness comprises determining the level of completeness is one of complete, partially complete, and not available.

[0046] In some embodiments, estimating the state of health of the used Li-ion battery comprises applying a machine learning algorithm to the operational history.

[0047] In some embodiments, estimating the state of health comprises using one or more of calendar aging and cycling aging based on the operational history.

[0048] In some embodiments, estimating the state of health is performed using a machine learning model, the machine learning model configured to receive a partially complete operational history and output a state of health. In some embodiments, the machine learning model is physics-based.

[0049] In some embodiments, the operational history comprises a most recent operational data segment and one or more partial charge and / or discharge curves.

[0050] In some embodiments, estimating the state of health is based on the most recent operational data and the one or more partial charge and / or discharge curves.

[0051] In some embodiments, generating the recommendation is based on a threshold of the state of health.

[0052] In some embodiments, the threshold is 20%.

[0053] In some embodiments, the state of health is above the threshold and the used Li-ion battery is categorized as suitable for a second life application.

[0054] In some embodiments, the method further comprises running a remaining useful life (RUL) simulation on the used Li-ion battery.

[0055] In some embodiments, the method further comprises determining, based on the RUL simulation, a second-life application of the recommendation.

[0056] In some embodiments, the RUL simulation predicts one or more second-life operational scenarios based on the state of health.- 6 -FH13093063.1Attomey Docket No.: BGW-00225

[0057] In some embodiments, the method further comprises testing the used Li-ion battery to measure the state of health when the level of completeness of the operational history is not available.

[0058] In some embodiments, the method further comprises extrapolating the state of health based on historical usage patterns associated with the serial number.

[0059] In some embodiments, the state of health is below the threshold and the battery is categorized as end of life.

[0060] In some embodiments, the method further comprises providing the used Li-ion battery to a battery recycling facility with a recommendation to recycle the used Li-ion battery.

[0061] In some embodiments, the method further comprises determining an amount of testing and / or characterization required to efficiently categorize the used Li-ion battery into second-life application or ready to recycle.

[0062] In an alternative embodiment, a system for categorizing a used Li-ion battery comprises a computing node comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor of the computing node to cause the processor to perform a method comprising: reading a serial number associated with the used Li-ion battery; loading, based on the serial number, an operational history of the used Li-ion battery; determining a level of completeness of the operational history; estimating a state of health for the used Li-ion battery, based on the operational history of the used Li-ion battery; generating, based on the state of health, a recommendation for an application of the used Li-ion battery, the application being one or more of a second-life application, recycling, and end-of-life; and providing the used Li-ion battery and an associated recommendation to a battery facility.

[0063] In another alternative embodiment, a computer program product for categorizing a used Li-ion battery comprises a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method comprising: reading a serial number associated with the used Li-ion battery; loading, based on the serial number, an operational history of the used Li-ion battery; determining a level of completeness of the operational history; estimating a state of health for the used Li-ion battery, based on the operational history of the used Li-ion battery; generating, based on the state of health, a recommendation for an application of the used Li-ion battery, the application being one or more of a second-life application, recycling,- 7 -FH13093063.1Attomey Docket No.: BGW-00225 and end-of-life; and providing the used Li-ion battery and an associated recommendation to a battery facility.BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The accompanying drawings, which are incorporated into and constitute a part of this specification, illustrate various exemplary embodiments and together with the description, serve to explain the principles of the disclosed embodiments.

[0065] FIG. 1A is a flowchart illustrating an exemplary method for categorizing a used battery, in accordance with one or more embodiments of the present disclosure.

[0066] FIG. IB is a flowchart illustrating an exemplary method for categorizing a used battery, in accordance with one or more embodiments of the present disclosure.

[0067] FIG. 2A-2B is a flowchart illustrating an exemplary method for categorizing a used battery, in accordance with one or more embodiments of this disclosure.

[0068] FIG. 3 is a series of graphs illustrating a workflow for predicting a state of health of a used Li-ion battery.

[0069] Fig. 4 is a block diagram illustrating a system for estimating battery state of health and generating a battery recommendation according to embodiments of the present disclosure.

[0070] Fig. 5 is a block diagram of training a machine learning model to estimate a state of health of a battery according to embodiments of the present disclosure.

[0071] FIG. 6 is a flowchart illustrating an exemplary method for training a machinelearning model according to embodiments of the present disclosure.

[0072] FIG. 7 is an exemplary computing node.DETAILED DESCRIPTION

[0073] Reference will now be made in detail to the exemplary embodiments of the present disclosure, examples of which are illustrated in the accompanying drawings. Wherever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts.

[0074] The systems, devices, and methods disclosed herein are described in detail by way of examples and with reference to the figures. The examples discussed herein are examples only and are provided to assist in the explanation of the apparatuses, devices, systems, and methods described herein. None of the features or components shown in the drawings or- 8 -FH13093063.1Attomey Docket No.: BGW-00225 discussed below should be taken as mandatory for any specific implementation of any of these devices, systems, or methods unless specifically designated as mandatory.

[0075] Also, for any methods described, regardless of whether the method is described in conjunction with a flow diagram, it should be understood that unless otherwise specified or required by context, any explicit or implicit ordering of steps performed in the execution of a method does not imply that those steps must be performed in the order presented but instead may be performed in a different order or in parallel.

[0076] As used herein, the term "exemplary" is used in the sense of "example," rather than "ideal." Moreover, the terms "a" and "an" herein do not denote a limitation of quantity, but rather denote the presence of one or more of the referenced items.

[0077] The term "about" means a range of values inclusive of the specified value that a person of ordinary skill in the art would reasonably consider to be comparable to the specified value. In some embodiments, about means within a standard deviation using measurements generally accepted by a person of ordinary skill in the art. In embodiments, about means ranging up to ±10% of the specified value. In embodiments, about means ranging up to ±5% of the specified value. In embodiments, about means the specified value.

[0078] The term "substantially similar" means a range of values inclusive of the specified value that a person of ordinary skill in the art would reasonably consider to be comparable to the specified value. In some embodiments, substantially similar means within a standard deviation using measurements generally accepted by a person of ordinary skill in the art. In embodiments, substantially similar means ranging up to ±10% of the specified value. In embodiments, substantially similar means ranging up to ±5% of the specified value. In embodiments, substantially similar means the specified value.

[0079] This disclosure pertains to the field of battery lifecycle management, specifically to methods for categorizing used lithium-ion (Li-ion) batteries based on their remaining usability or health. Alternative methods for sorting used Li-ion batteries often involve manual inspections or basic electronic testing, which can be time-consuming and inaccurate. These techniques may not effectively determine the Remaining Useful Life (RUL) or accurately categorize batteries into second-life application or ready-to-recycle categories. In circular economy models, seeking maximum value from each battery before determining the battery has reached its end of usable life is key to achieving sustainability and cost-efficiency goals.

[0080] Current methods to determine the health and reusability of batteries are limited in terms of their practicality, scalability, or accuracy in field conditions. Current methods such- 9 -FH13093063.1Attomey Docket No.: BGW-00225 as coulomb counting and incremental capacity / differential voltage analysis (ICA / DVA) are time-consuming and impractical for real-time monitoring of battery health. For example, Coulomb counting requires full charge and discharge cycles to estimate capacity fade, which is often infeasible in settings where uninterrupted service is expected. ICA and DVA require slow and stable charge / discharge profiles and high-resolution data. Therefore, ICA and DVA are unsuitable for most commercial or second-life applications.

[0081] Other current methods, such as internal resistance or impedance measurements, can suffer from variability due to temperature, state of charge, and measurement conditions. While internal resistance or impedance measurements can be fast and embedded in battery management systems, the variability described above limits their standalone reliability. Thermal behavior monitoring offers indirect insights into degradation, but it is coarse and imprecise, typically useful only when combined with other metrics. Current systems therefore lack the ability to optimize charging, discharging, and balancing based on both remaining capacity of a battery and a specific application for the battery, which can leave significant unutilized value.

[0082] Therefore, current estimation techniques are too slow, too sensitive to conditions, or too limited in scope to meet the demands of real-world, high-volume battery applications, particularly in repurposing, second-life, or fleet-scale monitoring contexts. There is a growing need for more adaptive, data-driven, and scalable approaches that can provide accurate insights without interrupting battery operation.

[0083] Accordingly, there is a need for an efficient method using minimal testing and manual inspection.

[0084] Embodiments of the present disclosure provide a process for categorizing used Li- ion batteries by determining their SoH and RUL. This process can involve minimal testing and characterization to efficiently sort batteries into one of at least two categories: batteries suitable for second-life applications and those ready to be recycled at their End of Life (EOL). In some embodiments, a partial-capacity test can be performed by charging or discharging the used Li-ion battery for a predetermined duration and incorporating resulting test data into the state of health estimation. In some examples, about six minutes of test data can result in 97% accuracy or higher.

[0085] When operational data about the battery is available, the SoH can be estimated based on the operational history of the battery without additional testing.

[0086] Batteries that need evaluation fall into one of three categories. The first category can comprise batteries for which all operational history is available through a battery health- 10 -FH13093063.1Attorney Docket No.: BGW-00225 monitoring system, (e.g. , a Bridge Green Upcycle (BGU) Battery Health Monitoring System). The operational history can include State of Charge (SoC) and State of Health (SoH). In other words, the first category can comprise batteries that have had their entire operational history recorded, and the battery health monitoring system can make that history accessible at a later time. Using calendar and cycling aging methods, a current state of health can be determined from that operational history. No testing is required for these batteries.

[0087] The second category of batteries can comprise batteries that have a partial operational history available. The partial history can comprise an incomplete operational history, or an operational history with some segments of the operational history available but other segments missing. For batteries in this second category, an artificial intelligence (AI)- based SoH estimation model can be employed to estimate state of health of the battery. An alternative example of an Al-based SoH estimation can be performed by the method outlined in Su et al., “Battery Charge Curve Prediction via Feature Extraction and Supervised Machine Learning,” DOI: 10.1002 / advs.202301737, hereinafter “Su,” incorporated herein by reference in its entirety. Partial charge / discharge curves obtained from the monitoring system can be input to the Al-based SoH model for estimating current SoH. In some embodiments, the models described herein are models that are trained on a training data set of partial charge / discharge curves. The resulting models can therefore receive partial charge / discharge curves and provide a state of health estimate in response.

[0088] Utilizing partial charge / discharge curves when training the Al-based SoH model significantly accelerates the battery screening process, as the model can rapidly estimate the Remaining Useful Life (RUL) without requiring complete testing cycles. This efficiency enables faster decision-making and throughput when evaluating large volumes of used batteries. Furthermore, the same predictive model can be integrated directly into the battery management system (BMS) to provide more accurate, real-time SoH estimations during battery operation. This in-situ capability supports optimized battery usage and prolongs lifecycle by adapting to changing conditions. When operational data from second-life battery systems is available, it can be incorporated to continually enrich and refine the model, improving its accuracy and robustness over time across diverse applications and usage profiles.

[0089] The third category can comprise batteries that have no operational data available. For these batteries, battery characterization tests are carried out to determine a current SoH.

[0090] Once a battery’s SoH is estimated, a second-life recommendation engine can be employed to recommend an application for the battery based on that estimated SoH or actual- 11 -FH13093063.1Attomey Docket No.: BGW-00225SoH. The second-life recommendation engine comprises simulation engine that models various scenarios of operation for the battery and determines a most appropriate application for the RUL of the battery.

[0091] FIG. 1A is a flowchart illustrating an exemplary method 100 for categorizing a used Li-ion battery according to embodiments of the present disclosure. Method 100 (i.e., steps 102-114) can be performed automatically or in response to a request from a user. In step 102, the method can include reading a serial number associated with the used Li-ion battery. The serial number can be read by scanning a barcode, or performing OCR on a printed label, for example.

[0092] In some embodiments, in step 104, the method can include loading, based on a serial number, an operational history and a representation of an electrolyte of the used Li-ion battery. The electrolytes can include lithium cobalt oxide, nickel manganese cobalt, lithium iron phosphate, and any electrolyte that can be employed by lithium ion battery implementations. The operational history can comprise usage data of the battery, parameters of the battery, or both. Usage data may include, but is not limited to, charge cycles, discharge cycles, voltage, current, and temperature associated with the battery. Parameters may include, but are not limited to diffusion coefficients, reaction kinetics, and thermal behavior. A database can be used to store the operational history of the used Li-ion battery, organized by serial number. For example, the database can be a cloud-based database such as an Amazon Web Services database, a Google Cloud database, or Mongo database. The database may include searchable fields such as date and / or timestamp, voltage, current, capacity, temperature, data relating to the operational history of the battery, parameters, etc.

[0093] In step 106, the method can include determining a level of completeness of the operational history. The level of completeness of the operational history can include a complete history, where all operational history is accessible, a partially complete history, where only some operational history is accessible, and a not available history, where none of the operational history is accessible.

[0094] When the level of completeness of the operational history is not available, the method can include further testing of the battery to measure the state of health (SoH). Testing methodologies, or battery testing methodologies, can include, but are not limited to, open circuit voltage testing, impedance testing, and capacity testing. The method can include determining an amount of testing and / or characterization needed to categorize the Li-ion battery. The method can include delivering the battery to a testing station, which can comprise a scanner configured to read the serial number of the battery, testing materials- 12 -FH13093063.1Attorney Docket No.: BGW-00225 necessary for one or more testing and / or characterization methods, and a computing unit capable of receiving resultant testing data. Upon the method determining that one testing methodology is necessary (e.g, impedance testing), the method can perform that testing methodology (e.g., impedance testing, such as by applying a predetermined alternating- current excitation to the battery and recording a resulting voltage response). The data resulting from performing that testing data can be provided to the SoH estimation model, which can thereby generate an SoH estimate. The resultant data can also be provided to the SoH estimation model as testing data for further refinement of the model. The method can also include extrapolating and estimating the operational history based on historical usage patterns associated with a serial number of the used Li-ion battery. For example, the serial number can provide battery properties such as the battery’s chemistry (e.g., electrolyte), use case, age, or size. The method can determine (e.g., extrapolate) an estimate of operational history by applying those factors loaded for the used Li-ion battery to known operational history of batteries having similar battery properties. Once the operational history is estimated, it can be provided to the SoH estimation model, which thereby provides an estimated SoH.

[0095] In step 110, the method can include estimating a SoH for the used Li-ion battery based on the operational history of the used Li-ion battery. Estimating the state of health can use calendar aging, cycling aging methods based on the operational history, a most recent operational data segment, one or more partial charge curves, and / or one or more partial discharge curves. Estimating the SoH can employ a physics-based machine learning model. Physics-based models can be trained based on electrochemical principles and degradation mechanics. A physics model can receive parameters such as diffusion coefficients, reaction kinetics, and thermal behavior and output metrics that are used to simulate a rate of battery degradation under certain operational conditions. In some embodiments, the physics-based machine learning model can be a generalist model trained on operational histories of batteries having multiple battery chemistries and states of health of those batteries. The resulting trained model can receive an operational history of a battery having any of the multiple battery chemistries and the electrolyte of that battery, and output a state of health of that battery. In some embodiments, the physics-based machine learning model can be a specialist machine learning model trained on operational histories of batteries having a specific battery chemistry and states of health of those batteries to output a state of health of a batteiy having that battery chemistry. The resulting trained model can receive an operational history of a- 13 -FH13093063.1Attomey Docket No.: BGW-00225 battery having any of the specific battery chemistry, and output a state of health of that battery.

[0096] The trained model can be a physics-based model. The trained physics-based model can simulate battery degradation by modeling solid electrolyte interphase (SEI) growth, lithium plating, and electrode degradation using electrochemical relationships. The physicsbased model can receive inputs comprising or equivalent to current profiles, temperature, state of charge (SoC), and material properties of the battery. The physics-based model can predict state of health (SoH) by tracking loss of lithium inventory or active material, impacting capacity and resistance. The hybrid approach of scalable machine learning assessments and the interpretive power of mechanistic simulations allows for the contextualization of battery health within the framework of a specific use case or application. A battery with 75% SoH can be ideal for a low-power application but unsuitable for one requiring high current bursts. For example, the battery may not be appropriate for an electric vehicle, but can be employed as a battery backup for a residential home. By combining machine learning outputs with simulation models that factor in chemistry-specific aging behaviors and load profiles, the simulation models provide a more nuanced understanding of long-term performance, safety, and economic viability under real-world conditions.

[0097] In step 112, the method can include generating, based on the SoH, a recommendation for an application of the used Li-ion battery, the application being one or more of a second life application, recycling, and end-of-life. Generating the recommendation can be based on the threshold percentage of the SoH. For example, the threshold SoH can be about 20%, or another predetermined percentage chosen by a user. Batteries with a SoH above the threshold can be categorized as suitable for a second life application. The method can categorize a battery as end-of- life when its SoH is below a threshold. If the SoH is below the threshold and the battery is recommended for end-of-life, the battery can be provided to a battery facility. The battery facility can be a recycling facility, and the recommendation may be to recycle the battery.

[0098] In step 114, the method can include providing the used Li-ion battery and an associated recommendation to the battery facility.

[0099] In some embodiments, a physics-based aging model can receive inputs such as stress factors. The stress factors can include temperature, state of charge (SoC) and charge / discharge current. The physics-based aging model can combine calendar aging of the battery (e.g., time based) and cycle-aging (e.g., usage based) effects. The model can then output a rate of degradation of the model. The rate of degradation can be expressed as a- 14 -FH13093063.1Attomey Docket No.: BGW-00225 capacity fade or a resistance growth over time or a number of cycles. For Remaining Useful Life (RUL), the physics-based model extends this by projecting future usage and stress conditions to estimate when SoH reaches end-of-life thresholds. The physics-based model for calculating RUL can be trained on or receive as inputs historical and forecasted load cycles, environmental data, and degradation parameters to simulate life expectancy.

[0100] Optionally, the method can further include running a Remaining Useful Life (RUL) simulation. The RUL simulation can include an operational scenario of a use case that the battery can be employed for in the future. The RUL simulation can comprise a coded sequence of simulated events or a model, where the user evaluates the battery performance against various simulations. The RUL simulation can receive inputs that comprise the rate of degradation received from the physics-based model, the parameters of the battery, operational history of the battery, results from testing the battery, and / or an estimate of the battery’s SoH. The RUL simulation can be partially or wholly based on a physics-based machine learning model.

[0101] Based on the results of the RUL simulation, a second-life application can be determined for inclusion in the recommendation. The second-life application can be based on a determined rate of degradation of the SoH of the battery. Further, the RUL simulation can predict one or more second life operational scenarios based on the SoH of the battery. Second life recommendations may include, but are not limited to, stationary grid storage, backup power system, or low-power mobility solutions. Batteries with substantially similar internal impedance values can be grouped together to create a battery pack, therefore extending the usefulness of the battery. In some embodiments, internal impedance values can be values that are substantially similar can be values that are within 1%, 2%, 3%, 4%, 5%, 10%, 15%, or 20%. In some embodiments, substantially similar impedance values can be determined by a clustering method.

[0102] In some embodiments, the method can also include determining an amount of testing and / or characterization required to efficiently sort batteries into second-life applications and ready-to-recycle categories, for batteries without sufficient operational history.

[0103] FIG. IB is a flowchart illustrating an exemplary method 150 for categorizing a used Li-ion battery according to embodiments of the present disclosure. Method 150 (z.e., steps 152-164) can be performed automatically or in response to a request from a user. In step 152, the method can include providing an at least partial set of an operational history and a representation of an electrolyte of a used Li-ion battery to a trained machine-learning- 15 -FH13093063.1Attomey Docket No.: BGW-00225 model, the trained machine-learning model trained to output a state of health in response to receiving an operational history and a representation of an electrolyte of a Li-ion battery.

[0104] In some embodiments, data of the at least partial set of the operational history of the Li-ion battery is derived from one or more of: open circuit voltage testing, impedance testing, and capacity testing. In some embodiments, the representation of the electrolyte is selected from one or more of: lithium cobalt oxide, nickel manganese cobalt, and lithium iron phosphate. In some embodiments, the operational history comprises usage data of the used Li-ion battery. In some embodiments, the usage data includes one or more of charge cycles, discharge cycles, voltage, current, and temperature. In some embodiments, the operational history comprises a most recent operational data segment and one or more partial charge and / or discharge curves.

[0105] In some embodiments, the method can further include reading a serial number associated with the used Li-ion battery and loading, from a memory and based on the serial number, the at least partial set of the operational history of the used Li-ion battery and a representation of an electrolyte of the used Li-ion battery. A database can be used to store the operational history of the used Li-ion battery, organized by serial number. For example, the database can be a cloud-based database such as an Amazon Web Services database, a Google Cloud database, or Mongo database. The database may include searchable fields such as date and / or timestamp, voltage, current, capacity, temperature, data relating to the operational history of the battery, parameters, etc.

[0106] In some embodiments, the method can further include measuring the at least partial set of the operational history of the used Li-ion battery using one or more open circuit voltage testing, impedance testing, and capacity testing. In some embodiments, the method can include determining a level of completeness of the operational history and measuring the at least partial set of operational history of the used Li-ion battery when the level of completeness of the operational history is below a threshold.

[0107] In some embodiments, the representation of the electrolyte is selected from one or more of: lithium cobalt oxide, nickel manganese cobalt, and lithium iron phosphate.

[0108] In some embodiments, in step 154, the method 150 can include receiving, from the trained machine-learning model, an estimate of a state of health of the used Li-ion battery.

[0109] In some embodiments, in step 156, the method 150 can include reading parameters of the used Li-ion battery, the parameters comprising one or more of a diffusion coefficient, reaction kinetics, and thermal behavior.- 16 -FH13093063.1Attomey Docket No.: BGW-00225

[0110] In some embodiments, in step 158, the method 150 can include providing the parameters and the estimate of the state of health of the used Li-ion battery to a physics-based machine-learning model, the physics-based machine-learning model trained to output a rate of degradation of the state of health of the used Li-ion battery in response to receiving parameters and a state of health. In some embodiments, estimating the state of health comprises using one or more of calendar aging and cycling aging based on the operational history.

[0111] In some embodiments, in step 160, the method includes receiving, from the physics-based machine-learning model, a rate of degradation of the state of health of the used Li-ion battery.

[0112] In some embodiments, in step 162, the method includes generating, based on the estimate of the state of health and the rate of degradation, a recommendation for an application of the used Li-ion battery, the application being one or more of a second-life application, recycling, and end-of-life. In some embodiments, generating the recommendation is based on a threshold of the state of health. In some embodiments, the state of health is above the threshold and the used Li-ion battery is categorized as suitable for a second life application. In some embodiments, the state of health is below the threshold and the used Li-ion battery is categorized as end of life.

[0113] Estimating the state of health can use calendar aging, cycling aging methods based on the operational history, a most recent operational data segment, one or more partial charge curves, and / or one or more partial discharge curves. Estimating the SoH can employ a physics-based machine learning model. Physics-based models can be trained based on electrochemical principles and degradation mechanics. A physics model can receive parameters such as diffusion coefficients, reaction kinetics, and thermal behavior and output metrics that are used to simulate a rate of battery degradation under certain operational conditions. In some embodiments, the physics-based machine learning model can be a generalist model trained on operational histories of batteries having multiple battery chemistries and states of health of those batteries. The resulting trained model can receive an operational history of a battery having any of the multiple battery chemistries and the electrolyte of that battery, and output a state of health of that battery. In some embodiments, the physics-based machine learning model can be a specialist machine learning model trained on operational histories of batteries having a specific battery chemistry and states of health of those batteries to output a state of health of a battery having that battery chemistry. The- 17 -FH13093063.1Attomey Docket No.: BGW-00225 resulting trained model can receive an operational history of a battery having any of the specific battery chemistry, and output a state of health of that battery.

[0114] The trained model can be a physics-based model. The trained physics-based model can simulate battery degradation by modeling solid electrolyte interphase (SEI) growth, lithium plating, and electrode degradation using electrochemical relationships. The physicsbased model can receive inputs comprising or equivalent to current profiles, temperature, state of charge (SoC), and material properties of the battery. The physics-based model can predict state of health (SoH) by tracking loss of lithium inventory or active material, impacting capacity and resistance. The hybrid approach of scalable machine learning assessments and the interpretive power of mechanistic simulations allows for the contextualization of battery health within the framework of a specific use case or application. A battery with 75% SoH can be ideal for a low-power application but unsuitable for one requiring high current bursts. For example, the battery may not be appropriate for an electric vehicle, but can be employed as a battery backup for a residential home. By combining machine learning outputs with simulation models that factor in chemistry-specific aging behaviors and load profiles, the simulation models provide a more nuanced understanding of long-term performance, safety, and economic viability under real-world conditions.

[0115] In some embodiments, the method further includes providing the used Li-ion battery to a battery recycling facility with a recommendation to recycle the used Li-ion battery.

[0116] In some embodiments, generating the recommendation for the application of the used Li-ion battery comprises running a remaining useful life (RUL) simulation on the used Li-ion battery. In some embodiments, the application is a second-life application. In some embodiments, the RUL simulation predicts one or more second-life operational scenarios based on the state of health.

[0117] Based on the results of the RUL simulation, a second-life application can be determined for inclusion in the recommendation. The second-life application can be based on a determined rate of degradation of the SoH of the battery. Further, the RUL simulation can predict one or more second life operational scenarios based on the SoH of the battery. Second life recommendations may include, but are not limited to, stationary grid storage, backup power system, or low-power mobility solutions. Batteries with substantially similar internal impedance values can be grouped together to create a battery pack, therefore extending the usefulness of the battery. In some embodiments, internal impedance values can be values that are substantial similar can be values that are within 1%, 2%, 3%, 4%, 5%, 10%, 15%, or- 18 -FH13093063.1Attomey Docket No.: BGW-0022520%. In some embodiments, substantial similar impedance values can be determined by a clustering method.

[0118] In some embodiments, in step 164, the method includes providing the used Li-ion battery and an associated recommendation to a battery facility.

[0119] In some embodiments, the method can further include determining a level of completeness of the usage data, wherein the level of completeness is one of complete, partially complete, and not available.

[0120] In relation to the models described herein, a physics based modal is a computational model that can receive information about a system and predict a future attribute of that system based on physical or chemical relationships.

[0121] FIG. 2A-2B is a flowchart illustrating an exemplary method 200 for categorizing a used battery. The method 200 provides a process for categorizing used Li-ion batteries by determining each battery’s SoH and RUL. The method 200 involves minimal testing and characterization to efficiently sort batteries into those suitable for second-life applications and those ready to be recycled. The method 200 can include one of more of the steps below. In step 202, the method can include receiving an incoming battery with a serial number. The serial number can be associated with battery life data, such as operational data and historical data. The serial number can also be associated with data such as voltage, current, and temperature, or operational history data.

[0122] In step 204, the method can include scanning the serial number to read the associated battery life data.

[0123] In step 206a, the method can include reading a full operational history of the battery. The full operational history can be determined by running a battery health monitoring system, including calendar and cycling aging methods. No testing is required for batteries with a full operational history available. The method can then estimate a SoH in step 206b from this full operational history.

[0124] In step 208a, the method can include reading a partial operational history and recent data associated with the battery. For these batteries, an artificial intelligence (AI)- based SoH estimation can be run to provide missing operational history data in step 208b. The Al-based SoH estimation can be performed by an Al model as described in FIG. 3. Data from the battery monitoring system can be provided to the Al-model as input for the AI- based estimation. The data can include partial charge / discharge curves.- 19 -FH13093063.1Attomey Docket No.: BGW-00225

[0125] In step 210a, a battery may have no operational history associated with the scanned serial number. These batteries can undergo additional testing to determine a SoH. Testing can include battery characterization tests in step 210b.

[0126] Once the battery SoH is estimated, the estimate can be compared to a threshold percentage in step 212. For example, the threshold can be a 20% SoH estimate or about a 20% SoH estimate. However, any suitable threshold can be used, or be predetermined by a user, administrator, etc..

[0127] When the SoH estimate is determined to be greater than the threshold percentage, the battery is categorized as having sufficient remaining life for reuse. The RUL can be based on battery characteristics, including the estimated SoH, historical data trends, and / or the results of running the RUL simulation engine. In step 214a, the method can include running a RUL simulation engine. In step 214b, the method can include estimating the RUL for various second life applications. The second life recommendation can be output in step 214c. Output recommendations can be communicated to a user via an associated screen display or. communicated to a device or processor.

[0128] When the SoH estimate is determined to be less than the threshold percentage, the battery can be categorized as ready to recycle in step 216. The system can generate a prompt or notification to the user to recycle the battery in a designated process or location.

[0129] FIG. 3 is a flowchart, having illustrative example charge curves, illustrating a workflow for predicting a SoH. Su, referenced above, illustrates the example of Fig. 3 and describes such a workflow. Such a workflow is described in further detail below. Fig. 3 illustrates 10066 charge curves of LiNiCh cells randomly divided into a training set and a testing set with a ratio of 8:2, according to embodiments of the present disclosure. As such, 20% of the training set can be used as the validation set to determine the hyper-parameters in the models. Five-fold cross-validation was conducted to avoid overfitting.

[0130] In the workflow of Su, features can be extracted from the training dataset using principal component analysis (PCA), Non-negative matrix factorization (NMF), and an autoencoder (AE). The feature extraction thereby can generate the matrix W. The matrix can contain p columns. Each column in matrix W can represent a feature extracted from the training set.

[0131] The workflow for predicting an entire charge curve of a battery based on a portion of the charge curve is shown as in Su. Both a continuous segment (e.g., Input Type 1) and multiple separated segments (e.g., Input type 2) can be used as the input. The model can apply multiple linear regression to either input type and generate a fitted weight vector h.- 20 -FH13093063.1Attorney Docket No.: BGW-00225Then, the output charge curve Q can be determined by performing the matrix multiplication Q = W • h. The output charge curve Q can derive many key states (SOC, SOH, and remaining energy) and even the aging mechanism of the battery using the dQ dV-1analysis.

[0132] Fig. 4 is a block diagram 400 illustrating a system for estimating battery state of health and generating a battery recommendation according to embodiments of the present disclosure. A trained ML model 402 is receives inputs of an operational history 416 of a battery and an electrolyte 418 of the battery. The operational history 416 of a battery and an electrolyte 418 of the battery can optionally be loaded from a database 414. Optionally, a scanning station 422 can read an identifier of the battery 424 and provide that identifier to the database 414 to load the operational history 416 of a battery and an electrolyte 418 of the battery. Optionally, a testing station 422 can perform tests on the battery 424 to determine an at least partial operational history 416.

[0133] In some embodiments, the trained machine-learning model 402 generates an estimate of a state of health 406. A trained physics based machine-learning model can then receive parameters of the battery 420 and the estimate of the state of health of the battery 406. Optionally, the parameters 420 can be loaded from the database 414. The trained physics based machine-learning model can then output a rate of degradation 408 of the state of health of the battery.

[0134] In some embodiments, a remaining useful life model 410 can then receive the estimate of the state of health 406 and the rate of degradation of the state of health 408. The remaining useful life model 410 can then provide a recommendation 412 of a use case for the battery. It can be recognized that the battery can then be used for the purpose indicated in the recommendation 412.

[0135] Fig. 5 is a block diagram 500 of training a machine learning model to estimate a state of health of a battery according to embodiments of the present disclosure. A training dataset 510 can be loaded from a database 514 of battery information, the training dataset 510 or 510a-n. The training dataset 510a-n comprises data associated with a plurality of batteries. An exemplary item of data 510b of the training dataset 510 comprises at least an operational history 516, an electrolyte 518, and a state of health 530. It can be recognized that training data 510b is representative of the training data set as a whole, and that 510a-n include similar data. The training dataset 510 is provided to the model training factory 502. The model training factory 502 then trains a trained machine-learning model 504 based on the training dataset 510. The trained ML model 504 is trained to receive an operational- 21 -FH13093063.1Attomey Docket No.: BGW-00225 history and electrolyte of a battery not included in the training dataset 510 and output an estimate of the state of health of that battery based on those factors.

[0136] FIG. 6 is a flowchart illustrating an exemplary method 600 for training a machinelearning model according to embodiments of the present disclosure. The method can include, at step 602, reading, for each of a plurality of used Li-ion batteries, (a) an operational history, (b) a representation of an electrolyte of that used Li-ion battery, and (c) a state-of-health of that used Li-ion battery. The operational history can comprise operational data of that battery. In some embodiments, the operational history comprises usage data of the battery. In some embodiments, the usage data comprises data fields of one or more of date, timestamp, voltage, current, capacity, and temperature. The state-of-health can represent a measure of electrical charge capacity of that Li-ion battery.

[0137] The method can include, at step 604, generating a training dataset comprising (a) the operational history of each of the used Li-ion batteries, (b) the representation of the electrolyte of each of the used Li-ion batteries, and (c) the state of health of each used Li-ion battery.

[0138] The method can include, at step 606, providing the training dataset to a machinelearning model.

[0139] The method can include, at step 608, training the machine-learning model on the training dataset. The training results in a trained machine-learning model trained to output a state of health in response to receiving an operational history and a representation of an electrolyte of a Li-ion battery.

[0140] In some embodiments, the method further includes determining a level of completeness of the usage data. The level of completeness can be complete, partially complete, or not available.

[0141] Referring now to FIG. 7, a schematic of an example of a computing node is shown. Computing node 10 is only one example of a suitable computing node and is not intended to suggest any limitation as to the scope of use or functionality of embodiments described herein. Regardless, computing node 10 is capable of being implemented and / or performing any of the functionality set forth hereinabove.

[0142] In computing node 10 there is a computer system / server 12, which is operational with numerous other general purpose or special purpose computing system environments or configurations. Examples of well-known computing systems, environments, and / or configurations that may be suitable for use with computer system / server 12 include, but are not limited to, personal computer systems, server computer systems, thin clients, thick- 22 -FH13093063.1Attorney Docket No.: BGW-00225 clients, handheld or laptop devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputer systems, mainframe computer systems, and distributed cloud computing environments that include any of the above systems or devices, and the like.

[0143] Computer system / server 12 may be described in the general context of computer system-executable instructions, such as program modules, being executed by a computer system. Generally, program modules may include routines, programs, objects, components, logic, data structures, and so on that perform particular tasks or implement particular abstract data types. Computer system / server 12 may be practiced in distributed cloud computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed cloud computing environment, program modules may be located in both local and remote computer system storage media including memory storage devices.

[0144] As shown in FIG. 7, computer system / server 12 in computing node 10 is shown in the form of a general-purpose computing device. The components of computer system / server 12 may include, but are not limited to, one or more processors or processing units 16, a system memory 28, and a bus 18 that couples various system components including system memory 28 to processor 16.

[0145] Bus 18 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, Enhanced ISA (EISA) bus, Video Electronics Standards Association (VESA) local bus, Peripheral Component Interconnect (PCI) bus, Peripheral Component Interconnect Express (PCIe), and Advanced Microcontroller Bus Architecture (AMBA).

[0146] Computer system / server 12 typically includes a variety of computer system readable media. Such media may be any available media that is accessible by computer system / server 12, and it includes both volatile and non-volatile media, removable and nonremovable media.

[0147] System memory 28 can include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. Computer system / server 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage- 23 -FH13093063.1Attomey Docket No.: BGW-00225 system 34 can be provided for reading from and writing to a non-removable, non-volatile magnetic media (not shown and typically called a "hard drive"). Although not shown, a magnetic disk drive for reading from and writing to a removable, non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading from or writing to a removable, non-volatile optical disk such as a CD-ROM, DVD-ROM or other optical media can be provided. In such instances, each can be connected to bus 18 by one or more data media interfaces. As will be further depicted and described below, memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the disclosure.

[0148] Program / utility 40, having a set (at least one) of program modules 42, may be stored in memory 28 by way of example, and not limitation, as well as an operating system, one or more application programs, other program modules, and program data. Each of the operating system, one or more application programs, other program modules, and program data or some combination thereof, may include an implementation of a networking environment. Program modules 42 generally carry out the functions and / or methodologies of embodiments as described herein.

[0149] Computer system / server 12 may also communicate with one or more external devices 14 such as a keyboard, a pointing device, a display 24, etc.; one or more devices that enable a user to interact with computer system / server 12; and / or any devices (e.g., network card, modem, etc.) that enable computer system / server 12 to communicate with one or more other computing devices. Such communication can occur via Input / Output (I / O) interfaces 22. Still yet, computer system / server 12 can communicate with one or more networks such as a local area network (LAN), a general wide area network (WAN), and / or a public network (e.g., the Internet) via network adapter 20. As depicted, network adapter 20 communicates with the other components of computer system / server 12 via bus 18. It should be understood that although not shown, other hardware and / or software components could be used in conjunction with computer system / server 12. Examples, include, but are not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archival storage systems, etc.

[0150] The present disclosure may be embodied as a system, a method, and / or a computer program product. The computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present disclosure.- 24 -FH13093063.1Attomey Docket No.: BGW-00225

[0151] The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD- ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiberoptic cable), or electrical signals transmitted through a wire.

[0152] Computer readable program instructions described herein can be downloaded to respective computing / processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and / or a wireless network. The network may comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and / or edge servers. A network adapter card or network interface in each computing / processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing / processing device.

[0153] Computer readable program instructions for carrying out operations of the present disclosure may be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++ or the like, and conventional procedural programming languages, such as the “C” programming language or similar programming languages. The computer readable program instructions may execute entirely on the user’s computer, partly on the user’s- 25 -FH13093063.1Attomey Docket No.: BGW-00225 computer, as a stand-alone software package, partly on the user’s computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user’s computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), or programmable logic arrays (PLA) may execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present disclosure.

[0154] Aspects of the present disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer readable program instructions.

[0155] These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and / or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including instructions which implement aspects of the function / act specified in the flowchart and / or block diagram block or blocks.

[0156] The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process, such that the instructions which execute on the computer, other programmable apparatus, or other device implement the functions / acts specified in the flowchart and / or block diagram block or blocks.- 26 -FH13093063.1Attomey Docket No.: BGW-00225

[0157] The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flowchart illustration, and combinations of blocks in the block diagrams and / or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts or carry out combinations of special purpose hardware and computer instructions.

[0158] The descriptions of the various embodiments of the present disclosure have been presented for purposes of illustration, but are not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein was chosen to best explain the principles of the embodiments, the practical application or technical improvement over technologies found in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.- 27 -FH13093063.1

Claims

Attomey Docket No.: BGW-00225CLAIMSWhat is claimed:

1. A method for categorizing a used lithium (Li)-ion battery, the method comprising: providing an at least partial set of an operational history and a representation of an electrolyte of a used Li-ion battery to a trained machine-learning model, the trained machinelearning model trained to output a state of health in response to receiving an operational history and a representation of an electrolyte of a Li-ion battery; receiving, from the trained machine-learning model, an estimate of a state of health of the used Li-ion battery; reading parameters of the used Li-ion battery, the parameters comprising one or more of a diffusion coefficient, reaction kinetics, and thermal behavior; providing the parameters and the estimate of the state of health of the used Li-ion battery to a physics-based machine-learning model, the physics-based machine-learning model trained to output a rate of degradation of the state of health of the used Li-ion battery in response to receiving parameters and a state of health; receiving, from the physics-based machine-learning model, a rate of degradation of the state of health of the used Li-ion battery; generating, based on the estimate of the state of health and the rate of degradation, a recommendation for an application of the used Li-ion battery, the application being one or more of a second-life application, recycling, and end-of-life; and providing the used Li-ion battery and an associated recommendation to a battery facility.

2. The method of claim 1, further comprising: reading a serial number associated with the used Li-ion battery; and- 28 -FH13093063.1Attomey Docket No.: BGW-00225 loading, from a memory and based on the serial number, the at least partial set of the operational history of the used Li-ion battery and a representation of an electrolyte of the used Li-ion battery.

3. The method of claim 1, wherein data of the at least partial set of the operational history of the Li-ion battery is derived from one or more of: open circuit voltage testing, impedance testing, and capacity testing.

4. The method of claim 1 , further comprising: measuring the at least partial set of the operational history of the used Li-ion battery using one or more open circuit voltage testing, impedance testing, and capacity testing.

5. The method of claim 3, further comprising: determining a level of completeness of the operational history; and measuring the at least partial set of operational history of the used Li-ion battery when the level of completeness of the operational history is below a threshold.

6. The method of claim 1 , wherein the representation of the electrolyte is selected from one or more of: lithium cobalt oxide, nickel manganese cobalt, and lithium iron phosphate.

7. The method of claim 1 , wherein the operational history includes data fields for one or more of date or timestamp, voltage, current, capacity, and temperature associated with the used Li-ion battery.- 29 -FH13093063.1Attomey Docket No.: BGW-002258. The method of claim 1 , wherein the operational history comprises usage data of the used Li-ion battery.

9. The method of claim 8, wherein the usage data includes one or more of charge cycles, discharge cycles, voltage, current, and temperature.

10. The method of claim 8, further comprising determining a level of completeness of the usage data, wherein the level of completeness is one of complete, partially complete, and not available.

11. The method of claim 1 , wherein estimating the state of health comprises using one or more of calendar aging and cycling aging based on the operational history.

12. The method of claim 1 , wherein the operational history comprises a most recent operational data segment and one or more partial charge and / or discharge curves.

13. The method of claim 1 , wherein generating the recommendation is based on a threshold of the state of health.

14. The method of claim 13, wherein the state of health is above the threshold and the used Li-ion battery is categorized as suitable for a second life application.

15. The method of claim 13, wherein the state of health is below the threshold and the used Li-ion battery is categorized as end of life.- 30 -FH13093063.1Attorney Docket No.: BGW-0022516. The method of claim 15, further comprising providing the used Li-ion battery to a battery recycling facility with a recommendation to recycle the used Li-ion battery.

17. The method of claim 1, wherein generating the recommendation for the application of the used Li-ion battery comprises running a remaining useful life (RUL) simulation on the used Li-ion battery.

18. The method of claim 17, wherein the application is a second-life application.

19. The method of claim 18, wherein the RUL simulation predicts one or more second- life operational scenarios based on the state of health.

20. The method of claim 1 , further comprising: reading a serial number associated with the used Li-ion battery; loading, from a memory and based on the serial number, metadata of the battery associated with the serial number, the metadata comprising one or more of the electrolyte of the used Li-ion battery, a use case of the Li-ion battery, an age of the Li-ion battery, a capacity of the Li-ion battery, or a physical size of the Li-ion battery; and determining the at least partial set of the operational history based on the metadata.

21. The method of claim 1 , further comprising determining an amount of testing and / or characterization required to efficiently categorize the used Li-ion battery into second-life application or ready to recycle.

22. The method of claim 1 , further comprising:- 31 -FH13093063.1Attomey Docket No.: BGW-00225 measuring an internal impedance of the used Li-ion battery; and providing the measured impedance to the trained machine-learning model, wherein the trained machine-learning model is further trained to output the state of health in response to receiving the operational history, the representation of the electrolyte, and an impedance measurement; wherein receiving the estimate of the state of health is further based on the measured impedance provided to the trained machine-learning model.

23. The method of claim 22, wherein the internal impedance is measured by applying a predefined alternating-current excitation to the battery and recording a resulting voltage response.

24. The method of claim 22, further comprising grouping the used Li-ion battery with one or more additional batteries having substantially similar internal impedance values as the internal impedance of the used Li-ion battery for integration into a battery pack.

25. The method of claim 1 , wherein the trained machine-learning model is a physicsbased simulation model; and wherein the operational history provided to the trained machine-learning model is free of battery capacity data.

26. The method of claim 1 , further comprising conducting a partial-capacity test by charging or discharging the used Li-ion battery for a predetermined duration and incorporating resulting test data into the state of health estimation.- 32 -FH13093063.1Attorney Docket No.: BGW-0022521. A system for categorizing a used lithium (Li)-ion battery, the system comprising: a computing node comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor of the computing node to cause the processor to perform the method of any of Claims 1-26.

28. A computer program product for categorizing a used lithium (Li)-ion battery, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform the method of any of Claims 1-26.

29. A method of training a machine-learning model to determine a state of health of a battery, the method comprising: reading, for each of a plurality of used Li-ion batteries: an operational history, the operational history comprising operational data of that battery, a representation of an electrolyte of that used Li-ion battery, and a state-of-health of that used Li-ion battery, the state-of-health representing a measure of electrical charge capacity of that Li-ion battery; generating a training dataset comprising: the operational history of each of the used Li-ion batteries, the representation of the electrolyte of each of the used Li-ion batteries, and the state of health of each used Li-ion battery, the generated training dataset comprising representations of at least two different electrolytes; providing the training dataset to a machine-learning model; and- 33 -FH13093063.1Attomey Docket No.: BGW-00225 training the machine-learning model on the training dataset, resulting in a trained machine-learning model trained to output a state of health in response to receiving an operational history and a representation of an electrolyte of a Li-ion battery.

30. The method of claim 29, wherein the operational history comprises usage data of the battery.

31. The method of claim 30, wherein the usage data comprises data fields of one or more of date, timestamp, voltage, current, capacity, and temperature.

32. The method of claim 31, further comprising determining a level of completeness of the usage data, wherein the level of completeness is one of complete, partially complete, and not available.

33. A system comprising: a scanner configured to read a serial number of a used Li-ion battery; a test station configured to perform at least one of an open-circuit-voltage test, an impedance measurement, or a partial-capacity test on the used Li-ion battery; and a computing unit including at least one processor and a memory storing instructions that, when executed, cause the computing unit to perform the method of any of claims 1-26.- 34 -FH13093063.1

Images