A battery retirement recycling quality traceability and grading disposal system and method

Abstract

This invention discloses a battery retirement recycling quality traceability and graded disposal system and method, belonging to the field of battery quality traceability technology. The system includes: acquiring and preprocessing data; constructing and solving a forward electrochemical model to obtain short-term high-temperature events and outputting thermal damage risk values; collecting external boundary measurements based on multimodal non-destructive testing and outputting chemical inhomogeneity through a multimodal coupling strategy; standardizing the recycling measurements, thermal damage risk values, and chemical inhomogeneity into recycling verification tags, calculating digest hashes, and submitting timestamps; integrating the above outputs and multidimensional battery recycling indicators into a comprehensive disposal score, outputting disposal recommendations for each batch according to set rules, and issuing sorting and execution instructions; and performing aggregate statistics and anomaly detection on the same manufacturing batch based on the recycling verification tags. This invention solves the problem of low material recovery rates caused by information gaps and fluctuations in recovery rates.

Description

Technical Field

[0001] This invention relates to the field of battery quality traceability technology, specifically to a battery retirement recycling quality traceability and graded disposal system and method. Background Technology

[0002] With the rapid development of electric transportation and large-scale energy storage, the production and retirement volume of lithium-ion batteries are growing exponentially. The key metals contained in batteries, such as lithium, nickel, and cobalt, are strategically important and of high value. Their recycling and reuse are not only crucial for resource security and the stability of the materials supply chain, but also directly affect environmental impact and carbon emission accounting. Therefore, building an efficient, traceable battery recycling system with both economic and environmental benefits has become a focus of attention for industry, regulators, and researchers.

[0003] Current battery recycling practices face two main constraints: First, the digitalization of identity and material information is inconsistent. There is a lack of unified and interoperable data standards and interfaces between the manufacturing, vehicle, operation, and recycling ends. Battery manufacturing information, chemical systems, and operating history cannot be seamlessly transferred, leading to frequent reliance on manual judgment in the recycling process due to information gaps, which hinders the realization of automated sorting and refined processing. Second, the output of recycling processes fluctuates greatly and material loss is significant. Dismantling accuracy, process selection, and impurity management directly determine the recyclability and material purity. Existing processes are not adaptable enough to complex high-nickel and mixed cathode systems, making it difficult to stably produce battery-grade materials and provide verifiable environmental performance data.

[0004] In summary, resolving the issues of information chain disruption and variability in recycling processes is key to achieving efficient recycling of battery resources. To this end, this invention provides a system and method for quality traceability and tiered disposal of decommissioned batteries. Summary of the Invention

[0005] The purpose of this invention is to provide a system and method for quality traceability and graded disposal of retired batteries to solve the existing problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a battery retirement recycling quality traceability and grading system, comprising:

[0007] The lifecycle data access module is used to acquire data from battery digital twin archives, battery management systems, IoT platforms, recycling centers and automated dismantling lines, and to clean, fill in missing data, align time and granularity, and map unique identifiers to the data.

[0008] The thermal history event inversion and quantification module is used to construct and solve a forward electrochemical model based on high-precision electrochemical impedance spectroscopy measurement and trace evidence chemical detection, to obtain short-term high-temperature events and their severity, and output thermal damage risk values.

[0009] The battery internal spatial gradient mapping module is used to acquire external boundary measurements based on multimodal nondestructive detection, and reconstruct the three-dimensional chemical distribution inside the cell through a multimodal coupling strategy, outputting chemical inhomogeneity.

[0010] The recycling verification label generation module is used to standardize recycling measurements, thermal damage risk values, and chemical inhomogeneities into recycling verification labels, calculate digest hashes and submit timestamps, and then write the recycling verification labels back to the lifecycle data access module.

[0011] The battery grading decision module is used to integrate the outputs of the thermal history event inversion quantification module and the battery internal spatial gradient mapping module with the battery recycling multidimensional indicators into a comprehensive disposal score, and output disposal suggestions for each batch according to the set rules, and issue sorting and execution instructions.

[0012] The batch aggregation analysis in the learning module is used to perform aggregation statistics and anomaly detection on the same manufacturing batch based on the recycling verification label, and generate batch alarms. At the same time, the verification label is returned to the recycling verification label generation module.

[0013] A further improvement of the present invention is that the thermal history event inversion quantification module includes an electrochemical impedance spectroscopy (EIS) acquisition unit, a forward electrochemical model construction unit, a thermal history parameter solution unit, and a thermal damage risk factor output unit.

[0014] The electrochemical impedance spectroscopy (EIS) acquisition unit is used to apply multi-frequency excitation to the battery cell under test and measure the impedance spectrum before the battery enters the factory. The thermal history parameter solving unit is used to define the high-temperature event hypothesis H and establish a set of electrochemical mapping parameters. The forward electrochemical model building unit is used to construct the forward operator for simulating impedance spectroscopy. The thermal damage risk factor output unit is used to perform SEI fragment analysis on the current batch of samples, and calculates the posterior probability by Bayesian fusion of chemical evidence and model inversion evidence, projecting the posterior probability into a normalized thermal damage risk factor. When the thermal damage risk factor When the damage threshold is exceeded, the batch is marked as having a high risk of thermal damage and the information is written into a field of the battery digital twin file.

[0015] A further improvement of this invention lies in that the thermal history parameter solving unit employs an iterative nonlinear least squares algorithm, and in each iteration, the model prediction error and measurement uncertainty are combined into the cost function weights; during the solution process, L1 regularization is applied to H to promote sparse solutions, and Bayesian posterior calibration is used on the inversion results to obtain the confidence interval; finally, the most severe short-term high-temperature event is obtained by solving the inversion objective function with the sparsification term in the following equation. :

[0016] ;

[0017] High-temperature events are included as risk factors in the battery grading decision module.

[0018] A further improvement of the present invention is that the battery internal spatial gradient mapping module includes a multimodal data acquisition unit, a scale feature decomposition unit, and a regularized inversion weight construction unit;

[0019] The multimodal data acquisition unit includes an electrical impedance tomography (EIT) data acquisition array and a microwave scanning device, which acquires boundary voltage measurement vectors through electrical impedance tomography. Electromagnetic scattering parameter vectors are obtained through a microwave scanning device. ;

[0020] The scale feature decomposition unit is used to perform scale decomposition on spatiotemporal indices at different granularities and calculate scale energy. It is used to set weights and select dominant scales in the inversion process;

[0021] The regularized inversion weight construction unit is used to execute a multimodal coupling strategy, reconstruct the weights of the boundary voltage measurement vector and electromagnetic scattering parameter vector in the inversion model, and obtain the chemical inhomogeneity risk factor by mapping the inverted data to ion concentration fields and byproduct concentration fields, respectively, and calculating their spatial variance and gradient index. .

[0022] A further improvement of this invention is that, when constructing weights, the multimodal coupling strategy uses the scale energy obtained from the scale feature decomposition unit multiplied by the time decay weight as the sample weight: , λ represents the current time during the mapping calculation, and λ represents the empirical decay rate.

[0023] A further improvement of the present invention is that the battery grading decision module includes a rule-based decision unit and a batch aggregation analysis unit;

[0024] The rule decision-making unit includes:

[0025] when and If all values ​​are below the corresponding threshold and the cathode material is of a type suitable for direct regeneration, then direct regeneration is recommended.

[0026] When the potential score for secondary use based on the current battery health and capacity prediction is greater than the set score threshold and If the score is less than the corresponding threshold, a comprehensive disposal score is calculated and sent to the recycling verification label.

[0027] when and If all values ​​are greater than or equal to the corresponding threshold, then strong hydrometallurgy is recommended.

[0028] In other cases, the standard wet method is recommended.

[0029] A further improvement of the present invention is that the batch aggregation analysis unit performs aggregation statistics on multiple recycling verification tags of the same manufacturing batch and calculates the statistical distance of the multidimensional residual components for anomaly judgment. When the statistical distance exceeds the threshold, a batch anomaly alarm is output and the aggregation evidence and suggestions are sent to the manufacturer or regulator for traceability.

[0030] A further improvement of this invention is that the comprehensive treatment score calculation formula in the battery grading decision module is:

[0031] ;

[0032] in, This indicates a potential score for secondary use based on current battery health and capacity predictions; This indicates the environmental cost obtained by process type.

[0033] A further improvement of this invention is that the potential score for secondary use based on the current battery health and capacity prediction is determined by the battery health state (SOH) expressed by the current energy ratio and the secondary use determination threshold. Typical depth of discharge for target secondary applications Relative power capability indicators Thermal damage risk value, chemical inhomogeneity risk value, and SOH loss parameter per complete cycle. Calculate the remaining number of cycles based on the equivalent complete cycle method With cycle score , among which when At that time, the cycle score 0, then , where the function This indicates that the result is limited to a range. , , , , This represents the weighting coefficient.

[0034] A further improvement of this invention is that the environmental cost is obtained by weighted summation after normalization of the following items: process energy consumption, chemical reagent and consumable consumption, labor and maintenance time allocation, equipment depreciation and facility occupation, transportation and logistics, and quality loss penalty.

[0035] A further improvement of this invention is that the potential score and its confidence level are used as structured fields to write the recycling verification tag of the recycling verification tag generation module and then written back to the battery digital twin file.

[0036] On the other hand, the present invention provides a method for quality traceability and graded disposal of retired batteries, comprising the following steps:

[0037] S1. Obtain data from battery digital twin archives, battery management system, IoT platform, recycling center and automated dismantling line, and clean, fill in missing data, align time and granularity and map unique identifiers to the data;

[0038] S2. Based on high-precision electrochemical impedance spectroscopy measurement and trace evidence chemical detection, a forward electrochemical model is constructed and solved to obtain short-term high-temperature events and their severity, and output thermal damage risk values.

[0039] S3. Based on multimodal nondestructive testing, external boundary measurements are collected, and the three-dimensional chemical distribution inside the cell is reconstructed through a multimodal coupling strategy, outputting chemical inhomogeneity.

[0040] S4. Standardize the recovery measurement, thermal damage risk value, and chemical inhomogeneity into recovery verification labels, calculate the digest hash and submit the timestamp, and then write the recovery verification labels back to S1.

[0041] S5. Integrate the outputs of steps S4 and S3 and the multi-dimensional indicators of battery recycling into a comprehensive disposal score, and output disposal recommendations for each batch according to the set rules, and issue sorting and execution instructions.

[0042] S6. Based on the recycling verification label, perform aggregate statistics and anomaly detection on the same manufacturing batch, generate batch alarm, and return the verification label to S4.

[0043] Compared with the prior art, the beneficial effects of the present invention are:

[0044] 1. This invention first solves the information gap problem between the manufacturing end, the user end, and the recycling end by establishing a unified and interoperable battery and data link. This enables the recycling end to obtain key information such as the cell chemical system, production batch, and historical charge and discharge spectrum in real time, thereby realizing automated and standardized sorting and disposal decisions, significantly reducing the cost of manual judgment and improving sorting accuracy and fine classification capabilities.

[0045] 2. By deploying automated dismantling and online or offline high-precision measurement on the recycling line, and supplementing it with process control and recycling-based optimization, the problem of unstable recovery rate caused by dismantling accuracy, process parameter fluctuations and impurity management is solved. This results in a significant improvement and consistency in the recovery rate and purity of key metals, reduces material loss and provides verifiable energy consumption accounting data. Attached Figure Description

[0046] Figure 1 This is a framework diagram of a battery retirement recycling quality traceability and graded disposal system according to the present invention;

[0047] Figure 2 This is a flowchart of a method for quality traceability and graded disposal of retired batteries according to the present invention. Detailed Implementation

[0048] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0049] The term "and / or" is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can represent three cases: A exists alone, A and B exist simultaneously, and B exists alone.

[0050] Example 1

[0051] Figure 1 This embodiment illustrates a framework diagram of a battery retirement recycling quality traceability and graded disposal system, including:

[0052] The lifecycle data access module is used to acquire data from battery digital twin archives, battery management systems, IoT platforms, recycling centers, and automated dismantling lines, and to clean, fill in missing data, align time and granularity, and map unique identifiers to the data. The data includes, but is not limited to: manufacturing batches, chemical formula declarations, theoretical element vectors, cycle counts, temperature profiles, SOH time series, incoming appearance records, XRF / ICP measurement and dismantling logs.

[0053] Data access uses a standardized interface and performs initial signature verification at the gateway; the edge gateway can perform initial noise reduction and compression; all access data is accompanied by a measurement timestamp and device ID.

[0054] Cross-validation is performed on the k-value, weighting function, and window size of the KNN interpolation to avoid overfitting or oversmoothing. A three-standard-deviation threshold can be adaptively adjusted for different signals (temperature, current). This ensures that subsequent inversion and judgment are not misled by noise or missing data, improving the stability and accuracy of the judgment; ID mapping ensures that the recovered evidence can be accurately traced back to the production batch and manufacturing information.

[0055] The thermal history event inversion and quantification module is used to construct and solve a forward electrochemical model based on high-precision electrochemical impedance spectroscopy (EIS) measurements and trace chemical evidence detection. This model yields short-duration high-temperature events and their severity, and outputs a thermal damage risk value after incorporating chemical evidence. Under reasonable physical model assumptions, it identifies the high-temperature event description that best explains the observed EIS spectrum. A high-temperature event represents the peak temperature and duration of a short-duration high-temperature event. By comparing the simulated impedance spectrum with the measured spectrum, the thermal event parameters with the smallest differences are identified, allowing for the inference of the most likely characteristics of the thermal event. To avoid interpreting noise as thermal events, a constraint encouraging sparse solutions is incorporated into the optimization process, promoting only a few significant thermal events rather than the frequent occurrence of many minor events.

[0056] The thermal history event inversion and quantification module includes an electrochemical impedance spectroscopy (EIS) acquisition unit, a forward electrochemical model construction unit, a thermal history parameter solution unit, and a thermal damage risk factor output unit.

[0057] The electrochemical impedance spectroscopy (EIS) acquisition unit is used to apply multi-frequency excitation to the battery cell under test and measure the impedance spectrum before the battery enters the factory. The thermal history parameter solving unit is used to define the high-temperature event hypothesis H and establish a set of electrochemical mapping parameters. The forward electrochemical model building unit is used to construct the forward operator for simulating impedance spectroscopy. This includes, but is not limited to, SEI resistance, solid-phase diffusion coefficient, and charge transfer resistance; the thermal damage risk factor output unit is used to perform SEI fragment analysis on the current batch of samples, and calculates the posterior probability by Bayesian fusion of chemical evidence and model inversion evidence, projecting the posterior probability into a normalized thermal damage risk factor. When the thermal damage risk factor When the damage threshold is exceeded, the batch is marked as having a high risk of thermal damage and the information is written into a field of the battery digital twin file.

[0058] (Model inversion evidence) refers to the inversion algorithm of the forward electrochemical model, based on observed impedance spectra. The obtained set of numerical evidence includes (but is not limited to) the most probable thermal history parameters and electrochemical parameters obtained through inversion. and The residual vectors of the simulated and observed spectra, the uncertainty estimate (covariance or confidence interval) of the inversion results, and the likelihood measure constructed based on the residuals are concatenated to form model evidence for the occurrence of high-temperature events. This evidence is then used in Bayesian fusion with chemical / physical evidence. Calculate the posterior probability together.

[0059] The thermal history parameter solving unit employs an iterative nonlinear least squares algorithm, merging model prediction error and measurement uncertainty into cost function weights in each iteration. L1 regularization is applied to H during the solution process to promote sparse solutions, and Bayesian posterior calibration is used on the inversion results to obtain confidence intervals. Finally, the most severe short-duration high-temperature event is obtained by solving the inversion objective function with sparsification terms. :

[0060] ;

[0061] Incorporate high-temperature events as a risk factor into RCI.

[0062] The difference between model predictions and observed data ensures that the selected thermal events accurately explain the observations. The L1 regularization term applies a preference to the sparsity of thermal event parameters, ensuring that the system prioritizes reporting a small number of significant thermal events rather than many weak spurious events. Through Bayesian fusion, chemical evidence and model evidence are combined, and the confidence level of thermal events is calculated probabilistically. If both types of evidence strongly point to the same conclusion—for example, electrochemical inversion indicates the presence of a high-temperature peak, and chemical analysis reveals high-temperature decomposition products or phase transitions—the posterior probability is high. If only one type of evidence supports the conclusion, while the other is irrelevant or contradictory, the posterior probability decreases accordingly. This application can non-destructively identify battery cells with intact surfaces but irreversible structural or chemical changes due to short-term high temperatures. Furthermore, the report provides not only a yes / no indication but also a confidence level (probability), facilitating subsequent decisions on whether further verification is needed or whether to proceed directly to higher-level processing.

[0063] The battery internal spatial gradient mapping module is used to acquire external boundary measurements based on multimodal nondestructive detection, and reconstruct the three-dimensional chemical distribution inside the cell through a multimodal coupling strategy, outputting chemical inhomogeneity.

[0064] The battery internal spatial gradient mapping module includes a multimodal data acquisition unit, a scale feature decomposition unit, and a regularized inversion weight construction unit;

[0065] The multimodal data acquisition unit includes an electrical impedance tomography (EIT) data acquisition array and a microwave scanning device, which acquires boundary voltage measurement vectors through electrical impedance tomography. Electromagnetic scattering parameter vectors are obtained through a microwave scanning device. It provides a measurable data source that is indirectly related to the ion concentration field and by-product distribution inside the battery, avoiding direct and destructive sampling.

[0066] This means that if there are m electrodes, following a certain measurement sequence (e.g., adjacent electrode measurement, span electrode measurement, or full-permutation electrode measurement), M measurement values ​​will be generated. Each measurement value can be a complex number (real / imaginary part or amplitude / phase), because what is measured under AC excitation is AC voltage / current.

[0067] Microwave / RF detection is used to scan the cell with an antenna or probe to measure the response of the reflection coefficient (e.g., S11) or transmission coefficient (e.g., S21) to changes in frequency or spatial location. It is sensitive to the internal dielectric constant and conductivity distribution and can detect density differences, cracks, or liquid / gas-containing areas.

[0068] The scale feature decomposition unit is used to perform scale decomposition on spatiotemporal indices at different granularities and calculate scale energy. It is used to set weights and select dominant scales in the inversion process;

[0069] The regularized inversion weight construction unit is used to execute a multimodal coupling strategy, reconstruct the weights of the boundary voltage measurement vector and electromagnetic scattering parameter vector in the inversion model, and obtain the chemical inhomogeneity risk factor by mapping the inverted data to ion concentration fields and byproduct concentration fields, respectively, and calculating their spatial variance and gradient index. .

[0070] Different NDT modes (e.g., EIT and microwave scanning) reflect different internal physical properties, but these properties often share common edges or discontinuities in the spatial structure (e.g., material interfaces, cracks, or high / low density regions). Joint regularized inversion couples the reconstruction problem of each mode to the same optimization objective and adds coupling terms that encourage consistency at the edges of different modes, allowing modes with complementary information to constrain each other, thereby significantly improving the stability and resolution of single-mode inversion under ill-posed conditions.

[0071] Compared to single-mode inversion, joint inversion can more accurately reconstruct the spatial heterogeneity of internal lithium distribution and SEI thickness, and can detect problems such as local failure / internal delamination / active material shedding earlier.

[0072] Scale-based eigendecomposition (SFD) and time decay weights make the system more sensitive to the latest measurements and enable robust up- and down-estimation across multi-granularity data.

[0073] The multimodal coupling strategy constructs weights by multiplying the scale energy obtained from the scale feature decomposition unit by the time decay weight as the sample weights. , λ represents the current time during the mapping calculation, and λ represents the empirical decay rate.

[0074] Furthermore, the regularization matrix L can be selected as a gradient operator or a multi-scale sparse transformation (e.g., an SFD base domain) to balance smoothness and structure preservation. This invention uses the scale energy obtained from SFD to assign weights to different spatial / temporal samples, making the data at the dominant scale dominant in the local mapping. A time decay weight is introduced to make newer measurements more influential on the inversion results. An adaptive regularization operator (such as gradient constraint or multi-scale sparse transformation domain regularization) is used to balance the smoothness of reconstruction and structure preservation. The reconstruction uncertainty is used as an output and applied to the subsequent confidence assessment of disposal decisions.

[0075] The recycling verification tag generation module is used to standardize recycling measurements into recycling verification tags, calculate digest hashes and submit timestamps, and then write the recycling verification tags back to the lifecycle data access module; recycling measurements include all data obtained in the aforementioned process.

[0076] A Recycling Validation Tag (RVT) is an evidence unit that stores on-site measurements, inversion results, disposal operations, and disposal results in a standardized structure. It records the cell ID, measured recycled elements / products, process parameters, their corresponding risk scores (which can be set by technicians), and measurement uncertainties. Summary hashing and timestamp storage ensure that this record is tamper-proof once written, thus forming an auditable chain of facts. This provides a provable factual basis for recycling decisions; it provides an immutable chain of evidence in case of subsequent disputes or requests for traceability; and it provides reliable data for batch quality analysis and manufacturer feedback. A Recycling Validation Tag includes at least the observation vector, mass loss, impurity rate, and dismantling success rate.

[0077] The battery grading decision module is used to integrate the outputs of the thermal history event inversion quantification module and the battery internal spatial gradient mapping module with the battery recycling multidimensional indicators into a comprehensive disposal score, and output disposal suggestions for each batch according to the set rules, and issue sorting and execution instructions.

[0078] The battery grading decision module includes a rule-based decision unit and a batch aggregation analysis unit;

[0079] The rule decision-making unit includes:

[0080] when and If all values ​​are below the corresponding threshold and the cathode material is of a type suitable for direct regeneration, then direct regeneration is recommended.

[0081] When the potential score for secondary use based on the current battery health and capacity prediction is greater than the set score threshold and If the score is less than the corresponding threshold, a comprehensive disposal score is calculated and sent to the recycling verification label.

[0082] when and If all values ​​are greater than or equal to the corresponding threshold, then strong hydrometallurgy is recommended.

[0083] In other cases, the standard wet method is recommended.

[0084] When recycling multiple battery cells from the same manufacturing batch, if the measured results of most components in that batch systematically deviate from their factory claims or historical baselines, it often indicates a systemic problem in manufacturing or raw materials (e.g., incorrect formulation, batch contamination). To detect such systematic deviations, the system aggregates and statistically analyzes the residual components (the difference between measured and claimed / expected values) at the batch level and calculates multidimensional anomaly distances. This distance comprehensively measures the degree of deviation of the residuals across different indicators and their covariance structure. When the distance exceeds a threshold, a batch alarm is triggered. Therefore, the batch aggregation analysis unit aggregates and statistically analyzes multiple recycling verification tags from the same manufacturing batch and calculates the statistical distance of the multidimensional residual components for anomaly detection. When the statistical distance exceeds a threshold, a batch anomaly alarm is output, and aggregated evidence and recommendations are sent to the manufacturer or regulator for traceability. This allows for early detection of manufacturing or material quality problems at the batch level, thereby achieving supply chain quality control and accountability; it also prevents individual anomalies from being overlooked or misunderstood.

[0085] The comprehensive handling score calculation formula in the battery grading decision module is as follows:

[0086] ;

[0087] in, This indicates a potential score for secondary use based on current battery health and capacity predictions; This indicates the environmental cost obtained by process type.

[0088] By using the time dimension (whether a short-term high-temperature event has occurred) and the spatial dimension (the non-uniformity of chemical distribution inside the cell) as two core "implicit risk factors" to supplement traditional explicit indicators such as cycle count, disposal decisions can be made to select more suitable, safer and more valuable processes that are differentiated based on the internal structure of the individual cell.

[0089] The potential score for secondary use based on the current battery health and capacity prediction is determined by the battery health state (SOH) expressed in terms of the current energy ratio and the secondary use determination threshold. Typical depth of discharge for target secondary applications Relative power capability indicators Thermal damage risk value, chemical inhomogeneity risk value, and SOH loss parameter per complete cycle. Calculate the remaining number of cycles based on the equivalent complete cycle method With cycle score , among which when At that time, the cycle score If it is limited to 0, then , where the function This indicates that the result is limited to a range. , , , , This represents the weighting coefficient.

[0090] The above scheme utilizes multiple physical indicators, including the current energy state of the cell, power capability, historical degradation information, and chemical risk, to quantify and normalize a comprehensive score for the reuse of a single cell. This score can be directly used for automated decision-making or to trigger manual review in the sorting and grading process, thereby reducing failures or safety risks caused by inappropriate reuse and improving resource utilization efficiency. Simultaneously, the score and confidence level are written back to the recycling verification tag, providing auditable factual evidence for subsequent traceability, statistical analysis, and model retraining.

[0091] The environmental cost is obtained by weighted summation after normalization of the following items: process energy consumption, chemical reagent and consumable consumption, labor and maintenance time allocation, equipment depreciation and facility occupation, transportation and logistics, and quality loss penalty.

[0092] The environmental cost is a quantitative comprehensive indicator of all direct resource consumption, pollution control burden, and safety compliance load required to complete the designated recycling and disposal of this batch of batteries. The components include, but are not limited to: process energy consumption (actual consumption of electricity, heat and fuel and their corresponding emission factors or energy scores), chemical reagent and consumable consumption (usage and disposal costs or substitution quantification of acids, solvents, extractants, filter materials, etc.), waste liquid / residue treatment and pollution control (processes required for neutralization, concentration, solidification, incineration or terminal disposal and the resulting environmental load), labor and maintenance time allocation, equipment depreciation and facility occupancy (allocated according to equipment life or annual processing energy), transportation and logistics, and quality loss penalties.

[0093] The original values ​​of each component are derived from on-site measurements (MES, LIMS, energy meters, chemical requisition records, transport documents, equipment ledgers, etc.), process templates, or industry standard databases (including life cycle assessment data and emission factors). When direct measurement is lacking, calibrated substitution coefficients or templates can be used for estimation. Each calculation should simultaneously produce an uncertainty estimate (based on measurement error, template confidence level, or statistical fluctuation), and the total... Uncertainty propagation or Monte Carlo sampling is performed to give confidence intervals.

[0094] To facilitate integration with other technical indicators, the aforementioned The sum or weighted sum is standardized to a fixed interval (e.g., 0–1) using linear / quantile / logarithmic mappings; the sum or weighted sum is then... When used as a recyclability RCI synthesis index input, the uncertainty threshold should be given priority: if If the uncertainty exceeds the set upper limit, a sampling review will be triggered;

[0095] By quantifying and structuring records according to process type It can enable unified measurement and comparable assessment of resource consumption, environmental impact and safety compliance burden in recycling, sorting and disposal decisions, reduce energy waste, pollution risks and recycling efficiency decline caused by improper process selection, and provide auditable data support for batch anomaly analysis, process template calibration and regulatory traceability.

[0096] The potential score for tiered utilization and its confidence level are obtained from the uncertainty of the input parameters through uncertainty propagation or Monte Carlo sampling, and are written as structured fields into the recycling verification label of the recycling verification label generation module and written back to the battery digital twin file.

[0097] The batch aggregation analysis in the learning module is used to perform aggregation statistics and anomaly detection on the same manufacturing batch based on the recycling verification label, and generate batch alarms. At the same time, the verification label is returned to the recycling verification label generation module.

[0098] The threshold and weight settings can be set by default according to the present invention, or they can be set by those skilled in the art.

[0099] Example 2

[0100] Figure 2 This invention presents a flowchart of a method for quality traceability and grading of battery recycling, based on the same inventive concept as Embodiment 1. The method includes the following steps:

[0101] S1. Obtain data from battery digital twin archives, battery management system, IoT platform, recycling center and automated dismantling line, and clean, fill in missing data, align time and granularity and map unique identifiers to the data;

[0102] S2. Based on high-precision electrochemical impedance spectroscopy measurement and trace evidence chemical detection, a forward electrochemical model is constructed and solved to obtain short-term high-temperature events and their severity, and output thermal damage risk values.

[0103] S3. Based on multimodal nondestructive testing, external boundary measurements are collected, and the three-dimensional chemical distribution inside the cell is reconstructed through a multimodal coupling strategy, outputting chemical inhomogeneity.

[0104] S4. Standardize the recovery measurement, thermal damage risk value, and chemical inhomogeneity into recovery verification labels, calculate the digest hash and submit the timestamp, and then write the recovery verification labels back to S1.

[0105] S5. Integrate the outputs of steps S4 and S3 and the multi-dimensional indicators of battery recycling into a comprehensive disposal score, and output disposal recommendations for each batch according to the set rules, and issue sorting and execution instructions.

[0106] S6. Based on the recycling verification label, perform aggregate statistics and anomaly detection on the same manufacturing batch, generate batch alarm, and return the verification label to S4.

[0107] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0108] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. 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 program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, 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, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0109] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0110] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0111] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A battery retirement recycling quality traceability and grading system, characterized in that: include: The lifecycle data access module is used to acquire data from battery digital twin archives, battery management systems, IoT platforms, recycling centers and automated dismantling lines, and to clean, fill in missing data, align time and granularity, and map unique identifiers to the data. The thermal history event inversion and quantification module is used to construct and solve a forward electrochemical model based on high-precision electrochemical impedance spectroscopy measurement and trace evidence chemical detection to obtain short-term high-temperature events and output thermal damage risk values. The battery internal spatial gradient mapping module is used to acquire external boundary measurements based on multimodal nondestructive detection, and reconstruct the three-dimensional chemical distribution inside the cell through a multimodal coupling strategy, outputting chemical inhomogeneity. The recycling verification label generation module is used to standardize recycling measurements, thermal damage risk values, and chemical inhomogeneities into recycling verification labels, calculate digest hashes and submit timestamps, and then write the recycling verification labels back to the lifecycle data access module. The battery grading decision module is used to integrate the outputs of the thermal history event inversion quantification module and the battery internal spatial gradient mapping module with the battery recycling multidimensional indicators into a comprehensive disposal score, and output disposal suggestions for each batch according to the set rules, and issue sorting and execution instructions. The batch aggregation analysis in the learning module is used to perform aggregation statistics and anomaly detection on the same manufacturing batch based on the recycling verification label, and generate batch alarms. At the same time, the verification label is returned to the recycling verification label generation module.

2. The battery retirement recycling quality traceability and graded disposal system according to claim 1, characterized in that: The thermal history event inversion and quantification module includes an electrochemical impedance spectroscopy (EIS) acquisition unit, a forward electrochemical model construction unit, a thermal history parameter solution unit, and a thermal damage risk factor output unit. The electrochemical impedance spectroscopy (EIS) acquisition unit is used to apply multi-frequency excitation to the battery cell under test and measure the impedance spectrum before the battery enters the factory. The thermal history parameter solving unit is used to define the high-temperature event hypothesis H and establish a set of electrochemical mapping parameters. The forward electrochemical model building unit is used to construct the forward operator for simulating impedance spectroscopy. The thermal damage risk factor output unit is used to perform SEI fragment analysis on the current batch of samples, and calculates the posterior probability by Bayesian fusion of chemical evidence and model inversion evidence, projecting the posterior probability into a normalized thermal damage risk factor. When the thermal damage risk factor When the damage threshold is exceeded, the batch is marked as having a high risk of thermal damage and the information is written into a field of the battery digital twin file.

3. The battery retirement recycling quality traceability and graded disposal system according to claim 2, characterized in that: The thermal history parameter solving unit employs an iterative nonlinear least squares algorithm, merging model prediction error and measurement uncertainty into cost function weights in each iteration. L1 regularization is applied to H during the solution process to promote sparse solutions, and Bayesian posterior calibration is used on the inversion results to obtain confidence intervals. Finally, the most severe short-duration high-temperature event is obtained by solving the inversion objective function with sparsification terms. : ； High-temperature events are included as risk factors in the battery grading decision module.

4. The battery retirement recycling quality traceability and graded disposal system according to claim 3, characterized in that: The battery internal spatial gradient mapping module includes a multimodal data acquisition unit, a scale feature decomposition unit, and a regularized inversion weight construction unit; The multimodal data acquisition unit includes an electrical impedance tomography (EIT) data acquisition array and a microwave scanning device, which acquires boundary voltage measurement vectors through electrical impedance tomography. Electromagnetic scattering parameter vectors are obtained through a microwave scanning device. ; The scale feature decomposition unit is used to perform scale decomposition on spatiotemporal indices at different granularities and calculate scale energy. It is used to set weights and select dominant scales in the inversion process; The regularized inversion weight construction unit is used to execute a multimodal coupling strategy, reconstruct the weights of the boundary voltage measurement vector and electromagnetic scattering parameter vector in the inversion model, and obtain the chemical inhomogeneity risk factor by mapping the inverted data to ion concentration fields and byproduct concentration fields, respectively, and calculating their spatial variance and gradient index. .

5. A battery retirement recycling quality traceability and grading system according to claim 4, characterized in that: The multimodal coupling strategy constructs weights by multiplying the scale energy obtained from the scale feature decomposition unit by the time decay weight as the sample weights. , λ represents the current time during the mapping calculation, and λ represents the empirical decay rate.

6. The battery retirement recycling quality traceability and graded disposal system according to claim 5, characterized in that: The battery grading decision module includes a rule-based decision unit and a batch aggregation analysis unit; The rule decision-making unit includes: when and If all values ​​are below the corresponding threshold and the cathode material is of a type suitable for direct regeneration, then direct regeneration is recommended. When the potential score for secondary use based on the current battery health and capacity prediction is greater than the set score threshold and If the score is less than the corresponding threshold, a comprehensive disposal score is calculated and sent to the recycling verification label. when and If all values ​​are greater than or equal to the corresponding threshold, then strong hydrometallurgy is recommended. In other cases, the standard wet method is recommended.

7. A battery retirement recycling quality traceability and grading system according to claim 6, characterized in that: The batch aggregation analysis unit aggregates and statistically analyzes multiple recycling verification tags of the same manufacturing batch and calculates the statistical distance of the multidimensional residual components for anomaly detection. When the statistical distance exceeds the threshold, it outputs a batch anomaly alarm and sends the aggregated evidence and recommendations to the manufacturer or regulator for traceability.

8. The battery retirement recycling quality traceability and grading system according to claim 7, characterized in that: The comprehensive handling score calculation formula in the battery grading decision module is as follows: ； in, This indicates a potential score for secondary use based on current battery health and capacity predictions; This indicates the environmental cost obtained by process type.

9. A battery retirement recycling quality traceability and grading system according to claim 8, characterized in that: The potential score for secondary use based on the current battery health and capacity prediction is determined by the battery health state (SOH) expressed in terms of the current energy ratio and the secondary use determination threshold. Typical depth of discharge for target secondary applications Relative power capability indicators Thermal damage risk value, chemical inhomogeneity risk value, and SOH loss parameter per complete cycle. Calculate the remaining number of cycles based on the equivalent complete cycle method With cycle score , among which when At that time, the cycle score 0, then , where the function This indicates that the result is limited to an interval. , , , , This represents the weighting coefficient.

10. A battery retirement recycling quality traceability and grading system according to claim 9, characterized in that: The environmental cost is obtained by weighted summation after normalization of the following items: process energy consumption, chemical reagent and consumable consumption, labor and maintenance time allocation, equipment depreciation and facility occupation, transportation and logistics, and quality loss penalty.

11. A battery retirement recycling quality traceability and grading system according to claim 10, characterized in that: The potential score and its confidence level are used as structured fields to write the recycling verification tag into the recycling verification tag generation module and then written back to the battery digital twin file.

12. A method for quality traceability and grading of retired battery recycling, used to implement a battery quality traceability and grading system as described in any one of claims 1-11, characterized in that: The specific steps include: S1. Obtain data from battery digital twin archives, battery management system, IoT platform, recycling center and automated dismantling line, and clean, fill in missing data, align time and granularity and map unique identifiers to the data; S2. Based on high-precision electrochemical impedance spectroscopy measurement and trace evidence chemical detection, a forward electrochemical model is constructed and solved to obtain short-term high-temperature events and their severity, and output thermal damage risk values. S3. Based on multimodal nondestructive testing, external boundary measurements are collected, and the three-dimensional chemical distribution inside the cell is reconstructed through a multimodal coupling strategy, outputting chemical inhomogeneity. S4. Standardize the recovery measurement, thermal damage risk value, and chemical inhomogeneity into recovery verification labels, calculate the digest hash and submit the timestamp, and then write the recovery verification labels back to S1. S5. Integrate the outputs of steps S4 and S3 and the multi-dimensional indicators of battery recycling into a comprehensive disposal score, and output disposal recommendations for each batch according to the set rules, and issue sorting and execution instructions. S6. Based on the recycling verification label, perform aggregate statistics and anomaly detection on the same manufacturing batch, generate batch alarm, and return the verification label to S4.

Images