Waste lithium battery cathode material repair system optimized by machine learning algorithm

Abstract

The present invention relates to the technical field of cathode material repair, and discloses a waste lithium battery cathode material repair system optimized by a machine learning algorithm, including a first repair module, a detection module, a repair strategy module, and a second repair module; the first repair module is used to perform pre-repair on the cathode material to be repaired to obtain a first repaired material; the detection module is used to perform state detection on the first repaired material and divide the first repaired material into different repair groups; the repair strategy module is used to control the repair process of the first repaired material; the second repair module is used to execute the repair strategy to obtain a second repaired material for each repair group; the repair strategy module is also used to perform grouping adjustment on the second repaired material of each repair group; through the method of grouped repair, the present invention realizes the efficient, low-cost and environmentally friendly repair of waste lithium battery cathode materials, and improves the quality and reuse rate of the repaired materials.

Description

Technical Field

[0001] The present invention relates to the technical field of cathode material repair, and particularly to a waste lithium battery cathode material repair system optimized by machine learning algorithms. Background Art

[0002] Currently, most waste lithium battery cathode material repair methods adopt a unified processing flow and fail to perform personalized repair according to the specific conditions of different materials. Due to the large differences in the initial states of different materials, a unified repair plan often fails to achieve the best effect, resulting in uneven performance of the repaired materials. Traditional waste lithium battery cathode material repair methods usually require multiple complex processing steps, such as deep cleaning, high-temperature drying, chemical precipitation, etc. These steps are not only time-consuming and laborious but also require a large amount of energy and chemicals, leading to a relatively high overall repair cost. In addition, due to the lack of effective grouping management, some repair steps may be unnecessary for certain materials, further increasing the cost. Existing repair methods often struggle to fully restore the original performance of waste lithium battery cathode materials, especially in terms of electrochemical performance. This is because during the repair process, the microstructure and surface characteristics of the materials may be damaged to varying degrees, and a single repair method is difficult to comprehensively repair these damages, resulting in the performance of the repaired materials falling short of expectations.

[0003] For example, the Chinese patent application with the authorization announcement number CN117594900B discloses a solid-phase repair method for waste lithium battery cathode materials. Aiming at problems such as how to improve the pole piece powder loss rate during the solid-phase repair of lithium batteries, reduce impurities, and improve the solid-phase recovery effect, the invention includes: discharging and disassembling waste lithium batteries to obtain positive pole pieces, performing heat treatment, then performing crushing, peeling, and screening, then performing calcination, adding a lithium source and performing calcination again to obtain the recycled lithium battery cathode material. The recovery method of the invention is simple to operate, has low equipment requirements, good recovery effect, a pole piece powder loss rate ≥ 99.5%, and an impurity content < 0.1%, and can recycle the cathode materials of waste lithium batteries at low cost, greenly and effectively, significantly reducing the production cost.

[0004] For example, the patent application with the publication number CN117239272A discloses a method for repairing lithium battery cathode materials. The method includes the following steps: Step 1, disassembling a waste lithium battery to obtain the cathode powder before repair; Step 2, uniformly mixing the cathode powder before repair with a lithium source, a binder, and a dispersant to obtain a slurry; Step 3, centrifugally drying the slurry to obtain a powder; Step 4, performing plasma treatment on the powder to obtain the repaired lithium battery cathode material. This technical solution aims at the recovered failed cathode materials, replenishes the lithium source, and regenerates the crystal structure under the action of a plasma jet to restore the electrochemical activity of the cathode powder material, and can directly repair the failed cathode materials under a short process.

[0005] All of the above technical solutions have the problems raised in this background art: they fail to perform personalized repair according to the specific conditions of different materials.

[0006] The information disclosed in this background art section is only intended to enhance the overall understanding of the present invention and should not be regarded as an admission or any form of implication that this information constitutes prior art already known to those of ordinary skill in the art. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to overcome the defects of the prior art and provide a repair system for waste lithium battery cathode materials optimized by machine learning algorithms. Through the method of grouped repair, efficient, low-cost and environmentally friendly repair of waste lithium battery cathode materials is achieved, and the quality and recycling rate of the repaired materials are improved.

[0008] To solve the above technical problems, the present invention provides the following technical solutions:

[0009] A repair system for waste lithium battery cathode materials optimized by machine learning algorithms, including a first repair module, a detection module, a repair strategy module, and a second repair module; wherein:

[0010] The first repair module is used to perform pre-repair on the cathode material to be repaired to obtain a first repaired material;

[0011] The detection module is used to detect the state of the first repaired material, extract state parameters, and divide the first repaired material into different repair groups based on the state parameters;

[0012] The repair strategy module includes a first strategy unit, and the first strategy unit is configured with a repair strategy, and the repair strategy is used to control the repair process of the first repaired material in each repair group;

[0013] The second repair module is used to execute the repair strategy and repair the first repaired material in different repair groups respectively to obtain a second repaired material for each repair group;

[0014] The repair strategy module further includes a second strategy unit, and the second strategy unit is configured with an adjustment strategy, and the adjustment strategy is used to perform grouped adjustment on the second repaired material of each repair group.

[0015] As a preferred solution of the repair system for waste lithium battery cathode materials optimized by machine learning algorithms of the present invention, wherein: the pre-repair includes cleaning, drying, and grinding;

[0016] The first repair module includes a first cleaning unit, a first heat treatment unit, and a mechanical treatment unit; wherein, the first cleaning unit is used to clean the cathode material to be repaired;

[0017] The first heat treatment unit is used to dry the cathode material to be repaired;

[0018] The mechanical treatment unit is used to grind the cathode material to be repaired.

[0019] As a preferred solution of the waste lithium battery cathode material repair system optimized by the machine learning algorithm according to the present invention, wherein: the repair group includes a first repair group, a second repair group, and a third repair group;

[0020] The state parameters include defect density, specific surface area, pore volume, key metal content, and total impurity content;

[0021] The detection module includes a first detection unit, a second detection unit, a third detection unit, and a grouping unit; wherein:

[0022] The first detection unit is equipped with a scanning electron microscope for extracting the defect density;

[0023] The second detection unit is equipped with a specific surface area measuring device for extracting the specific surface area and pore volume;

[0024] The third detection unit is equipped with an ICP-ES composition analysis device for extracting the key metal content and total impurity content.

[0025] As a preferred solution of the waste lithium battery cathode material repair system optimized by the machine learning algorithm according to the present invention, wherein: the grouping unit divides the first repair material into different repair groups based on the state parameters, and the method is as follows:

[0026] Perform standardization processing on each state index of the first repair material. The standardization processing includes normalization and dimensionless, and calculate the state index of the first repair material. The formula is as follows:

[0027] ;

[0028] Wherein, S represents the state index of the first repair material; represents the content of the i-th key metal; the value range of i is 1, 2,..., n, and n is the number of types of key metals; represents the weight coefficient of the content of the i-th key metal, represents the defect density; represents the specific surface area; represents the pore volume; represents the total impurity content; , , , are all weight coefficients;

[0029] The grouping unit is configured with a first threshold and a second threshold for the status indicator; the first threshold is less than the second threshold; for any first repair material, if the status indicator is lower than the first threshold, the first repair material is classified into the third repair group; if the status indicator is not lower than the first threshold and lower than the second threshold, the first repair material is classified into the second repair group; if the status indicator is not lower than the second threshold, the first repair material is classified into the first repair group.

[0030] As a preferred solution of the waste lithium battery cathode material repair system optimized by the machine learning algorithm according to the present invention, wherein: the repair strategy includes a first repair strategy; the first repair strategy is used to control the repair process of the first repair material in the first repair group, specifically as follows:

[0031] Perform deep cleaning and material purification on the first repair material;

[0032] Perform heating and drying and crystal structure optimization on the first repair material;

[0033] Perform surface modification treatment on the first repair material;

[0034] Perform electrochemical activation treatment on the first repair material.

[0035] As a preferred solution of the waste lithium battery cathode material repair system optimized by the machine learning algorithm according to the present invention, wherein: the repair strategy further includes a second repair strategy; the second repair strategy is used to control the repair process of the first repair material in the second repair group, specifically as follows:

[0036] Perform conventional cleaning on the first repair material; perform heating and drying on the first repair material; perform passivation layer removal treatment on the first repair material.

[0037] The repair strategy further includes a third repair strategy; the third repair strategy is used to control the repair process of the first repair material in the third repair group, specifically as follows: Perform conventional cleaning on the first repair material and perform heating and drying.

[0038] As a preferred solution of the waste lithium battery cathode material repair system optimized by the machine learning algorithm according to the present invention, wherein: the second repair module includes a second cleaning unit, a second heat treatment unit, a surface repair unit, and an activation unit; wherein, the second cleaning unit is used to perform conventional cleaning, deep cleaning, and material purification on the first repair material; the second heat treatment unit is used to perform heating and drying and crystal structure optimization on the first repair material; the surface repair unit is used to perform surface modification treatment on the first repair material; the activation unit is used to perform electrochemical activation treatment and passivation layer removal treatment on the first repair material.

[0039] As a preferred solution of the waste lithium battery cathode material repair system optimized by the machine learning algorithm described in the present invention, wherein: the detection module is further configured to detect the state of the second repair material of each repair group, and extract the state parameters of the second repair material in each repair group; wherein, the state parameters extracted from the second repair materials of the first repair group and the second repair group include defect density, specific surface area, pore volume, key metal content, and total impurity content; the state parameters extracted from the third repair group include key metal content;

[0040] The second strategy unit is also respectively configured with a threshold range corresponding to each state parameter of the second repair material in each repair group; the adjustment strategy includes: for the first repair group, if each state parameter of any second repair material is within the corresponding threshold range, no grouping adjustment is made; otherwise, the second repair material is divided into the second repair group;

[0041] For the second repair group, if each state parameter of any second repair material is within the corresponding threshold range, no grouping adjustment is made; otherwise, the second repair material is divided into the third repair group;

[0042] For the third repair group, if each state parameter of any second repair material is within the corresponding threshold range, no grouping adjustment is made; otherwise, the second repair material is removed from the third repair group.

[0043] As a preferred solution of the waste lithium battery cathode material repair system optimized by the machine learning algorithm described in the present invention, wherein: the second strategy unit is also configured with a machine learning algorithm for clustering; the adjustment strategy further includes using the machine learning algorithm to modularly package the second repair materials of the second repair group, specifically as follows:

[0044] S1: Standardize the state parameters of each second repair material in the second repair group and form a state parameter vector;

[0045] S2: Determine the number K of clustering clusters for the division of the second repair materials;

[0046] S3: Randomly select the state parameter vectors of K second repair materials as the clustering centers of the K clustering clusters;

[0047] S4: Divide each second repair material into the clustering cluster with the closest state distance;

[0048] S5: Update the clustering center of each clustering cluster to the mean vector of the state parameter vectors of all second repair materials in the clustering cluster;

[0049] S6: Repeat steps S4 - S5 until the change rate of the cluster center of any cluster is less than a preset change rate threshold;

[0050] The change rate of the cluster center is the Euclidean distance between the current cluster center and the cluster center of the previous iteration; those skilled in the art can set the change rate threshold according to actual needs.

[0051] S7: Combine the second repair materials in each cluster and perform modular packaging.

[0052] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0053] The present invention conducts a detailed state detection on the spent lithium - ion battery cathode material through a detection module, extracts various state parameters, and divides the materials into different repair groups based on these parameters. Each repair group adopts a targeted repair strategy to ensure that the repair process is more accurate and effective, thereby significantly improving the performance consistency of the repaired material. For the materials in different repair groups, different repair strategies are adopted. This personalized repair strategy can more effectively restore the electrochemical performance of the material and improve the overall quality of the repaired material.

[0054] Through the grouping repair method, the most suitable repair process is selected according to the actual state of different materials, avoiding unnecessary processing steps and reducing the repair cost. This flexible repair strategy greatly improves the resource utilization rate and reduces the overall repair cost. The repaired cathode material can be applied to different scenarios according to its quality and performance.

[0055] By using a machine - learning algorithm to modularly package the second repair materials in the second repair group, the performance of each group of modularly packaged second repair materials is similar and can be used for the production of the same energy - storage device. Automatically matching the optimal cathode materials for modular packaging combination can minimize the performance differences of the cathode materials inside the energy - storage device package and maintain the consistency and stability of the overall performance of the energy - storage device. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. Among them:

[0057] Figure 1 is a schematic structural diagram of a spent lithium - ion battery cathode material repair system optimized by a machine - learning algorithm provided by the present invention;

[0058] Figure 2Schematic diagram of the working principle of the waste lithium battery cathode material repair system optimized by machine learning algorithm provided by the present invention;

[0059] Figure 3 Flow chart of the machine learning algorithm for modular packaging of the second repair material provided by the present invention. Specific embodiments

[0060] 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 specific features in the embodiments of the present invention are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. Without conflict, the technical features in the embodiments of the present invention and the embodiments can be combined with each other.

[0061] This embodiment introduces a waste lithium battery cathode material repair system optimized by a machine learning algorithm. Referring to Figure 1 , the system includes a first repair module, a detection module, a repair strategy module, and a second repair module; the working principles of each module are as Figure 2 shown.

[0062] The first repair module is used to pre-repair the cathode material to be repaired to obtain a first repair material;

[0063] The pre-repair includes cleaning, drying, and grinding; the pre-repair is mainly to remove pollutants and impurities on the surface of the cathode material to be repaired and prepare for subsequent state detection and further repair.

[0064] The first repair module includes a first cleaning unit, a first heat treatment unit, and a mechanical treatment unit; among them, the first cleaning unit is used to clean the cathode material to be repaired; the first cleaning unit uses deionized water, organic solvents (such as ethanol) or weak acidic solutions (such as diluted hydrochloric acid) for ultrasonic cleaning to remove electrolyte residues and other pollutants attached to the surface of the cathode material to be repaired.

[0065] The first heat treatment unit is used to dry the cathode material to be repaired; the first heat treatment unit dries or calcines the cathode material to be repaired at a certain temperature to remove moisture and volatile substances.

[0066] The mechanical treatment unit is used to grind the cathode material to be repaired. The mechanical treatment unit removes large particle impurities through mechanical means such as screening and grinding and ensures that the particle size distribution of the cathode material to be repaired is uniform.

[0067] The detection module is used to detect the state of the first repair material, extract state parameters, and classify the first repair material into different repair groups based on the state parameters;

[0068] The repair groups include a first repair group, a second repair group, and a third repair group;

[0069] The state parameters include defect density, specific surface area, pore volume, key metal content, and total impurity content;

[0070] After pre - repair is completed, it is necessary to conduct a state detection on the first repair material. The above - mentioned state parameters can be used to evaluate the repair difficulty and potential recycling value of the first repair material.

[0071] The detection module includes a first detection unit, a second detection unit, a third detection unit, and a grouping unit; where:

[0072] The first detection unit is equipped with a scanning electron microscope for extracting the defect density; through the scanning electron microscope, the material morphology can be observed, and the surface element composition and distribution can be analyzed. Observe and record the surface morphology features such as cracks and holes, and calculate the density of defects as the defect density.

[0073] The second detection unit is equipped with a specific surface area measurement device for extracting the specific surface area and pore volume; the second detection unit can accurately measure the specific surface area and pore volume of the first repair material by the BET method. These two state parameters reflect the pore structure characteristics of the material and are important indicators for measuring the material activity.

[0074] The third detection unit is equipped with an ICP - ES composition analysis device for extracting the key metal content and total impurity content; among them, the key metal content includes the percentage contents of lithium, cobalt, nickel, manganese, etc.; the total impurity content includes the total percentage contents of impurity elements such as iron, copper, aluminum, etc. that may exist.

[0075] The grouping unit divides the first repair material into different repair groups based on the state parameters, and the method is as follows:

[0076] Perform standardization processing on each state index of the first repair material. The standardization processing includes normalization and dimensionless processing, and calculate the state index of the first repair material. The formula is as follows:

[0077] ;

[0078] Among them, S represents the state index of the first repair material; represents the content of the i - th key metal; represents the weight coefficient of the i - th key metal content, which is set by those skilled in the art based on actual needs; the value range of i is 1, 2, ……, n, and n is the number of types of key metals; represents the defect density; represents the specific surface area; represents the pore volume; Represents the total impurity content; , , , are all weight coefficients, which are set by those skilled in the art based on actual requirements.

[0079] The grouping unit is configured with a first threshold and a second threshold for the status indicator; the first threshold is less than the second threshold; for any first repair material, if the status indicator is lower than the first threshold, the first repair material is classified into the third repair group; if the status indicator is not lower than the first threshold and lower than the second threshold, the first repair material is classified into the second repair group; if the status indicator is not lower than the second threshold, the first repair material is classified into the first repair group.

[0080] The repair strategy module includes a first strategy unit, and the first strategy unit is configured with a repair strategy, and the repair strategy is used to control the repair process of the first repair material in each repair group;

[0081] The repair strategy includes a first repair strategy; the first repair strategy is used to control the repair process of the first repair material in the first repair group, specifically as follows:

[0082] Perform deep cleaning and material purification on the first repair material; perform multiple cleanings with a high-concentration cleaning solvent, and use the chemical precipitation method to remove trace impurities on the surface of the first repair material.

[0083] Perform heating and drying and crystal structure optimization on the first repair material; perform annealing treatment on the first repair material by precisely controlling the temperature curve to restore or optimize its crystal structure.

[0084] Perform surface modification treatment on the first repair material; use coating technology or doping treatment to improve the interfacial properties and electrochemical stability of the first repair material.

[0085] Perform electrochemical activation treatment on the first repair material; activate the first repair material through charge / discharge cycles under specific conditions to improve its initial capacity.

[0086] The repair strategy further includes a second repair strategy; the second repair strategy is used to control the repair process of the first repair material in the second repair group, specifically as follows:

[0087] Perform conventional cleaning on the first repair material; compared with deep cleaning, the concentration of the cleaning solvent and the number of cleanings are reduced in conventional cleaning;

[0088] Perform heating and drying on the first repair material; the purpose is to dry the surface of the cleaned first repair material.

[0089] Perform a passivation layer removal treatment on the first repair material; by performing a limited number of charge and discharge cycles, remove the passivation layer on the surface of the first repair material to improve its electrochemical performance.

[0090] The repair strategy further includes a third repair strategy; the third repair strategy is used to control the repair process of the first repair material in the third repair group, specifically as follows: perform a conventional cleaning on the first repair material and then heat it for drying.

[0091] Through such meticulous grouping and targeted repair process design, while maximizing the recycling rate of the waste lithium battery cathode material, the repair cost can be saved, and it can be ensured that the repaired material can meet the requirements of different application scenarios.

[0092] The second repair module is used to execute the repair strategy, and repair the first repair material in different repair groups respectively to obtain the second repair material of each repair group;

[0093] The second repair module includes a second cleaning unit, a second heat treatment unit, a surface repair unit, and an activation unit; among them, the second cleaning unit is used to perform conventional cleaning, deep cleaning, and material purification on the first repair material; the second heat treatment unit is used to heat and dry the first repair material and optimize its crystal structure; the surface repair unit is used to perform surface modification treatment on the first repair material; the activation unit is used to perform electrochemical activation treatment and passivation layer removal treatment on the first repair material.

[0094] The repaired cathode material can be applied to different scenarios according to its quality and performance. Specifically as follows: The cathode material in the first repair group is reused to manufacture lithium batteries. The cathode material in the first repair group has good quality and can be directly used in the manufacture of new lithium batteries, especially in electronic products and electric vehicle batteries with high performance requirements. The performance of the cathode material in the second repair group has slightly decreased but can still be used in energy storage batteries. They can be used in fixed energy storage systems, such as home energy storage units or grid energy storage facilities. The cathode material in the third repair group is difficult to repair and has low potential for reuse, and is not suitable for direct use in lithium battery production, but the metal components in it still have recycling value and can be supplied as raw materials to other industries, such as alloy manufacturing, chemical products, etc.

[0095] The repair strategy module further includes a second strategy unit, and the second strategy unit is configured with an adjustment strategy, and the adjustment strategy is used to perform grouping adjustment on the second repair material of each repair group.

[0096] The detection module is also used to detect the status of the second repair materials of each repair group and extract the status parameters of the second repair materials in each repair group; among them, the status parameters extracted from the second repair materials of the first repair group and the second repair group include defect density, specific surface area, pore volume, key metal content, and total impurity content; the status parameters extracted from the third repair group include key metal content;

[0097] The second strategy unit is also respectively configured with a threshold range corresponding to each status parameter of the second repair materials in each repair group; for example: for the first repair group, the threshold range corresponding to each status parameter includes: Defect density: <100 pieces / μm²; the materials of the first repair group need to have as few defects as possible to ensure good electrochemical performance and cycle stability. Specific surface area: 5 - 20 m² / g; an appropriate specific surface area helps to improve the diffusion rate of lithium ions, but too high a specific surface area may lead to an increase in side reactions. Pore volume: 0.2 - 0.5 cm³ / g; an appropriate pore volume helps the insertion and extraction of lithium ions while maintaining good structural stability. Key metal content: Lithium: ≥5 wt%; Cobalt: ≥15 wt%; Nickel: ≥10 wt%; Manganese: ≥10 wt%; these contents meet the standard specifications for new battery production, ensuring that the materials have good electrochemical performance. Total impurity content: <0.1 wt%; high purity is the key to high-quality materials, and low impurity content helps to improve the stability and electrochemical performance of the materials.

[0098] For the second repair group, the threshold range corresponding to each status parameter includes: Defect density: <300 pieces / μm²; Specific surface area: 10 - 30 m² / g; Pore volume: 0.3 - 0.8 cm³ / g; Key metal content: Lithium: ≥4 wt%; Cobalt ≥10 wt%; Nickel: ≥8 wt%; Manganese: ≥8 wt%; Total impurity content: <0.2 wt%;

[0099] For the third repair group, the threshold range corresponding to each status parameter includes: Lithium content ≥5 wt%, Cobalt content ≥2 wt%, Nickel content ≥2 wt%. Since the third repair group is subsequently used for metal recycling, therefore, it is only necessary to confirm whether the metal content meets the recycling standards.

[0100] The adjustment strategy includes: for the first repair group, if each status parameter of any second repair material is within the corresponding threshold range, no grouping adjustment is made; otherwise, the second repair material is classified into the second repair group;

[0101] For the second repair group, if each status parameter of any second repair material is within the corresponding threshold range, no grouping adjustment is made; otherwise, the second repair material is classified into the third repair group;

[0102] For the third repair group, if each state parameter of any second repair material is within the corresponding threshold range, no grouping adjustment is made; otherwise, the second repair material is removed from the third repair group; the second repair material removed from the third repair group has too low metal content and has no recycling value.

[0103] The second policy unit is also configured with a machine learning algorithm for clustering; the adjustment policy further includes modularly packing the second repair materials in the second repair group using the machine learning algorithm; referring to Figure 3 , specifically as follows:

[0104] S1: Standardize each state parameter of the second repair materials in the second repair group and form a state parameter vector;

[0105] S2: Determine the number K of clustering clusters for the second repair materials; determine a suitable number of clusters according to the requirements of modular packing and expert experience.

[0106] S3: Randomly select the state parameter vectors of K second repair materials as the clustering centers of the K clustering clusters;

[0107] S4: Assign each second repair material to the clustering cluster with the closest state distance;

[0108] The state distance is the Euclidean distance between the state parameter vector of any second repair material and the clustering center of any clustering cluster;

[0109] S5: Update the clustering center of each clustering cluster to the mean vector of the state parameter vectors of all second repair materials in the clustering cluster;

[0110] S6: Repeat steps S4 - S5 until the change rate of the clustering center of any clustering cluster is less than the preset change rate threshold;

[0111] The change rate of the clustering center is the Euclidean distance between the current clustering center and the clustering center of the previous iteration; those skilled in the art can set the change rate threshold according to actual needs.

[0112] S7: Combine the second repair materials in each clustering cluster together for modular packing.

[0113] By performing modular packing on the second repair materials in the second repair group through the above steps, the performance of each group of modularly packed second repair materials is similar and can be used for the production of the same energy storage device; automatically matching the optimal cathode materials for modular packing combination can minimize the performance differences of the cathode materials inside the energy storage device package and maintain the consistency and stability of the overall performance of the energy storage device.

[0114] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0115] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the purpose and scope of the present invention. All of these are within the protection scope of the present invention.

Claims

1. A waste lithium battery positive electrode material repair system optimized by machine learning algorithm, characterized by: It includes a first repair module, a detection module, a repair strategy module, and a second repair module; wherein: The first repair module is used to pre-repair the positive electrode material to be repaired to obtain a first repair material; The detection module is used to perform state detection on the first repair material, extract state parameters, and classify the first repair material into different repair groups based on the state parameters; The repair group includes a first repair group, a second repair group, and a third repair group; The state parameters include defect density, specific surface area, pore volume, key metal content, and total impurity content; The detection module includes a first detection unit, a second detection unit, a third detection unit, and a grouping unit; wherein: The first detection unit is configured with a scanning electron microscope for extracting the defect density; The second detection unit is equipped with a specific surface area measuring device for extracting the specific surface area and pore volume; The third detection unit is equipped with an ICP-ES component analysis device for extracting key metal content and total impurity content; The repair strategy module includes a first strategy unit, wherein the first strategy unit is configured with a repair strategy, wherein the repair strategy is used to control the repair process of the first repair material in each repair group; The second repair module is used to execute the repair strategy, and repair the first repair materials in different repair groups respectively to obtain the second repair material of each repair group; The repair strategy module further includes a second strategy unit, wherein the second strategy unit is configured with an adjustment strategy, and the adjustment strategy is used to perform group adjustment on the second repair material of each repair group; The detection module is also used to perform state detection on the second repair material of each repair group and extract the state parameters of the second repair material in each repair group; wherein the state parameters extracted from the second repair materials of the first repair group and the second repair group include defect density, specific surface area, pore volume, key metal content, and total impurity content; and the state parameters extracted from the third repair group include key metal content; The second strategy unit is also respectively configured with a threshold range corresponding to each state parameter of the second repair material in each repair group; the adjustment strategy includes: for the first repair group, if each state parameter of any second repair material is within the corresponding threshold range, no group adjustment is performed; otherwise, the second repair material is divided into the second repair group; For the second repair group, if each state parameter of any second repair material is within the corresponding threshold range, no grouping adjustment is made; otherwise, the second repair material is divided into the third repair group; For the third repair group, if each state parameter of any second repair material is within the corresponding threshold range, no grouping adjustment is made; otherwise, the second repair material is removed from the third repair group; The second strategy unit is also configured with a machine learning algorithm for clustering; the adjustment strategy also includes modularizing and packaging the second repair materials of the second repair group using the machine learning algorithm, as follows: S1: standardizing each state parameter of the second restoration material in the second restoration group and forming a state parameter vector; S2: Determine the number K of clusters into which the second restoration material is divided; S3: randomly selecting K state parameter vectors of the second restorative material as cluster centers of the K clusters; S4: Divide each second restorative material into a cluster with the closest state distance; S5: updating the cluster center of each cluster to the mean vector of the state parameter vectors of all the second restoration materials in the cluster; S6: Repeat steps S4 to S5 until the change rate of the cluster center of any cluster cluster is less than a preset change rate threshold; the change rate of the cluster center is the Euclidean distance between the current cluster center and the cluster center of the previous iteration; S7: The second repair materials in each cluster are combined together for modular packaging.

2. The waste lithium battery positive electrode material repair system optimized by machine learning algorithm as claimed in claim 1, characterized in that: The pre-repairing includes cleaning, drying and grinding; The first repair module includes a first cleaning unit, a first heat treatment unit, and a mechanical treatment unit; wherein the first cleaning unit is used to clean the positive electrode material to be repaired; The first heat treatment unit is used to dry the positive electrode material to be repaired; The mechanical processing unit is used for grinding the positive electrode material to be repaired.

3. The waste lithium battery positive electrode material repair system optimized by machine learning algorithm as claimed in claim 2, characterized in that: The grouping unit divides the first repair material into different repair groups based on the state parameter in the following manner: Performing standardization processing on each state index of the first repair material, wherein the standardization processing includes normalization and dimension removal, and calculating the state index of the first repair material; the state index of the first repair material is obtained by weighted summation of each key metal content, defect density, specific surface area, pore volume, and total impurity content; The grouping unit is configured with a first threshold and a second threshold of a state indicator; the first threshold is less than the second threshold; for any first repair material, if the state indicator is lower than the first threshold, the first repair material is classified into a third repair group; If the state indicator is not lower than the first threshold and lower than the second threshold, classifying the first repair material into the second repair group; If the state indicator is not lower than the second threshold, the first repair material is classified into the first repair group.

4. The waste lithium battery positive electrode material repair system optimized by machine learning algorithm as claimed in claim 3, characterized in that: The repair strategy includes a first repair strategy; the first repair strategy is used to control the repair process of the first repair material in the first repair group, specifically as follows: Deep cleaning and material purification of the first repair material; Heating, drying and optimizing the crystal structure of the first repair material; performing surface modification treatment on the first repair material; The first repair material is subjected to electrochemical activation treatment.

5. The waste lithium battery positive electrode material repair system optimized by machine learning algorithm as claimed in claim 4, characterized in that: The repair strategy also includes a second repair strategy; the second repair strategy is used to control the repair process of the first repair material in the second repair group, as follows: Performing routine cleaning on the first repair material; heating and drying the first repair material; and removing the passivation layer of the first repair material; The repair strategy also includes a third repair strategy; the third repair strategy is used to control the repair process of the first repair material in the third repair group, specifically as follows: the first repair material is routinely cleaned and heated and dried.

6. The waste lithium battery positive electrode material repair system optimized by machine learning algorithm as claimed in claim 5, characterized in that: The second repair module includes a second cleaning unit, a second heat treatment unit, a surface repair unit, and an activation unit; wherein the second cleaning unit is used to perform conventional cleaning, deep cleaning, and material purification on the first repair material; the second heat treatment unit is used to heat and dry the first repair material and optimize the crystal structure; the surface repair unit is used to perform surface modification treatment on the first repair material; and the activation unit is used to perform electrochemical activation treatment and passivation layer removal treatment on the first repair material.

Images