A battery shredding and sorting method, system, device, and storage medium

Abstract

The application provides a battery crushing and sorting method, system, device and storage medium, and relates to the field of lithium battery recycling. The method comprises the following steps: acquiring a material image and crushing treatment parameters after lithium battery crushing and sorting, wherein the material image comprises a first metal particle image and a second metal particle image; calculating the proportions of the two metal particles; judging whether the crushing and sorting is qualified according to the comparison result of the proportions and respective preset thresholds; if at least one kind of metal particle is unqualified, determining the unqualified reason; inputting the metal particle image, the crushing treatment parameters and the unqualified reason into a pre-trained process parameter adjustment model to obtain a parameter adjustment threshold; and obtaining target process parameters according to the parameter adjustment threshold and the crushing treatment parameters, so that subsequent sorting is qualified. In this way, the parameter adjustment threshold obtained from the material image, the crushing unqualified reason and the crushing treatment parameters is used to adjust the crushing treatment parameters, replacing manual adjustment, ensuring stable production line and improving sorting quality.

Description

Technical Field

[0001] This application relates to the field of lithium battery recycling technology, and more specifically, to a battery crushing and sorting method, system, equipment, and storage medium. Background Technology

[0002] When recycling waste lithium-ion batteries, it is necessary to perform fine sorting of the lithium battery materials after crushing and pyrolysis in order to achieve efficient separation and separate recycling of copper foil, aluminum foil and black powder (containing nickel, cobalt, manganese, lithium and carbon materials).

[0003] Currently, the industry commonly adopts a multi-stage physical separation process that combines crushing, pyrolysis, screening, airflow separation, and gravity separation. Among them, the airflow separation equipment initially separates light black powder from heavy metal particles by controlling the pressure difference between 800 Pa and 2500 Pa. Then, the remaining copper-aluminum mixture is sent to a circular vibrating screen for classification according to particle size, and then enters a gravity separator. The different settling behaviors of copper and aluminum due to density differences at the same particle size are utilized to complete the final separation under the combined action of bottom fan blowing and mechanical vibration.

[0004] In the production process, if the copper-aluminum separation obtained in the final gravity separation stage is not up to standard, the parameters of each preceding process need to be adjusted to ensure that the copper-aluminum separation obtained in the final gravity separation is up to standard. However, since there are many equipment production parameters that need to be adjusted in each process, relying on manual adjustment may affect the stability of the production line, while not making adjustments may affect the separation quality and even create a chain of risks. This needs to be improved. Summary of the Invention

[0005] In view of this, the purpose of this application is to provide a battery crushing and sorting method, system, equipment and storage medium to solve the problem that the current method requires manual adjustment of various process parameters after the specific gravity sorting is unqualified, which can easily affect the stability of the production line and reduce the sorting quality.

[0006] To achieve the above objectives, in a first aspect, embodiments of this application provide a battery crushing and sorting method, the method comprising: acquiring material images and crushing processing parameters of the crushed and sorted lithium battery, wherein the material images include a first metal particle image and a second metal particle image; calculating the proportion of a first metal particle in the first metal particle image and the proportion of a second metal particle in the second metal particle image; determining whether the first metal particle and the second metal particle are qualified for crushing and sorting based on a comparison result of the proportion of the first metal particle with a first preset threshold and a comparison result of the proportion of the second metal particle with a second preset threshold; determining the reason for the unqualified crushing and sorting if at least one of the first metal particle and the second metal particle is unqualified; inputting the first metal particle image, the second metal particle image, the crushing processing parameters, and the reason for the unqualified crushing and sorting into a pre-trained process parameter adjustment model to obtain a parameter adjustment threshold; and obtaining a target process parameter based on the parameter adjustment threshold and the crushing processing parameters, wherein the target process parameter is used to ensure that the first metal particle and the second metal particle after the crushing and sorting of the lithium battery are qualified for crushing and sorting.

[0007] In this embodiment, by analyzing the material images after lithium battery crushing and sorting, and simultaneously obtaining the crushing and sorting parameters in the crushing and sorting process, the proportion of the first metal particles in the first metal particle image and the proportion of the second metal particles in the second metal particle image are calculated. Then, based on the proportions of the first and second metal particles, the reasons for crushing and sorting failures are determined. This allows the first metal particle image, the second metal particle image, the crushing and sorting parameters, and the reasons for crushing and sorting failures to be input into a pre-trained process parameter adjustment model to obtain parameter adjustment thresholds. These thresholds are then used to adjust the crushing and sorting parameters for those failing the crushing and sorting process, ensuring that the first and second metal particles pass the crushing and sorting process. Thus, by using material images, the reasons for crushing failures derived from the analysis of the first and second metal particles, and the parameter adjustment thresholds derived from the crushing and sorting parameters, the various crushing and sorting parameters in the crushing and sorting process can be adjusted using these thresholds, replacing manual parameter adjustments, maintaining production line stability, and improving the quality of metal particle crushing and sorting.

[0008] In some embodiments, the first metal particle image and the second metal particle image further include black particles. The calculation of the proportion of the first metal particle in the first metal particle image and the proportion of the second metal particle in the second metal particle image includes: obtaining the brightness values ​​of the first metal particle image and the second metal particle image, and comparing the brightness values ​​of the first metal particle image and the second metal particle image with preset brightness thresholds respectively; if the brightness value of the first metal particle image is greater than or equal to the preset brightness threshold, determining that the black particles in the first metal particle image are sorted successfully; if the brightness value of the second metal particle image is greater than or equal to the preset brightness threshold, determining that the black particles in the second metal particle image are sorted successfully; and if both the first metal particle image and the second metal particle image are sorted successfully, calculating the proportion of the first metal particle in the first metal particle image and the proportion of the second metal particle in the second metal particle image.

[0009] This setup allows for the calculation of the proportion of metal particles based on whether the black powder sorting in the first and second metal particle images is qualified, thereby reducing the impact of black powder on the calculation of the proportion of metal particles.

[0010] In some embodiments, determining whether the first metal particle and the second metal particle are qualified for crushing and sorting based on the comparison result of the proportion of the first metal particle with a first preset threshold and the comparison result of the proportion of the second metal particle with a second preset threshold includes: determining that the first metal particle is unqualified for crushing and sorting when the proportion of the first metal particle is less than the first preset threshold; determining that the first metal particle is qualified for crushing and sorting when the proportion of the first metal particle is greater than or equal to the first preset threshold; determining that the second metal particle is unqualified for crushing and sorting when the proportion of the second metal particle is less than the second preset threshold; and determining that the second metal particle is qualified for crushing and sorting when the proportion of the second metal particle is greater than or equal to the second preset threshold.

[0011] This setting allows for the determination of whether the first and second metal particles in the current crushing and sorting process are qualified by comparing the proportion of the first metal particles with the first preset threshold and the proportion of the second metal particles with the second preset threshold. This helps to determine whether the current crushing and processing parameters have been adjusted properly.

[0012] In some embodiments, determining the reason for the failure of the crushing and sorting includes: determining that the failure of the crushing and sorting of the first metal particles is due to failure of black powder sorting when the brightness of the first metal particle image is less than a preset brightness threshold; determining that the failure of the crushing and sorting of the second metal particles is due to failure of black powder sorting when the brightness of the second metal particle image is less than a preset brightness threshold; determining that the failure of the crushing and sorting of the first metal particles is due to failure of proportion when the proportion of the first metal particles is less than the first preset threshold; determining that the failure of the crushing and sorting of the second metal particles is due to failure of proportion when the proportion of the second metal particles is less than the second preset threshold; determining that the failure of the crushing and sorting of the first metal particles is due to failure of particle size when the particle size of the first metal particles is greater than a preset particle size threshold; and determining that the failure of the crushing and sorting of the second metal particles is due to failure of particle size when the particle size of the second metal particles is greater than a preset particle size threshold.

[0013] This setup is designed to help determine whether the failure to properly crush and separate the first and second metal particles is due to issues such as improper black powder separation, incorrect proportion of black powder, or incorrect particle size.

[0014] In some embodiments, the process parameter adjustment model is trained through the following steps: obtaining a training dataset, which includes multiple training data and corresponding real parameter adjustment thresholds, wherein the training data includes a first metal particle image, a second metal particle image, crushing processing parameters, and reasons for crushing and sorting failures corresponding to each metal particle image; inputting the training data into a pre-constructed process parameter adjustment model to obtain corresponding predicted parameter adjustment thresholds; calculating a loss function based on the predicted parameter adjustment thresholds corresponding to the training data and the real parameter adjustment thresholds, and updating the model parameters of the process parameter adjustment model based on the loss function to obtain a trained process parameter adjustment model.

[0015] This setup allows for the calculation of the loss function using the predicted parameters and the actual parameters obtained from training to adjust the threshold. This loss function then updates the model parameters of the process parameter adjustment model, resulting in a well-trained process parameter adjustment model. This helps to adjust the crushing and sorting parameters of the first and second metal particles using the target process parameters.

[0016] In some embodiments, obtaining the target process parameters based on the parameter adjustment threshold and the crushing treatment parameters includes: adding the parameter adjustment threshold to the crushing treatment parameters to obtain the target process parameters.

[0017] This setting allows for the adjustment of thresholds based on crushing parameters to obtain target process parameters, thereby adjusting the first and second metal particles to meet the crushing and sorting requirements.

[0018] In some embodiments, the lithium battery crushing and sorting step includes: crushing and heat-treating the lithium battery to obtain first metal particles, second metal particles and black powder; separating the first metal particles, second metal particles and black powder by airflow to obtain the first metal particles and second metal particles after recovering the black powder; and separating the first metal particles and second metal particles after recovering the black powder by gravity separation to obtain the first metal particles and second metal particles that are separated from each other.

[0019] This setup facilitates the separation of the first and second metal particles obtained from the lithium battery after crushing, heat treatment, airflow separation, and gravity separation, which helps to separate the first and second metal particles after the lithium battery is recycled.

[0020] Secondly, embodiments of this application provide a battery crushing and sorting system, the system comprising: a preprocessing module, configured to acquire material images and crushing processing parameters after lithium battery crushing and sorting, the material images including a first metal particle image and a second metal particle image; a comparison module, configured to calculate the proportion of a first metal particle in the first metal particle image and the proportion of a second metal particle in the second metal particle image; determine whether the first metal particle and the second metal particle are qualified for crushing and sorting based on the comparison result of the proportion of the first metal particle with a first preset threshold and the comparison result of the proportion of the second metal particle with a second preset threshold; determine the reason for the unqualified crushing and sorting if at least one of the first metal particle and the second metal particle is unqualified; and a processing module, configured to input the first metal particle image, the second metal particle image, the crushing processing parameters, and the reason for the unqualified crushing and sorting into a pre-trained process parameter adjustment model to obtain a parameter adjustment threshold; and obtain a target process parameter based on the parameter adjustment threshold and the crushing processing parameters, the target process parameter being used to ensure that the first metal particle and the second metal particle after lithium battery crushing and sorting are qualified for crushing and sorting.

[0021] Thirdly, embodiments of this application provide an electronic device, including a processor and a memory, wherein the memory stores a computer program executable by the processor, and the processor can execute the computer program to implement the battery breakage sorting method as described in the first aspect.

[0022] Fourthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the battery breakage sorting method as described in the first aspect.

[0023] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 A flowchart of a battery breakage and sorting method provided in this application embodiment; Figure 2 This is a flowchart of the lithium battery crushing and sorting process provided in the embodiments of this application; Figure 3 This is a schematic diagram of the functional modules of the battery crushing and sorting system provided in the embodiments of this application; Figure 4 This is a block diagram of an electronic device provided in an embodiment of this application.

[0026] Reference numerals: 1000-Battery crushing and sorting system; 1100-Pre-processing module; 1200-Comparison module; 1300-Processing module; 2000-Electronic equipment; 2100-Processor; 2200-Memory; 2300-Bus; 2400-Communication interface. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can be arranged and designed in various different configurations.

[0028] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0029] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0030] As described in the background section, traditional copper-aluminum separation methods require on-site inspections. Operators are prone to negligence, delays in inspections, or lack of experience, leading to ineffective adjustments to crushing parameters and resulting in poor copper-aluminum separation quality, necessitating reprocessing. Furthermore, if the final gravity separation fails to meet standards, adjustments to the parameters of preceding processes are required to achieve satisfactory separation. However, given the numerous equipment parameters that need adjustment across various processes, relying on manual adjustments may disrupt production line stability, while neglecting adjustments could compromise product quality and even create cascading risks. Therefore, improvements are needed.

[0031] Therefore, this application provides a battery crushing and sorting method. By analyzing material images, the reasons for crushing defects derived from the analysis of first and second metal particles, and parameter adjustment thresholds derived from crushing processing parameters, the method allows for the adjustment of various crushing parameters in the crushing process. This replaces manual parameter adjustment, maintains production line stability, and improves the quality of metal particle crushing and sorting. Simultaneously, it achieves copper-aluminum sorting identification and automatic adjustment, avoiding situations where human error leads to improper adjustment of crushing processing parameters, resulting in poor copper-aluminum sorting quality and the need for reprocessing. This shortens the copper-aluminum sorting process and equipment lifespan, achieving energy savings through low-carbon practices.

[0032] See Figure 1 , Figure 1 This application provides a flowchart of a battery breakage and sorting method, which includes steps S100 to S600: S100: Obtain material images and crushing processing parameters after lithium battery crushing and sorting. The material images include a first metal particle image and a second metal particle image.

[0033] In this embodiment, after the lithium battery is crushed and sorted, a high-speed camera captures real-time images of the material at the discharge port of the gravity separator. The captured images include images of the first and second metal particles, and crushing parameters are simultaneously acquired. These parameters include the granulator gap, granulation frequency, airflow separation volume, vibrating screen frequency, screen aperture, gravity separator frequency, and airflow and pressure. It should be noted that the first metal particle can be copper, and the second metal particle can be aluminum.

[0034] S200. Calculate the proportion of the first metal particle in the first metal particle image and the proportion of the second metal particle in the second metal particle image.

[0035] In this embodiment, the acquired material image is processed using a visual recognition algorithm. First, the material image undergoes grayscale processing, noise reduction, edge detection, and color segmentation. Then, color features of the first metal particle (copper) and the second metal particle (aluminum) are extracted, and particle size distribution is statistically analyzed. After image processing, the entire material image is assessed at the pixel level, and the number of the first and second metal particles in the image is counted. The proportions of the first and second metal particles in the entire image are calculated, respectively.

[0036] S300. Based on the comparison results of the proportion of the first metal particles with the first preset threshold and the comparison results of the proportion of the second metal particles with the second preset threshold, determine whether the first metal particles and the second metal particles are qualified for crushing and sorting.

[0037] In this embodiment, the calculated percentage of first metal particles is compared with a first preset threshold, and the percentage of second metal particles is compared with a second preset threshold, wherein the first preset threshold is greater than the second preset threshold. Specifically, the qualified threshold for the first metal particles (copper) is set to 98%, and the qualified threshold for the second metal particles (aluminum) is set to 95%. When the percentage of copper particles is greater than or equal to 98%, the first metal particles are determined to be qualified in crushing and sorting; otherwise, the first metal particles are determined to be unqualified in crushing and sorting. When the percentage of aluminum particles is greater than or equal to 95%, the second metal particles are determined to be qualified in crushing and sorting; otherwise, the second metal particles are determined to be unqualified in crushing and sorting.

[0038] S400, if at least one of the first metal particles and the second metal particles fails to meet the crushing and sorting requirements, determine the cause of the failure.

[0039] In this embodiment, if at least one of the first metal particle and the second metal particle is determined to be unqualified in crushing and sorting, then both the first metal particle and the second metal particle are unqualified in crushing and sorting, and the reasons for the unqualification are further determined. Specifically, defect analysis is performed on the material image after image processing, and the reasons for unqualified crushing and sorting are determined based on brightness information, proportion information, and particle size information. The reasons for unqualified crushing and sorting include unqualified black powder sorting, unqualified proportion, and unqualified particle size, thereby completing the labeling of the reasons for unqualification.

[0040] S500: Input the first metal particle image, the second metal particle image, the crushing processing parameters, and the reasons for unqualified crushing and sorting into the pre-trained process parameter adjustment model to obtain the parameter adjustment threshold.

[0041] In this embodiment, the first metal particle image, the second metal particle image, the current crushing processing parameters, and the determined reasons for the failure of crushing and sorting are input into a pre-trained process parameter adjustment model. The process parameter adjustment model calculates based on the learned weight values ​​and outputs the parameter adjustment threshold corresponding to each crushing processing parameter.

[0042] S600: Adjust the threshold and crushing parameters according to the parameters to obtain the target process parameters.

[0043] In this embodiment, the target process parameters are used to ensure that the first and second metal particles after lithium battery crushing and sorting meet the required crushing and sorting standards. The target process parameters are obtained by adding the parameter adjustment threshold to the crushing and processing parameters. These target process parameters are then transmitted to a distributed control system (DCS) to automatically decide and precisely adjust the parameters of each device in the lithium battery crushing and sorting process, ensuring that the copper and aluminum particles after crushing and sorting meet the required crushing and sorting standards.

[0044] In some embodiments, the first metal particle image and the second metal particle image further include black powder, which includes nickel, cobalt, and lithium. For step S200, one possible implementation provided in this application embodiment includes sub-steps S201~S204: S201. Obtain the brightness values ​​of the first metal particle image and the second metal particle image, and compare the brightness values ​​of the first metal particle image and the second metal particle image with preset brightness thresholds respectively.

[0045] S202. If the brightness value of the first metal particle image is greater than or equal to the preset brightness threshold, the black powder in the first metal particle image is determined to be qualified for sorting.

[0046] S203. If the brightness value of the second metal particle image is greater than or equal to the preset brightness threshold, the black powder in the second metal particle image is determined to be qualified for sorting.

[0047] In steps S201-S203, the brightness values ​​of the first metal particle image and the second metal particle image are calculated, and the calculated brightness values ​​are compared with preset brightness thresholds respectively. When the image brightness value is lower than the set threshold, it indicates that there is a lot of diffuse black powder in the image, and it is directly determined that the black powder sorting is unqualified; when the image brightness value is greater than or equal to the set threshold, the black powder sorting is determined to be qualified.

[0048] S204. If both the first metal particle image and the second metal particle image are qualified for black powder sorting, calculate the proportion of the first metal particle in the first metal particle image and the proportion of the second metal particle in the second metal particle image.

[0049] In this embodiment, only after both the first and second metal particle images are determined to be qualified for black powder sorting are pixel-level statistical calculations performed on them. The number of copper particles in the first image and the number of aluminum particles in the second image are counted respectively. The proportions of copper and aluminum particles in the entire image are then calculated according to formulas (1) and (2), thereby completing the detection of the sorting results. Formulas (1) and (2) are as follows: (1) (2) In formula (1), P cu This represents the purity of the copper particles, i.e., the percentage; N is the total number of particles detected; p is the index of a single particle; Is Cu It is a decision function; H p It is the hue feature of the p-th particle; S p V is the saturation characteristic of the p-th particle; p It is the brightness characteristic of the p-th particle; the same applies to Al in formula (2).

[0050] In some embodiments, for step S300, one possible implementation provided by the present application is that step S300 includes sub-steps S301 to S304: S301. If the proportion of the first metal particles is less than the first preset threshold, the first metal particles are determined to be unqualified for crushing and sorting.

[0051] S302. If the proportion of the first metal particles is greater than or equal to the first preset threshold, the first metal particles are determined to be qualified for crushing and sorting.

[0052] In steps S301 to S302, the calculated percentage of the first metal particles (copper) is compared with a first preset threshold (i.e., 98%). If the percentage of copper particles is less than 98%, the copper particle crushing and sorting is determined to be unqualified; if the percentage of copper particles is greater than or equal to 98%, the copper particle crushing and sorting is determined to be qualified.

[0053] S303. If the proportion of the second metal particles is less than the second preset threshold, the second metal particles are determined to be unqualified for crushing and sorting.

[0054] S304. If the proportion of the second metal particles is greater than or equal to the second preset threshold, the second metal particles are determined to be qualified for crushing and sorting.

[0055] In steps S303-S304, the calculated percentage of the second metal particles (aluminum) is compared with the second preset threshold (i.e., 95%). If the percentage of aluminum particles is less than 95%, the aluminum particle crushing and sorting is determined to be unqualified; if the percentage of aluminum particles is greater than or equal to 95%, the aluminum particle crushing and sorting is determined to be qualified.

[0056] In some embodiments, for step S400, one possible implementation provided by the present application is that step S400 includes sub-steps S401 to S406: S401. If the brightness of the first metal particle image is less than the preset brightness threshold, the reason for the failure of the first metal particle to be broken and sorted is determined to be the failure of the black powder sorting.

[0057] S402. If the brightness of the second metal particle image is less than the preset brightness threshold, the reason for the failure of the second metal particle crushing and sorting is determined to be the failure of the black powder sorting.

[0058] In steps S401 to S402, when the brightness value of the first metal particle image is less than the preset brightness threshold, it indicates that there is a lot of black powder in the first metal particle image, and the reason for the failure of the first metal particle crushing and sorting is determined to be the failure of black powder sorting; when the brightness value of the second metal particle image is less than the preset brightness threshold, it indicates that there is a lot of black powder in the second metal particle image, and the reason for the failure of the second metal particle crushing and sorting is determined to be the failure of black powder sorting.

[0059] S403. If the proportion of the first metal particles is less than the first preset threshold, the reason for the failure of the crushing and sorting of the first metal particles is determined to be that the proportion is unqualified.

[0060] S404. If the proportion of the second metal particles is less than the second preset threshold, the reason for the failure of the crushing and sorting of the second metal particles is determined to be that the proportion is unqualified.

[0061] In steps S403 to S404, when the proportion of the first metal particles is less than the first preset threshold (i.e., 98%), the reason for the failure of the crushing and sorting of the first metal particles is determined to be the proportion failure; when the proportion of the second metal particles is less than the second preset threshold (i.e., 95%), the reason for the failure of the crushing and sorting of the second metal particles is determined to be the proportion failure.

[0062] S405. If the particle size of the first metal particle is greater than the preset particle size threshold, the reason for the failure of crushing and sorting of the first metal particle is determined to be that the particle size is unqualified.

[0063] S406. If the particle size of the second metal particle is greater than the preset particle size threshold, the reason for the failure of crushing and sorting of the second metal particle is determined to be that the particle size is unqualified.

[0064] In steps S405-S406, the particle size distribution information of the first metal particle (copper) and the second metal particle (aluminum) is obtained through image processing. When the particle size of the copper particle is greater than the preset particle size threshold, it is determined that the reason for the failure of the crushing and sorting of the copper particle is that the particle size is unqualified. When the particle size of the aluminum particle is greater than the preset particle size threshold, it is determined that the reason for the failure of the crushing and sorting of the aluminum particle is that the particle size is unqualified.

[0065] In some embodiments, the process parameter adjustment model in step S500 includes a regression model. One possible implementation provided in this application embodiment is that the training step of the process parameter adjustment model in step S500 includes sub-steps S501~S503: S501. Obtain the training dataset. The training dataset includes multiple training data and the corresponding real parameter adjustment thresholds. The training data includes the first metal particle image, the second metal particle image, the crushing processing parameters, and the reasons for the failure of crushing and sorting corresponding to each metal particle image.

[0066] In this embodiment, during the system debugging phase, in order to obtain a large dataset for training the regression model, the crushing parameters such as the granulator gap, granulation frequency, airflow separator air volume, vibrating screen frequency, screen aperture, gravity separator frequency, and airflow and pressure were continuously adjusted during the debugging process. At the same time, an industrial camera was used to continuously photograph the output material. After the photographs, the OpenCV algorithm was used to perform statistical calculations on the output material to determine whether the crushing and sorting were qualified. Materials that failed the crushing and sorting were manually marked and the reasons for the failure were recorded. The corresponding crushing parameters were recorded simultaneously. Multiple sets of valid data were collected and calibrated to construct a dataset containing the correspondence between crushing parameters and metal particle sorting quality. Each set of training data includes a first metal particle image, a second metal particle image, crushing parameters, and the reasons for the failure of crushing and sorting corresponding to each metal particle image, and the corresponding real parameter adjustment thresholds are labeled.

[0067] S502. Input the training data into the pre-built process parameter adjustment model to obtain the corresponding predicted parameter adjustment threshold.

[0068] In this embodiment, training data is input into a pre-constructed process parameter adjustment model, also known as a regression model. The regression model processes the input first metal particle image, second metal particle image, crushing parameters, and reasons for unqualified crushing and sorting. Based on the weight values ​​learned by the regression model, it calculates and outputs the corresponding predicted parameter adjustment threshold. The predicted parameter adjustment threshold is the amount of parameter change that the regression model predicts needs to be adjusted based on the input features.

[0069] S503. Calculate the loss function based on the predicted parameter adjustment threshold and the actual parameter adjustment threshold corresponding to the training data, and update the model parameters of the process parameter adjustment model according to the loss function to obtain the trained process parameter adjustment model.

[0070] In this embodiment, the mean squared error loss function is used as the loss function for training the regression model. The predicted parameter adjustment thresholds and the actual parameter adjustment thresholds corresponding to the training data are substituted into the mean squared error loss function for calculation. The mean squared error loss function is used to quantify the deviation between the predicted parameter adjustment thresholds and the actual parameter adjustment thresholds of the regression model. The regression model training process aims to minimize the loss function value. The model parameters (W matrix and C matrix biases) are iteratively updated through backpropagation to gradually improve the prediction accuracy. When the loss function value converges, the regression model training is complete, resulting in a well-trained process parameter adjustment model.

[0071] For example, the trained regression model is applied to production line production, where it is used to make judgments and adjustments based on the learned weight values. In the following formula (4), The x matrix represents the parameters to be adjusted (parameter adjustment threshold) after being judged by the process parameter adjustment model. The W matrix represents the weights learned by the process parameter adjustment model. The U matrix represents the input vector, which is the current crushing and processing parameters. The C matrix represents the bias. It can be understood that W and C are the weight values ​​continuously learned by the process parameter adjustment model, i.e., the model parameters.

[0072] (4) After calculating the parameter adjustment threshold, the adjusted target process parameters are obtained. Target process parameter: x 目标 =x 旧 + x(5).

[0073] In some embodiments, there are two ways to address the reasons for unqualified crushing and sorting in the regression model training data. One method, as mentioned above, involves manually marking and recording the reasons for unqualified crushing and sorting of materials. The other method is to use a classification model. That is, before training the regression model and applying it in practice, the classification model can identify the corresponding reasons for unqualified crushing and sorting in the images. Specifically, the classification model can identify the specific reasons for unqualified crushing and sorting based on the first and second metal particle images. This allows the classification model to replace the manual marking and recording of unqualified crushing and sorting reasons. The unqualified crushing and sorting reasons output by the classification model are then used as training data for the regression model. This allows the regression model to use the unqualified crushing and sorting reasons identified by the classification model and the crushing processing parameters to output true parameters and adjust the threshold.

[0074] For example, the classification model uses the multi-class cross-entropy loss function as the training loss constraint function. The classification loss constraint function is as follows: (3); where N represents the total number of samples in the classification, and N=7 in this scheme; yi is the true label value of the i-th category, which takes the value of 0 or 1; pi is the predicted probability that the image output by the model belongs to the i-th category.

[0075] Specifically, for the training of the classification model, based on the reasons for image failure in sorting and breaking, the breaking processing parameters are divided into seven categories. The network's output is configured with seven neurons, employing the Softmax activation function. The Softmax activation function normalizes the output of each neuron through exponential operations, mapping each output result to a probability range of 0-1. The sum of all probabilities is 1, directly reflecting the confidence level of assigning an image to each reason for failure in sorting and breaking. During the training phase, a cross-entropy loss function is used to calculate the error between the predicted probability of belonging to a particular reason for failure in sorting and breaking ...

[0076] After learning through neural networks, the classification model has a preliminary classification and judgment ability. In actual use, it can identify the output screen in real time. It can skip over qualified cases that it cannot identify. For unqualified material images and parameter inputs, it outputs the corresponding reasons for unqualified crushing and sorting in real time, and completes the collection of training data for the regression model.

[0077] During the system debugging phase, after collecting the training data for the regression model, a mapping table is first manually created. Parameters are adjusted based on this mapping relationship, establishing corresponding parameter tuning ranges for each cause of non-compliance. This reduces the human error rate. Experienced employees then manually adjust the parameters based on their judgment. After adjustment, the above process is repeated for data collection. If a qualified product is produced, the current adjusted parameters are recorded, along with the actual adjustment amount that ultimately restores the product to a qualified state for that cause. This amount serves as the label for that sample, i.e., the true parameter adjustment threshold. This method completes the collection of training data for the regression model.

[0078] Finally, the training data is input into the regression model, and the mean squared error loss function is used to iteratively optimize the weight matrix and bias within the model, enabling the regression model to determine how much adjustment to make for specific reasons for non-compliance and the current parameter state.

[0079] In some embodiments, see Figure 2 , Figure 2 This is a flow chart of a lithium battery crushing and sorting process provided in this application embodiment. In the production of crushed lithium batteries, the black powder from battery crushing that cannot be processed by simple sieving is treated using the above method. The black powder to be treated has a copper content of 35.21%, an aluminum content of 22.51%, and a cobalt + nickel content of 10.17%. The black powder is calcined at 400℃ for 1 hour to remove residual separator debris. Then, the heat-treated material is fed into a three-stage series airflow separator. The pressure difference of the airflow separator is controlled at 1500 Pa, separating the outer shell. Fine black powder is collected by a bag filter. The third-stage airflow separator output enters a granulator for grinding, controlling the particle size to approximately 150 mesh. The ground material then re-enters a two-stage series airflow separator, with the pressure difference controlled at 1100 Pa. The remaining material after the airflow separator is sieved through a 150-mesh circular vibrating screen. The oversize material is returned to the granulator for circulation, while the undersize material is separated into copper and aluminum by a gravity separator. For difficult-to-process materials in gravity separators, visual identification of the output material and statistical analysis of the proportions of copper and aluminum allow for timely adjustments to the sorting frequency, thereby improving the recovery rate of copper and aluminum. Ultimately, this achieves the separation of copper particles, aluminum particles, and black powder. The recovered copper particles contain over 98% copper and <2% cobalt + nickel; the recovered aluminum particles contain approximately 95% aluminum and <3% cobalt + nickel; and the recovered black powder contains <3% copper and <3% aluminum.

[0080] In some embodiments, for the battery breakage sorting in step S100, one possible implementation of the present application embodiment is that the battery breakage sorting step in step S100 includes sub-steps S101~S103: S101. The lithium battery is subjected to crushing and heat treatment to obtain first metal particles, second metal particles and black powder.

[0081] In this embodiment, the lithium battery is first fed into the crushing module for crushing into small pieces. The organic matter is volatilized by drying. After being sorted by the casing, the residual membrane debris on the surface of the material is removed by pyrolysis, causing the membrane to decompose into various organic substances that are volatilized and removed. At the same time, the binders such as PVDF are decomposed, which promotes the complete separation of black powder and metal particles, thereby obtaining a mixture containing copper particles, aluminum particles and black powder.

[0082] S102. The first metal particles, the second metal particles, and the black powder are separated by airflow to obtain the first metal particles and the second metal particles after the black powder is recovered.

[0083] In this embodiment, the heat-treated mixture is fed into an air classifier. The air classifier is set up with multiple stages connected in series, and the pressure difference is controlled within a preset range. The black powder that falls off is uniformly sucked away and collected through air separation. The remaining large pieces of material are mainly copper and aluminum metal particles. Then, after being granulated by a granulator, the copper and aluminum metal particles are further crushed. After being separated by air separation, a mixture of copper particles and aluminum particles after the black powder is recovered is obtained.

[0084] S103. The first and second metal particles after the black powder is recovered are separated by gravity separation to obtain the first and second metal particles that are separated from each other.

[0085] In this embodiment, the mixture of copper and aluminum particles after the recovery of black powder is screened by a circular vibrating screen. The material oversizes is returned to the granulator for circulation, while the material undersizes is separated into copper and aluminum by a gravity separator. The copper-aluminum mixture undersizes is then fed into the gravity separator. Using a blower at the bottom of the gravity separator and the separator's vibration, the material is continuously dispersed. Under gravity, aluminum particles move downwards and are collected from the aluminum outlet, while copper particles move upwards and are collected from the copper outlet, thus obtaining separated copper and aluminum particles.

[0086] Based on the above method, embodiments of this application also provide a system corresponding to the above method, such as... Figure 3 As shown, Figure 3This is a schematic diagram of the functional modules of the battery crushing and sorting system provided in this embodiment. It should be noted that the basic principle and technical effects of the battery crushing and sorting system 1000 provided in this embodiment are the same as those in the above method embodiments. For the sake of brevity, parts not mentioned in this embodiment can be referred to the corresponding content in the method embodiments.

[0087] In this embodiment, the battery crushing and sorting system 1000 includes a preprocessing module 1100, a comparison module 1200, and a processing module 1300.

[0088] The preprocessing module 1100 is used to acquire material images and crushing parameters after the lithium battery has been crushed and sorted. The material images include a first metal particle image and a second metal particle image. It can be understood that the preprocessing module 1100 is used to perform the above-mentioned step S100.

[0089] The comparison module 1200 is used to calculate the proportion of the first metal particle in the first metal particle image and the proportion of the second metal particle in the second metal particle image; based on the comparison result of the proportion of the first metal particle with a first preset threshold and the comparison result of the proportion of the second metal particle with a second preset threshold, it determines whether the first metal particle and the second metal particle are qualified for crushing and sorting; if at least one of the first metal particle and the second metal particle is unqualified for crushing and sorting, it determines the reason for the unqualified crushing and sorting. It can be understood that the comparison module 1200 is used to perform the above steps S200~S400.

[0090] The processing module 1300 is used to input the first metal particle image, the second metal particle image, the crushing processing parameters, and the reasons for unqualified crushing and sorting into a pre-trained process parameter adjustment model to obtain a parameter adjustment threshold; based on the parameter adjustment threshold and the crushing processing parameters, it obtains the target process parameters, which are used to ensure that the first and second metal particles after crushing and sorting of the lithium battery are qualified. It can be understood that the processing module 1300 is used to execute the above steps S500~S600.

[0091] In some embodiments, the first metal particle image and the second metal particle image further include black particles. The comparison module 1200 is used to obtain the brightness values ​​of the first metal particle image and the second metal particle image, and compare the brightness values ​​of the first metal particle image and the second metal particle image with preset brightness thresholds respectively; if the brightness value of the first metal particle image is greater than or equal to the preset brightness threshold, it is determined that the black particles in the first metal particle image are sorted successfully; if the brightness value of the second metal particle image is greater than or equal to the preset brightness threshold, it is determined that the black particles in the second metal particle image are sorted successfully; if the black particles in both the first metal particle image and the second metal particle image are sorted successfully, the proportion of the first metal particles in the first metal particle image and the proportion of the second metal particles in the second metal particle image are calculated. It can be understood that the comparison module 1200 is used to perform the above steps S201~S204.

[0092] In some embodiments, the comparison module 1200 is used to determine that the first metal particle is unqualified for crushing and sorting when the proportion of the first metal particle is less than a first preset threshold; to determine that the first metal particle is qualified for crushing and sorting when the proportion of the first metal particle is greater than or equal to the first preset threshold; to determine that the second metal particle is unqualified for crushing and sorting when the proportion of the second metal particle is less than a second preset threshold; and to determine that the second metal particle is qualified for crushing and sorting when the proportion of the second metal particle is greater than or equal to the second preset threshold. It can be understood that the comparison module 1200 is used to perform the above steps S301 to S304.

[0093] In some embodiments, the comparison module 1200 is used to determine that the reason for the failure of the first metal particle crushing and sorting is failure of black powder sorting when the brightness of the first metal particle image is less than a preset brightness threshold; to determine that the reason for the failure of the second metal particle crushing and sorting is failure of black powder sorting when the brightness of the second metal particle image is less than a preset brightness threshold; to determine that the reason for the failure of the first metal particle crushing and sorting is failure of proportion when the proportion of the first metal particle is less than a first preset threshold; to determine that the reason for the failure of the second metal particle crushing and sorting is failure of proportion when the proportion of the second metal particle is less than a second preset threshold; to determine that the reason for the failure of the first metal particle crushing and sorting is failure of proportion when the particle size of the first metal particle is greater than a preset particle size threshold; and to determine that the reason for the failure of the second metal particle crushing and sorting is failure of particle size when the particle size of the second metal particle is greater than a preset particle size threshold. It can be understood that the comparison module 1200 is used to perform the above steps S401 to S406.

[0094] In some embodiments, the processing module 1300 is used to acquire a training dataset, which includes multiple training data and corresponding real parameter adjustment thresholds. The training data includes a first metal particle image, a second metal particle image, crushing processing parameters, and reasons for crushing and sorting failures corresponding to each metal particle image. The training data is input into a pre-constructed process parameter adjustment model to obtain the corresponding predicted parameter adjustment thresholds. A loss function is calculated based on the predicted parameter adjustment thresholds and real parameter adjustment thresholds corresponding to the training data, and the model parameters of the process parameter adjustment model are updated based on the loss function to obtain a trained process parameter adjustment model. It can be understood that the processing module 1300 is used to perform the above steps S501~S503.

[0095] In some embodiments, the processing module 1300 is used to add the parameter adjustment threshold to the crushing processing parameters to obtain the target process parameters.

[0096] In some embodiments, the pretreatment module 1100 is used to perform a crushing heat treatment on the lithium battery to obtain first metal particles, second metal particles, and black powder; the first metal particles, second metal particles, and black powder are then separated by airflow to obtain first metal particles and second metal particles after black powder recovery; the first metal particles and second metal particles after black powder recovery are then separated by gravity separation to obtain mutually separated first metal particles and second metal particles. It can be understood that the pretreatment module 1100 is used to perform the above steps S101~S103.

[0097] Based on the same inventive concept disclosed above, this application also provides a block diagram of an electronic device 2000 performing the above method. Please refer to... Figure 4 , Figure 4 This is a block diagram of an electronic device provided in an embodiment of this application. The electronic device 2000 includes a processor 2100, a memory 2200, a bus 2300, and a communication interface 2400. The processor 2100 and the memory 2200 are connected via the bus 2300, and the processor 2100 communicates with external devices via the communication interface 2400.

[0098] Processor 2100 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed through integrated logic circuits in the hardware of processor 2100 or through software instructions. The processor 2100 may be a general-purpose processor 2100, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0099] The memory 2200 is used to store computer programs. For example, the battery breakage sorting system 1000 in this embodiment includes at least one software function module that can be stored in the memory 2200 in the form of software or firmware. After receiving the execution instruction, the processor 2100 executes the program to implement the battery breakage sorting method in this embodiment.

[0100] The memory 2200 may include high-speed random access memory (RAM) or non-volatile memory. Optionally, the memory 2200 may be a storage device built into the processor 2100 or a storage device independent of the processor 2100.

[0101] Bus 2300 can be ISA bus 2300, PCI bus 2300 or EISA bus 2300, etc. Figure 4 It is indicated by only one double-headed arrow, but does not mean that there is only one bus 2300 or one type of bus 2300.

[0102] Electronic devices 2000 can be mobile phones, tablets, laptops, desktop computers, and other computer devices.

[0103] Based on the same inventive concept, this application also provides a computer-readable storage medium storing a computer program thereon. When executed by processor 2100, the computer program implements the battery breakage and sorting method described above. This computer-readable storage medium may include various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0104] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for sorting and breaking down batteries, characterized in that, The method includes: Acquire material images and crushing processing parameters after lithium battery crushing and sorting, wherein the material images include a first metal particle image and a second metal particle image; Calculate the proportion of the first metal particle in the first metal particle image and the proportion of the second metal particle in the second metal particle image; Based on the comparison results of the proportion of the first metal particles with the first preset threshold and the comparison results of the proportion of the second metal particles with the second preset threshold, it is determined whether the first metal particles and the second metal particles are qualified for crushing and sorting. If at least one of the first metal particles and the second metal particles fails to be crushed and sorted, the reason for the failure to be crushed and sorted shall be determined. The first metal particle image, the second metal particle image, the crushing processing parameters, and the reasons for the crushing and sorting failures are input into a pre-trained process parameter adjustment model to obtain the parameter adjustment threshold. The target process parameters are obtained by adjusting the threshold and the crushing process parameters according to the parameters. The target process parameters are used to ensure that the first metal particles and the second metal particles after the lithium battery crushing and sorting are qualified.

2. The method according to claim 1, characterized in that, The first and second metal particle images also include black particles. The calculation of the proportion of the first metal particles in the first metal particle image and the proportion of the second metal particles in the second metal particle image includes: Obtain the brightness values ​​of the first metal particle image and the second metal particle image, and compare the brightness values ​​of the first metal particle image and the second metal particle image with a preset brightness threshold respectively; If the brightness value of the first metal particle image is greater than or equal to a preset brightness threshold, the black powder in the first metal particle image is determined to be qualified for sorting. If the brightness value of the second metal particle image is greater than or equal to a preset brightness threshold, the black powder in the second metal particle image is determined to be qualified for sorting. If both the first and second metal particle images are qualified for black powder sorting, calculate the proportion of the first metal particle in the first metal particle image and the proportion of the second metal particle in the second metal particle image.

3. The method according to claim 1, characterized in that, The step of determining whether the first metal particles and the second metal particles are qualified for crushing and sorting based on the comparison result of the proportion of the first metal particles with the first preset threshold and the comparison result of the proportion of the second metal particles with the second preset threshold includes: If the proportion of the first metal particles is less than the first preset threshold, the first metal particles are determined to be unqualified for crushing and sorting. If the proportion of the first metal particles is greater than or equal to the first preset threshold, the first metal particles are determined to be qualified for crushing and sorting. If the proportion of the second metal particles is less than the second preset threshold, the second metal particles are determined to be unqualified for crushing and sorting. If the proportion of the second metal particles is greater than or equal to the second preset threshold, the second metal particles are determined to be qualified for crushing and sorting.

4. The method according to claim 2 or 3, characterized in that, The determination of the reasons for the failure of the crushing and sorting includes: If the brightness of the first metal particle image is less than a preset brightness threshold, the reason for the failure of the first metal particle to be broken and sorted is determined to be the failure of black powder sorting. If the brightness of the second metal particle image is less than a preset brightness threshold, the reason for the failure of the second metal particle crushing and sorting is determined to be the failure of black powder sorting. If the proportion of the first metal particles is less than the first preset threshold, the reason for the failure of the crushing and sorting of the first metal particles is determined to be that the proportion is unqualified. If the proportion of the second metal particles is less than the second preset threshold, the reason for the failure of the crushing and sorting of the second metal particles is determined to be that the proportion is unqualified. If the particle size of the first metal particle is greater than the preset particle size threshold, the reason for the failure of crushing and sorting of the first metal particle is determined to be that the particle size is unqualified. If the particle size of the second metal particle is greater than the preset particle size threshold, the reason for the failure of crushing and sorting of the second metal particle is determined to be that the particle size is unqualified.

5. The method according to claim 1, characterized in that, The process parameter adjustment model is trained through the following steps: Obtain a training dataset, which includes multiple training data and real parameter adjustment thresholds corresponding to the training data. The training data includes a first metal particle image, a second metal particle image, crushing processing parameters, and reasons for unqualified crushing and sorting corresponding to each metal particle image. The training data is input into a pre-built process parameter adjustment model to obtain the corresponding predicted parameter adjustment thresholds; The loss function is calculated based on the predicted parameter adjustment threshold and the actual parameter adjustment threshold corresponding to the training data, and the model parameters of the process parameter adjustment model are updated based on the loss function to obtain the trained process parameter adjustment model.

6. The method according to claim 1, characterized in that, The step of adjusting the threshold and the crushing parameters according to the parameters to obtain the target process parameters includes: The target process parameters are obtained by adding the parameter adjustment threshold to the crushing and processing parameters.

7. The method according to claim 1, characterized in that, The lithium battery crushing and sorting steps include: The lithium battery is subjected to crushing and heat treatment to obtain first metal particles, second metal particles and black powder; The first metal particles, the second metal particles, and the black powder are separated by airflow to obtain the first metal particles and the second metal particles after the black powder has been recovered. The first and second metal particles after the black powder is recovered are separated by gravity separation to obtain mutually separated first and second metal particles.

8. A battery crushing and sorting system, characterized in that, The system includes: The preprocessing module is used to acquire material images and crushing parameters after lithium battery crushing and sorting. The material images include a first metal particle image and a second metal particle image. The comparison module is used to calculate the proportion of the first metal particle in the first metal particle image and the proportion of the second metal particle in the second metal particle image; based on the comparison result of the proportion of the first metal particle with a first preset threshold and the comparison result of the proportion of the second metal particle with a second preset threshold, it determines whether the first metal particle and the second metal particle are qualified for crushing and sorting; if at least one of the first metal particle and the second metal particle is unqualified for crushing and sorting, it determines the reason for the unqualified crushing and sorting. The processing module is used to input the first metal particle image, the second metal particle image, the crushing processing parameters, and the reasons for the crushing and sorting failure into a pre-trained process parameter adjustment model to obtain a parameter adjustment threshold; and to obtain a target process parameter based on the parameter adjustment threshold and the crushing processing parameters, wherein the target process parameter is used to ensure that the first metal particle and the second metal particle after the lithium battery crushing and sorting are qualified.

9. An electronic device, characterized in that, It includes a processor and a memory, the memory storing a computer program that can be executed by the processor to implement the battery breakage and sorting method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the battery breakage sorting method as described in any one of claims 1-7.

Images