Recycling and treatment method based on lithium battery powder

Abstract

The present application provides a recycling and treatment method based on lithium-extracted battery powder, which relates to the field of battery recycling. The method includes the following steps: collecting historical recycling data to construct a first neural network model; performing chemical and physical analyses on the lithium-extracted waste battery powder, and inputting the analysis results into the first neural network model to generate optimal processing parameters; adjusting the processing method based on the optimal processing parameters to maximize the resource recycling efficiency. In addition, the method also includes determining the key chemical elements in the battery powder, evaluating the metal content and composition, and optimizing the model prediction accuracy through feature engineering. This method can effectively improve the resource recycling efficiency and reduce the environmental impact, providing an efficient and intelligent solution for the battery recycling field at a time when the battery design and materials are becoming increasingly complex with the rapid development of the electric vehicle and energy storage industries.

Description

Technical Field

[0001] This application relates to the field of battery recycling, and particularly to a recycling and treatment method for lithium battery powder extraction. Background Art

[0002] With the rapid development of the electric vehicle and energy storage industries, the design and materials of lithium batteries are becoming increasingly complex, and various metals and composite materials are used in common batteries. These emerging batteries contain different types of cathode materials and more complex electrolyte components, making recycling and treatment more difficult. However, traditional lithium battery powder recycling methods, including techniques such as oxygen pressure acid leaching, two-stage impurity removal, and extraction, are difficult to cope with this increasingly complex battery structure and materials, resulting in low metal recovery rates and increased pollution risks. The industry needs more intelligent solutions.

[0003] For example, a Chinese patent with the authorization announcement number CN107591584B discloses a method for recycling and utilization of waste lithium-ion battery cathode powder, including the following steps: oxygen pressure acid leaching, two-stage impurity removal, extraction impurity removal, alkaline sedimentation, and evaporation crystallization, finally obtaining high-purity lithium sulfate. This invention considers the metal content and recycling value of each component, and separates and prepares ternary precursor raw materials and high-purity lithium sulfate from waste lithium-ion battery cathode powder. However, it still has a narrow scope of application. In emerging batteries, there are different types of cathode materials and more complex electrolyte components, which makes recycling and treatment more difficult. Such traditional recycling methods are difficult to fully utilize the valuable metal resources in the batteries.

[0004] Therefore, it is necessary to develop an intelligent analysis method based on a neural network model. By analyzing the chemical and physical data of battery powder, it can effectively cope with the diversity of complex materials, accurately predict the optimal recycling parameters, improve the metal recovery efficiency, reduce the environmental impact, and provide a more intelligent and efficient solution for the recycling of complex lithium battery powder. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, this application proposes a recycling and treatment method for lithium battery powder extraction, which is used to construct a neural network model for characterizing the recycling parameters of lithium battery powder extraction based on the historical recycling data of lithium battery powder extraction. After chemically analyzing and physically analyzing the extracted lithium battery powder, the neural network model is used to generate the optimal processing parameters, and the recycling and treatment method and parameters of the current lithium battery powder extraction are adjusted based on the optimal processing parameters to optimize the recycling efficiency of valuable metals, plastics, and organic materials in waste battery powder.

[0006] The method includes:

[0007] Obtaining historical recycling data; constructing a first neural network model based on the historical recycling data;

[0008] Collect the waste battery powder from which lithium has been extracted, and conduct chemical analysis and physical analysis on the waste battery powder from which lithium has been extracted to generate a first analysis result; input the first analysis result into the first neural network model to generate optimal processing parameters;

[0009] Based on the optimal processing parameters, adjust the processing parameters of the waste battery powder from which lithium has been extracted, and recover valuable metals, plastics, and organic materials in the waste battery powder from which lithium has been extracted based on the adjusted processing parameters.

[0010] As a possible implementation manner, the historical recovery data includes: physical information of the battery powder, chemical information of the battery powder, processing parameter information, and recovery efficiency information.

[0011] As a possible implementation manner, the processing parameters include at least one of the following: processing method, temperature, pressure, pH value, reaction time, type and dosage of additives.

[0012] As a possible implementation manner, constructing the first neural network model based on the historical recovery data includes:

[0013] Preprocess the historical recovery data;

[0014] Based on the preprocessed historical recovery data, perform feature engineering to generate a first target feature;

[0015] Construct the first neural network model based on the first target feature and the processing parameters.

[0016] As a possible implementation manner, when performing feature engineering based on the preprocessed historical recovery data to generate a first target feature, the following steps are performed:

[0017] Based on the historical recovery data, determine the key chemical elements in the battery powder;

[0018] Based on the key chemical elements in the battery powder, create a metal content index to determine the importance and potential economic value of each batch of waste battery powder from which lithium has been extracted, and determine the recovery priority and resource allocation for each batch of waste battery powder from which lithium has been extracted. The mathematical expression is:

[0019]

[0020] Among them, MCI is the metal content index, n is the type of key chemical elements, and the content of each element i in the waste battery powder from which lithium has been extracted is c i , ω i is the weight of each element, and the weights of each chemical component in the metal content index are preset according to the contribution and value of each metal in the battery performance;

[0021] Calculate the chemical combination ratios of each key chemical element in the battery powder, determine the relative content relationship between different chemical components, and the mathematical expression for determining the chemical combination ratio is:

[0022]

[0023] Among them, Ratio a,b represents the combination ratio of chemical component a and chemical component b; c a is the content of chemical component a in the battery powder, and c b is the content of chemical component b in the battery powder;

[0024] Introduce the density-particle size index characterizing the physical characteristics, and determine the influence of the density-particle size index on the recovery efficiency;

[0025] Index D,S = D × S

[0026] Among them, Index D,S is the density-particle size index; D is the density of the battery powder; S is the particle size of the battery powder;

[0027] Introduce the square term of the content of the key chemical element in the battery powder and the cubic term of the content of the key chemical element in the battery powder as polynomials to determine the non-linear influence of the change in the content of the key chemical element in the battery powder on the recovery effect;

[0028] Introduce the temperature-pressure interaction term to determine the influence of temperature and pressure on the recovery efficiency. The mathematical expression is:

[0029] Interaction T,P = T × P

[0030] Among them, Interaction T,P is the temperature-pressure interaction term; T is the temperature value during the treatment of the battery powder; P is the pressure value during the treatment of the battery powder;

[0031] Introduce the pH-reaction time interaction term to determine the relationship between the pH value and the reaction time. The mathematical expression is:

[0032] Interaction pH,t = pH × t

[0033] Among them, pH represents the acidity and alkalinity of the solution, t represents the reaction time, and Interaction pH,t is the pH-reaction time interaction term;

[0034] Based on the above feature engineering steps, generate the first target feature; wherein, the target first feature includes at least one of the following: metal content index, various chemical composition ratios, density-particle size index, polynomial established from the content of key chemical elements, temperature-pressure interaction term, pH-value reaction time interaction term.

[0035] As a possible implementation manner, the first neural network model includes: an input layer, a first hidden layer, a batch normalization layer, a second hidden layer, a dropout layer, and an output layer; constructing the first neural network model based on the first target feature and the processing parameters includes:

[0036] Based on the first target feature, set the number of neurons in the input layer, the first hidden layer, and the second hidden layer in the first neural network model.

[0037] Based on the processing parameters, set the number of neurons in the output layer in the first neural network model.

[0038] As a possible implementation manner, the first analysis result includes: element content, chemical state, electrochemical properties, particle size distribution, morphology and surface structure, thermal stability and organic material content, crystal structure, potential recycling value, and environmental and safety assessment; collecting the lithium-extracted waste battery powder and performing chemical analysis and physical analysis on the lithium-extracted waste battery powder to generate the first analysis result includes at least one of the following steps:

[0039] Use X-ray fluorescence spectrometry to determine the content of various elements in the battery powder.

[0040] Accurately measure the concentration of specific metal elements by atomic absorption spectrometry.

[0041] Apply inductively coupled plasma mass spectrometry to detect trace and ultra-trace elements.

[0042] Perform electrochemical analysis to evaluate the electrochemical performance of the residual chemical substances in the battery powder.

[0043] Determine the particle size distribution of the battery powder by particle size analysis.

[0044] Use a scanning electron microscope to observe the surface morphology and microstructure of the battery powder.

[0045] Perform thermogravimetric analysis to evaluate the thermal stability and organic material content of the material.

[0046] Use X-ray diffraction analysis to identify the mineral phase and crystalline phase in the battery powder.

[0047] As a possible implementation, the treatment methods include: chemical treatment method, physical treatment method, and comprehensive treatment method; based on the optimal treatment parameters, adjust the treatment parameters of the spent battery powder after lithium extraction, and recycle the valuable metals, plastics, and organic materials in the spent battery powder after lithium extraction based on the adjusted treatment parameters, including:

[0048] Based on the optimal treatment parameters, adjust the treatment parameters for the current spent battery powder after lithium extraction;

[0049] For the current spent battery powder after lithium extraction, based on the adjusted treatment parameters, recover valuable metals, plastics, and organic materials from the current spent battery powder after lithium extraction, and monitor the extraction effect;

[0050] Record all the above adjustment data and recovery result data, and update the adjustment data and the recovery result data to the historical recovery data.

[0051] As a possible implementation, the adjusting of the treatment parameters for the current spent battery powder after lithium extraction based on the optimal treatment parameters includes:

[0052] In response to determining that the treatment method for the current spent battery powder after lithium extraction is the chemical treatment method, when adjusting the treatment parameters of the battery powder, perform the following steps:

[0053] Based on the optimal treatment parameters, adjust the temperature setting in the reactor;

[0054] Based on the optimal treatment parameters, use an acid or a base to adjust the pH value of the solution to the recommended pH range;

[0055] Based on the optimal treatment parameters, set the reaction time of the reactor;

[0056] Based on the optimal treatment parameters, add chemical additives of specified types and dosages;

[0057] In response to determining that the treatment method for the current spent battery powder after lithium extraction is the physical treatment method, when adjusting the treatment parameters of the battery powder, perform the following steps:

[0058] Based on the optimal treatment parameters, select a separation technology; wherein, the separation technology includes: screening, magnetic separation, centrifugal separation, and flotation;

[0059] Based on the optimal treatment parameters, adjust the temperature and pressure settings of the separator;

[0060] Based on the optimal treatment parameters, adjust the equipment operation parameters.

[0061] Compared with the prior art, the beneficial effects of the present application are as follows: This method uses a neural network model to analyze the chemical and physical data in waste battery powder, which not only improves the accuracy of the processing parameters during the recycling of lithium battery powder, but also optimizes the recycling process, and can effectively cope with the complexity and diversity of battery materials. In addition, this method also introduces various relevant features in the process of recycling lithium battery powder through feature engineering, optimizes the model prediction accuracy, further improves the efficiency and economy of the recycling process, and provides a more advanced technical path for the battery recycling field, especially in the context of the rapid development of the electric vehicle and energy storage industries. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] To more clearly illustrate the technical solutions of the embodiments of the present application, the accompanying drawings required for use in the embodiments will be briefly introduced below. The accompanying drawings herein are incorporated into the specification and constitute a part of this specification. These drawings show the embodiments that conform to the present application and are used together with the specification to illustrate the technical solutions of the present application. It should be understood that the following drawings only show some embodiments of the present application and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without creative efforts.

[0063] Figure 1 It is a flowchart of a recycling and treatment method based on lithium battery powder provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0064] In order to make the objectives, technical solutions, and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings. The specific operation methods in the method embodiments can also be applied to the device embodiments or system embodiments. It should be noted that in the description of the present application, "at least one" means one or more, and "a plurality" means two or more. In view of this, "a plurality" in the embodiments of the present application can also be understood as "at least two". "And / or" describes the association relationship of associated objects and indicates that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " generally represents an "or" relationship between the associated objects before and after, unless otherwise specified. In addition, it should be understood that in the description of the present application, terms such as "first" and "second" are only used for the purpose of distinguishing descriptions and cannot be understood as indicating or implying relative importance, nor can they be understood as indicating or implying order.

[0065] Refer to Figure 1 as shown in Figure 1 It is a flowchart of a recycling and treatment method based on lithium battery powder provided by an embodiment of the present application. The method includes steps S101 to S103, where:

[0066] S101: Obtain historical recycling data; based on the historical recycling data, construct a first neural network model;

[0067] S102: Collect the spent battery powder after lithium extraction, and perform chemical analysis and physical analysis on the spent battery powder after lithium extraction to generate a first analysis result; input the first analysis result into the first neural network model to generate optimal processing parameters;

[0068] S103: Based on the optimal processing parameters, adjust the processing parameters of the spent battery powder after lithium extraction, and recycle valuable metals, plastics, and organic materials in the spent battery powder after lithium extraction based on the adjusted processing parameters.

[0069] In the embodiments of the present disclosure: After obtaining the historical recycling data of the lithium-extracted battery powder, based on the historical recycling data of the lithium-extracted battery powder, construct a first neural network model for characterizing the recycling parameters of the lithium-extracted battery powder; after performing physical and chemical analysis on the currently collected spent battery powder after lithium extraction, input the analysis result into the first neural network model to generate optimal processing parameters; after obtaining the optimal processing parameters, the processing parameters of the currently spent battery powder after lithium extraction can be adjusted with reference to the optimal processing parameters to optimize the recycling efficiency of valuable metals, plastics, and organic materials in the spent battery powder.

[0070] The following will separately describe the above S101 to S103.

[0071] Regarding the above S101: The historical recycling data can be collected from different battery recycling manufacturers, laboratories, or institutions. This data may include information on battery powder samples of different batches.

[0072] As a possible implementation manner, the historical recycling data includes: physical information of the battery powder, chemical information of the battery powder, processing parameter information, and recycling efficiency information.

[0073] In specific implementation, the historical recycling data may include but is not limited to the following: Physical information of the battery powder: particle size, density, etc.; Chemical information of the battery powder: content of various metals and elements, composition of compounds, etc.; Recycling processing information: processing method, processing parameters, temperature, pressure, time, etc.; Recycling effect information: metal recovery rate, recovery purity, output, etc.

[0074] As a possible implementation manner, when constructing the first neural network model based on the historical recycling data, the following steps 11 to 13 can be executed:

[0075] Step 11: Preprocess the historical recycling data;

[0076] Step 12: Based on the preprocessed historical recycling data, perform feature engineering to generate the first target feature;

[0077] Step 13: Based on the first target feature and the processing parameters, construct the first neural network model;

[0078] In a specific implementation:

[0079] For Step 11, historical data of waste lithium battery powder can be obtained from battery recycling enterprises and research institutions. And according to specific recycling needs, determine the required data types, such as: physical information of the battery powder: density, particle size distribution of the battery powder; chemical information of the battery powder: contents of metals such as lithium, cobalt, nickel, copper, aluminum, manganese, etc.; processing parameters during the battery powder recycling process: processing temperature, pressure, processing time, type and dosage of additives; recycling effect: metal recovery rate, purity, etc. After determining the required data types, data cleaning can be carried out, deleting records with missing key fields such as the density and metal content of the battery powder and dealing with outliers. For example, if the cobalt content in a record is negative, this abnormal record can be deleted or replaced with a reasonable value. Convert all data into the same format (such as a CSV file) and ensure that the fields are consistent. Convert numerical data such as temperature and pressure into floating-point numbers, and convert metal types into integers or categories. Normalize numerical features such as temperature, pressure, and time to the range of 0 - 1, and standardize the metal content so that its mean is 0 and the standard deviation is 1.

[0080] Among them, the main reason for converting numerical data such as temperature and pressure into floating-point numbers is that physical quantities such as temperature and pressure often involve decimals. Using floating-point numbers can ensure that these subtle changes will not be lost during integer rounding, thus ensuring the accuracy and reliability of the data.

[0081] For Step 12, as a possible implementation manner, when performing feature engineering based on the preprocessed historical recycling data to generate the first target feature, the following steps 121 to 128 can be executed:

[0082] Step 121: Based on the historical recycling data, determine the key chemical elements in the battery powder;

[0083] In a specific implementation, to determine the key chemical elements in the battery powder, it can be based on historical recycling data and experience and analysis during the recycling process.

[0084] Exemplarily, the correlation between the content of each element and the recycling effect can be analyzed, and the relationship between each element and the target variable (such as the recovery rate, purity) can be quantified by calculating the correlation coefficient (such as the Pearson correlation coefficient). Elements with a stronger correlation can be considered more critical to the recycling process; it is also possible to determine which elements have higher economic value based on the metal market value. Generally, metals with high value are key elements in battery powder recycling; it is also possible to check the occurrence frequency and content of each element in historical data. If certain elements appear in most samples and have a high content, they may be key elements that need to be focused on during the recycling process; if certain elements require special attention due to their impact on the environment and safety. For example, waste batteries containing heavy metals may require special treatment to prevent environmental pollution and harm to health. By using the above methods, the most valuable and important elements for the recycling process can be identified from historical data and used as the key for subsequent analysis and feature engineering.

[0085] Step 122: Based on the key chemical elements in the battery powder, create a metal content index to determine the importance and potential economic value of each batch of lithium-extracted waste battery powder, and determine the recycling priority and resource allocation for each batch of lithium-extracted waste battery powder. The mathematical expression is:

[0086] MCI = ∑ i n ＝1 ω i ×c i

[0087] where MCI is the metal content index, n is the type of key chemical elements, the content of each element i in the lithium-extracted waste battery powder is c i ,ω i is the weight of each element. The weights of each chemical component in the metal content index are preset according to the contribution and value of each metal in the battery performance.

[0088] In a specific implementation, based on the key chemical elements in the battery powder, a metal content index reflecting the contribution or value of each chemical component in the battery performance is constructed. This index is achieved by assigning different weights to different chemical components, and the determination of the weights can be based on the market value of each metal and its role in the battery.

[0089] Exemplarily, the economic value and recycling priority of the batch of extracted lithium battery powder are evaluated by calculating the Metal Content Index (MCI). Based on the chemical analysis results, the remaining key metal elements in the battery powder are listed, such as cobalt (Co), nickel (Ni), copper (Cu), aluminum (Al), etc.; weights are set for each element according to its market value, contribution to recycling economy, and importance in the battery; according to the chemical analysis results of the battery powder, the content of each key metal element (percentage or mass per unit weight) is determined.

[0090] For example, the extracted lithium battery powder contains three main metals: cobalt (Co), nickel (Ni), and copper (Cu), and their corresponding weights and contents are as follows:

[0091] Cobalt: weight ω Co = 0.5, content c Co = 4%;

[0092] Nickel: weight ω Ni = 0.3, content c Ni = 6%;

[0093] Copper: weight ω Cu = 0.2, content c Cu = 5%;

[0094] Substitute into the above formula for calculating the Metal Content Index MCI:

[0095] MCI = (0.5 × 4) + (0.3 × 6) + (0.2 × 5) = 2 + 1.8 + 1 = 4.8

[0096] The above is the Metal Content Index of an exemplary batch of extracted lithium battery powder. In specific implementation, the Metal Content Indexes of multiple batches of extracted lithium battery powder can be compared to determine the importance of each batch of extracted lithium battery powder and decide the strategy of priority recycling and resource allocation.

[0097] Step 123: Calculate the chemical combination ratios of the key chemical elements in the battery powder, determine the relative content relationship between different chemical components. The mathematical expression for determining the chemical combination ratio is:

[0098]

[0099] where Ratio a,b represents the combination ratio of chemical component a and chemical component b; c a is the content of chemical component a in the battery powder, and c b is the content of chemical component b in the battery powder;

[0100] In a specific implementation, the main chemical elements in the battery powder, such as cobalt, nickel, copper, and aluminum, are obtained from chemical analysis. The content of each element in the battery powder is determined. For example, methods such as X-ray Fluorescence Spectrometry (XRF) or Inductively Coupled Plasma Mass Spectrometry (ICP-MS) can be used. Assume that the following element contents are measured by the above methods:

[0101] Cobalt (Co): c Co = 3%

[0102] Nickel (Ni): c Ni = 5%

[0103] Copper (Cu): c Cu = 2%

[0104] Aluminum (Al): c AI = 4%

[0105] Calculate the combination ratio: For example, calculate the combination ratio of cobalt and nickel:

[0106]

[0107] Calculate the combination ratio of nickel and copper:

[0108]

[0109] The combination ratio (e.g., the ratio of nickel to cobalt) can help predict the metal recovery efficiency more accurately mainly for the following reasons:

[0110] As a combined feature, the combination ratio directly represents the relative content relationship between two elements in a sample. The ratio between metals can provide insights into the interrelationship between metals. For example, during the recycling process, the extraction efficiencies of nickel and cobalt may affect each other, and the model can capture this interrelationship through the ratio, thereby improving the prediction accuracy. The combination ratio can help simplify the model by combining two or more features into one, thus reducing the number of input features. This helps prevent overfitting of the model and also improves the training efficiency and prediction speed of the model. When analyzing the chemical element combinations in battery powder, the combination ratio can identify which metal combinations are more valuable. For example, the ratio of nickel to cobalt can show the abundance of nickel relative to cobalt in the battery powder. If the ratio indicates that the nickel content is much higher than that of cobalt, it means that preferentially extracting nickel would be more economical and efficient because cobalt is scarcer and more valuable. Different metal combination ratios may require different separation techniques. For example, if the ratio of copper to aluminum in the battery powder is high, the separation process may need to pay more attention to the extraction of copper. However, if the aluminum content is higher, techniques targeting aluminum are more suitable. This can help select the appropriate separation techniques and processes, reduce resource waste during the recycling process, and improve efficiency.

[0111] Exemplarily, a high ratio of copper (Cu) to aluminum (Al) in the battery powder means that the copper content is much higher than that of aluminum. Due to the high surface activity of copper, the flotation method can use chemical agents to make copper particles float, while aluminum is more difficult to be floated. This difference makes the flotation method an effective method for separating copper from battery powder with a high ratio of copper. If the ratio of copper to aluminum is high, the electrolysis method can also be used to purify copper because the electrochemical behavior of copper in the electrolyte solution is different from that of aluminum. Copper can be deposited by electrolytic deposition, while aluminum remains in the solution.

[0112] Exemplarily, if the aluminum content in the battery powder is much higher than that of copper. If the ratio of aluminum is high, and the copper content is low, and there are other magnetic metals such as iron, magnetic separation technology can be used to remove these impurity metals. Although copper is not a magnetic metal, magnetic separation can be used as a preliminary treatment means. In the case of a high aluminum content, the battery powder can be treated with acid or alkali to dissolve aluminum without affecting copper. Sodium hydroxide (alkali) or dilute acid can effectively dissolve aluminum and separate it from copper.

[0113] When constructing a machine learning model for predicting metal recycling efficiency, the combination ratio can be used as a useful feature to capture the potential impact of specific element combinations on the recycling efficiency. For example, the ratio of nickel to cobalt as a feature input into the model can more accurately predict which recycling method can maximize the efficiency.

[0114] Step 124: Introduce the density-particle size index characterizing physical features and determine the influence of the density-particle size index on the recycling efficiency;

[0115] IndexD,S = D × S

[0116] where Index D,S is the density - particle size index; D is the density of the battery powder; S is the particle size of the battery powder;

[0117] In a specific implementation, the density - particle size index Index of the battery powder D,S can help the neural network model better understand the behavior of the battery powder during the recycling process and its impact on the recycling efficiency.

[0118] Exemplarily, in the case of high density - small particle size (high Index D,S ), the density of the battery powder is high and the particle size is small, resulting in a high Index D,S value. A high density - particle size index means that the battery powder particles are heavy and small in size, so they settle faster in the liquid medium. This speeds up the filtration process in the chemical recycling process. The small particle size increases the specific surface area of the powder, making its surface activity high in chemical reactions, which is conducive to fully reacting with the recycling reagent, thus improving the recycling efficiency. Battery powder with a high Index D,S settles quickly in the liquid phase and can be processed using techniques such as centrifugal separation or filtration separation.

[0119] Exemplarily, in the case of low density - large particle size (low Index D,S ), the density of the battery powder is low and the particle size is large, resulting in a low Index D,S value. A low density - particle size index causes the battery powder particles to float or suspend in the liquid, especially when the processing medium is water. Low - density particles are more difficult to separate from metals and require more complex physical or chemical separation methods. For example, battery powder with a low Index D,S is suitable for separation by flotation because they are more likely to float on the surface. By analyzing the density - particle size index of the battery powder, the impact of its physical properties on the recycling efficiency can be better understood, and the most appropriate treatment method can be selected.

[0120] Step 125: Introduce the square term and the cube term of the content of the key chemical elements in the battery powder as polynomials to determine the non - linear impact of the change in the content of the key chemical elements in the battery powder on the recycling effect;

[0121] In a specific implementation, chemical elements at different concentrations may affect the recycling process in different ways. A simple linear model cannot capture this non - linear relationship, while the square and cube terms can effectively express this.

[0122] Exemplarily, when recovering copper and cobalt, an increase in the copper content may promote the extraction efficiency within a certain range, but if the content is too high, it may instead lead to a decrease in efficiency due to complexing agent saturation. By introducing the square and cubic terms of the copper content, the model can accurately predict this non-linear effect. In addition, during the actual recovery process, there may be interactions between multiple chemical elements, and this relationship is not always a simple linear one. The polynomial form of the features can capture these complex interactions.

[0123] Among them, complexing agent saturation and competitive complexation are concepts related to chemical complexation reactions. Complexing agent saturation refers to the state in a solution where the ability of the complexing agent to bind with metal ions reaches its upper limit. At this time, even if more metal ions are added, no new complexes will be formed with the complexing agent because all the complexing agents have been occupied by metal ions. In the case of complexing agent saturation, the excess metal ions cannot be further bound by the complexing agent, which will reduce the efficiency of recovery or separation; Competitive complexation refers to the process in a solution where multiple different metal ions compete for binding with the complexing agent. Different metal ions have different affinities for the complexing agent. Therefore, in the case of limited complexing agent, the metal ions with stronger affinity will preferentially form complexes. If there are two metal ions, copper and nickel, in the solution, when a complexing agent is added, they will compete for binding with the complexing agent. If the complexing agent binds more easily with copper, then the complexing agent in the solution will preferentially form complexes with copper ions, while nickel ions will be restricted by the competition and form fewer complexes. Competitive complexation can significantly affect the selectivity and efficiency of metal recovery. To improve the recovery efficiency of a certain metal, it may be necessary to use a specific complexing agent, or to change the affinity between metal ions and the complexing agent by controlling conditions such as the pH value and temperature of the solution.

[0124] Exemplarily, when extracting nickel and cobalt, if the contents of both nickel and cobalt are low, the extraction efficiency is usually high; but if the nickel content is very high and the cobalt content is also very high, the extraction efficiency in this case may decrease due to competitive complexation. This interaction may be manifested by introducing the square or cubic terms of the nickel and cobalt contents.

[0125] Exemplarily, when preprocessing and separating battery powder, the relationship between the contents of cobalt and copper and the recovery efficiency may not be linear. For example, after the copper content reaches a certain concentration, its influence on the recovery efficiency begins to show a decreasing trend. By introducing the square term or cubic term of the copper content into the model, the actual situation can be better predicted.

[0126] In specific implementation, in waste battery powder, the content of cobalt affects the extraction efficiency of metals, but this relationship is usually not linear. For this reason, a polynomial can be introduced as an input feature. Exemplarily, the polynomial is:

[0127] E = α×c Co +β×cCo 2 +γ×c Co 3

[0128] Among them, E represents the recovery efficiency of cobalt metal, which is the target variable predicted by the model; c Co represents the cobalt content in the waste battery powder; α is the coefficient of the linear term, indicating the influence of the linear change in cobalt content on the recovery efficiency. For example, if α is positive, it means that an increase in cobalt content will linearly increase the recovery efficiency; β is the coefficient of the quadratic term, indicating the influence of the quadratic term of cobalt content on the recovery efficiency. If β is positive, it shows that the recovery efficiency increases quadratically with the increase of cobalt content; if β is negative, it means that the recovery efficiency may decrease when the cobalt content is relatively high. γ is the coefficient of the cubic term, representing the influence of the cubic term of cobalt content on the recovery efficiency, capturing more complex non-linear relationships. For example, if γ is positive, it means that the efficiency may increase after the cobalt content continues to increase, but if it is negative, it may decrease.

[0129] Exemplarily, if the cobalt content is 5%, 10%, 15%, 20%, 25%, and the recovery efficiency is 70%, 75%, 80%, 77%, 72%. Using the polynomial model to fit these data, it can be found that: when the cobalt content increases from 5% to 15%, the recovery efficiency increases (mainly linear relationship); when the cobalt content continues to increase to 20%, the recovery efficiency decreases (influence of the quadratic term); when the cobalt content increases to 25%, the recovery efficiency further decreases (influence of the cubic term).

[0130] By introducing the quadratic term and cubic term of cobalt content into the model, the complex non-linear relationship between cobalt content and recovery efficiency can be explained. This non-linear relationship stems from factors such as the interaction between cobalt and other metals, the saturation of complexing agents, and the pH value of the solution. By introducing the polynomial model, this complex relationship can be more accurately described and guide the actual recovery treatment parameters.

[0131] Step 126: Introduce the temperature-pressure interaction term to determine the influence of temperature and pressure on the recovery efficiency. The mathematical expression is:

[0132] Interaction T,P =T×P

[0133] Among them, Interaction T,P is the temperature-pressure interaction term; T is the temperature value during the treatment of the battery powder; P is the pressure value during the treatment of the battery powder;

[0134] In the specific implementation, the reason for introducing the temperature-pressure interaction term in the embodiment of the present application is that the influences of the two parameters of temperature and pressure on the recovery efficiency are often not independent but interact with each other.

[0135] Exemplarily, during the process of extracting cobalt from waste battery powder, chemical treatment such as leaching or dissolution is usually required under specific temperature and pressure. Adjusting the temperature and pressure can optimize the recovery efficiency of cobalt. At higher temperatures, the solubility of cobalt increases and the leaching reaction is faster. Therefore, increasing the temperature generally can improve the extraction efficiency of cobalt. Increasing the pressure can accelerate the reaction rate and increase the solubility of chemicals, thus extracting metals more effectively. When both the temperature and pressure increase, they have a combined promoting effect on the reaction rate and the solubility of cobalt. Therefore, the interaction effect between temperature and pressure can amplify or reduce the impact of individual temperature or pressure on the recovery efficiency. For example: at a lower temperature (60°C) and a higher pressure (10 atm), the recovery efficiency of cobalt may be 60%. At a higher temperature (80°C) and a higher pressure (10 atm), the recovery efficiency of cobalt may increase to 90%, showing the interaction between temperature and pressure.

[0136] Exemplarily, in the process of battery powder recovery treatment, considering the interaction between temperature and pressure is because these two parameters jointly affect the recovery efficiency and safety. In the basic setting, if the theoretical temperature is 95°C and the theoretical pressure is 1.5 atm, when chemically extracting element C from waste battery powder at this time, the interaction between temperature and pressure on solubility and reaction rate needs to be considered.

[0137] For example, in this basic setting, although the dissolution rate of element C is good, the high temperature causes the system pressure to rise to an unsafe level. To reduce the safety risk without sacrificing the extraction efficiency of element C, it is decided to lower the temperature and slightly increase the pressure at the same time. The temperature can be lowered to 90°C and the pressure increased to 1.7 atm.

[0138] Among them, lowering the temperature helps to reduce the overall pressure of the system, making the treatment process safer, and by appropriately increasing the pressure, the decrease in dissolution rate that may be caused by the temperature reduction can be compensated, thus maintaining the extraction efficiency of element C.

[0139] By introducing the temperature-pressure interaction term, the neural network model can better understand the combined impact of temperature and pressure on the recovery efficiency, and this relationship can be used to adjust the processing parameters, thereby optimizing the recovery efficiency of cobalt.

[0140] Step 127: Introduce the pH-reaction time interaction term to determine the relationship between pH value and reaction time. The mathematical expression is:

[0141] Interaction pH,t = pH × t

[0142] Among them, pH represents the acidity and alkalinity of the solution, t represents the reaction time, and Interaction pH,tis the interaction term of pH value and reaction time;

[0143] In a specific implementation, the purpose of this interaction term is to capture the combined effect of the two variables of pH value and reaction time and explain the relationship between them.

[0144] Exemplarily, during the process of extracting nickel from waste battery powder, the acidity (pH) and reaction time (t) both affect the dissolution rate of nickel and the final recovery efficiency. Generally, the lower the pH value, the more acidic the solution, and the faster the dissolution rate of nickel. However, a long-term acidic environment may cause excessive dissolution of impurities as well.

[0145] At a lower pH value (more acidic), the acidity in the solution is stronger and the dissolution reaction is faster, which helps to accelerate the nickel extraction process. But if the pH value is too low and the acidity is too strong, it may affect the recovery of other metals. In addition, the longer the reaction time, the more sufficient the time for dissolution and extraction, so the recovery efficiency of nickel can be improved within an appropriate time range. But if the time is too long, the extraction efficiency of nickel will decrease instead. The combination of pH value and reaction time has an important impact on the extraction efficiency of nickel. For example: at a lower pH value (pH = 2) and a short reaction time (30 minutes), the extraction efficiency of nickel may be relatively low (about 60%); at a lower pH value (pH = 2) and a longer reaction time (2 hours), the extraction efficiency of nickel may increase to 80%, indicating the combined effect of acidity and time; at a higher pH value (pH = 5) and a longer reaction time (2 hours), the extraction efficiency of nickel may decrease again, showing that a higher pH value is not conducive to long-term extraction.

[0146] By introducing the interaction term of pH and reaction time, the comprehensive influence of them on the nickel extraction efficiency can be better understood, so that the process parameters can be adjusted to achieve the efficient recovery of nickel.

[0147] Step 128: Based on the above feature engineering steps, generate the first target feature; wherein, the target first feature includes at least one of the following: metal content index, various chemical composition ratios, density - particle size index, polynomial established by key chemical element contents, temperature - pressure interaction term, pH value - reaction time interaction term;

[0148] Regarding step 13, as a possible implementation manner, the processing parameters include at least one of the following: processing method, temperature, pressure, pH value, reaction time, type and dosage of additives.

[0149] As a possible implementation manner, the first neural network model includes: an input layer, a first hidden layer, a batch normalization layer, a second hidden layer, a dropout layer, and an output layer; when constructing the first neural network model based on the first target feature and the processing parameters, the following steps 131 to 132 can be executed:

[0150] Step 131: Based on the first target feature, set the number of neurons in the input layer, the first hidden layer, and the second hidden layer of the first neural network model;

[0151] Step 132: Based on the processing parameter, set the number of neurons in the output layer of the first neural network model.

[0152] In a specific implementation:

[0153] Regarding the above Step 131: The number of neurons in the input layer of the first neural network model is equal to the number of input features. Exemplarily, if there are 10 input features, the number of neurons in the input layer is 10. Among them, the input feature can be one or more of the target first features described above. In addition, the above features can be split and combined according to specific recovery goals. However, regardless of how the input features change, the number of neurons in the input layer should be consistent with the number of input features.

[0154] Set a relatively large number of neurons in the first hidden layer to ensure that complex features of the data can be captured. A relatively large number of neurons helps the representational ability of the model.

[0155] Exemplarily, the number of neurons in the first hidden layer can be set to 32, and the activation function is ReLU; among them, 32 is a relatively common initial setting, and the specific value can be further adjusted according to the performance of the model. ReLU is a commonly used activation function in current deep learning, which is computationally efficient and can introduce non-linear relationships, and can effectively solve the problem of gradient disappearance, making the training of deep networks more stable.

[0156] Set a relatively small number of neurons in the second hidden layer to reduce the complexity of the model and prevent overfitting.

[0157] Exemplarily, the number of neurons in the second hidden layer can be set to 16, and the activation function is ReLU; among them, 16 neurons is a moderate setting, allowing further processing of the extracted features.

[0158] Regarding the above Step 132: Set the output layer of the first neural network model. The number of output neurons should be equal to the number of parameters to be predicted. For example, for the optimal temperature, pH value, and pressure, the output layer should be set with 3 neurons, and each neuron corresponds to one parameter. If it is to predict a single parameter, such as the optimal pH value, the output layer can have only 1 neuron. In a specific implementation, the number of neurons can be set according to the specific recovery task.

[0159] Example 1, in the case of setting multiple neurons. In the embodiments of the present application, the processing parameters include: processing method, temperature, pressure, pH value, reaction time, type and dosage of additives. Then, the number of neurons mapped to the output layer is 7. In a multi-classification problem, each class has one neuron, and each output neuron represents the probability of the corresponding class. If predicting the best processing method for waste battery powder, there may be three different methods A, B, and C. The output layer should have 3 neurons to predict the probability of each method respectively, and the Softmax activation function can be used to normalize the probabilities.

[0160] Example 2, in the case of setting only one output neuron, predicting the metal recovery rate after treating waste battery powder. The recovery rate is a continuous value. The output layer should have 1 neuron and use a linear activation function. In a binary classification problem, only the probability of one of the two classes needs to be predicted, for example, predicting whether the waste battery powder is suitable for a certain treatment method. If this method is suitable, the output is 1, otherwise the output is 0. The output layer should have 1 neuron, and the Sigmoid activation function is used to convert the output into a probability in the range of [0,1]. For a prediction problem of continuous values (regression problem), such as predicting the extraction efficiency of cobalt in waste battery powder, since the efficiency is a continuous percentage, the output layer should have 1 neuron and use a linear activation function.

[0161] In summary, the number of output neurons can be determined according to the goal and output type of the problem. When predicting a single target or continuous value, usually one output neuron is sufficient. When dealing with multi-target, multi-classification, and sequence prediction, multiple output neurons are often required.

[0162] Regarding the above step S102: Collect the waste battery powder after lithium extraction, and perform chemical analysis and physical analysis on the waste battery powder after lithium extraction to generate a first analysis result; input the first analysis result into the first neural network model to generate optimal processing parameters.

[0163] As a possible implementation manner, the first analysis result includes: element content, chemical state, electrochemical properties, particle size distribution, morphology and surface structure, thermal stability and organic material content, crystal structure, potential recovery value, and environmental and safety assessment. The step of collecting the waste battery powder after lithium extraction and performing chemical analysis and physical analysis on the waste battery powder after lithium extraction to generate a first analysis result includes at least one of the following steps:

[0164] Use X-ray fluorescence spectrometry to determine the content of various elements in the battery powder.

[0165] Precisely measure the concentration of specific metal elements by atomic absorption spectrometry.

[0166] Apply inductively coupled plasma mass spectrometry to detect trace and ultra-trace elements.

[0167] Electrochemical analysis is performed to evaluate the electrochemical properties of the residual chemicals in the battery powder.

[0168] The particle size distribution of the battery powder is determined by particle size analysis;

[0169] The surface morphology and microstructure of the battery powder are observed using a scanning electron microscope;

[0170] Thermogravimetric analysis is carried out to evaluate the thermal stability and organic material content of the material;

[0171] X-ray diffraction analysis is used to identify the mineral phases and crystalline phases in the battery powder.

[0172] In a specific implementation, it is necessary to comprehensively analyze the collected lithium-extracted waste battery powder to determine its composition and related characteristics, so as to guide the subsequent recycling process. The following are the specific functions of each step in the embodiments of this application:

[0173] 1. Using X-ray fluorescence spectrometry: mainly used to determine the content of various elements in the battery powder. This analysis method can effectively identify the main chemical elements in the powder. Exemplarily, the battery powder sample can be placed in an XRF analyzer, and the type and content of elements in the sample are determined through the interaction of X-rays.

[0174] 2. Atomic absorption spectrometry: used to accurately measure the concentration of specific metal elements, especially for precise quantification of some key metals. Exemplarily, after the sample is properly processed, it is evaporated into a gas and passed through a high-temperature flame, and the content of the element is quantified by measuring the absorption of light.

[0175] 3. Inductively coupled plasma mass spectrometry: detects trace and ultra-trace elements and provides necessary information for the extraction process. Exemplarily, the sample is dissolved and injected into the plasma, and the plasma ionizes the atoms in the sample, and the mass / charge ratio is analyzed by a mass spectrometer to identify and quantify the elements.

[0176] 4. Electrochemical analysis: evaluates the electrochemical properties of the residual chemicals in the battery powder to determine its stability and reactivity in subsequent processing. Exemplarily, cyclic voltammetry tests or charge-discharge tests can be carried out using an electrochemical workstation to measure the electrochemical stability and performance of the material.

[0177] 5. Particle size analysis: used to determine the particle size distribution of the battery powder, thereby inferring its physical properties and affecting the subsequent processing process. Exemplarily, a laser particle size analyzer or a screening method can be used to measure the particle size and distribution.

[0178] 6. Scanning Electron Microscope: Observe the surface morphology and microstructure of the battery powder, and further analyze its physical properties. Exemplarily, place the sample under high vacuum conditions and scan it with an electron beam, collect the generated secondary electrons and reflected electrons, and form an image.

[0179] 7. Thermogravimetric Analysis: Evaluate the thermal stability and organic material content of the material to determine the heat resistance performance of the battery powder and its performance in a high-temperature environment. Exemplarily, the sample can be gradually heated in a controlled temperature environment while recording the mass change.

[0180] 8. X-ray Diffraction Analysis: Identify the mineral phases and crystalline phases in the battery powder to provide further analysis of the chemical composition of the battery powder. Exemplarily, the crystal structure of the material can be determined by analyzing the diffraction pattern of the sample to X-rays.

[0181] The first analysis results obtained through these chemical and physical analysis methods can be input into the first neural network model described above to output the most suitable recycling strategy and treatment method. For example, based on the metal content and particle size data, the metal recycling process can be optimized, and the most suitable separation and purification technologies can be selected. In addition, these data can also be used to update and optimize the neural network model to improve the accuracy of predicting future processing parameters.

[0182] Among them, valuable metals and other materials mainly refer to the metal and non-metal materials used in battery manufacturing. These substances remain in the battery powder after lithium is extracted and have important economic value and environmental significance.

[0183] Exemplarily, the following are some valuable metals and other materials that can be recycled through the embodiments of the present application:

[0184] 1. Cobalt: Cobalt is an important component in lithium-ion batteries, especially in lithium cobalt oxide batteries. Cobalt is not only expensive but also in limited supply, and has important recycling value.

[0185] 2. Nickel: Nickel is used in a variety of battery chemistries, including nickel cobalt aluminum (NCA) and nickel cobalt manganese (NCM) batteries. The recycling of nickel helps to reduce dependence on raw materials and environmental impact.

[0186] 3. Aluminum: Aluminum is commonly used in the battery housing and some internal structures. It has a high recycling rate and relatively low reprocessing cost.

[0187] 4. Copper: Copper is usually used as a conductive material in batteries, especially on the anode. The recycling of copper is very important because it is a valuable metal and is widely used in the electronics and electrical industries.

[0188] 5. Manganese: Manganese is a key material in certain types of batteries, such as lithium manganese oxide batteries. The recycling of manganese contributes to the sustainable use of resources.

[0189] 6. Graphite: Graphite is commonly used as the anode material in batteries. Although it is not a metal, its recycling is equally important for battery manufacturing.

[0190] 7. Rare earth elements and other rare metals: Certain high-performance batteries may contain rare earth elements or other rare metals, which also have recycling value.

[0191] 8. Plastics and organic materials: Various plastics and organic materials used in battery manufacturing, although their economic value may be lower than that of metals, proper recycling can reduce environmental pollution.

[0192] In order to effectively recycle valuable metals and other materials from the spent battery powder after lithium extraction, different physical, chemical or comprehensive methods can be adopted. Considering the characteristics of various materials and their environmental impacts, the following are some exemplary recycling methods:

[0193] Recycling methods for cobalt, nickel, copper, and manganese:

[0194] 1. Chemical leaching method: Acid leaching extraction, using sulfuric acid, hydrochloric acid or nitric acid to leach the treated battery powder. These acids can effectively dissolve metal ions; Chelating agent-assisted extraction: Introducing chelating agents such as EDTA (ethylenediaminetetraacetic acid) to enhance the dissolution rate of specific metal ions such as cobalt and nickel.

[0195] 2. Electrochemical method: Using the electrolysis process to deposit pure metals such as copper and nickel at the cathode, which requires operating in an appropriate electrolyte solution with adjusted current and voltage.

[0196] 3. Solvent extraction: Using organic solvents to selectively extract specific metals from solutions containing multiple metals.

[0197] For the recycling of aluminum, physical separation methods can be adopted. Since aluminum is non-magnetic, it can be separated from iron-containing materials through magnetic separation technology, or separated using particle size differences and specific gravity differences.

[0198] For the recycling of graphite, physical screening can be used. Since graphite usually exists in powder form in batteries, it can be separated from other heavier metal powders through fine sieving and air separation technology.

[0199] For the recycling of plastics and organic materials, appropriate solvents such as acetone or other organic solvents can be used to dissolve the plastics from other insoluble materials. Further, heating the plastics under anaerobic conditions can convert them into fuel oil and carbon black, and this method can convert organic materials into useful energy.

[0200] In a specific implementation, the first neural network model constructed by the embodiments of the present application through historical recovery data can determine the optimal processing parameters for the current batch of lithium-extracted waste battery powder based on the chemical analysis results and physical analysis results of the input lithium-extracted waste battery powder.

[0201] Exemplarily, if there is a sample of lithium-extracted waste battery powder, this sample has the following characteristics: high content of element A, small particle size, and contains a certain amount of impurities. Inputting this information into the first neural network model, the first neural network model will predict the optimal processing parameters based on these characteristics and historical data.

[0202] Example 1:

[0203] 1. Input data: Content of element A: high; Particle size distribution: small; Impurity content: medium.

[0204] 2. Neural network analysis: The network learns through analyzing historical recovery data and the characteristics of the current sample that small particle size and high content of element A may be more suitable for a certain specific processing method.

[0205] 3. Output optimal processing parameters:

[0206] Processing method: It is recommended to use a chemical treatment method combined with a physical treatment method.

[0207] Temperature: The recommended processing temperature is 90 °C to promote the chemical reaction rate.

[0208] pH value: The recommended pH value is 10, corresponding to the selected chemical treatment method.

[0209] Reaction time: The predicted reaction time is 30 minutes, which is sufficient for the ions of element A to be released from the battery powder.

[0210] Dosage of additive: It is recommended to use 10% of a certain chemical additive to enhance the extraction effect.

[0211] 4. Further processing suggestions:

[0212] Use screening technology after chemical treatment to further separate and purify the lithium-extracted waste battery powder. Example 2:

[0213] When processing a batch of spent battery powder from which lithium has been extracted, after preliminary chemical and physical analyses, the analysis results are input into the first neural network model. The first neural network model can generate optimal processing parameters based on the input data, including the metal content, particle size distribution, chemical state, etc. of the battery powder. For example: The optimal reaction temperature recommended by the model is 95 °C to optimize the extraction efficiency of element B and reduce side reactions; to maximize the solubility of element B and other valuable metals, the model recommends adjusting the pH value of the reaction solution to 9.5; according to the calculations of the model, the reaction time is set to 120 minutes to ensure the integrity of the chemical reaction and the extraction efficiency; the model suggests adding 0.5% by volume of citric acid as a complexing agent to help extract element B from the battery powder more effectively; the model recommends using centrifugal separation technology and setting the rotational speed of the centrifuge to 2000 rpm to more efficiently separate valuable metals from the battery powder.

[0214] The above parameters are the results of the neural network model's learning and optimization of a large amount of historical data and the characteristics of the current battery powder sample. Using these optimal processing parameters, the processing process can be adjusted according to the specific characteristics of the battery powder and processing requirements, thereby achieving an efficient and environmentally friendly recycling process.

[0215] In the above example, the neural network not only provides a specific processing method but also provides detailed operating parameters, which help optimize the entire element B extraction process and ensure efficiency and product quality. In this way, the neural network makes the processing decision more data-driven and accurate, thereby improving resource utilization and environmental sustainability.

[0216] Regarding the above S103: Based on the optimal processing parameters, adjust the processing parameters of the spent battery powder from which lithium has been extracted, and recycle the valuable metals, plastics, and organic materials in the spent battery powder from which lithium has been extracted according to the adjusted processing parameters.

[0217] As a possible implementation method, the processing methods include: chemical processing method, physical processing method, and comprehensive processing method; based on the optimal processing parameters, adjusting the processing parameters of the spent battery powder from which lithium has been extracted and recycling the valuable metals, plastics, and organic materials in the spent battery powder from which lithium has been extracted according to the adjusted processing parameters includes:

[0218] Based on the optimal processing parameters, adjust the processing parameters for the current spent battery powder from which lithium has been extracted;

[0219] For the current spent battery powder from which lithium has been extracted, based on the adjusted processing parameters, recycle the valuable metals, plastics, and organic materials from the spent battery powder from which lithium has been extracted, and monitor the extraction effect;

[0220] Record all the above adjustment data and recovery result data, and update the adjustment data and the recovery result data to the historical recovery data.

[0221] As a possible implementation manner, adjusting the processing parameters for the currently lithium-extracted waste battery powder based on the optimal processing parameters includes:

[0222] In response to determining that the processing method for the currently lithium-extracted waste battery powder is the chemical processing method, when adjusting the processing parameters of the battery powder, the following steps are performed:

[0223] Based on the optimal processing parameters, adjust the temperature setting in the reactor;

[0224] Based on the optimal processing parameters, use an acid or a base to adjust the pH value of the solution to the recommended pH range;

[0225] Based on the optimal processing parameters, set the reaction time of the reactor;

[0226] Based on the optimal processing parameters, add chemical additives of specified types and dosages;

[0227] In response to determining that the processing method for the currently lithium-extracted waste battery powder is the physical processing method, when adjusting the processing parameters of the battery powder, the following steps are performed:

[0228] Based on the optimal processing parameters, select a separation technique; wherein, the separation technique includes: screening, magnetic separation, centrifugal separation, and flotation;

[0229] Based on the optimal processing parameters, adjust the temperature and pressure settings of the separator;

[0230] Based on the optimal processing parameters, adjust the equipment operation parameters.

[0231] In a specific implementation, the optimal processing parameters in the embodiments of the present application are the results predicted by a neural network model based on historical recovery data and the current chemical and physical characteristics of the battery powder. These parameters include, but are not limited to: processing method, temperature, pressure, pH value, reaction time, types and dosages of additives. Although the optimal processing parameters provide a theoretically optimal setting, in actual applications, it may be necessary to adjust according to specific circumstances to ensure the processing effect and cope with unforeseen variables. For example: dynamically adjust the temperature and pressure settings according to the real-time monitored chemical reaction conditions and external environment changes to maintain the reaction efficiency and safety; as the reaction progresses, adjust the pH value and extend or shorten the reaction time according to the dissolution rate of elements in the solution and the behavior of other chemical substances.

[0232] Exemplarily, an example of optimal processing parameters: Chemical treatment method, recommended to be set at 95°C, recommended to be set at 9.5, recommended to be set for 120 minutes, recommended to use citric acid at a volume ratio of 0.5%, recommended to use centrifugal separation, the separator temperature is set at 25°C, and the separator pressure is set at 1 atm.

[0233] Examples of actual adjustments: If the following situations are encountered during the execution process, adjustments need to be made to the optimal processing parameters:

[0234] It is found that at 95°C, volatile components contained in some battery powders pose a safety risk. In actual operation, the temperature is adjusted to 90°C to ensure the safety of the operation.

[0235] During the processing, it is monitored in real time that the pH value rises too fast, which may damage the equipment or affect the recovery efficiency of the final metal. Therefore, the addition rate of the alkali used is adjusted to reach the ideal pH value more smoothly.

[0236] According to the online monitored progress of the chemical reaction, the actual reaction rate is faster than expected. Therefore, the reaction time is shortened to 100 minutes to avoid overreaction and waste of resources.

[0237] During the reaction process, it is found that the effect of the additive is not as expected. Through small-scale laboratory tests, it is decided to increase the proportion of citric acid to 0.7% to optimize the effect.

[0238] During the centrifugal separation process, it is found that a lower temperature (20°C) can improve the separation efficiency, and at the same time, the pressure needs to be increased to 1.2 atm to accelerate the separation speed.

[0239] It can be understood that although there is a set of theoretically optimal processing parameters, factors such as the specific performance of the equipment, the unique reactions of the materials, and changes in environmental conditions may all lead to the need to adjust these optimal processing parameters.

[0240] Through the above steps in the embodiments of the present application, the processing parameters of the currently lithium-extracted waste battery powder can be adjusted according to the optimal processing parameters, and valuable metals, plastics, and organic materials can be effectively recovered according to the adjusted processing parameters, and the extraction effect can be monitored.

[0241] Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0242] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described systems and units can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0243] In several embodiments provided by the present application, it should be understood that the disclosed system can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces, and the coupling or communication connection between devices or units can be in an electrical or other form.

[0244] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A recycling and treatment method for lithium battery powder, characterized in that, Including: Obtain historical recycling data; Based on the historical recycling data, construct a first neural network model; Collect the spent battery powder after lithium extraction, and perform chemical analysis and physical analysis on the spent battery powder after lithium extraction to generate a first analysis result; input the first analysis result into the first neural network model to generate optimal processing parameters; Based on the optimal processing parameters, adjust the processing parameters of the spent battery powder after lithium extraction, and recycle the valuable metals and organic materials in the spent battery powder after lithium extraction based on the adjusted processing parameters; The constructing the first neural network model based on the historical recycling data includes: Preprocess the historical recycling data; Based on the preprocessed historical recycling data, perform feature engineering to generate a first target feature; Based on the first target feature and the processing parameters, construct the first neural network model; The performing feature engineering steps based on the preprocessed historical recycling data to generate a first target feature includes: Based on the historical recycling data, determine the key chemical elements in the battery powder; Based on the key chemical elements in the battery powder, create a metal content index to determine the importance and potential economic value of each batch of spent battery powder after lithium extraction, and determine the recycling priority and resource allocation for each batch of spent battery powder after lithium extraction. The mathematical expression is: Among them, MCI is the metal content index, n is the type of key chemical elements, and the content of each element i in the lithium-extracted waste battery powder is c i , ω i is the weight of each element, and the weights of each chemical component in the metal content index are preset according to the contribution and value of each metal in battery performance; Calculate the chemical combination ratios of each chemical combination among the key chemical elements in the battery powder to determine the relative content relationship between different chemical components. The mathematical expression for determining the chemical combination ratio is: Among them, Ratio a,b represents the combined ratio of chemical component a and chemical component b; c a is the content of chemical component a in the battery powder, and c b is the content of chemical component b in the battery powder; Introduce a density-particle size index characterizing physical characteristics to determine the influence of the density-particle size index on the recycling efficiency; Index D,S = D × S Among them, Index D,S is the density-particle size index; D is the density of the battery powder; S is the particle size of the battery powder; Introduce the square term of the content of the key chemical elements in the battery powder and the cubic term of the content of the key chemical elements in the battery powder as polynomials to determine the non-linear influence of the change in the content of the key chemical elements in the battery powder on the recycling effect; Introduce a temperature-pressure interaction term to determine the influence of temperature and pressure on the recycling efficiency. The mathematical expression is: Interaction T,P = T × P Among them, Interaction T,P is the temperature-pressure interaction term; T is the temperature value during the battery powder treatment; P is the pressure value during the battery powder treatment; Introduce a pH-reaction time interaction term to determine the relationship between the pH value and the reaction time. The mathematical expression is: Interaction pH,t = pH × t Among them, pH represents the acidity and alkalinity of the solution, t represents the reaction time, and Interaction pH,t is the interaction term of pH value and reaction time; Based on the above feature engineering steps, generate the first target feature.

2. The recycling method based on lithium battery powder according to claim 1, wherein The historical recycling data includes: physical information of the battery powder, chemical information of the battery powder, processing parameter information, and recycling efficiency information.

3. The recycling and treatment method based on lithium battery powder according to claim 1, wherein, The processing parameters include at least one of the following: processing method, temperature, pressure, pH value, reaction time, type and dosage of additives.

4. The recycling method based on lithium battery powder according to claim 1, wherein The first neural network model includes: an input layer, a first hidden layer, a batch normalization layer, a second hidden layer, a dropout layer, and an output layer; the constructing the first neural network model based on the first target feature and the processing parameters includes: Based on the first target feature, set the number of neurons in the input layer, the first hidden layer, and the second hidden layer in the first neural network model; Based on the processing parameters, set the number of neurons in the output layer in the first neural network model.

5. The recycling method based on lithium battery powder according to claim 1, wherein The first analysis results include: elemental content, chemical state, electrochemical properties, particle size distribution, morphology and surface structure, thermal stability and organic material content, crystal structure, potential recycling value, and environmental and safety assessment. Collecting the spent battery powder after lithium extraction and performing chemical and physical analyses on the spent battery powder after lithium extraction to generate the first analysis results includes at least one of the following steps: Using X-ray fluorescence spectrometry to determine the various elemental contents in the battery powder; Precisely measuring the concentration of specific metal elements by atomic absorption spectrometry; Applying inductively coupled plasma mass spectrometry to detect trace and ultra-trace elements; Performing electrochemical analysis to evaluate the electrochemical performance of the residual chemicals in the battery powder; Determining the particle size distribution of the battery powder by particle size analysis; Observing the surface morphology and microstructure of the battery powder using a scanning electron microscope; Performing thermogravimetric analysis to evaluate the thermal stability and organic material content of the material; Using X-ray diffraction analysis to identify the crystalline phases in the battery powder.

6. The recycling method based on lithium battery powder according to claim 3, characterized in that, The treatment methods include: chemical treatment method, physical treatment method, and comprehensive treatment method. Based on the optimal treatment parameters, adjusting the treatment parameters of the spent battery powder after lithium extraction and recovering the valuable metals and organic materials in the spent battery powder after lithium extraction based on the adjusted treatment parameters includes: Based on the optimal treatment parameters, adjusting the treatment parameters for the current spent battery powder after lithium extraction; For the current spent battery powder after lithium extraction, based on the adjusted treatment parameters, recovering the valuable metals and organic materials from the current spent battery powder after lithium extraction and monitoring the extraction effect; Recording all the above adjustment data and recovery result data and updating the adjustment data and the recovery result data to the historical recovery data.

7. The recycling method based on lithium battery powder according to claim 6, characterized in that, Based on the optimal treatment parameters, adjusting the treatment parameters for the current spent battery powder after lithium extraction includes: In response to determining that the treatment method for the current spent battery powder after lithium extraction is the chemical treatment method, when adjusting the treatment parameters of the battery powder, perform the following steps: Based on the optimal treatment parameters, adjusting the temperature setting in the reactor; Based on the optimal treatment parameters, using an acid or a base to adjust the pH value of the solution to the recommended pH range; Based on the optimal treatment parameters, setting the reaction time of the reactor; Based on the optimal treatment parameters, adding chemical additives of specified types and dosages; In response to determining that the treatment method for the current spent battery powder after lithium extraction is the physical treatment method, when adjusting the treatment parameters of the battery powder, perform the following steps: Based on the optimal treatment parameters, selecting a separation technique; wherein, the separation techniques include: screening, magnetic separation, centrifugal separation, and flotation; Based on the optimal treatment parameters, adjusting the temperature and pressure settings of the separator; based on the optimal treatment parameters, adjusting the equipment operation parameters.

Images