Method for repairing and remanufacturing lithium ion batteries

Abstract

The application provides a repairing and remanufacturing method of a lithium ion battery, relates to the technical field of lithium ion battery manufacturing, and obtains a positive electrode waste powder by performing multistage pretreatment on a waste lithium ion battery, performs hydrothermal repairing and remanufacturing searching according to powder detection data, builds a positive electrode regeneration searching space to perform global confidence mining optimization, and determines a regeneration control scheme; a regeneration positive electrode evaluation mechanism is introduced to evaluate the scheme, if the evaluation result does not satisfy the constraint, the scheme is feedback adjusted to obtain a first positive electrode regeneration adjustment group, and iterative breeding optimization is performed according to a positive electrode regeneration optimization fitness function to obtain a positive electrode regeneration optimization result, and the positive electrode material is remanufactured in combination with the positive electrode waste powder. The application solves the technical problems that, due to the lack of multidimensional data driving and global optimization optimization mechanism, the repairing and remanufacturing scheme has poor adaptability, and the performance of the regenerated material is unstable in the prior art, and improves the performance and quality stability of the regenerated positive electrode material.

Description

Technical Field

[0001] The present invention relates to the technical field of lithium ion battery manufacturing, and in particular to a method for repairing and remanufacturing lithium ion batteries. Background Art

[0002] In recent years, lithium-ion batteries have been widely used in consumer electronics, new energy vehicles, and energy storage systems. As their service life increases, the number of used lithium-ion batteries continues to rise. Repairing and remanufacturing used lithium-ion batteries not only effectively reduces environmental pollution and enables resource recycling, but also reduces dependence on virgin lithium-ion battery materials.

[0003] Existing methods for repairing and remanufacturing lithium-ion batteries mainly include physical sorting, chemical extraction, and resynthesis, which usually rely on fixed processing procedures and single process conditions. However, the sources of waste lithium-ion batteries are complex, resulting in large differences in the composition and structure of their positive electrode materials. Traditional repair and remanufacturing processes are difficult to adapt to different powder properties and remanufacturing needs in a timely manner. This not only leads to poor adaptability of repair and remanufacturing solutions, but also easily causes fluctuations in the performance of recycled positive electrode materials, making it difficult to ensure the high performance and quality stability of remanufactured materials, thereby limiting the large-scale promotion and application of lithium-ion battery repair and remanufacturing technologies. Summary of the Invention

[0004] The present invention provides a method for repairing and remanufacturing lithium-ion batteries, which solves the technical problems in the prior art caused by the lack of multi-dimensional data drive and global optimization mechanism, resulting in poor adaptability of lithium-ion battery repair and remanufacturing solutions and unstable performance of recycled materials, thereby achieving the technical effect of improving the performance and quality stability of recycled positive electrode materials.

[0005] In view of the above problems, the present invention provides a method for repairing and remanufacturing lithium-ion batteries, which includes: performing multi-stage pretreatment on waste lithium-ion batteries to obtain positive electrode waste powder, and performing hydrothermal repair and remanufacturing search based on powder detection data of the positive electrode waste powder to build a positive electrode regeneration search space; performing global confidence mining optimization based on the positive electrode regeneration search space to determine a positive electrode regeneration control scheme; introducing a regenerated positive electrode evaluation mechanism to evaluate the positive electrode regeneration control scheme to obtain a regenerated positive electrode evaluation result; if the regenerated positive electrode evaluation result does not meet the regenerated positive electrode evaluation constraint, feedback-adjusting the positive electrode regeneration control scheme according to the regenerated positive electrode evaluation mechanism and the positive electrode regeneration search space to obtain a first positive electrode regeneration adjustment group that meets the regenerated positive electrode evaluation constraint; iteratively breeding and optimizing the first positive electrode regeneration adjustment group according to the positive electrode regeneration optimality function to obtain a positive electrode regeneration optimization result, and remanufacturing the positive electrode material in combination with the positive electrode waste powder.

[0006] Preferably, the regenerated positive electrode evaluation mechanism includes: predicting the regenerated positive electrode characteristics according to the positive electrode regeneration control scheme and the positive electrode waste powder to obtain the regenerated positive electrode structural characteristics, regenerated positive electrode electrochemical characteristics and regenerated positive electrode impurity characteristics; activating the regenerated positive electrode evaluation model, the regenerated positive electrode evaluation model includes an input layer, a structural defect evaluation layer, an electrochemical performance evaluation layer, an impurity impact evaluation layer and an output layer; inputting the regenerated positive electrode structural characteristics into the structural defect evaluation layer to obtain a structural defect evaluation coefficient; inputting the regenerated positive electrode electrochemical characteristics into the electrochemical performance evaluation layer to obtain an electrochemical performance evaluation coefficient; inputting the regenerated positive electrode impurity characteristics into the impurity impact evaluation layer to obtain an impurity impact evaluation coefficient, and combining the structural defect evaluation coefficient and the electrochemical performance evaluation coefficient to generate the regenerated positive electrode evaluation result.

[0007] Preferably, the positive electrode regeneration control scheme is feedback-adjusted according to the regenerative positive electrode evaluation mechanism and the positive electrode regeneration search space to obtain a first positive electrode regeneration adjustment group that satisfies the regenerative positive electrode evaluation constraint, including: performing adjustment constraint analysis according to the positive electrode regeneration search space to establish a multi-node adjustment constraint domain; performing random adjustment on the positive electrode regeneration control scheme according to the multi-node adjustment constraint domain to obtain a first positive electrode regeneration adjustment space; evaluating the first positive electrode regeneration adjustment space according to the regenerative positive electrode evaluation mechanism to obtain a regenerative positive electrode evaluation distribution; based on the regenerative positive electrode evaluation distribution, the first positive electrode regeneration adjustment space is optimized and screened according to the regenerative positive electrode evaluation constraint to generate the first positive electrode regeneration adjustment group.

[0008] Preferably, regulation constraint analysis is performed based on the positive electrode regeneration search space to establish a multi-node regulation constraint domain, including: screening the precursor synthesis search space based on a predetermined global confidence level to establish a precursor synthesis preferred space; screening the hydrothermal reaction search space based on the predetermined global confidence level to establish a hydrothermal reaction preferred space; screening the high-temperature sintering search space based on the predetermined global confidence level to establish a high-temperature sintering synthesis preferred space; performing adaptive variable triggering interval analysis based on the precursor synthesis preferred space, the hydrothermal reaction preferred space and the high-temperature sintering synthesis preferred space to generate the multi-node regulation constraint domain.

[0009] Preferably, performing iterative propagation optimization on the first positive electrode regeneration adjustment group according to the positive electrode regeneration optimality function to obtain the positive electrode regeneration optimization result includes: constructing the positive electrode regeneration optimality function, wherein the positive electrode regeneration optimality function is: ; wherein, the CRS represents the positive electrode regeneration optimality, the exp represents the exponential function with the natural constant as the base, the CEX represents the normalized electrochemical performance evaluation coefficient, the EW, the KW and the PW represent the predetermined weight conditions, the sum of the EW, the KW and the PW is 1, the CKX represents the normalized impurity influence evaluation coefficient, and the CPX represents the normalized structure defect evaluation coefficient; the first positive electrode regeneration adjustment group is subjected to positive electrode regeneration optimality maximization optimization according to the positive electrode regeneration optimality function, to obtain a first positive electrode regeneration optimization strategy; the first positive electrode regeneration adjustment group is subjected to reproductive optimization according to the regenerated positive electrode evaluation mechanism and the regenerated positive electrode evaluation constraint, to obtain a second positive electrode regeneration adjustment group; the second positive electrode regeneration adjustment group is subjected to positive electrode regeneration optimality maximization optimization according to the positive electrode regeneration optimality function, to obtain a second positive electrode regeneration optimization strategy; the first positive electrode regeneration optimization strategy and the second positive electrode regeneration optimization strategy are compared according to the positive electrode regeneration optimality, to obtain a current optimal regeneration optimization strategy; and the current optimal regeneration optimization strategy is subjected to iterative reproductive optimization according to the regenerated positive electrode evaluation mechanism and the regenerated positive electrode evaluation constraint based on the positive electrode regeneration optimality function, until the positive electrode regeneration optimization result meeting the number of iterations is obtained.

[0010] Preferably, the first positive electrode regeneration adjustment group is subjected to reproductive optimization according to the regenerated positive electrode evaluation mechanism and the regenerated positive electrode evaluation constraint, to obtain a second positive electrode regeneration adjustment group, including: the first positive electrode regeneration adjustment group is subjected to a first reproductive number distribution according to a predetermined reproductive number, to obtain a first reproductive number distribution result; the first positive electrode regeneration adjustment group is adjusted according to a multi-node adjustment constraint domain based on the first reproductive number distribution result, to obtain a positive electrode regeneration adjustment second space; and the positive electrode regeneration adjustment second space is subjected to optimization analysis according to the regenerated positive electrode evaluation mechanism and the regenerated positive electrode evaluation constraint, to generate the second positive electrode regeneration adjustment group.

[0011] Preferably, the hydrothermal repair remanufacturing search is performed according to the powder detection data of the positive electrode waste powder, and a positive electrode regeneration search space is built, including: a precursor synthesis scheme retrieval is performed according to the powder detection data, to obtain a precursor synthesis search space; a hydrothermal reaction scheme retrieval is performed according to the powder detection data, to obtain a hydrothermal reaction search space; a high-temperature sintering scheme retrieval is performed according to the powder detection data, to obtain a high-temperature sintering search space; and the precursor synthesis search space and the hydrothermal reaction search space are combined, to generate the positive electrode regeneration search space.

[0012] Preferably, global confidence mining is performed on the positive electrode regeneration search space to determine the positive electrode regeneration control scheme, including: randomly numbering according to the precursor synthesis search space to obtain a first synthesis matching scheme, a second synthesis matching scheme...a Kth synthesis matching scheme, where K is a positive integer; performing global confidence calculation on the first synthesis matching scheme, the second synthesis matching scheme...the Kth synthesis matching scheme respectively to obtain a first global confidence, a second global confidence...the Kth global confidence; performing global confidence maximization optimization on the first synthesis matching scheme, the second synthesis matching scheme...the Kth synthesis matching scheme according to the first global confidence, the second global confidence...the Kth global confidence to obtain a global confidence precursor synthesis scheme; continuing to perform global confidence maximization optimization on the hydrothermal reaction search space and the high-temperature sintering search space to obtain a global confidence hydrothermal reaction scheme and a global confidence high-temperature sintering scheme, and combining the global confidence precursor synthesis scheme to generate the positive electrode regeneration control scheme.

[0013] Preferably, global confidence calculations are performed on the first synthetic matching scheme, the second synthetic matching scheme...the Kth synthetic matching scheme, respectively, including: performing support analysis on the first synthetic matching scheme according to the positive pole regeneration search space to obtain a first synthetic global support; performing support analysis on each precursor synthetic parameter in the first synthetic matching scheme according to the positive pole regeneration search space to obtain multiple synthetic parameter support, and taking the sum of the multiple synthetic parameter support as the first synthetic trigger degree; performing a ratio calculation based on the first synthetic trigger degree and the first synthetic global support degree to generate the first global confidence.

[0014] Preferably, the multi-stage pretreatment includes discharge disassembly, screening extraction, positive electrode powder desorption and ultrasonic impurity removal.

[0015] One or more technical solutions provided in the present invention have at least the following beneficial effects:

[0016] By multi-stage pretreatment of waste lithium ion batteries, positive electrode waste powder is obtained, and according to the powder detection data of the positive electrode waste powder, hydrothermal repair remanufacturing search is carried out, the positive electrode regeneration search space is built, and accurate and comprehensive data support and customized search range are provided for the subsequent repair process. According to the positive electrode regeneration search space, global confidence mining optimization is carried out to determine the positive electrode regeneration control scheme, which avoids local optimization and human experience limitation, and ensures the global optimality of the remanufacturing scheme. By introducing the regenerated positive electrode evaluation mechanism, the regenerated positive electrode evaluation result is obtained, and the defects in the scheme are found in time to ensure that the performance of the regenerated positive electrode material meets the expected requirements and provides a clear direction and target for subsequent feedback adjustment. If the regenerated positive electrode evaluation result does not meet the regenerated positive electrode evaluation constraint, the regenerated positive electrode evaluation mechanism and the positive electrode regeneration search space are used to carry out feedback adjustment on the positive electrode regeneration control scheme to obtain a first positive electrode regeneration adjustment group that meets the regenerated positive electrode evaluation constraint, thereby enhancing the adaptability and flexibility of the regeneration scheme. According to the positive electrode regeneration optimal fitness function, the first positive electrode regeneration adjustment group is iteratively reproduced and optimized to obtain a positive electrode regeneration optimization result, which further refines and improves the performance of the positive electrode regeneration control scheme. Finally, the positive electrode material is remanufactured in combination with the positive electrode waste powder to form a complete and sustainable lithium ion battery repair and remanufacturing process chain.

[0017] In summary, the present application realizes efficient adaptation of positive electrode waste powder and dynamic optimization of the remanufacturing process by multi-stage pretreatment and search space establishment driven by powder detection data, determining the positive electrode regeneration control scheme by global confidence mining optimization, introducing the regenerated positive electrode evaluation mechanism to form a closed-loop feedback adjustment, and using the optimal fitness function for group iterative reproduction and optimization. The whole scheme not only solves the performance fluctuation problem caused by poor process solidification and adaptability in the prior art, but also improves the quality stability and performance consistency of the regenerated positive electrode material, realizes intelligent and self-adaptive lithium ion battery repair and remanufacturing, and promotes the scale application and green and sustainable development of lithium ion battery repair and remanufacturing.

[0018] The above description is only a summary of the technical scheme of the present application, in order to more clearly understand the technical means of the present application, the content of the specification can be implemented, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application are described as follows. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 The flowchart of the lithium ion battery repair and remanufacturing method provided by the embodiment of the present application is shown.

[0020] Figure 2A flowchart of a mechanism for evaluating regenerated positive electrodes in a lithium ion battery repair and remanufacturing method according to an embodiment of the present application is shown.

[0021] Figure 3 A flowchart of obtaining a first positive electrode regeneration adjustment group in a lithium ion battery repair and remanufacturing method according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0022] The embodiments of the present application provide a lithium ion battery repair and remanufacturing method, which solves the technical problem of poor adaptability of the lithium ion battery repair and remanufacturing scheme and unstable performance of regenerated materials due to lack of multi-dimensional data driving and global optimization optimization mechanism in the prior art, and achieves the technical effect of improving the performance and quality stability of regenerated positive electrode materials.

[0023] As shown in Figure 1 The embodiments of the present application provide a lithium ion battery repair and remanufacturing method, which includes:

[0024] Step S100: performing multi-stage pretreatment on the waste lithium ion battery to obtain positive electrode waste powder, and performing hydrothermal repair and remanufacturing search according to the powder detection data of the positive electrode waste powder to build a positive electrode regeneration search space.

[0025] Further, the multi-stage pretreatment includes discharging and disassembling, screening and extracting, positive electrode powder desorption, and ultrasonic impurity removal.

[0026] Specifically, multi-stage pretreatment involves a multi-step process for waste lithium-ion batteries, including discharge, disassembly, screening, cathode powder desorption, and ultrasonic impurity removal, to obtain high-quality cathode waste powder. First, the waste lithium-ion batteries are discharged and disassembled: the batteries are depleted using specialized discharge equipment (such as a resistive load box) to ensure safe disassembly. After disassembly, the cathode sheets are separated using physical or chemical methods and screened to extract cathode powder with a target particle size (e.g., 10-50 microns). Subsequently, the cathode powder is desorbed using a chemical cleaning agent (such as dilute acid or deionized water) or heat treatment to remove residual electrolyte and active materials. The powder is then ultrasonically treated at an appropriate frequency (e.g., 40 kHz) using an ultrasonic cleaner to further remove fine impurities, resulting in pure cathode waste powder. After obtaining the cathode waste powder, the composition and physicochemical properties of the powder are tested using instruments such as X-ray diffraction and scanning electron microscopy. This data is then collected, including information on metal element ratios, impurity content, particle size distribution, and morphology. Based on these powder test results, a data-driven search is conducted to identify multi-dimensional parameter combinations suitable for the powder properties, including precursor synthesis conditions, hydrothermal reaction conditions, and high-temperature sintering conditions. This results in a corresponding search space for precursor synthesis, hydrothermal reaction, and high-temperature sintering. These search spaces are then integrated to construct a complete cathode regeneration search space.

[0027] Through multi-stage pretreatment, the purity and uniformity of the positive electrode powder are significantly improved. Based on the powder detection data, a positive electrode regeneration search space is constructed, and adaptive adaptation to the characteristics of different waste positive electrode powders is achieved, providing a search framework for subsequent global optimization and remanufacturing processes.

[0028] Step S200: performing global confidence mining optimization based on the positive electrode regeneration search space to determine a positive electrode regeneration control scheme.

[0029] Specifically, global confidence mining optimization evaluates the global confidence of different solutions within the cathode regeneration search space to identify highly reliable and feasible cathode regeneration control solutions. First, within the cathode regeneration search space constructed in step S100, the different solutions within each search space, such as precursor synthesis, hydrothermal reaction, and high-temperature sintering, are randomly numbered, and a global confidence calculation is performed for each solution. During the calculation, a support analysis is first performed on each matching solution to obtain the global support (frequency of occurrence of the solution) and the synthesis parameter triggering degree (frequency of occurrence of each process parameter). A global confidence score is then generated based on the ratio of the two. Then, based on the global confidence score, the solution with the highest global confidence score is selected within each search space and combined to form the overall cathode regeneration control solution.

[0030] This step achieves efficient screening of complex parameter spaces through global confidence mining, ensuring the reliability and adaptability of the selected positive electrode regeneration control scheme in the repair and remanufacturing process, and improving the overall controllability and success rate of the remanufacturing process.

[0031] Step S300: introducing a regenerated positive electrode evaluation mechanism to evaluate the positive electrode regeneration control scheme to obtain a regenerated positive electrode evaluation result.

[0032] Specifically, the regenerated positive electrode evaluation mechanism refers to a mechanism that comprehensively evaluates the structure, electrochemical performance, and impurity characteristics of the regenerated positive electrode based on the positive electrode powder detection data and the parameters of the positive electrode regeneration control plan. The regenerated positive electrode characteristics are predicted based on the positive electrode regeneration control plan and the positive electrode waste powder to obtain the structural characteristics, electrochemical characteristics, and impurity characteristics of the regenerated positive electrode. The regenerated positive electrode evaluation model is then activated, and the structural characteristics, electrochemical characteristics, and impurity characteristics of the regenerated positive electrode are respectively input into the structural defect evaluation layer, electrochemical performance evaluation layer, and impurity impact evaluation layer of the regenerated positive electrode evaluation model to obtain the corresponding evaluation coefficients. Finally, these evaluation coefficients are combined to generate the regenerated positive electrode evaluation results.

[0033] Through this regenerated positive electrode evaluation mechanism, the positive electrode regeneration effect can be comprehensively and quantitatively analyzed to ensure the scientific nature and practical feasibility of the positive electrode regeneration control scheme, and provide key guiding data for subsequent feedback adjustment and remanufacturing decisions.

[0034] Step S400: If the regenerated positive electrode evaluation result does not meet the regenerated positive electrode evaluation constraint, feedback-adjust the positive electrode regeneration control scheme according to the regenerated positive electrode evaluation mechanism and the positive electrode regeneration search space to obtain a first positive electrode regeneration adjustment group that meets the regenerated positive electrode evaluation constraint.

[0035] Specifically, the regenerated positive electrode evaluation constraints refer to the minimum requirements that the performance of the regenerated positive electrode must meet, including impurity impact evaluation constraints, structural defect evaluation constraints, and electrochemical performance evaluation constraints. The first positive electrode regeneration adjustment group refers to a series of alternative regeneration process parameter combinations that meet the regenerated positive electrode performance requirements after feedback adjustment. When the regenerated positive electrode evaluation results do not meet the standards, the feedback adjustment mechanism is activated. First, a multi-node adjustment constraint domain is parsed based on the regenerated positive electrode search space. For example, the adjustable hydrothermal temperature, sintering time and other sensitive parameter ranges are further screened by global confidence. Subsequently, the control scheme is randomly or distributedly perturbed within the adjustment constraint domain to form the first positive electrode regeneration adjustment space. The parameter combinations in the first positive electrode regeneration adjustment space are re-entered into the regenerated positive electrode evaluation mechanism to obtain a regenerated positive electrode evaluation distribution map. Then, based on the preset regenerated positive electrode evaluation constraints, all combinations that meet the conditions are screened to form the first positive electrode regeneration adjustment group.

[0036] Step S500: performing iterative multiplication optimization on the first positive electrode regeneration adjustment group according to the positive electrode regeneration optimality function to obtain a positive electrode regeneration optimization result, and performing positive electrode material remanufacturing in combination with the positive electrode waste powder.

[0037] Specifically, the cathode regeneration optimization function measures the regeneration optimization (overall suitability) of different process parameter combinations. It is a weighted exponential function based on regenerative electrochemical performance, impurity effects, and structural defects. The cathode regeneration optimization result is the optimal remanufacturing process parameter combination obtained after multiple iterations and optimization comparisons.

[0038] First, based on a predefined positive electrode regeneration optimality function, the optimality score of each process parameter combination in the first positive electrode regeneration adjustment group is calculated. Subsequently, a genetic algorithm is used to prioritize the best-performing combination (the first optimization strategy) based on the optimality score. The first positive electrode regeneration adjustment group is then multiplied (i.e., replicated, crossover, mutated, etc.) to form a second positive electrode regeneration adjustment group. The second adjustment group is then optimized to maximize the positive electrode regeneration optimality, resulting in a second optimization strategy. The first and second strategies are compared to select the optimal positive electrode regeneration optimization strategy. This iterative multiplication process is repeated until the maximum number of iterations is reached, ultimately resulting in the optimal positive electrode regeneration optimization result. This optimal result is combined with the positive electrode waste powder to remanufacture the positive electrode material. This involves adding the positive electrode waste powder and a repair agent, such as lithium hydroxide, to a hydrothermal reactor. Parameters such as reaction temperature, pressure, time, and solution concentration are controlled based on the optimal result, allowing the hydrothermal reaction to recrystallize the positive electrode waste powder under the hydrothermal conditions, repairing its crystal structure and replenishing lithium, thereby producing a positive electrode material for lithium-ion batteries.

[0039] This step obtains the optimal solution for the positive electrode material remanufacturing process through multiple rounds of intelligent optimization, significantly improving the comprehensive performance of the regenerated products and the consistency of repair and remanufacturing, reducing the cost of manual trial and error, and ensuring the optimal adaptability of the remanufacturing solution.

[0040] Furthermore, step S100 includes:

[0041] Step S110: searching for precursor synthesis schemes based on the powder detection data to obtain a precursor synthesis search space.

[0042] Step S120: performing a hydrothermal reaction scheme search based on the powder detection data to obtain a hydrothermal reaction search space.

[0043] Step S130: searching for a high-temperature sintering solution based on the powder detection data to obtain a high-temperature sintering search space, and combining the precursor synthesis search space and the hydrothermal reaction search space to generate the positive electrode regeneration search space.

[0044] Specifically, based on the powder detection data of the cathode waste powder, a search is performed in the process database to screen the historical precursor synthesis schemes, historical hydrothermal reaction schemes and historical high-temperature sintering schemes corresponding to the powder detection data. Among them, the process database is a collection of a large amount of experimental and production process conditions-product performance data accumulated during the recycling and preparation of lithium-ion batteries, including but not limited to: different precursor synthesis schemes (temperature, pH, reaction time, etc.) and the corresponding product composition, morphology, purity, etc.; different hydrothermal reaction conditions (temperature, time, additive concentration, etc.) and the corresponding morphology and crystal form of the regenerated products; different sintering parameters (temperature, atmosphere, time) and the corresponding final electrochemical properties of the cathode materials. The data in the process database comes from historical production data, literature, experimental or simulation data, etc. The specific search process is as follows:

[0045] Based on the powder test data of cathode waste powder, a matching search is performed in the process database, and historical precursor synthesis scheme records under the same powder test data are screened to form a precursor synthesis search space containing several matching precursor synthesis schemes. Among them, the precursor synthesis scheme refers to the precursor mixing system scheme that can be used for hydrothermal repair obtained through precursor ratio and pretreatment, including precursor ratio, solvent type, and auxiliary agent addition order.

[0046] Using powder test data as an index, the corresponding historical hydrothermal reaction parameter combinations are searched in the process database to obtain several hydrothermal reaction schemes, forming a hydrothermal reaction search space. The hydrothermal reaction scheme refers to the combination of process parameters for the hydrothermal treatment of cathode waste powder after precursor pretreatment, including temperature, time, reaction pH value, and additive system.

[0047] Using powder detection data as an index, the corresponding historical high-temperature sintering conditions are further retrieved from the process database to obtain several high-temperature sintering schemes and a high-temperature sintering search space. Among them, the high-temperature sintering scheme refers to the final high-temperature solid-phase treatment process parameters for the regeneration of the positive electrode material, including temperature, time, and atmosphere type. Subsequently, the precursor synthesis search space, hydrothermal reaction search space, and high-temperature sintering search space are multi-dimensionally fused to generate a positive electrode regeneration search space. This space supports subsequent global confidence mining and multi-round optimization. By introducing independent search spaces for precursor synthesis, hydrothermal reaction, and high-temperature sintering, the positive electrode regeneration search space is further formed, which greatly enriches the diversity and selectivity of process parameter combinations, ensuring a high degree of flexibility and adjustability in the remanufacturing process, thereby achieving higher positive electrode regeneration performance and stable process adaptability.

[0048] Furthermore, step S200 includes:

[0049] Step S210: Random number is generated according to the precursor synthesis search space, and first synthesis matching scheme, second synthesis matching scheme, and Kth synthesis matching scheme are obtained, where K is a positive integer.

[0050] Step S220: Global confidence calculation is performed on the first synthesis matching scheme, the second synthesis matching scheme, and the Kth synthesis matching scheme, and first global confidence, second global confidence, and Kth global confidence are obtained.

[0051] Step S230: Global confidence maximization optimization is performed on the first synthesis matching scheme, the second synthesis matching scheme, and the Kth synthesis matching scheme according to the first global confidence, the second global confidence, and the Kth global confidence, and a global confidence precursor synthesis scheme is obtained.

[0052] Step S240: Global confidence maximization optimization is continuously performed on the hydrothermal reaction search space and the high-temperature sintering search space, and a global confidence hydrothermal reaction scheme and a global confidence high-temperature sintering scheme are obtained, which are combined with the global confidence precursor synthesis scheme to generate the positive electrode regeneration control scheme.

[0053] Specifically, in the precursor synthesis search space obtained after the process database retrieval, a plurality of matching precursor synthesis schemes are contained. In order to facilitate subsequent global confidence mining and comparative analysis, the schemes are randomly numbered and marked as first synthesis matching scheme, second synthesis matching scheme, and Kth synthesis matching scheme, where K is a positive integer. The numbering can be assisted by a hash function or a random permutation algorithm to ensure the fairness and randomness of the scheme arrangement.

[0054] For each numbered synthesis matching scheme, global confidence calculation is performed in the precursor synthesis search space, and the global confidence of the scheme is determined according to the appearance frequency of the synthesis matching scheme and its respective process parameters in the entire precursor synthesis search space, and first global confidence, second global confidence, and Kth global confidence are obtained. According to the global confidence values, all synthesis matching schemes are optimized using the maximization criterion, and the synthesis matching scheme with the highest global confidence is selected as the global confidence precursor synthesis scheme.

[0055] Using the same global confidence calculation and maximization optimization idea, global confidence calculation and optimization are continuously performed on the hydrothermal reaction search space and the high-temperature sintering search space, and a global confidence hydrothermal reaction scheme and a global confidence high-temperature sintering scheme are obtained. Finally, the global confidence precursor synthesis scheme, the global confidence hydrothermal reaction scheme, and the global confidence high-temperature sintering scheme are integrated to generate a complete positive electrode regeneration control scheme.

[0056] Further, step S220 includes:

[0057] Step S221: performing support analysis on the first synthetic matching solution according to the positive electrode regeneration search space to obtain a first synthetic global support.

[0058] Step S222: performing support analysis on each precursor synthesis parameter in the first synthesis matching scheme according to the positive electrode regeneration search space to obtain multiple synthesis parameter support degrees, and taking the sum of the multiple synthesis parameter support degrees as the first synthesis trigger degree.

[0059] Step S223: Calculate a ratio between the first synthetic triggering degree and the first synthetic global support degree to generate the first global confidence degree.

[0060] Specifically, in the cathode regeneration search space, the number of occurrences of the combination conditions of the first synthetic matching solution is counted as the first synthetic global support. This support value indicates the activity and frequency of occurrence of the synthetic solution in historical data and simulation results, and can be considered the basic credibility of the solution. Next, a support analysis is performed on each precursor synthesis parameter in the first synthetic matching solution. The number of occurrences of each precursor synthesis parameter in the cathode regeneration search space is counted to obtain multiple synthesis parameter support values. These support values ​​are summed to obtain the first synthetic triggering degree. The ratio of the first synthetic triggering degree to the first synthetic global support value is calculated to quantify the global advantage of the matching solution in the entire multidimensional process database and generate the first global confidence value. The first global confidence value = first synthetic triggering degree / first synthetic global support value. The larger this ratio, the more frequently the multidimensional parameters of the first synthetic matching solution appear in the search space, the more stable they are, and thus the higher the confidence value. Similarly, the same analysis and confidence calculation are performed for the other matching solutions (the second through the Kth synthetic matching solutions) using this method.

[0061] Further, such as Figure 2 As shown, the regenerated positive electrode evaluation mechanism in step S300 includes:

[0062] Step 310: predicting the characteristics of the regenerated positive electrode according to the positive electrode regeneration control scheme and the positive electrode waste powder to obtain the structural characteristics, electrochemical characteristics and impurity characteristics of the regenerated positive electrode.

[0063] Step 320: activating a regenerated positive electrode evaluation model, wherein the regenerated positive electrode evaluation model includes an input layer, a structural defect evaluation layer, an electrochemical performance evaluation layer, an impurity impact evaluation layer, and an output layer.

[0064] Step 330: Input the structural characteristics of the regenerated positive electrode into the structural defect evaluation layer to obtain a structural defect evaluation coefficient.

[0065] Step 340: Input the electrochemical characteristics of the regenerated positive electrode into the electrochemical performance evaluation layer to obtain an electrochemical performance evaluation coefficient.

[0066] Step 350: Input the regenerated positive electrode impurity characteristics into the impurity impact evaluation layer to obtain the impurity impact evaluation coefficient, and combine the structural defect evaluation coefficient and the electrochemical performance evaluation coefficient to generate the regenerated positive electrode evaluation result.

[0067] Specifically, the regenerated cathode characteristic prediction is a process based on the cathode regeneration control scheme and the characteristics of the cathode waste powder, using machine learning and other technologies to predict the structural, electrochemical, and impurity characteristics of the regenerated cathode material, providing input data for the regenerated cathode evaluation model. The regenerated cathode evaluation model is composed of an input layer, a structural defect evaluation layer, an electrochemical performance evaluation layer, an impurity impact evaluation layer, and an output layer. It is used to comprehensively evaluate the performance and quality of the regenerated cathode material and generate the regenerated cathode evaluation results.

[0068] Predicting the characteristics of regenerated cathodes based on the cathode regeneration control scheme and cathode scrap powder involves collecting a large amount of known cathode material data, including structural characteristics (such as crystal structure parameters and particle size distribution), electrochemical characteristics (such as charge-discharge curves, specific capacity, and cycle life), and impurity characteristics (such as impurity type and content). This data can be obtained from experimental tests, literature reports, or industrial production records. Corresponding data for the cathode regeneration control scheme, including precursor synthesis parameters, hydrothermal reaction conditions, and high-temperature sintering parameters, are also collected. The collected data undergoes preprocessing operations such as cleaning and normalization to eliminate dimensionality differences and data noise, making it suitable for the input requirements of a machine learning model. An appropriate machine learning algorithm, such as a neural network or support vector machine, is selected to construct a predictive model. For example, a neural network is constructed, consisting of an input layer, hidden layers, and an output layer. The preprocessed data is divided into a training set and a test set. The neural network is trained using the training set. By adjusting parameters such as the network weights and bias, the model is able to accurately predict the structural, electrochemical, and impurity characteristics of the regenerated cathode based on the cathode regeneration control scheme and cathode scrap powder inputs. During the training process, an error backpropagation algorithm is used to continuously optimize the model's performance to minimize the discrepancy between predicted and actual values. The trained model is validated and evaluated using a test set to ensure good generalization and prediction accuracy, thereby obtaining a regenerated positive electrode characteristic prediction model. The current positive electrode regeneration control scheme and powder inspection data from scrap positive electrode powder are input into the trained regenerated positive electrode characteristic prediction model. Based on the characteristics of the input data, the model uses internal neuron calculations and activation functions to predict the structural characteristics (grain size, crystal plane orientation, crystal defect density, etc.), electrochemical characteristics (theoretical specific capacity, expected cycle life, etc.), and impurity characteristics (residual impurity element content and its distribution characteristics in the material, etc.) of the regenerated positive electrode.

[0069] Activate the regenerated positive electrode evaluation model. The regenerated positive electrode evaluation model adopts a multi-layer neural network structure, including an input layer, a structural defect evaluation layer, an electrochemical performance evaluation layer, an impurity impact evaluation layer and an output layer. The input layer is used to receive data such as the structural characteristics, electrochemical characteristics and impurity characteristics of the regenerated positive electrode; the structural defect evaluation layer is used to analyze the structural characteristics of the regenerated positive electrode, evaluate its crystal structure integrity, defect degree, etc., and output the structural defect evaluation coefficient; the electrochemical performance evaluation layer evaluates the electrochemical characteristics of the regenerated positive electrode, such as charge and discharge capacity, cycle stability, etc., and generates an electrochemical performance evaluation coefficient; the impurity impact evaluation layer analyzes the impurity characteristics of the regenerated positive electrode, determines the degree of influence of impurities on performance, and gives an impurity impact evaluation coefficient; the output layer integrates the structural defect evaluation coefficient, electrochemical performance evaluation coefficient and impurity impact evaluation coefficient to generate the final regenerated positive electrode evaluation result. The regenerated cathode evaluation model is trained based on a large amount of data from regenerated cathode materials from different batches. This data includes, but is not limited to, historical production data, publicly available literature, and experimental and simulation data. It covers the regenerated cathode characteristics (structural, electrochemical, and impurity characteristics) of each batch of regenerated cathode material, as well as the structural defect evaluation coefficients, electrochemical performance evaluation coefficients, and impurity impact evaluation coefficients of the corresponding regenerated cathode materials, annotated based on expert experience or a rule engine. After preprocessing, this historical data is divided into a training set (80%), a validation set (10%), and a test set (10%) according to predetermined proportions. The regenerated cathode evaluation model is trained using the training set. Training data is fed into the model batch by batch, and the output is calculated using forward propagation. A loss function is then used to calculate the error between the predicted and actual values. Backpropagation is used to update the model's weights and bias parameters using the Adam optimizer based on the gradient of the loss function. During training, the model is validated using the validation set to monitor performance and determine the optimal model weights. A final performance evaluation is conducted on the test set to obtain a regenerated cathode evaluation model that meets the expected performance requirements. Finally, the trained regenerated positive electrode evaluation model is integrated into the regenerated positive electrode evaluation mechanism to output the comprehensive evaluation results of the regenerated positive electrode material in real time, supporting the adjustment and feedback optimization of the subsequent positive electrode regeneration control scheme.

[0070] The input layer of the regenerated positive electrode evaluation model receives the regenerated positive electrode structural characteristics, regenerated positive electrode electrochemical characteristics, and regenerated positive electrode impurity characteristics output by the regenerated positive electrode characteristic prediction model and transmits the regenerated positive electrode structural characteristic data to the structural defect evaluation layer. Neurons in the structural defect evaluation layer perform weighted summation and nonlinear transformation on the structural characteristic data based on preset weights and activation functions to analyze information such as crystal structure integrity, defect type, and severity. For example, by analyzing the degree of deviation of crystal structure parameters and the intensity of defect characteristic peaks, a structural defect evaluation coefficient is calculated. This structural defect evaluation coefficient reflects the severity of structural defects in the regenerated positive electrode material; lower values ​​indicate a more complete structure and fewer defects. Similarly, the input layer transmits the regenerated positive electrode electrochemical characteristic data to the electrochemical performance evaluation layer. Neurons in the electrochemical performance evaluation layer process the electrochemical characteristic data and evaluate the electrochemical performance of the regenerated positive electrode material based on indicators such as charge-discharge capacity and cycle stability. For example, by comparing the predicted charge-discharge curve with the ideal curve and calculating the cycle life decay rate, the electrochemical performance evaluation coefficient is calculated; a higher coefficient indicates better electrochemical performance. The input layer transmits the regenerated positive electrode impurity characteristic data to the impurity impact evaluation layer. Neurons in the impurity impact evaluation layer analyze information such as the type and content of impurities, determine the potential impact of impurities on the positive electrode material performance, and calculate the impurity impact evaluation coefficient. The lower the coefficient value, the smaller the impact of the impurity on performance. Finally, the output layer receives and outputs the structural defect evaluation coefficient, electrochemical performance evaluation coefficient, and impurity impact evaluation coefficient, generating the regenerated positive electrode evaluation result. This regenerated positive electrode evaluation result, including the impurity impact evaluation coefficient, structural defect evaluation coefficient, and electrochemical performance evaluation coefficient, is used to determine whether the regenerated positive electrode control plan meets the regenerated positive electrode evaluation constraints.

[0071] Further, such as Figure 3 As shown, step S400 includes:

[0072] Step S410: performing regulation constraint analysis according to the positive electrode regeneration search space to establish a multi-node regulation constraint domain.

[0073] Step S420: Randomly adjust the positive electrode regeneration control scheme according to the multi-node adjustment constraint domain to obtain a first positive electrode regeneration adjustment space.

[0074] Step S430: Evaluate the positive electrode regeneration adjustment first space according to the regenerated positive electrode evaluation mechanism to obtain a regenerated positive electrode evaluation distribution.

[0075] Step S440: Based on the regenerated positive electrode evaluation distribution, the first positive electrode regeneration adjustment space is optimized and screened according to the regenerated positive electrode evaluation constraint to generate the first positive electrode regeneration adjustment group.

[0076] Specifically, based on the multidimensional nature of the cathode regeneration search space, a multi-node adjustment constraint analysis is performed on the precursor synthesis parameters, hydrothermal reaction parameters, and high-temperature sintering parameters. Within the search spaces for each process node (precursor synthesis search space, hydrothermal reaction search space, and high-temperature sintering search space), solutions with a global confidence level greater than or equal to a predetermined value are screened to form three corresponding optimal spaces. The adjustable parameter ranges for each process node are then determined based on the parameter values ​​of historical solutions in each optimal space, thereby constructing a multi-node adjustment constraint domain. This multi-node adjustment constraint domain covers the parameter value ranges of the control schemes for multiple key steps in the cathode regeneration process (precursor synthesis, hydrothermal reaction, high-temperature sintering, etc.), and is used to guide the adjustment of the cathode regeneration control scheme.

[0077] Under the guidance of the multi-node regulation constraint domain, random perturbation regulation is performed on the existing positive electrode regeneration control schemes (including precursor synthesis scheme, hydrothermal reaction scheme, and high-temperature sintering scheme): multi-dimensional sampling methods such as Monte Carlo random sampling or Latin hypercube sampling are used to generate a series of possible regulation strategy sets, forming the first positive electrode regeneration regulation space that contains multiple feasible positive electrode regeneration control schemes. Each positive electrode regeneration control scheme is a complete process parameter combination.

[0078] Based on the positive electrode regeneration adjustment first space generated by the above-mentioned random adjustment, the regenerated positive electrode evaluation mechanism is called to perform a systematic evaluation to obtain the predicted distribution of the regenerated positive electrode performance: each positive electrode regeneration control scheme is input into the regenerated positive electrode evaluation model, and its corresponding structural defect evaluation coefficient, electrochemical performance evaluation coefficient and impurity influence evaluation coefficient are calculated to generate the regenerated positive electrode evaluation result of each positive electrode regeneration control scheme, thereby obtaining the regenerated positive electrode evaluation distribution, that is, the distribution map of the performance evaluation results of each positive electrode regeneration control scheme in the first space.

[0079] Using the regenerated positive electrode evaluation constraint as a screening threshold, positive electrode regeneration control schemes that do not meet the regenerated positive electrode evaluation constraint are eliminated. The remaining positive electrode regeneration control schemes form the first positive electrode regeneration adjustment group. This group provides an initial set of high-quality strategies for subsequent positive electrode regeneration optimization, ensuring that the remanufacturing process has an efficient and feasible adjustment path.

[0080] Furthermore, step S410 includes:

[0081] The precursor synthesis search space is screened according to a predetermined global confidence level to establish a precursor synthesis preferred space; the hydrothermal reaction search space is screened according to the predetermined global confidence level to establish a hydrothermal reaction preferred space; the high-temperature sintering search space is screened according to the predetermined global confidence level to establish a high-temperature sintering synthesis preferred space; adaptive variable trigger interval analysis is performed based on the precursor synthesis preferred space, the hydrothermal reaction preferred space and the high-temperature sintering synthesis preferred space to generate the multi-node adjustment constraint domain.

[0082] Specifically, based on the global confidence distribution of the precursor synthesis search space in the positive electrode regeneration search space, screening is performed according to a predetermined global confidence threshold (for example, greater than 80%), retaining only those precursor synthesis schemes with high global confidence to construct a precursor synthesis optimization space. Each scheme in the precursor synthesis optimization space has been proven to have higher regenerative material compatibility and process feasibility in historical production or simulation results. Using the same predetermined global confidence screening criteria, the hydrothermal reaction search space is screened, retaining high-confidence hydrothermal reaction process parameter combinations to form a hydrothermal reaction optimization space, including high-quality parameter combinations such as temperature, reaction time, and additive concentration under hydrothermal reaction conditions. Continuing in the high-temperature sintering search space, based on the predetermined global confidence, high-temperature sintering schemes with a global confidence greater than or equal to the predetermined global confidence are screened to obtain a high-temperature sintering synthesis optimization space.

[0083] The above-mentioned optimal spaces for precursor synthesis, hydrothermal reaction, and high-temperature sintering are multi-dimensionally integrated, and adaptive variable trigger interval analysis is performed to determine the key parameter value ranges of the control schemes for different process nodes (precursor synthesis, hydrothermal reaction, and high-temperature sintering), generating a multi-node adjustment constraint domain. An example of a multi-node adjustment constraint domain is shown in Table 1:

[0084] Table 1 Example of multi-node adjustment constraint domain

[0085]

[0086] By performing global confidence screening on the three search spaces of precursor synthesis, hydrothermal reaction, and high-temperature sintering, an optimal space is established, and a multi-node adjustment constraint domain is further constructed through adaptive variable trigger interval analysis. This effectively narrows the search scope of the positive electrode regeneration control scheme, focusing on the scheme combination that is more likely to obtain high-performance regenerated positive electrode materials, improving the efficiency of subsequent adjustment screening, and reducing unnecessary experimental and computational costs.

[0087] Furthermore, step S500 includes:

[0088] Step S510: constructing the positive electrode regeneration optimality function, wherein the positive electrode regeneration optimality function is: Among them, CRS represents the quality of positive electrode regeneration, exp represents the exponential function with natural constant as the base, CEX represents the normalized electrochemical performance evaluation coefficient, EW, KW and PW represent the predetermined weight conditions, the sum of EW, KW and PW is 1, CKX represents the normalized impurity impact evaluation coefficient, and CPX represents the normalized structural defect evaluation coefficient.

[0089] Step S520: maximizing the positive electrode regeneration fitness of the first positive electrode regeneration adjustment group according to the positive electrode regeneration fitness function to obtain a first positive electrode regeneration optimization strategy.

[0090] Step S530: breeding optimization of the first positive electrode regeneration adjustment group according to the regenerated positive electrode evaluation mechanism and the regenerated positive electrode evaluation constraint to obtain a second positive electrode regeneration adjustment group.

[0091] Step S540: maximizing the positive electrode regeneration fitness of the second positive electrode regeneration adjustment group according to the positive electrode regeneration fitness function to obtain a second positive electrode regeneration optimization strategy.

[0092] Step S550: comparing the positive electrode regeneration fitness according to the first positive electrode regeneration optimization strategy and the second positive electrode regeneration optimization strategy to obtain a current optimal regeneration optimization strategy.

[0093] Step S560: based on the positive electrode regeneration fitness function, continuing to iteratively breed and optimize the current optimal regeneration optimization strategy according to the regenerated positive electrode evaluation mechanism and the regenerated positive electrode evaluation constraint until the positive electrode regeneration optimization result that meets the number of iterations is obtained.

[0094] Specifically, in order to objectively measure the pros and cons of different positive electrode regeneration adjustment groups, a positive electrode regeneration fitness function is constructed Wherein: CRS represents the positive electrode regeneration fitness, the higher the value, the better the regeneration process scheme. exp represents the exponential function with natural constant as the base, which is used to amplify the differentiated results. CEX represents the normalized electrochemical performance evaluation coefficient. EW, KW and PW represent predetermined weight conditions, the sum of EW, KW and PW is 1, which can be set according to actual needs, for example, EW=0.5, KW=0.3, PW=0.2. CKX represents the normalized impurity influence evaluation coefficient. CPX represents the normalized structure defect evaluation coefficient. This function can comprehensively quantify the evaluation of the positive electrode regeneration control scheme in the first positive electrode regeneration adjustment group in terms of regeneration electrochemical performance, impurity influence and structure defect influence.

[0095] In the first positive electrode regeneration adjustment group, based on the above-mentioned positive electrode regeneration fitness function, the positive electrode regeneration fitness value of each positive electrode regeneration control scheme is calculated by exponential weighting, and the positive electrode regeneration control scheme with the maximum positive electrode regeneration fitness value is identified to form the first positive electrode regeneration optimization strategy. This strategy has the best comprehensive performance in terms of electrochemical performance, impurity control and structure integrity, etc.

[0096] A genetic algorithm is used to perform crossover and mutation operations on the first positive electrode regeneration adjustment group based on the regeneration positive electrode evaluation mechanism and regeneration positive electrode evaluation constraints to generate a second positive electrode regeneration adjustment group, achieving multi-dimensional evolutionary optimization of the regeneration process. Similarly, the positive electrode regeneration fitness function is used to maximize the regeneration fitness of the second positive electrode regeneration adjustment group, and the second positive electrode regeneration optimization strategy with the largest positive electrode regeneration fitness value in the second positive electrode regeneration adjustment group is selected.

[0097] The first positive electrode regeneration optimization strategy is compared with the second positive electrode regeneration optimization strategy in terms of positive electrode regeneration quality. The strategy with the higher quality is selected as the current winning regeneration optimization strategy, providing the optimal starting point for the next round of reproduction iteration. Based on the current winning regeneration optimization strategy, iterative reproduction optimization is continued under the regeneration positive electrode evaluation mechanism and the regeneration positive electrode evaluation constraints. Each round of iteration is based on the first and second group optimization strategies. The reproduction optimization, positive electrode regeneration quality maximization optimization, and quality comparison operations are repeated until the predetermined number of iterative reproductions is reached, and the positive electrode regeneration optimization result is finally obtained.

[0098] Furthermore, step S530 includes:

[0099] Step S531: performing a reproduction number allocation on the first positive electrode regeneration adjustment group according to a predetermined reproduction number to obtain a first reproduction number allocation result.

[0100] Step S532: Based on the first reproduction number allocation result, the first positive electrode regeneration adjustment group is adjusted according to the multi-node adjustment constraint domain to obtain a second positive electrode regeneration adjustment space.

[0101] Step S533: Based on the regenerative positive electrode evaluation mechanism, the second positive electrode regeneration adjustment space is optimized and analyzed according to the regenerative positive electrode evaluation constraint to generate the second positive electrode regeneration adjustment group.

[0102] Specifically, in the process of performing multiplication optimization on the first positive regeneration adjustment group to obtain the second positive regeneration adjustment group, first, based on the predetermined multiplication number allocation mechanism, the multiplication number of each positive regeneration control scheme in the first positive regeneration adjustment group is allocated to obtain the first multiplication number allocation result. The specific process is as follows: the predetermined multiplication number is set to M (for example, M=100) as the total number of positive regeneration control schemes to be generated in this round of iteration. For each positive regeneration adjustment scheme in the first positive regeneration adjustment group, its positive regeneration optimality value is calculated. (i=1, 2, ..., N; N is the total number of solutions of the first positive electrode regeneration adjustment group). Calculate the total optimality sum of the first positive electrode regeneration adjustment group, For each positive electrode regeneration regulation scheme, its reproduction number is allocated based on its optimal proportion The round function is used to round off the results, ensuring that each cathode regeneration control solution receives a number of propagations proportional to its optimality. This allocation method ensures that solutions with greater optimal cathode regeneration receive a greater number of propagations, increasing the proportion of excellent solutions in the next generation and strengthening the evolutionary optimization capabilities of the regenerative cathode process.

[0103] Based on the first reproduction number allocation result and combined with the multi-node regulation constraint domain, the first positive electrode regeneration regulation group is regenerated and generated: for each positive electrode regeneration control scheme, according to its reproduction number, the multi-node regulation constraint domain is used to perform parameter perturbations (for example: random perturbations within the range of ±5°C for temperature regulation, perturbations within the range of ±0.1 for pH regulation, etc.), to generate new positive electrode regeneration control schemes and construct the second positive electrode regeneration regulation space, in which the number of schemes is equal to the predetermined reproduction number.

[0104] Based on the regenerative positive electrode evaluation mechanism and combined with the regenerative positive electrode evaluation constraints, the second space of positive electrode regeneration regulation is optimized and analyzed: each positive electrode regeneration control scheme in the second space of positive electrode regeneration regulation is evaluated using the regenerative positive electrode evaluation mechanism to determine the regenerative positive electrode evaluation result, and combined with the predetermined global confidence level, the scheme with excellent performance is selected to form the second positive electrode regeneration regulation group, which serves as the input group for further optimization of the positive electrode regeneration optimum maximization.

[0105] The above steps generate new strategies through optimal proportion reproduction and parameter perturbation regeneration, and complete the generation of the second positive electrode regeneration adjustment group in combination with the evaluation constraint mechanism, ensuring that the positive electrode regeneration process continues to evolve towards the direction of the highest optimality.

[0106] In summary, the lithium-ion battery repair and remanufacturing method provided by the embodiment of the present invention has the following beneficial effects:

[0107] By performing multi-stage pretreatment on spent lithium-ion batteries to obtain positive electrode waste powder, a hydrothermal remanufacturing search is conducted based on the powder testing data of the positive electrode waste powder. This constructs a positive electrode regeneration search space, providing accurate and comprehensive data support and a customized search scope for the subsequent remanufacturing process. A global confidence mining optimization search is performed within the positive electrode regeneration search space to determine a positive electrode regeneration control scheme, avoiding local optima and human experience constraints, and ensuring the global optimality of the remanufacturing scheme. A regenerated positive electrode evaluation mechanism is introduced to evaluate the positive electrode regeneration control scheme, obtaining regenerated positive electrode evaluation results. This allows for timely identification of deficiencies and defects in the scheme, ensuring that the performance of the regenerated positive electrode material meets expected requirements, and providing clear direction and goals for subsequent feedback adjustments. If the regenerated positive electrode evaluation results do not meet the regenerated positive electrode evaluation constraints, the positive electrode regeneration control scheme is feedback-adjusted based on the regenerated positive electrode evaluation mechanism and the positive electrode regeneration search space to obtain a first positive electrode regeneration adjustment group that meets the regenerated positive electrode evaluation constraints, enhancing the adaptability and flexibility of the regeneration scheme. The first positive electrode regeneration control group is iteratively propagated and optimized based on the positive electrode regeneration optimality function to obtain a positive electrode regeneration optimization result, further refining and improving the performance of the positive electrode regeneration control scheme. Finally, the positive electrode waste powder is combined with the positive electrode material to remanufacture the positive electrode material, forming a complete and sustainable lithium-ion battery repair and remanufacturing process chain.

[0108] Overall, the embodiment of the present invention establishes a search space driven by multi-level preprocessing and powder detection data, combines global confidence mining to determine the positive electrode regeneration control scheme, and then introduces a regenerated positive electrode evaluation mechanism to form a closed-loop feedback regulation. It also uses the optimal fitness function to perform group iterative reproduction optimization, ultimately achieving efficient adaptation of positive electrode waste powder and dynamic optimization of the remanufacturing process. The entire solution not only solves the performance fluctuation problem caused by process curing and poor adaptability in the existing technology, but also improves the quality stability and performance consistency of the regenerated positive electrode material, realizes intelligent and adaptive lithium-ion battery repair and remanufacturing, and promotes the large-scale application and green and sustainable development of lithium-ion battery repair and remanufacturing.

[0109] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for repairing and remanufacturing a lithium-ion battery, characterized in that: The method comprises: Perform multi-stage pretreatment on waste lithium-ion batteries to obtain positive electrode waste powder, and conduct hydrothermal repair and remanufacturing search based on powder detection data of the positive electrode waste powder to build a positive electrode regeneration search space; Performing global confidence mining optimization based on the positive electrode regeneration search space to determine a positive electrode regeneration control scheme; Introducing a regenerated positive electrode evaluation mechanism to evaluate the positive electrode regeneration control scheme and obtain a regenerated positive electrode evaluation result; If the regenerated positive electrode evaluation result does not satisfy the regenerated positive electrode evaluation constraint, feedback adjustment is performed on the positive electrode regeneration control scheme according to the regenerated positive electrode evaluation mechanism and the positive electrode regeneration search space to obtain a first positive electrode regeneration adjustment group that satisfies the regenerated positive electrode evaluation constraint, including: Performing regulation constraint analysis based on the positive electrode regeneration search space to establish a multi-node regulation constraint domain; Randomly adjusting the positive electrode regeneration control scheme according to the multi-node adjustment constraint domain to obtain a first positive electrode regeneration adjustment space; Evaluating the positive electrode regeneration adjustment first space according to the regenerated positive electrode evaluation mechanism to obtain a regenerated positive electrode evaluation distribution; Based on the regenerated positive electrode evaluation distribution, the first positive electrode regeneration adjustment space is optimized and screened according to the regenerated positive electrode evaluation constraint to generate the first positive electrode regeneration adjustment group; Iteratively propagate and optimize the first positive electrode regeneration adjustment group according to the positive electrode regeneration optimality function to obtain a positive electrode regeneration optimization result, and remanufacture the positive electrode material in combination with the positive electrode waste powder; The regenerated positive electrode evaluation mechanism includes: Predicting the characteristics of the regenerated positive electrode according to the positive electrode regeneration control scheme and the positive electrode waste powder to obtain the structural characteristics, electrochemical characteristics and impurity characteristics of the regenerated positive electrode; activating a regenerative positive electrode evaluation model, wherein the regenerative positive electrode evaluation model includes an input layer, a structural defect evaluation layer, an electrochemical performance evaluation layer, an impurity impact evaluation layer, and an output layer; Inputting the structural characteristics of the regenerated positive electrode into the structural defect evaluation layer to obtain a structural defect evaluation coefficient; Inputting the electrochemical characteristics of the regenerated positive electrode into the electrochemical performance evaluation layer to obtain an electrochemical performance evaluation coefficient; Inputting the impurity characteristics of the regenerated positive electrode into the impurity impact evaluation layer to obtain the impurity impact evaluation coefficient, and combining the structural defect evaluation coefficient and the electrochemical performance evaluation coefficient to generate the regenerated positive electrode evaluation result; The step of performing regulation constraint analysis based on the positive electrode regeneration search space to establish a multi-node regulation constraint domain includes: Screening the precursor synthesis search space according to a predetermined global confidence level to establish an optimal precursor synthesis space; screening the hydrothermal reaction search space according to the predetermined global confidence level to establish an optimal hydrothermal reaction space; screening a high-temperature sintering search space according to the predetermined global confidence level to establish a high-temperature sintering synthesis optimal space; Adaptive variable triggering interval analysis is performed based on the precursor synthesis preferred space, the hydrothermal reaction preferred space, and the high-temperature sintering synthesis preferred space to generate the multi-node regulation constraint domain.

2. The method for repairing and remanufacturing a lithium-ion battery according to claim 1, wherein: Iteratively reproducing and optimizing the first positive electrode regeneration adjustment group according to the positive electrode regeneration optimality function to obtain a positive electrode regeneration optimization result, including: Construct the positive electrode regeneration optimality function, which is: ； Among them, CRS represents the positive electrode regeneration quality, exp represents the exponential function with the natural constant as the base, CEX represents the normalized electrochemical performance evaluation coefficient, EW, KW and PW represent the predetermined weight conditions, the sum of EW, KW and PW is 1, CKX represents the normalized impurity impact evaluation coefficient, and CPX represents the normalized structural defect evaluation coefficient; maximizing the positive electrode regeneration optimum of the first positive electrode regeneration adjustment group according to the positive electrode regeneration optimum function to obtain a first positive electrode regeneration optimization strategy; Performing multiplication and optimization on the first positive electrode regeneration adjustment group according to the regeneration positive electrode evaluation mechanism and the regeneration positive electrode evaluation constraint to obtain a second positive electrode regeneration adjustment group; maximizing the positive electrode regeneration optimum of the second positive electrode regeneration adjustment group according to the positive electrode regeneration optimum function to obtain a second positive electrode regeneration optimization strategy; Comparing the positive electrode regeneration optimization strategy with the first positive electrode regeneration optimization strategy and the second positive electrode regeneration optimization strategy to obtain a current winning regeneration optimization strategy; Based on the positive electrode regeneration optimality function, the current winning regeneration optimization strategy is continuously iteratively optimized according to the regenerative positive electrode evaluation mechanism and the regenerative positive electrode evaluation constraint until the positive electrode regeneration optimization result that satisfies the iterative reproduction times is obtained.

3. The method for repairing and remanufacturing a lithium-ion battery according to claim 2, wherein: The first positive electrode regeneration adjustment group is multiplied and optimized according to the regeneration positive electrode evaluation mechanism and the regeneration positive electrode evaluation constraint to obtain a second positive electrode regeneration adjustment group, including: Allocating the reproduction number of the first positive electrode regeneration adjustment group according to the predetermined reproduction number to obtain a first reproduction number allocation result; Based on the first reproduction number allocation result, adjusting the first positive electrode regeneration adjustment group according to the multi-node adjustment constraint domain to obtain a second positive electrode regeneration adjustment space; Based on the regenerative positive electrode evaluation mechanism, the second positive electrode regeneration adjustment space is optimized and analyzed according to the regenerative positive electrode evaluation constraints to generate the second positive electrode regeneration adjustment group.

4. The method for repairing and remanufacturing a lithium-ion battery according to claim 1, wherein: Based on the powder detection data of the positive electrode waste powder, a hydrothermal repair and remanufacturing search is performed to build a positive electrode regeneration search space, including: Performing a precursor synthesis scheme search based on the powder detection data to obtain a precursor synthesis search space; Performing a hydrothermal reaction scheme search based on the powder detection data to obtain a hydrothermal reaction search space; A high-temperature sintering scheme is retrieved based on the powder detection data to obtain a high-temperature sintering search space, and the positive electrode regeneration search space is generated by combining the precursor synthesis search space and the hydrothermal reaction search space.

5. The method for repairing and remanufacturing a lithium-ion battery according to claim 1, wherein: A global confidence mining optimization is performed based on the positive electrode regeneration search space to determine a positive electrode regeneration control scheme, including: Randomly numbering the precursor synthesis search space to obtain a first synthesis matching solution, a second synthesis matching solution, and a Kth synthesis matching solution, where K is a positive integer; performing global confidence calculations on the first synthetic matching solution, the second synthetic matching solution, ..., the Kth synthetic matching solution, respectively, to obtain a first global confidence, a second global confidence, ..., a Kth global confidence; performing global confidence maximization optimization on the first synthesis matching scheme, the second synthesis matching scheme, ... the Kth synthesis matching scheme according to the first global confidence, the second global confidence, ... the Kth global confidence, to obtain a globally confident precursor synthesis scheme; Continue to perform global confidence maximization optimization on the hydrothermal reaction search space and the high-temperature sintering search space to obtain a global confidence hydrothermal reaction scheme and a global confidence high-temperature sintering scheme, and combine the global confidence precursor synthesis scheme to generate the positive electrode regeneration control scheme.

6. The method for repairing and remanufacturing a lithium-ion battery according to claim 5, wherein: Performing global confidence calculations on the first synthetic matching solution, the second synthetic matching solution, ..., the Kth synthetic matching solution, respectively, including: performing support analysis on the first synthetic matching scheme according to the positive electrode regeneration search space to obtain a first synthetic global support; performing support analysis on each precursor synthesis parameter in the first synthesis matching scheme according to the positive electrode regeneration search space to obtain multiple synthesis parameter support degrees, and taking the sum of the multiple synthesis parameter support degrees as a first synthesis trigger degree; A ratio calculation is performed based on the first synthetic triggering degree and the first synthetic global support degree to generate the first global confidence degree.

7. The method for repairing and remanufacturing a lithium-ion battery according to claim 1, wherein: The multi-stage pretreatment includes discharge disassembly, screening extraction, positive electrode powder desorption and ultrasonic impurity removal.

Images