Photocatalysis-cooperated battery repair management and control method and system

Abstract

The invention provides a photocatalytic synergistic battery repair control method and system, and the method comprises the steps: obtaining the detection information of a to-be-repaired battery, including the molar weight of trivalent Fe at the M1 site, the battery model, the dielectric constant of electrolyte and the temperature of a repair environment; judging whether repairing is needed or not and calculating the molar weight of to-be-reduced trivalent Fe; counting the historical experiment high-energy electron number of the matching condition; randomly configuring a plurality of groups of photocatalytic cooperative parameters and generating a simulated high-energy electron number through a simulator; and performing parameter optimization by taking the deviation between a simulation result and historical data as a fitness parameter, and obtaining a photocatalytic cooperative target parameter for battery repair control. And through comparison of historical data and a simulation result, photocatalysis cooperation parameters are optimized, the ferric iron reduction process is accurately controlled, the lithium ion battery repair efficiency is improved, and the service life of the battery is prolonged.

Description

Technical Field

[0001] The present invention relates to the technical field of battery repair, and particularly to a battery repair control method and system with photocatalytic synergy. Background Art

[0002] With the rapid development of electric vehicles and portable electronic devices, lithium-ion batteries, as key energy supply devices, have received extensive attention. However, during long-term cyclic use of lithium-ion batteries, especially lithium iron phosphate batteries, there will be a serious problem of capacity attenuation. Research shows that one of the main reasons for this attenuation is the structural change caused by the migration of iron ions. During the use of the battery, divalent Fe at the M1 site is oxidized to trivalent Fe. Since the migration energy barrier of trivalent Fe is relatively high, it is difficult for it to return to the FeO6 octahedral site, and at the same time, it hinders the re-insertion of Li+, ultimately resulting in battery capacity attenuation.

[0003] Currently, the repair technologies for lithium-ion batteries mainly focus on heat treatment, electrochemical assistance, etc., but these methods have disadvantages such as high energy consumption and low efficiency. Recent research has proposed a method of ultraviolet photocatalysis, using ultraviolet light to excite high-energy electrons to reduce trivalent Fe to divalent Fe, thereby reducing the migration energy barrier of Fe atoms, promoting the return of Fe atoms to the FeO6 octahedral site, and at the same time creating vacancies for the re-insertion of Li+, to improve the lifespan of lithium-ion batteries. However, the influence of ultraviolet light on the repair of lithium-ion batteries is multi-stage and non-linear, and its repair effect is related to many factors, such as the addition amount of DA polymer, ultraviolet light band, ultraviolet light irradiation intensity, ultraviolet light irradiation frequency, aqueous solution temperature, and annealing temperature curve, etc. It is difficult to accurately quantify and determine a specific repair control scheme, resulting in inaccurate parameter settings of the existing photocatalytic synergy parameters in practical applications, unable to fully exert the potential of photocatalytic repair, and ultimately affecting the battery repair efficiency and battery lifespan. Summary of the Invention

[0004] Aiming at the technical problem that the inaccurate setting of photocatalytic synergy parameters used in battery repair in the prior art leads to low repair efficiency and short battery lifespan of lithium-ion batteries, the present invention provides a battery repair control method and system with photocatalytic synergy to solve this problem.

[0005] The technical solution of the present invention to solve the above technical problems is as follows: In a first aspect, the present invention provides a method for controlling battery repair through photocatalytic synergy, including: obtaining detection information of a battery to be repaired, where the detection information of the battery to be repaired includes the molar amount of trivalent Fe at the M1 site, the battery model, the electrolyte dielectric constant, and the repair environment temperature; when the molar amount of trivalent Fe at the M1 site is greater than the target molar amount of trivalent Fe at the M1 site, calculating the molar amount difference to obtain the molar amount of trivalent Fe to be reduced; statistically counting the historical experimental high-energy electron numbers that satisfy the molar amount of trivalent Fe to be reduced, the repair environment temperature, and the electrolyte dielectric constant; randomly configuring multiple sets of photocatalytic synergy parameters, processing them through a high-energy electron excitation simulator bound to the battery model to generate multiple simulated high-energy electron numbers; using the deviation between the simulated high-energy electron numbers and the historical experimental high-energy electron numbers as a fitness parameter, combining the multiple simulated high-energy electron numbers and the multiple sets of photocatalytic synergy parameters, performing optimization of photocatalytic synergy parameters, and obtaining photocatalytic synergy target parameters for photocatalytic control of battery repair.

[0006] In a second aspect, the present invention provides a system for controlling battery repair through photocatalytic synergy, including: a data acquisition module for obtaining detection information of a battery to be repaired, where the detection information of the battery to be repaired includes the molar amount of trivalent Fe at the M1 site, the battery model, the electrolyte dielectric constant, and the repair environment temperature; a difference calculation module for calculating the molar amount difference to obtain the molar amount of trivalent Fe to be reduced when the molar amount of trivalent Fe at the M1 site is greater than the target molar amount of trivalent Fe at the M1 site; a historical data statistics module for statistically counting the historical experimental high-energy electron numbers that satisfy the molar amount of trivalent Fe to be reduced, the repair environment temperature, and the electrolyte dielectric constant; a simulation processing module for randomly configuring multiple sets of photocatalytic synergy parameters and generating multiple simulated high-energy electron numbers through processing by a high-energy electron excitation simulator bound to the battery model; an optimization control module for using the deviation between the simulated high-energy electron numbers and the historical experimental high-energy electron numbers as a fitness parameter, combining the multiple simulated high-energy electron numbers and the multiple sets of photocatalytic synergy parameters, performing optimization of photocatalytic synergy parameters, and obtaining photocatalytic synergy target parameters for photocatalytic control of battery repair.

[0007] The beneficial effects of the present invention are: Obtain the detection information of the battery to be repaired. Among them, the detection information of the battery to be repaired includes the molar amount of trivalent Fe at the M1 site, the battery model, the dielectric constant of the electrolyte, and the repair environment temperature. By detecting the battery to be repaired, obtain the key parameters required for repair, laying a data foundation for accurately judging the repair requirements and formulating a repair plan in the follow-up. When the molar amount of trivalent Fe at the M1 site is greater than the target molar amount of trivalent Fe at the M1 site, calculate the molar amount difference to obtain the molar amount of trivalent Fe to be reduced. By comparing the difference between the actual molar amount of trivalent Fe and the target value, judge whether the battery needs to be repaired and the specific amount of trivalent Fe that needs to be reduced. If it is less than or equal to the target molar amount, reject the repair, indicating that the battery can continue to be used. Statistically count the historical experimental high-energy electron numbers that meet the molar amount of trivalent Fe to be reduced, the repair environment temperature, and the dielectric constant of the electrolyte, providing a reference benchmark for optimizing the photocatalytic parameters in the follow-up. Randomly configure multiple groups of photocatalytic cooperation parameters, process them through a high-energy electron excitation simulator bound to the battery model to generate multiple simulated high-energy electron numbers. By simulating the influence of different photocatalytic cooperation parameters (including the addition amount of DA polymer, ultraviolet light band, ultraviolet light irradiation intensity, ultraviolet light irradiation frequency, aqueous solution temperature, and annealing temperature curve) on the generation of high-energy electrons, generate multiple groups of simulated data, providing multiple possible solutions for parameter optimization. Using the deviation between the simulated high-energy electron number and the historical experimental high-energy electron number as the fitness parameter, combine multiple simulated high-energy electron numbers and multiple groups of photocatalytic cooperation parameters, perform optimization of the photocatalytic cooperation parameters, obtain the target photocatalytic cooperation parameters for photocatalytic control of battery repair, evaluate the advantages and disadvantages of each group of photocatalytic cooperation parameters by calculating the deviation between the simulation results and the historical experimental results, use an optimization algorithm to find the best parameter combination, and finally determine the target photocatalytic cooperation parameters to guide the photocatalytic control in the actual battery repair process.

[0008] Through the reverse derivation method, first determine the required number of high-energy electrons, and then determine the optimal photocatalytic parameters through the method of simulating and comparing with historical data to find the best. It solves the problem that it is difficult to quantitatively determine the multi-stage non-linear characteristics of the influence of ultraviolet light on the repair of lithium-ion batteries, realizes precise control of the repair process, and effectively improves the repair efficiency of lithium-ion batteries and extends the battery life. Description of the Drawings

[0009] Figure 1 It is a schematic flow chart of a method for controlling battery repair with photocatalytic cooperation provided by the present invention; Figure 2 It is a schematic structural diagram of a system for controlling battery repair with photocatalytic cooperation provided by the present invention.

[0010] In the drawings, the components represented by each reference numeral are as follows: Data acquisition module 11, difference calculation module 12, historical data statistics module 13, simulation processing module 14, optimization control module 15. Detailed implementation manners

[0011] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present invention.

[0012] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality of" means two or more, unless otherwise specifically defined.

[0013] In the description of the present invention, the term "for example" is used to mean "serving as an example, illustration, or explanation". Any embodiment described as "for example" in the present invention is not necessarily construed as being more preferred or having more advantages than other embodiments. In order for any person skilled in the art to implement and use the present invention, the following description is given. In the following description, details are set forth for the purpose of explanation. It should be understood that those skilled in the art can recognize that the present invention can be implemented without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid unnecessary details from obscuring the description of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope that conforms to the principles and features disclosed in the present invention.

[0014] Embodiment 1, as Figure 1 shown, the embodiment of the present invention provides a battery repair control method with photocatalytic synergy, including: S100: Obtain the detection information of the battery to be repaired, where the detection information of the battery to be repaired includes the molar amount of trivalent Fe at the M1 site, the battery model, the dielectric constant of the electrolyte, and the repair environment temperature.

[0015] Specifically, first, obtain the detection information of the battery to be repaired. The detection information of the battery to be repaired includes the molar amount of trivalent Fe at the M1 site, the battery model, the dielectric constant of the electrolyte, and the repair environment temperature.

[0016] Among them, the molar amount of trivalent Fe at the M1 site characterizes the degree of disorder in the internal structure of the battery and the level of capacity decay, and is an indicator for evaluating the feasibility of battery repair. For example, the molar content of trivalent Fe ions at the M1 site inside the battery is determined by electrochemical impedance spectroscopy and X-ray absorption fine structure analysis. The battery model is the complete model information of the battery to be repaired, including but not limited to parameters such as manufacturer code, chemical system identification, nominal capacity, and geometric dimensions. Different models of batteries have specific internal structures and material compositions, which play an important restrictive role in the selection of subsequent repair strategies. The dielectric constant of the electrolyte characterizes the ability of the electrolyte to affect charge transport, which is directly related to the transport efficiency of high-energy electrons and the reduction kinetic characteristics of trivalent Fe ions during the photocatalytic process. The real-time dielectric constant of the electrolyte inside the battery can be measured by a dielectric impedance analyzer. Since the electrochemical reaction and the photocatalytic process are sensitive to temperature, the repair ambient temperature directly affects the optimization direction of subsequent photocatalytic cooperation parameters, and the ambient temperature where the current battery is located can be recorded by a precision temperature sensor.

[0017] By obtaining the molar amount of trivalent Fe at the M1 site, the battery model, the dielectric constant of the electrolyte, and the repair ambient temperature, it lays a necessary data foundation for the subsequent steps, thus realizing personalized repair for batteries in different states and improving the pertinence and success rate of the repair.

[0018] S200: When the molar amount of trivalent Fe at the M1 site is greater than the target molar amount of trivalent Fe at the M1 site, calculate the molar amount difference to obtain the molar amount of trivalent Fe to be reduced.

[0019] Specifically, determine whether the molar amount of trivalent Fe at the M1 site is greater than the target molar amount of trivalent Fe at the M1 site. When the molar amount of trivalent Fe at the M1 site is less than or equal to the target molar amount of trivalent Fe at the M1 site, the repair is rejected, indicating that the battery can continue to be used.

[0020] When the molar amount of trivalent Fe at the M1 site is greater than the target molar amount of trivalent Fe at the M1 site, calculate the molar amount difference, that is, subtract the target molar amount of trivalent Fe at the M1 site from the molar amount of trivalent Fe at the M1 site to obtain the molar amount of trivalent Fe to be reduced. The molar amount of trivalent Fe to be reduced is the molar amount of trivalent Fe that needs to be reduced to divalent Fe through photocatalysis, and this value determines the number of high-energy electrons required in the subsequent photocatalytic repair process. Among them, the target molar amount of trivalent Fe at the M1 site is a threshold parameter preset based on the battery model, usage environment, and performance requirements, which characterizes the maximum content of trivalent Fe at the M1 site allowed for the normal operation of the battery. When the measured molar amount of trivalent Fe at the M1 site exceeds this threshold, it means that the migration disorder of Fe ions inside the battery has reached the level that needs to be repaired.

[0021] By calculating the molar amount of trivalent Fe to be reduced, the theoretical reduction amount required for battery repair can be accurately quantified, providing a clear target indicator for subsequent optimization of photocatalytic repair parameters, avoiding the problems of over-repair or under-repair in traditional repair methods, and improving repair efficiency and battery life.

[0022] S300: Counting the number of high-energy electrons in historical experiments that meet the molar amount of trivalent Fe to be reduced, the repair environment temperature, and the dielectric constant of the electrolyte.

[0023] Specifically, based on the three parameters of the molar amount of trivalent Fe to be reduced, the repair environment temperature and the electrolyte dielectric constant, historical experimental records matching the current repair conditions are retrieved from the battery repair database, and the corresponding number of high-energy electrons is counted.

[0024] First, a battery repair database containing multiple sets of historical experimental data is constructed. Each set of historical experimental data contains experimental condition parameters (molar amount of trivalent Fe to be reduced, repair environment temperature and electrolyte dielectric constant) and corresponding experimental results (number of high-energy electrons). Through data matching, historical experimental records with the same or closest parameters to the current repair condition parameters are retrieved, and the corresponding high-energy electron values are extracted to form the number of high-energy electrons in historical experiments. The number of high-energy electrons in historical experiments refers to the number of high-energy electrons used to reduce trivalent Fe generated by ultraviolet photocatalytic reactions under specific experimental conditions. It reflects the photocatalytic energy input required to achieve the reduction of a specific amount of trivalent Fe under given conditions, and is a quantitative indicator of the photocatalytic repair effect.

[0025] By counting the number of high-energy electrons in historical experiments that meet the current remediation conditions, a goal-oriented remediation parameter optimization benchmark was established, which provided a reference target for the configuration and optimization of subsequent photocatalytic synergistic parameters. This allows the remediation process to directly tune parameters based on the most effective remediation effect in history, avoiding blind exploration and inefficient iteration in traditional methods and improving remediation efficiency and accuracy.

[0026] S400: randomly configuring a plurality of sets of photocatalytic synergistic parameters, and generating a plurality of simulated high-energy electron numbers through processing by a high-energy electron excitation simulator bound to the battery model.

[0027] Specifically, first, multiple sets of photocatalytic synergistic parameters are randomly configured. Photocatalytic synergistic parameters include repair control parameters such as DA polymer addition amount, UV light band, UV light irradiation intensity, UV light irradiation frequency, aqueous solution temperature and annealing temperature curve. A random sampling strategy is adopted to generate multiple sets of parameter combinations within the effective value range of each parameter to form multiple sets of photocatalytic synergistic parameters to cover a wider parameter space and provide sufficient candidate solutions for subsequent parameter optimization.

[0028] Subsequently, multiple sets of randomly configured photocatalytic synergy parameters are input into a high-energy electron excitation simulator bound to the battery model for processing. The high-energy electron excitation simulator is a customized neural network model trained for specific battery models, which can predict the number of high-energy electrons generated according to the input photocatalytic synergy parameters. Each battery model is bound to an exclusive high-energy electron excitation simulator to ensure the accuracy and pertinence of the simulation results.

[0029] Through the processing of the high-energy electron excitation simulator, multiple simulated high-energy electron numbers for multiple sets of photocatalytic synergy parameters can be quickly generated. These simulated high-energy electron numbers reflect the theoretically possible high-energy electron numbers under different photocatalytic synergy parameters. Compared with the traditional experimental test method, it reduces the time cost and material consumption of repair parameter screening and improves the efficiency of repair parameter optimization.

[0030] The generation of simulated high-energy electron numbers provides a theoretical basis for subsequent parameter optimization, enabling the pre-evaluation of the effects of different photocatalytic synergy parameters without actual repair operations, thereby screening out the parameter combinations most likely to achieve the ideal repair effect and further improving the accuracy and controllability of the repair process.

[0031] S500: Using the deviation between the simulated high-energy electron number and the historical experimental high-energy electron number as the fitness parameter, combining the multiple simulated high-energy electron numbers and the multiple sets of photocatalytic synergy parameters, perform photocatalytic synergy parameter optimization to obtain the photocatalytic synergy target parameters for battery repair photocatalytic control.

[0032] Specifically, first, calculate the deviation between the simulated high-energy electron number and the historical experimental high-energy electron number, and use this deviation as the fitness parameter to evaluate the quality of the photocatalytic synergy parameters. The smaller the fitness parameter, the closer the simulated high-energy electron number generated by this set of photocatalytic synergy parameters is to the best effect in the historical experiment, and the more ideal the repair effect is. Subsequently, combine multiple simulated high-energy electron numbers and the corresponding multiple sets of photocatalytic synergy parameters to construct a parameter-effect mapping relationship, perform the optimization process of photocatalytic synergy parameters, conduct a directional search in the parameter space, and gradually approach the optimal photocatalytic synergy parameters. During the optimization process, first cluster multiple sets of photocatalytic synergy parameters to obtain several clusters of photocatalytic synergy parameters and their centroid parameters; then calculate the fitness values corresponding to each centroid parameter, extract the centroid parameter corresponding to the minimum fitness value as the interference photocatalytic synergy parameter, and extract the cluster where the centroid parameter corresponding to the maximum fitness value is located as the interfered catalytic synergy parameter; then, guide the interfered catalytic synergy parameter to mutate in the direction of the interference photocatalytic synergy parameter to generate derivative photocatalytic synergy parameters; afterwards, continuously iterate and optimize based on the derivative photocatalytic synergy parameters until the preset number of iterations is satisfied, and output the photocatalytic synergy parameters with the global minimum fitness value, which are the photocatalytic target parameters. After obtaining the photocatalytic target parameters, apply them to the photocatalytic control process of battery repair to precisely regulate parameters such as the addition amount of DA polymer, ultraviolet light band, ultraviolet light irradiation intensity, ultraviolet light irradiation frequency, aqueous solution temperature, and annealing temperature curve to achieve efficient repair of the battery.

[0033] Through the above parameter optimization, the bottleneck of difficult parameter determination in traditional photocatalytic repair is broken through, realizing efficient and accurate repair of the battery, and improving the repair efficiency and battery cycle life.

[0034] Furthermore, statistically analyze the historical experimental high-energy electron numbers that satisfy the molar amount of trivalent Fe to be reduced, the repair environment temperature, and the electrolyte dielectric constant, including: S310: Statistically analyze the first set of historical experimental high-energy electron numbers that satisfy the molar amount of trivalent Fe to be reduced, the repair environment temperature, and the electrolyte dielectric constant; S320: Perform heterogeneous cleaning on the set of historical experimental high-energy electron numbers to obtain the second set of historical experimental high-energy electron numbers; S330: Take the maximum value of the second set of historical experimental high-energy electron numbers and set it as the historical experimental high-energy electron number.

[0035] In a feasible implementation, first, all historical experimental records that match or are close to the current repair condition parameters (molar amount of trivalent Fe to be reduced, repair environment temperature, and electrolyte dielectric constant) are retrieved from the battery repair database. These historical experimental records contain the number of high-energy electrons generated when the battery is repaired under similar conditions. These high-energy electron data are extracted to form the first historical experimental high-energy electron number set, which contains all historical experimental data that meet the search conditions, but may contain outliers or deviations, which need to be further processed to improve the accuracy of the reference data.

[0036] Then, a heterogeneous cleaning operation is performed on the first historical experimental high-energy electron number set. Heterogeneous cleaning is a data preprocessing technology that aims to identify and remove outliers or outliers in the data to improve data quality. By calculating the pairwise electron number deviations between the historical experimental high-energy electron numbers, evaluating the heterogeneity of each data point, and eliminating those data points with a large degree of heterogeneity, a cleaned second historical experimental high-energy electron number set is obtained. Heterogeneous cleaning can effectively remove abnormal data caused by experimental errors, operational errors, or equipment failures, ensuring that the reference benchmark for subsequent parameter optimization is more reliable.

[0037] Subsequently, the maximum value is extracted from the set of high-energy electron numbers of the second historical experiment after heterogeneous cleaning, and is set as the high-energy electron number of the historical experiment. The maximum value is selected instead of the average or median in order to pursue the best repair effect. The maximum value represents the best high-energy electron generation effect that can be achieved in historical experiments under similar conditions. As the target benchmark for parameter optimization, it can guide the development towards a more efficient repair direction.

[0038] Through the above three steps, high-quality reference benchmarks are extracted from historical data, which provides a reliable target orientation for the subsequent photocatalytic synergistic parameter optimization and improves the accuracy of restoration parameter optimization and the stability of restoration effects.

[0039] Further, the historical experimental high-energy electron number set is subjected to heterogeneous cleaning to obtain a second historical experimental high-energy electron number set, including: S321: Obtain the pairwise electron number deviations of the historical experimental high-energy electron number set, and obtain the historical experimental high-energy electron number difference set; S322: Obtaining the first historical experiment high energy electron number of the historical experiment high energy electron number set; S323: Taking the first historical experiment high-energy electron number as a starting point, based on the historical experiment high-energy electron number difference set, extracting a preset number of first selected historical experiment high-energy electron number sets from the historical experiment high-energy electron number set from near to far; S324: until the Qth selected historical experimental high-energy electron number set is obtained, where Q represents the total number of historical experimental high-energy electron number sets; S325: Extract a first reverse selected historical high-energy electron number set including the first historical high-energy electron number from the second selected historical high-energy electron number set to the Qth selected historical high-energy electron number set; S326: Take the union of the first selected historical high-energy electron number set and the first reverse selected historical high-energy electron number set to obtain a heterogeneous evaluation historical high-energy electron number set, and calculate the average value of the difference in the number of electrons between the heterogeneous evaluation historical high-energy electron number set and the first historical high-energy electron number, which is set as the first heterogeneity coefficient; S327: When the ratio of the first heterogeneity coefficient to the average heterogeneity coefficient is greater than or equal to the heterogeneity ratio threshold, perform heterogeneous cleaning on the first historical high-energy electron number; S328: When the traversal of the historical high-energy electron number set is completed, output the second historical high-energy electron number set.

[0040] In a preferred implementation manner, when performing heterogeneous cleaning on the historical high-energy electron number set to obtain the second historical high-energy electron number set, first, calculate the absolute difference between any two historical high-energy electron numbers in the historical high-energy electron number set to form a historical high-energy electron number difference set. This historical high-energy electron number difference set comprehensively describes the distance relationship between data points and provides basic data for subsequent heterogeneous evaluation. Secondly, traverse the historical high-energy electron numbers in the historical high-energy electron number set, and each time take out a historical high-energy electron number, which is set as the first historical high-energy electron number. Evaluate each historical high-energy electron number in the historical high-energy electron number set as an object in turn until the entire historical high-energy electron number set is traversed.

[0041] Next, starting from the first historical high-energy electron number, that is, taking the first historical high-energy electron number as the center point, based on the difference magnitude in the historical high-energy electron number difference set, in the principle of from near to far, that is, in the order of the difference from small to large, select a preset number of data points from the historical high-energy electron number set to form a first selected historical high-energy electron number set, realizing the preliminary screening of the data closest to the first historical high-energy electron number. Among them, the preset number is usually set to 20% to 30% of the total data volume and can be adjusted according to specific application scenarios. Repeat the above selection process, starting from different historical high-energy electron numbers, and respectively select the corresponding adjacent data points until all possible selected sets are formed, that is, from the first selected historical high-energy electron number set to the Qth selected historical high-energy electron number set, where Q is equal to the total number of data points of the historical high-energy electron numbers in the historical high-energy electron number set.

[0042] In the previous steps S323 - S324, forward selection has been completed, that is, starting from the high - energy electron number of the first historical experiment, the historical experiment high - energy electron numbers closest to it are selected to form the first selected historical experiment high - energy electron number set. The forward selection mechanism calculates the similarity relationship between data points based on the electron number difference, identifies the data subset closest to the target data point from the historical experiment high - energy electron number set, and provides an initial reference set for subsequent heterogeneity evaluation. A local similarity structure between data points is established through distance - metric methods, laying a data foundation for outlier identification.

[0043] Subsequently, a reverse selection strategy is implemented. Check all sets containing the high - energy electron number of the first historical experiment from the second selected historical experiment high - energy electron number set to the Q - th selected historical experiment high - energy electron number set, and merge these sets to form the first reverse - selected historical experiment high - energy electron number set. Reverse selection establishes a data mutual - selection relationship verification mechanism by checking whether other data points include the high - energy electron number of the first historical experiment in their neighboring - point sets, providing a way to evaluate data relevance from different perspectives. Then, by taking the union of the first selected historical experiment high - energy electron number set and the first reverse - selected historical experiment high - energy electron number set, a heterogeneous - evaluation historical experiment high - energy electron number set is obtained, which synthesizes the selection results of both positive and negative dimensions. Subsequently, calculate the mean value of the electron - number difference between each data point in the heterogeneous - evaluation historical experiment high - energy electron number set and the high - energy electron number of the first historical experiment, and define this value as the first heterogeneity coefficient. The first heterogeneity coefficient quantifies the average deviation degree of the target data point from its related data group and is an index for evaluating data heterogeneity.

[0044] After that, compare the ratio of the first heterogeneity coefficient to the mean heterogeneity coefficient (i.e., the arithmetic mean of all calculated heterogeneity coefficients) with a preset heterogeneity - ratio threshold. The heterogeneity - ratio threshold is a judgment criterion preset according to data characteristics and application requirements, used to distinguish normal data from outliers. When the ratio of the first heterogeneity coefficient to the mean heterogeneity coefficient is greater than or equal to the heterogeneity - ratio threshold, the high - energy electron number of the first historical experiment is determined as an outlier and heterogeneous cleaning is performed; when the ratio of the first heterogeneity coefficient to the mean heterogeneity coefficient is less than the heterogeneity - ratio threshold, the data point is retained. Then, repeat the above evaluation process for each historical experiment high - energy electron number in the historical experiment high - energy electron number set until the heterogeneity evaluation of all historical experiment high - energy electrons is completed, and output the second historical experiment high - energy electron number set after heterogeneous cleaning.

[0045] Through the forward - reverse two - way selection mechanism, multi - dimensional data heterogeneity evaluation is achieved, improving the accuracy of outlier identification and the effectiveness of data cleaning. This two - way verification strategy can capture the mutual relationship between data more comprehensively and provide highly reliable data support for subsequent parameter optimization.

[0046] Further, randomly configure multiple sets of photocatalytic synergy parameters, and process them through a high-energy electron excitation simulator bound to the battery model to generate multiple simulated high-energy electron numbers, including: S410: Collect the experimental quantities of the photocatalytic synergy parameters of the battery model, and then collect the high-energy electron excitation quantities of several experiments with the same experimental quantities of the photocatalytic synergy parameters, perform mean annotation, and obtain the label identifying the high-energy electron number; S420: Collect multiple sets of corresponding labels of the identified high-energy electron numbers and the experimental quantities of the photocatalytic synergy parameters, train the network parameters of the feedforward neural network until convergence, generate the high-energy electron excitation simulator, and bind it to the battery model; S430: Input the multiple sets of photocatalytic synergy parameters into the high-energy electron excitation simulator respectively, and output the multiple simulated high-energy electron numbers.

[0047] In a preferred embodiment, first, for a specific battery model, collect the experimental quantities of the photocatalytic synergy parameters, including the experimental set values of parameters such as the addition amount of DA polymer, the ultraviolet light band, the ultraviolet light irradiation intensity, the ultraviolet light irradiation frequency, the aqueous solution temperature, and the annealing temperature curve. Subsequently, under the condition of the same experimental quantities of the photocatalytic synergy parameters, conduct multiple repeated experiments, and collect the high-energy electron excitation quantities generated in each experiment. To eliminate the influence of random fluctuations, perform mean annotation processing on these high-energy electron excitation quantities, that is, calculate the arithmetic mean of multiple experiments as the label identifying the high-energy electron number. This mean annotation mechanism improves the reliability of the training data and reduces the influence of accidental errors in a single experiment.

[0048] Then, collect multiple sets of corresponding labels of the identified high-energy electron numbers and the experimental quantities of the photocatalytic synergy parameters to form the input-output pairs for training the high-energy electron excitation simulator. Based on these training data, adopt a feedforward neural network structure to iteratively train the network parameters until the loss function converges below a preset threshold. After training, generate a high-energy electron excitation simulator for a specific battery model, and bind the high-energy electron excitation simulator to the corresponding battery model to establish a mapping relationship from the battery model to the high-energy electron excitation simulator. This binding mechanism ensures the high applicability of the simulator to a specific model of battery and avoids the generalization error of the general model. Subsequently, input the randomly configured multiple sets of photocatalytic synergy parameters into the high-energy electron excitation simulator bound to the battery model respectively, and through forward calculation, output the corresponding multiple simulated high-energy electron numbers. These simulated high-energy electron numbers reflect the theoretically possible high-energy electron quantities under different combinations of photocatalytic synergy parameters, providing diverse candidate solutions for subsequent parameter optimization.

[0049] Through the construction and application of the above high-energy electron excitation simulator, an efficient mapping from experimental data to theoretical models is achieved, reducing the experimental costs and time overheads for optimizing battery repair parameters and improving the efficiency and accuracy of screening repair parameters.

[0050] Furthermore, multiple sets of tags corresponding one-to-one to the identified high-energy electron numbers and experimental quantities of the photocatalytic synergy parameters are collected to train the network parameters of the feedforward neural network until convergence to generate the high-energy electron excitation simulator. It further includes: S421: Construct a residual fitting training period; S422: Collect multiple sets of tags corresponding one-to-one to the identified high-energy electron numbers and experimental quantities of the photocatalytic synergy parameters, and train the network parameters of the feedforward neural network. When the residual fitting training period is satisfied, obtain a first high-energy electron excitation simulator, and statistically calculate the centroid error vector of the error vector set within the training period; S423: When the magnitude of the centroid error vector is greater than or equal to the error threshold, based on the centroid error vector, construct a first output corrector for the first high-energy electron excitation simulator to obtain a second high-energy electron excitation simulator, where the output correction method is to sum the output of the first high-energy electron excitation simulator and the centroid error vector as the final output; S424: Return to the training step, and debug the second high-energy electron excitation simulator until the high-energy electron excitation simulator is obtained; S425: When the magnitude of the centroid error vector is less than the error threshold, set the first high-energy electron excitation simulator as the high-energy electron excitation simulator.

[0051] In a feasible implementation, by correcting the systematic bias in the output of the high-energy electron excitation simulator, the convergence efficiency and prediction accuracy of the high-energy electron excitation simulator are improved.

[0052] First, construct a residual fitting training cycle. A residual fitting training cycle refers to the time interval during the neural network training process when, at every preset number of iterations, statistical analysis and correction are performed on the prediction error of the model. By setting the residual fitting training cycle, error evaluation can be carried out at specific stages of model training rather than only at the end of training, which is beneficial for timely capturing and correcting systematic biases during the training process. Then, collect multiple groups of input-output pairs (labels indicating the number of high-energy electrons and experimental quantities of photocatalytic cooperation parameters) to train the parameters of the feedforward neural network. When the number of training iterations reaches the preset residual fitting training cycle, a first high-energy electron excitation simulator after preliminary training is obtained. Subsequently, statistically analyze the set of error vectors within this training cycle, that is, the set of difference vectors between the model prediction values and the actual label values, and calculate the centroid error vector of these error vectors. The centroid error vector characterizes the average systematic bias of the model prediction and serves as the basis for subsequent residual correction.

[0053] Next, determine whether the norm value of the centroid error vector is greater than or equal to a preset error threshold. When the norm value is greater than or equal to the error threshold, it indicates that the model has significant systematic biases and needs residual correction. At this time, based on the centroid error vector, construct a first output corrector for the first high-energy electron excitation simulator to obtain a second high-energy electron excitation simulator after residual correction. The correction method uses an additive residual compensation mechanism, that is, add the original output of the first high-energy electron excitation simulator and the centroid error vector as the final output, thereby offsetting the systematic bias in the model prediction and improving the prediction accuracy. Subsequently, continue to debug and optimize the parameters of the second high-energy electron excitation simulator. Since the first output corrector has eliminated most of the systematic biases, subsequent training will focus more on capturing subtle non-linear relationships in the data, accelerating the model's convergence to a better solution. Repeat the residual evaluation and correction process until a high-energy electron excitation simulator that meets the accuracy requirements is obtained.

[0054] When it is detected that the norm value of the centroid error vector is less than the preset error threshold, it indicates that there are no significant systematic biases in the model prediction. At this time, set the first high-energy electron excitation simulator as the final high-energy electron excitation simulator to complete the model training process.

[0055] By introducing the residual principle for model training and optimization, the accurate identification and effective correction of the prediction error of the neural network are achieved. Compared with traditional training methods, it can significantly reduce the number of iterations in model training and accelerate the convergence process. Secondly, through the statistics and compensation of the centroid error vector, it can effectively overcome the noise interference in the training data and improve the robustness of the model. At the same time, the additive residual compensation mechanism makes the model easier to capture the subtle non-linear relationships in the data and improves the prediction accuracy. These advantages enable the high-energy electron excitation simulator to more accurately predict the high-energy electron generation effect under different photocatalytic synergy parameters, providing a highly reliable theoretical support for the optimization of battery repair parameters.

[0056] Further, taking the deviation between the simulated high-energy electron number and the historical experimental high-energy electron number as the fitness parameter, combining the multiple simulated high-energy electron numbers and the multiple sets of photocatalytic synergy parameters, perform photocatalytic synergy parameter optimization to obtain the photocatalytic synergy target parameters, including: S510: Cluster the multiple sets of photocatalytic synergy parameters to obtain several clusters of photocatalytic synergy parameters, where the several clusters of photocatalytic synergy parameters have several sets of centroid photocatalytic synergy parameters; S520: Traverse the several sets of centroid photocatalytic synergy parameters, and calculate several centroid fitness parameters based on the multiple simulated high-energy electron numbers; S530: Extract the centroid photocatalytic synergy parameter corresponding to the minimum value of the several centroid fitness parameters, and set it as the interference photocatalytic synergy parameter; S540: Extract all the photocatalytic synergy parameters of the centroid photocatalytic synergy parameter corresponding to the maximum value of the several centroid fitness parameters, and set it as the interfered catalytic synergy parameter; S550: Take the interference photocatalytic synergy parameter as the target, mutate the interfered catalytic synergy parameter to obtain the derivative photocatalytic synergy parameter; S560: Continuously iterate and optimize based on the derivative photocatalytic synergy parameter until the preset number of iterations is met, and output the photocatalytic synergy parameter with the minimum global fitness value, which is set as the photocatalytic synergy target parameter.

[0057] In a preferred embodiment, first, perform cluster analysis on multiple randomly configured sets of photocatalytic synergy parameters to divide the parameter space into several clusters of photocatalytic synergy parameters. Among them, the cluster analysis can adopt an improved K-means algorithm. By calculating the Euclidean distance between parameter vectors, similar parameter combinations are aggregated into the same cluster. Each cluster contains multiple sets of similar photocatalytic synergy parameters, and at the same time, calculate the centroid photocatalytic synergy parameter of each cluster as the representative of the cluster. Through clustering, the huge parameter space is effectively compressed, reducing the computational complexity of subsequent optimization.

[0058] Subsequently, all centroid photocatalytic synergy parameters are traversed, input into the high-energy electron excitation simulator to obtain the corresponding simulated high-energy electron numbers, and the deviation from the historical experimental high-energy electron numbers is calculated to obtain the centroid fitness parameters. The centroid fitness parameters are calculated using the weighted Euclidean distance method. By performing a weighted sum of the squares of the differences between the simulated high-energy electron numbers and the historical experimental high-energy electron numbers, and then taking the square root. Preferably, an adaptive weight allocation mechanism can be adopted to determine the weight coefficients of each parameter in the weighted Euclidean distance calculation through sensitivity analysis. The specific method is as follows: make a small perturbation to each photocatalytic synergy parameter and observe the change rate of the simulated high-energy electron numbers. The larger the change rate, the more significant the impact of this parameter on the result, and accordingly, a higher weight is assigned. For example, when a 1% change in the ultraviolet light irradiation intensity parameter results in a 5% change in the high-energy electron numbers, while a 1% change in the DA polymer addition amount only results in a 1% change in the high-energy electron numbers, a higher weight coefficient is assigned to the ultraviolet light irradiation intensity parameter. This weight allocation method can make the optimization process pay more attention to key parameters and improve the optimization efficiency and accuracy.

[0059] Subsequently, the centroid photocatalytic synergy parameter corresponding to the minimum value of the centroid fitness parameter is extracted and set as the interference photocatalytic synergy parameter. The minimum centroid fitness parameter indicates that the simulated high-energy electron numbers under the corresponding centroid photocatalytic synergy parameter are closest to the historical experimental high-energy electron numbers, representing the optimal parameter direction. At the same time, the entire cluster of photocatalytic synergy parameters where the centroid photocatalytic synergy parameter corresponding to the maximum value of the centroid fitness parameter is located is extracted and set as the interfered catalytic synergy parameter. The maximum centroid fitness parameter indicates that the simulation effect under the corresponding centroid photocatalytic synergy parameter is the worst and needs to be optimized and adjusted.

[0060] After that, with the interference photocatalytic synergy parameter as the target, a mutation operation is performed on the interfered catalytic synergy parameter. The mutation adopts a directional mutation mechanism, that is, in the parameter space, the interfered parameter is moved towards the interference parameter direction, and at the same time, a certain degree of random perturbation is introduced to avoid falling into local optima. The derivative photocatalytic synergy parameters generated by the mutation operation retain some characteristics of the original parameters and introduce the characteristics of high-quality parameters, providing a new candidate solution for finding a better parameter combination. Subsequently, with the derivative photocatalytic synergy parameters as the new starting point, the above optimization process is continuously executed until the preset number of iterations is met. Finally, from all the parameter combinations generated by the iterations, the photocatalytic synergy parameter corresponding to the minimum global fitness value is selected and set as the photocatalytic synergy target parameter, which is used to guide the actual battery repair process.

[0061] Reducing the search space dimension through clustering analysis, achieving efficient parameter optimization through directed mutation, and introducing an adaptive weight allocation mechanism to enhance the optimization accuracy significantly improve the screening efficiency and repair effect of battery repair parameters. Compared with traditional exhaustive search or random search methods, it can converge to the global optimal solution faster and provide more accurate parameter guidance for battery repair.

[0062] Furthermore, extract the full-cluster photocatalytic synergy parameters of the centroid photocatalytic synergy parameters corresponding to the maximum value of the several centroid fitness parameters, and set them as the interfered catalytic synergy parameters, including: S541: Configure the minimum search step size, cluster the full-cluster photocatalytic synergy parameters, and obtain multiple groups of photocatalytic synergy parameters within the cluster; S542: Set the multiple centroid photocatalytic synergy parameters within the cluster of the multiple groups of photocatalytic synergy parameters within the cluster as the interfered catalytic synergy parameters.

[0063] In a preferred embodiment, first, configure the minimum search step size, which defines the minimum effective search distance in the parameter space. When the distance between two parameter combinations is less than the minimum search step size, these two groups of parameters are considered substantially equivalent and do not need to be repeatedly evaluated. Subsequently, perform secondary clustering on the full-cluster photocatalytic synergy parameters, group the parameter combinations with closer distances into the same sub-cluster, and form multiple groups of photocatalytic synergy parameters within the cluster. This hierarchical clustering strategy effectively avoids the waste of resources caused by repeated calculation of similar parameters and improves the optimization efficiency.

[0064] Subsequently, calculate the centroid of each sub-cluster of the multiple groups of photocatalytic synergy parameters within the cluster to obtain multiple centroid photocatalytic synergy parameters within the cluster, and set these centroid parameters as the interfered catalytic synergy parameters. By selecting the sub-cluster centroid as the interfered parameter instead of using all parameters, the number of parameters that need to perform mutation operations is effectively reduced, while ensuring parameter diversity and avoiding redundancy and inefficiency of mutation operations.

[0065] By adopting this two-level clustering and screening mechanism, defining the minimum effective distance with the minimum search step size, and using hierarchical clustering technology to select representative parameter combinations, the computational complexity is reduced, while ensuring the comprehensiveness and effectiveness of the parameter space search, providing a more computationally efficient technical means for battery repair parameter optimization.

[0066] Example 2, as Figure 2 shown, based on the same inventive concept as the photocatalytic synergy-based battery repair control method provided in Example 1, the embodiment of the present invention also provides a photocatalytic synergy-based battery repair control system, including: A data acquisition module 11, configured to obtain battery detection information to be repaired, where the battery detection information to be repaired includes the molar amount of trivalent Fe at the M1 site, battery model, electrolyte dielectric constant, and repair environment temperature; A difference calculation module 12, configured to calculate a molar amount difference and obtain a molar amount of trivalent Fe to be reduced when the molar amount of trivalent Fe at the M1 site is greater than the target molar amount of trivalent Fe at the M1 site; A historical data statistics module 13, configured to count the number of high-energy electrons in historical experiments that meet the molar amount of trivalent Fe to be reduced, the repair environment temperature, and the electrolyte dielectric constant; A simulation processing module 14, configured to randomly configure multiple sets of photocatalytic cooperation parameters, and generate multiple simulated high-energy electron numbers through a high-energy electron excitation simulator bound to the battery model; An optimization control module 15, configured to use the deviation between the simulated high-energy electron number and the high-energy electron number in historical experiments as a fitness parameter, and combine the multiple simulated high-energy electron numbers and the multiple sets of photocatalytic cooperation parameters to perform optimization of photocatalytic cooperation parameters, and obtain photocatalytic cooperation target parameters for photocatalytic control of battery repair.

[0067] Further, the historical data statistics module 13 includes the following execution steps: Count a first set of high-energy electron numbers in historical experiments that meet the molar amount of trivalent Fe to be reduced, the repair environment temperature, and the electrolyte dielectric constant; Perform heterogeneous cleaning on the set of high-energy electron numbers in historical experiments to obtain a second set of high-energy electron numbers in historical experiments; Take the maximum value of the second set of high-energy electron numbers in historical experiments and set it as the high-energy electron number in historical experiments.

[0068] Further, the historical data statistics module 13 further includes the following execution steps: Obtain the pairwise electron number deviations of the set of high-energy electron numbers in historical experiments to obtain a set of historical experimental high-energy electron quantity differences; Obtain the first high-energy electron number in the set of high-energy electron numbers in historical experiments; Starting from the first high-energy electron number in historical experiments, based on the set of historical experimental high-energy electron quantity differences, from near to far, extract a preset number of first selected sets of high-energy electron numbers in historical experiments from the set of high-energy electron numbers in historical experiments; Until the Qth selected set of high-energy electron numbers in historical experiments is obtained, where Q represents the total number of sets of high-energy electron numbers in historical experiments; From the second selected set of high-energy electron numbers in historical experiments to the Qth selected set of high-energy electron numbers in historical experiments, extract a first reverse selected set of high-energy electron numbers in historical experiments that includes the first high-energy electron number in historical experiments; Take the union of the first selected historical experimental high-energy electron number set and the first reverse-selected historical experimental high-energy electron number set to obtain a heterogeneous evaluation historical experimental high-energy electron number set, and calculate the average value of the electron number difference between the heterogeneous evaluation historical experimental high-energy electron number set and the first historical experimental high-energy electron number, which is set as the first heterogeneous coefficient; When the ratio of the first heterogeneous coefficient to the average value of the heterogeneous coefficients is greater than or equal to the heterogeneous ratio threshold, perform heterogeneous cleaning on the first historical experimental high-energy electron number; When the traversal of the historical experimental high-energy electron number set is completed, output the second historical experimental high-energy electron number set.

[0069] Further, the simulation processing module 14 includes the following execution steps: Collect the experimental quantity of the photocatalytic synergy parameters of the battery model, and then collect the high-energy electron excitation quantities of several experiments with the same experimental quantity of the photocatalytic synergy parameters, and perform mean annotation to obtain the label for identifying the high-energy electron number; Collect multiple groups of the labels for identifying the high-energy electron number and the experimental quantity of the photocatalytic synergy parameters that correspond one by one, and train the network parameters of the feedforward neural network until convergence to generate the high-energy electron excitation simulator and bind it to the battery model; Input the multiple groups of photocatalytic synergy parameters into the high-energy electron excitation simulator respectively, and output the multiple simulated high-energy electron numbers.

[0070] Further, the simulation processing module 14 includes the following execution steps: Construct a residual fitting training cycle; Collect multiple groups of the labels for identifying the high-energy electron number and the experimental quantity of the photocatalytic synergy parameters that correspond one by one, and train the network parameters of the feedforward neural network. When the residual fitting training cycle is satisfied, obtain the first high-energy electron excitation simulator, and statistically calculate the centroid error vector of the error vector set within the training cycle; When the modulus value of the centroid error vector is greater than or equal to the error threshold, construct the first output corrector of the first high-energy electron excitation simulator based on the centroid error vector to obtain the second high-energy electron excitation simulator, where the output correction method is to add the output of the first high-energy electron excitation simulator and the centroid error vector as the final output; Return to the training step and debug the second high-energy electron excitation simulator until the high-energy electron excitation simulator is obtained; When the modulus value of the centroid error vector is less than the error threshold, set the first high-energy electron excitation simulator as the high-energy electron excitation simulator.

[0071] Further, the optimization control module 15 includes the following execution steps: Cluster the multiple sets of photocatalytic synergy parameters to obtain several clusters of photocatalytic synergy parameters, where the several clusters of photocatalytic synergy parameters have several sets of centroid photocatalytic synergy parameters; Traverse the several sets of centroid photocatalytic synergy parameters and calculate several centroid fitness parameters based on the multiple simulated high-energy electron numbers; Extract the centroid photocatalytic synergy parameter corresponding to the minimum value of the several centroid fitness parameters, and set it as the interference photocatalytic synergy parameter; Extract the full-cluster photocatalytic synergy parameter of the centroid photocatalytic synergy parameter corresponding to the maximum value of the several centroid fitness parameters, and set it as the interfered catalytic synergy parameter; Take the interference photocatalytic synergy parameter as the target and mutate the interfered catalytic synergy parameter to obtain a derivative photocatalytic synergy parameter; Based on the derivative photocatalytic synergy parameter, continuously iterate and optimize until the preset number of iterations is satisfied, and output the photocatalytic synergy parameter with the minimum global fitness, which is set as the photocatalytic synergy target parameter.

[0072] Further, the optimization control module 15 includes the following execution steps: Configure the minimum value of the search step size, cluster the full-cluster photocatalytic synergy parameters to obtain multiple sets of photocatalytic synergy parameters within the cluster; Set the multiple centroid photocatalytic synergy parameters within the cluster of the multiple sets of photocatalytic synergy parameters within the cluster as the interfered catalytic synergy parameter.

[0073] It should be noted that in the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0074] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0075] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and combinations of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded computers, or other programmable data processing devices to generate a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices produce means for implementing the functions specified in the Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0076] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufacture including instruction means that implement the functions specified in the Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0077] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in the Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0078] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they learn the basic inventive concept.

[0079] Obviously, those skilled in the art can make various changes and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. A battery repair control method with photocatalytic synergy, characterized in that, Including: Obtaining the detection information of the battery to be repaired, where the detection information of the battery to be repaired includes the molar amount of trivalent Fe at the M1 site, the battery model, the dielectric constant of the electrolyte, and the repair environment temperature; When the molar amount of trivalent Fe at the M1 site is greater than the target molar amount of trivalent Fe at the M1 site, calculate the molar amount difference to obtain the molar amount of trivalent Fe to be reduced; Statistically count the historical experimental high-energy electron numbers that satisfy the molar amount of trivalent Fe to be reduced, the repair environment temperature, and the dielectric constant of the electrolyte; Randomly configure multiple sets of photocatalytic synergy parameters, and process them through a high-energy electron excitation simulator bound to the battery model to generate multiple simulated high-energy electron numbers; Taking the deviation between the simulated high-energy electron number and the historical experimental high-energy electron number as the fitness parameter, combining the multiple simulated high-energy electron numbers and the multiple sets of photocatalytic synergy parameters, perform optimization of the photocatalytic synergy parameters to obtain the target photocatalytic synergy parameters for photocatalytic control of battery repair.

2. The method according to claim 1, wherein Statistically counting the historical experimental high-energy electron numbers that satisfy the molar amount of trivalent Fe to be reduced, the repair environment temperature, and the dielectric constant of the electrolyte includes: Statistically counting the first set of historical experimental high-energy electron numbers that satisfy the molar amount of trivalent Fe to be reduced, the repair environment temperature, and the dielectric constant of the electrolyte; Performing heterogeneous cleaning on the set of historical experimental high-energy electron numbers to obtain a second set of historical experimental high-energy electron numbers; Taking the maximum value of the second set of historical experimental high-energy electron numbers as the historical experimental high-energy electron number.

3. The method according to claim 2, wherein Performing heterogeneous cleaning on the set of historical experimental high-energy electron numbers to obtain a second set of historical experimental high-energy electron numbers, including: Obtaining the pairwise electron number deviation of the set of historical experimental high-energy electron numbers to obtain a set of historical experimental high-energy electron quantity differences; Obtaining the first historical experimental high-energy electron number of the set of historical experimental high-energy electron numbers; Starting from the first historical experimental high-energy electron number, based on the set of historical experimental high-energy electron quantity differences, from near to far, extract a preset number of first selected sets of historical experimental high-energy electron numbers from the set of historical experimental high-energy electron numbers; Until the Qth selected set of historical experimental high-energy electron numbers is obtained, where Q represents the total number of sets of historical experimental high-energy electron numbers; From the second selected set of historical experimental high-energy electron numbers to the Qth selected set of historical experimental high-energy electron numbers, extract a first reverse selected set of historical experimental high-energy electron numbers that includes the first historical experimental high-energy electron number; Taking the union of the first selected set of historical experimental high-energy electron numbers and the first reverse selected set of historical experimental high-energy electron numbers to obtain a set of heterogeneous evaluation historical experimental high-energy electron numbers, and calculating the average value of the electron quantity difference between the set of heterogeneous evaluation historical experimental high-energy electron numbers and the first historical experimental high-energy electron number as the first heterogeneous coefficient; When the ratio of the first heterogeneous coefficient to the average value of the heterogeneous coefficients is greater than or equal to the heterogeneous ratio threshold, perform heterogeneous cleaning on the first historical experimental high-energy electron number; When the traversal of the set of historical experimental high-energy electron numbers is completed, output the second set of historical experimental high-energy electron numbers.

4. The method according to claim 1, wherein Randomly configure multiple sets of photocatalytic synergy parameters, and process them through a high-energy electron excitation simulator bound to the battery model to generate multiple simulated high-energy electron numbers, including: Collect the experimental amounts of photocatalytic synergy parameters of the battery model, then collect the experimental high-energy electron excitation amounts of several experiments with the same experimental amounts of photocatalytic synergy parameters, perform mean annotation to obtain the labels identifying the high-energy electron numbers; Collect multiple sets of corresponding labels of the identified high-energy electron numbers and the experimental amounts of photocatalytic synergy parameters, and train the network parameters of the feedforward neural network until convergence to generate the high-energy electron excitation simulator and bind it to the battery model; Input the multiple sets of photocatalytic synergy parameters into the high-energy electron excitation simulator respectively, and output the multiple simulated high-energy electron numbers.

5. The method according to claim 4, wherein Collect multiple sets of corresponding labels of the identified high-energy electron numbers and the experimental amounts of photocatalytic synergy parameters, and train the network parameters of the feedforward neural network until convergence to generate the high-energy electron excitation simulator, which also includes: Construct a residual fitting training cycle; Collect multiple sets of corresponding labels of the identified high-energy electron numbers and the experimental amounts of photocatalytic synergy parameters, and train the network parameters of the feedforward neural network. When the residual fitting training cycle is satisfied, obtain the first high-energy electron excitation simulator, and statistically calculate the centroid error vector of the error vector set within the training cycle; When the modulus of the centroid error vector is greater than or equal to the error threshold, based on the centroid error vector, construct the first output corrector of the first high-energy electron excitation simulator to obtain the second high-energy electron excitation simulator, where the output correction method is to sum the output of the first high-energy electron excitation simulator and the centroid error vector as the final output; Return to the training step, and debug the second high-energy electron excitation simulator until the high-energy electron excitation simulator is obtained; When the modulus of the centroid error vector is less than the error threshold, set the first high-energy electron excitation simulator as the high-energy electron excitation simulator.

6. The method according to claim 1, wherein Using the deviation between the simulated high-energy electron numbers and the historical experimental high-energy electron numbers as the fitness parameter, combine the multiple simulated high-energy electron numbers and the multiple sets of photocatalytic synergy parameters to perform photocatalytic synergy parameter optimization to obtain the photocatalytic synergy target parameters, including: Cluster the multiple sets of photocatalytic synergy parameters to obtain several clusters of photocatalytic synergy parameters, where the several clusters of photocatalytic synergy parameters have several sets of centroid photocatalytic synergy parameters; Traverse the several sets of centroid photocatalytic synergy parameters, and calculate several centroid fitness parameters based on the multiple simulated high-energy electron numbers; Extract the centroid photocatalytic synergy parameter corresponding to the minimum value of the several centroid fitness parameters and set it as the interference photocatalytic synergy parameter; Extract all the photocatalytic synergy parameters of the centroid photocatalytic synergy parameter corresponding to the maximum value of the several centroid fitness parameters and set it as the interfered catalytic synergy parameter; Using the interference photocatalytic synergy parameter as the target, mutate the interfered catalytic synergy parameter to obtain the derivative photocatalytic synergy parameter; Continuously iteratively optimize based on the derived photocatalytic synergy parameters until the preset number of iterations is met, and output the photocatalytic synergy parameters with the minimum global fitness, which are set as the photocatalytic synergy target parameters.

7. The method according to claim 6, characterized in that, Extract the full cluster of photocatalytic synergy parameters of the centroid photocatalytic synergy parameters corresponding to the maximum value of the several centroid fitness parameters, which are set as the interfered catalytic synergy parameters, including: Configure the minimum search step size, cluster the full cluster of photocatalytic synergy parameters, and obtain multiple groups of photocatalytic synergy parameters within the cluster; Set the multiple centroid photocatalytic synergy parameters within the cluster of the multiple groups of photocatalytic synergy parameters within the cluster as the interfered catalytic synergy parameters.

8. A battery repair control system with photocatalytic synergy, characterized in that For implementing a battery repair control method for photocatalytic synergy according to any one of claims 1 to 7, including: A data acquisition module for obtaining the detection information of the battery to be repaired, wherein the detection information of the battery to be repaired includes the molar amount of trivalent Fe at the M1 site, the battery model, the electrolyte dielectric constant, and the repair environment temperature; A difference calculation module for calculating the molar amount difference and obtaining the molar amount of trivalent Fe to be reduced when the molar amount of trivalent Fe at the M1 site is greater than the target molar amount of trivalent Fe at the M1 site; A historical data statistics module for statistically calculating the number of high-energy electrons in historical experiments that meet the molar amount of trivalent Fe to be reduced, the repair environment temperature, and the electrolyte dielectric constant; A simulation processing module for randomly configuring multiple groups of photocatalytic synergy parameters and generating multiple simulated high-energy electron numbers through a high-energy electron excitation simulator bound to the battery model; An optimization control module for using the deviation between the simulated high-energy electron number and the historical experimental high-energy electron number as the fitness parameter, combining the multiple simulated high-energy electron numbers and the multiple groups of photocatalytic synergy parameters, performing optimization of the photocatalytic synergy parameters, and obtaining the photocatalytic synergy target parameters for photocatalytic control of battery repair.

Images