Retired lithium battery electrochemical system identification method, electronic equipment and medium
By constructing an equivalent circuit model library and matching the EIS test data of retired lithium batteries, its electrochemical system is determined, solving the problem of difficulty identifying the type of retirement lithium battery, which causes difficulties in recycling and utilization. Essence
Patent Information
- Application Number
- CN202411325681.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-23
- Publication Date
- 2025-01-21
AI Technical Summary
Retired lithium batteries are difficult to identify due to their difficult types, which makes them difficult to recycle.
By constructing an equivalent circuit integer-order model library and an equivalent circuit fractional-order model library, the EIS test data of retired lithium batteries is obtained, and the target equivalent circuit model is matched to determine the electrochemical system.
The electrochemical system identification of multiple types of retired lithium batteries has been realized, and the problem that cannot be recognized due to label lack is solved. The operation is simple and easy to promote.
Smart Images
Figure CN119337580A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of battery technology, and in particular to a method for identifying an electrochemical system of a retired lithium battery, an electronic device and a medium. Background Art
[0002] With the advent of the retirement tide of power batteries for electric vehicles and low-speed vehicles, a large number of power batteries have been retired. However, the vast majority of retired power batteries still have 80% of remaining available capacity. By recycling their monomers and reassembling them into groups, the remaining value can be fully utilized. At present, electric vehicle batteries are mainly divided into lead-acid batteries, nickel-metal hydride batteries, sodium-sulfur batteries, secondary lithium batteries, air batteries and ternary lithium batteries. Considering the different sources of batteries, with the advent of the retired battery tide, various batteries are piled up in the factory and it is difficult to distinguish the types of models. In addition, there may be a large number of battery labels missing, and it is impossible to identify the type of battery, making it difficult to recycle and process in actual use, which brings difficulties to recycling and cascade utilization. Summary of the invention
[0003] In view of this, it is necessary to provide a method, electronic device and medium for identifying the electrochemical system of retired lithium batteries, so as to solve the problem that retired power batteries are difficult to recycle due to the difficulty in identifying their types.
[0004] In order to solve the above problems, in a first aspect, the present invention provides a method for identifying an electrochemical system of a retired lithium battery, comprising: Constructing an equivalent circuit integer-order model library and an equivalent circuit fractional-order model library; the equivalent circuit integer-order model library includes a plurality of equivalent circuit integer-order models of different orders, and the equivalent circuit fractional-order model library includes a plurality of different types of equivalent circuit fractional-order models; Obtain EIS test data of retired lithium batteries; Determine a target equivalent circuit integer-order model from the equivalent circuit integer-order model library according to the EIS test data, and determine a target equivalent circuit fractional-order model from the equivalent circuit fractional-order model library; The electrochemical system of the retired lithium battery is determined according to the target equivalent circuit integer-order model and the target equivalent circuit fractional-order model.
[0005] Optionally, determining a target equivalent circuit integer-order model from the equivalent circuit integer-order model library according to the EIS test data includes: Analyzing the EIS test data by a relaxation time distribution method to obtain a relaxation time distribution graph corresponding to the EIS test data; wherein the relaxation time distribution graph includes a plurality of characteristic peaks; Determine a target characteristic peak and the number of extreme value points in the target characteristic peak; wherein the target characteristic peak is the remaining characteristic peak after removing the first characteristic peak from the multiple characteristic peaks; Determine whether the number of extreme value points is an even number or an odd number; If the number of extreme value points is an even number, the target equivalent circuit integer-order model is determined by the following formula: ,in, is the order of the integer-order model of the target equivalent circuit, N is the number of extreme points; If the number of extreme value points is an odd number, the target equivalent circuit integer-order model is determined by the following formula: .
[0006] Optionally, each of the equivalent circuit fractional-order models includes initial parameters, and determining a target equivalent circuit fractional-order model from the equivalent circuit fractional-order model library includes: Inputting the EIS test data into each of the equivalent circuit fractional-order models, so that each of the equivalent circuit fractional-order models performs fitting processing according to the EIS test data; Obtaining first fitting parameters used in fitting each of the equivalent circuit fractional-order models; Determining a parameter fitting error of each of the equivalent circuit fractional-order models according to the initial parameters of each of the equivalent circuit fractional-order models and the first fitting parameters; According to the parameter fitting error, the equivalent circuit fractional-order models are sorted, and the first M equivalent circuit fractional-order models with the highest order are used as target equivalent circuit fractional-order models.
[0007] Optionally, the constructing of an equivalent circuit integer-order model library and an equivalent circuit fractional-order model library includes: Establishing a plurality of initial equivalent circuit models; wherein the plurality of initial equivalent circuit models include a plurality of initial equivalent circuit integer-order models of different orders, and a plurality of initial equivalent circuit fractional-order models of different types; Acquire sample EIS test data, and input the sample EIS test data into each of the initial equivalent circuit models, so that each of the initial equivalent circuit models performs fitting processing according to the sample EIS test data; Obtaining second fitting parameters used in fitting each of the initial equivalent circuit models, and fitting EIS test data; Determining the EIS fitting error of each of the initial equivalent circuit models according to the sample EIS test data and each of the fitted EIS test data; When the EIS fitting error of the initial equivalent circuit model meets a preset condition, using the second fitting parameters of the initial equivalent circuit model as initial parameters of the initial equivalent circuit model; According to each of the initial equivalent circuit models and the initial parameters corresponding to the initial equivalent circuit models, an equivalent circuit integer-order model library and an equivalent circuit fractional-order model library are constructed.
[0008] Optionally, the method further includes: When the EIS fitting error of the initial equivalent circuit model satisfies the first preset condition, obtaining a first evaluation index of the initial equivalent circuit model; Determining whether a first evaluation index of the initial equivalent circuit model meets a second preset condition; If the first evaluation index of the initial equivalent circuit model satisfies the second preset condition, constructing an equivalent circuit integer-order model library and an equivalent circuit fractional-order model library according to each of the initial equivalent circuit models and the initial parameters corresponding to the initial equivalent circuit model; If the first evaluation index of the initial equivalent circuit model does not satisfy the second preset condition, updating the initial equivalent circuit model until the first evaluation index of the initial equivalent circuit model satisfies the second preset condition; The updating process includes: adjusting at least one of a second fitting parameter of the initial equivalent circuit model, adjusting the order of the initial equivalent circuit model, and adjusting the sample EIS test data.
[0009] Optionally, the method further includes: When the EIS fitting error of the initial equivalent circuit model meets the preset condition, determining the root mean square error and the Matthews correlation coefficient when each of the initial equivalent circuit models is fitted; According to the root mean square error and the Matthews correlation coefficient when each of the initial equivalent circuit models is fitted, a comprehensive evaluation index of each of the initial equivalent circuit models is determined.
[0010] Optionally, the method further includes: When the ranking of the equivalent circuit fractional-order models continues to be the last K in a preset time period, the equivalent circuit fractional-order models ranked last are deleted from the equivalent circuit fractional-order model library.
[0011] Optionally, before obtaining the EIS test data of the retired lithium battery, the method further includes: Measuring the terminal voltage of the retired lithium battery; If the terminal voltage is less than a preset voltage threshold, charging the retired lithium battery; When the terminal voltage rising speed of the retired lithium battery is less than a preset speed threshold, charging is stopped.
[0012] In a second aspect, the present invention also provides an electronic device comprising a memory and a processor, wherein the memory is used to store programs; and the processor is coupled to the memory and is used to execute the programs stored in the memory to implement the steps in any one of the above-mentioned methods for identifying the electrochemical system of retired lithium batteries.
[0013] In a third aspect, the present invention further provides a computer-readable storage medium for storing a computer-readable program, wherein the program or instruction, when executed by a processor, can implement the steps in any one of the above-mentioned methods for identifying the electrochemical system of retired lithium batteries.
[0014] The beneficial effects of the present invention are: The present invention first constructs an equivalent circuit integer-order model library and an equivalent circuit fractional-order model library; then obtains the EIS test data of the retired lithium battery of the electrochemical system to be determined; and matches the target equivalent circuit integer-order model from the equivalent circuit integer-order model library through the EIS test data, and matches the target equivalent circuit fractional-order model from the equivalent circuit fractional-order model library; finally, the electrochemical system of the retired lithium battery is determined according to the target equivalent circuit integer-order model and the target equivalent circuit fractional-order model. The present invention can identify the electrochemical system of various types of retired lithium batteries, and solve the problem that the type of lithium battery system cannot be identified due to the lack of labels in the lithium battery recycling application, and then it cannot be recycled. And the present invention is simple to operate and easy to promote in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 A schematic diagram of a process flow of an embodiment of a method for identifying an electrochemical system of a retired lithium battery provided by the present invention; Figure 2 A schematic diagram of a third-order equivalent circuit model provided by the present invention; Figure 3 A schematic diagram of a second type fractional-order equivalent circuit model provided by the present invention; Figure 4 A schematic diagram of a first type fractional-order equivalent circuit model provided by the present invention; Figure 5 A schematic diagram of a third type fractional-order equivalent circuit model provided by the present invention; Figure 6 A technical process flow chart of a method for identifying an electrochemical system of a retired lithium battery provided by the present invention. DETAILED DESCRIPTION
[0016] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.
[0017] In the description of the embodiments of the present invention, unless otherwise specified, "multiple" means two or more than two. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" may mean: A exists alone, A and B exist at the same time, and B exists alone.
[0018] The descriptions of "first", "second", etc. involved in the embodiments of the present invention are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the technical features defined as "first" and "second" may explicitly or implicitly include at least one of the features.
[0019] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.
[0020] Reference Figure 1 , shows a schematic flow chart of an embodiment of a method for identifying an electrochemical system of a retired lithium battery provided by the present invention, the method comprising: Step S101, constructing an equivalent circuit integer-order model library and an equivalent circuit fractional-order model library; the equivalent circuit integer-order model library includes a plurality of equivalent circuit integer-order models of different orders, and the equivalent circuit fractional-order model library includes a plurality of different types of equivalent circuit fractional-order models.
[0021] The equivalent circuit integer model library can include multiple equivalent circuit integer models of different orders. The equivalent circuit integer model can be an n-order RC equivalent circuit model, where n is a positive integer. The model is composed of basic components such as resistors (R) and capacitors (C) connected in series or in parallel, and each R and C parallel link represents the electrochemical process inside the battery.
[0022] Since it is difficult to accurately fit the EIS images of the same battery at different temperatures and different states of charge (SOC) using an n-order RC equivalent circuit model in many cases, this embodiment also constructs an equivalent circuit fractional-order model library, which includes multiple different types of equivalent circuit fractional-order models, and different types can refer to different orders or the same order but different model structures. Various types of equivalent circuit fractional-order models can more accurately describe the complex concentration polarization reaction inside the battery and more accurately fit the EIS images of the same battery at different temperatures and different SOCs.
[0023] The equivalent circuit fractional-order model can be composed of a constant phase angle element (CPE element) with a fractional order and a Warburg element, and the order of the Warburg element is fixed at 0.5.
[0024] Step S102, obtaining EIS test data of retired lithium batteries.
[0025] According to the electrochemical system, retired lithium batteries are generally divided into lithium iron phosphate batteries, ternary lithium batteries and lithium titanate batteries. This embodiment can perform an electrochemical impedance spectroscopy (EIS) test on retired lithium batteries of unknown type or unknown electrochemical system, and obtain EIS test data of the retired lithium batteries.
[0026] The retired lithium batteries used for testing should have no damage on the outside, and should not have any problems such as bloating, leakage, or deformation.
[0027] In one embodiment, before performing an EIS test on a retired lithium battery, the terminal voltage of the retired lithium battery can be measured first; if the terminal voltage is less than a preset voltage threshold, the retired lithium battery is charged; when the rising speed of the terminal voltage of the retired lithium battery is less than a preset speed threshold, charging is stopped and then the battery is allowed to stand for a long time.
[0028] After sufficient rest, the EIS test can be performed. The EIS test can be performed using an electrochemical impedance test device. The test process includes: connecting the retired lithium battery to the voltage and current sensors of the test equipment, and then testing to obtain the impedance data at each frequency point, which is the EIS test data. The impedance data at each frequency point includes: frequency distribution , the real part of impedance and the negative imaginary part of the impedance ;in, represents the frequency of the i-th point, represents the real part of impedance at the ith frequency, Represents the negative imaginary part of the impedance at the i-th frequency.
[0029] Furthermore, the point data with a negative imaginary part less than 0 in the EIS data can be eliminated. The eliminated point data represents the influence of the inductance of the test cable. The data below the 0 axis in the EIS curve, that is, the data to the lower left of the dotted line, is eliminated, and the remaining point data is used as valid EIS test data for subsequent steps.
[0030] Step S103 , determining a target equivalent circuit integer-order model from an equivalent circuit integer-order model library according to the EIS test data, and determining a target equivalent circuit fractional-order model from an equivalent circuit fractional-order model library.
[0031] After obtaining the EIS test data, the target equivalent circuit integer-order model can be matched from the equivalent circuit integer-order model library according to the EIS test data, and the target equivalent circuit fractional-order model can be matched from the equivalent circuit fractional-order model library. The target equivalent circuit integer-order model can be one or more equivalent circuit integer-order models in the equivalent circuit integer-order model library that best fit the EIS test data. Similarly, the target equivalent circuit fractional-order model can be one or more equivalent circuit fractional-order models in the equivalent circuit fractional-order model library that best fit the EIS test data.
[0032] Step S104, determining the electrochemical system of the retired lithium battery according to the target equivalent circuit integer-order model and the target equivalent circuit fractional-order model.
[0033] One equivalent circuit corresponds to one electrochemical system, and different equivalent circuits may correspond to the same electrochemical system or different electrochemical systems. After determining the target equivalent circuit integer-order model and the target equivalent circuit fractional-order model in this embodiment, the electrochemical system of the retired lithium battery may be determined based on the electrochemical system corresponding to the target equivalent circuit integer-order model and the electrochemical system corresponding to the target equivalent circuit fractional-order model.
[0034] The method of this embodiment can be used to identify the electrochemical system of various types of retired lithium batteries, solving the problem that the type of lithium battery system cannot be identified due to missing labels in lithium battery recycling applications, and thus cannot be recycled. In addition, the present invention is simple to operate and easy to promote in practical applications.
[0035] In addition, in the prior art, there are the following methods for identifying the type of battery: (1) Physical appearance feature analysis. This method can only roughly estimate the battery type. For power batteries such as ternary lithium batteries and lithium iron phosphate batteries, it is impossible to judge from the appearance.
[0036] (2) Electrochemical performance test. Electrochemical performance test includes single EIS test and cyclic voltammetry (CV) curve test. The single EIS test method requires manual comparison of electrochemical impedance spectra to confirm the type of battery. It is too dependent on the experience of relevant personnel and is not generalizable to meet the requirements of mass production. The results of CV curve test are affected by many factors, including electrode materials, electrolytes, electrode surface conditions, etc. Therefore, in-depth professional knowledge and experience are required to interpret and analyze the test results. It is not generalizable enough, and different types of batteries may have overlapping or similar features on the CV curve, which makes it difficult to accurately distinguish different electrochemical systems. It is not suitable for accurately identifying electrochemical systems of retired batteries with a large number of missing labels.
[0037] (3) Machine learning and artificial intelligence technologies. This type of method usually uses machine learning and artificial intelligence technologies to build a classification model based on battery characteristic data. By training data on known battery types, the characteristic data of unknown batteries is input into the model to automatically identify the electrochemical system of the battery. This method requires a large number of tedious experiments in the early stage, and requires frequent charging and discharging, consuming a large amount of electricity.
[0038] Compared with the physical appearance characteristic analysis method, the present invention can non-destructively and accurately inspect the battery electrochemical system to which the retired lithium batteries belong by performing EIS tests on retired lithium batteries and analyzing the EIS test data.
[0039] Compared with the single EIS test and analysis method, the present invention pre-constructs an equivalent circuit integer-order model library and an equivalent circuit fractional-order model library, and determines the electrochemical system of retired lithium batteries by automatically matching the EIS test data of retired lithium batteries with the models in the model library. The operation is simpler and easy to promote in practical applications.
[0040] Compared with machine learning and artificial intelligence recognition methods, the present invention does not require a large number of tests to obtain training samples, thereby reducing the investment in time and financial resources and having high cost-effectiveness.
[0041] In one embodiment, step S103 may specifically include: analyzing the EIS test data by a relaxation time distribution method (Distribution of Relaxation Times, DRT) to obtain a relaxation time distribution diagram corresponding to the EIS test data; wherein the relaxation time distribution diagram includes multiple characteristic peaks; determining a target characteristic peak and the number of extreme value points in the target characteristic peak; wherein the target characteristic peak is a remaining characteristic peak after removing the first characteristic peak from the multiple characteristic peaks; determining whether the number of extreme value points is an even number or an odd number; if the number of extreme value points is an even number, determining the target equivalent circuit integer order model by the following formula: ,in, is the order of the integer-order model of the target equivalent circuit, N is the number of extreme points. If the number of extreme points is an odd number, the target equivalent circuit integer-order model is determined by the following formula: .
[0042] The expression of the DRT function is:
[0043] in: For Gaussian distribution function centered at ; is the shape factor of the Gaussian function, is the weight of the mth basis function.
[0044] The EIS test data can be analyzed by the DRT function to obtain the relaxation time distribution diagram corresponding to the EIS test data. There are generally 3-5 peaks in the relaxation time distribution diagram. The first peak is considered to be related to the ohmic internal resistance of the battery. The definition starts from the second peak of the relaxation time distribution diagram and calculates the extreme points. Then, the order of the integer-order model of the target equivalent circuit can be determined by combining the number of extreme points and the above formula. For example, when the number of extreme points is 5, the above formula can be used to determine that the order of the integer-order model of the target equivalent circuit is 1st order, then the integer-order model of the target equivalent circuit can be a 1st-order RC equivalent circuit.
[0045] This embodiment can quickly and accurately identify the target equivalent circuit integer-order model corresponding to the EIS test data through the DRT function.
[0046] In one embodiment, each equivalent circuit fractional-order model includes initial parameters, and step S103 may specifically include: inputting EIS test data into each equivalent circuit fractional-order model so that each equivalent circuit fractional-order model is fitted according to the EIS test data; obtaining the first fitting parameters used when fitting each equivalent circuit fractional-order model; determining the parameter fitting error of each equivalent circuit fractional-order model according to the initial parameters and the first fitting parameters of each equivalent circuit fractional-order model; and sorting each equivalent circuit fractional-order model according to the parameter fitting error, and taking the top M equivalent circuit fractional-order models as the target equivalent circuit fractional-order models.
[0047] The determination of the target equivalent circuit fractional-order model requires fitting the EIS test data with each equivalent circuit fractional-order model in the equivalent circuit fractional-order model library to determine the first fitting parameters of each equivalent circuit fractional-order model, and then compare the parameter fitting errors of each equivalent circuit fractional-order model.
[0048] The first fitting parameter of each equivalent circuit fractional-order model is , The function defining the parameter fitting error is: , is the parameter fitting error of the model. The equivalent circuit fractional-order models in the equivalent circuit fractional-order model library are sorted according to the numerical value of the parameter fitting error, and the first M models with the highest ranking are used as the target equivalent circuit fractional-order models, where M can be 2.
[0049] In one embodiment, step S101 may specifically include the following sub-steps S11-S16: Sub-step S11, establishing a plurality of initial equivalent circuit models; wherein the plurality of initial equivalent circuit models include a plurality of initial equivalent circuit integer-order models of different orders, and a plurality of initial equivalent circuit fractional-order models of different types.
[0050] The initial equivalent circuit integer-order model may include an initial first-order RC equivalent circuit model, an initial first-order RC equivalent circuit model, an initial first-order RC equivalent circuit model, etc. The initial equivalent circuit integer-order model is an equivalent circuit integer-order model whose model parameters have not been determined.
[0051] Reference Figure 2 , showing a schematic diagram of a third-order equivalent circuit model provided by the present invention. Figure 2 Middle U OC represents the open circuit voltage, R0 represents the ohmic internal resistance; U EC Represents the terminal voltage of the battery, R Pi and C Pi Constitute an RC network, representing the i-th order RC network, R Pi Represents the i-th order polarization resistance of the battery, C Pi Represents the i-th order polarization capacitance of the battery.
[0052] According to Kirchhoff's current-voltage law:
[0053]
[0054] First-order and second-order equivalent circuit modeling and Figure 2 The third-order equivalent circuit model in is similar to that in , except that there is an i-th-order RC network.
[0055] The initial equivalent circuit integer-order model may include multiple initial equivalent circuit fractional-order models of different types. Figure 3 , showing a schematic diagram of a second type of fractional-order equivalent circuit model provided by the present invention. Figure 3 In the example, CPE1 represents the order The constant phase angle element, Warburg represents The order of the Warburg impedance element is 1 / 2. The electrical characteristic equation of the second type of fractional-order equivalent circuit model can be expressed as:
[0056]
[0057] The differential equation in the parallel circuit is expressed as:
[0058]
[0059] in The CPE constant phase angle element The fractional-order model, is the coulombic efficiency, Q is the current battery charge, and S(t) is the current state of charge of the battery.
[0060] Reference Figure 4 , showing a schematic diagram of a first type fractional-order equivalent circuit model provided by the present invention. Figure 5 , shows a schematic diagram of a third type of fractional-order equivalent circuit model provided by the present invention. The model-related electrical characteristic equations corresponding to the first and third types of models are similar to the electrical characteristic equations of the second type of fractional-order equivalent circuit model, and only the parameters in the same components need to be replaced.
[0061] Sub-step S12, acquiring sample EIS test data, and inputting the sample EIS test data into each initial equivalent circuit model, so that each initial equivalent circuit model performs fitting processing according to the sample EIS test data.
[0062] After obtaining the sample EIS test data, the point data with a negative imaginary part less than 0 in the sample EIS test data can be eliminated, and then the data at each point can be normalized and then input into each initial equivalent circuit model for fitting.
[0063] The normalization formula is:
[0064] in is the maximum value in the sample EIS test data, is the minimum value in the sample EIS test data.
[0065] Sub-step S13, obtaining the second fitting parameters used in fitting each initial equivalent circuit model, and fitting EIS test data.
[0066] During the fitting process, the parameters (second fitting parameters) in each initial equivalent circuit model can be identified by an intelligent algorithm, and then their specific values can be determined. For the initial equivalent circuit integer-order model, its model parameters can be open-circuit voltage, ohmic internal resistance, polarization resistance and capacitance in each i-order RC network, terminal voltage, etc. For the initial equivalent circuit fractional-order model, its model parameters can be , , R i and S(t), etc.
[0067] In order to enable the intelligent algorithm to better identify the parameters in the model, a model fitting parameter level set can be formed:
[0068] Y represents the level set, Represents the g-th equivalent circuit model parameter in the l-th model in the model library Z, and G represents each model in the model library.
[0069] The intelligent algorithm is trained through the model fitting parameter level set to improve its recognition effect.
[0070] Sub-step S14, determining the EIS fitting errors of the respective initial equivalent circuit models according to the sample EIS test data and the respective fitted EIS test data.
[0071] The EIS fitting error can be expressed by the following formula:
[0072] Where i represents all frequency points after data processing, and Respectively represent the real and imaginary parts of the impedance measured at the i-th frequency in the sample EIS test data, and They represent the real and imaginary parts of the impedance in the fitted EIS test data calculated by the equivalent circuit model.
[0073] Sub-step S15, when the EIS fitting error of the initial equivalent circuit model meets the first preset condition, taking the second fitting parameters of the initial equivalent circuit model as the initial parameters of the initial equivalent circuit model.
[0074] When the EIS fitting error of the initial equivalent circuit model reaches a minimum, the second fitting parameters used to minimize the EIS fitting error of the initial equivalent circuit model are used as initial parameters of the initial equivalent circuit model.
[0075] Sub-step S16, constructing an equivalent circuit integer-order model library and an equivalent circuit fractional-order model library according to each initial equivalent circuit model and the initial parameters corresponding to the initial equivalent circuit model.
[0076] An equivalent circuit integer-order model library is constructed according to each initial equivalent circuit integer-order model and its corresponding initial parameters, and an equivalent circuit fractional-order model library is constructed according to each initial equivalent circuit fractional-order model and its corresponding initial parameters.
[0077] In one embodiment, sub-step S13 is iterated multiple times until the EIS fitting error of the EIS test data obtained according to sub-step S13 satisfies a preset condition. The number of iterations of sub-step S13 can be determined according to the actual discreteness of the fitting curve, which can be determined by , , When the discreteness is less than the given threshold, it is a good fit and serves as the end condition of the iteration.
[0078] By iteratively calculating the algorithm to find the optimal parameters (i.e., the second fitting parameters used when the EIS fitting error is minimized), the model fitting error after using the optimal parameters can be calculated:
[0079] Where M is the total number of data points, is the actual impedance value of the jth data point, is the predicted impedance value, and i is the i-th model in the fractional-order equivalent circuit model library.
[0080] In one embodiment, the method for identifying the electrochemical system of retired lithium batteries also includes: when the EIS fitting error of the initial equivalent circuit model meets the first preset condition, obtaining a first evaluation index of the initial equivalent circuit model; judging whether the first evaluation index of the initial equivalent circuit model meets the second preset condition; if the first evaluation index of the initial equivalent circuit model meets the second preset condition, constructing an equivalent circuit integer-order model library and an equivalent circuit fractional-order model library according to the initial parameters corresponding to each initial equivalent circuit model and the initial equivalent circuit model; if the first evaluation index of the initial equivalent circuit model does not meet the second preset condition, updating the initial equivalent circuit model until the first evaluation index of the initial equivalent circuit model meets the second preset condition; wherein the updating process includes: adjusting at least one of the second fitting parameter of the initial equivalent circuit model, adjusting the order of the initial equivalent circuit model, and adjusting the sample EIS test data.
[0081] The first evaluation index can be any index for evaluating the fitting effect of the initial equivalent circuit model. Judging whether the first evaluation index of the initial equivalent circuit model meets the second preset condition can be judging whether the fitting effect of the initial equivalent circuit model is good enough. Specifically, it can be judging whether the first evaluation index is greater than the index threshold or less than the index threshold. For example, the first evaluation index can be the root mean square error (RMSE), which can be determined based on the sample EIS test data and the fitted EIS test data. The root mean square error can be expressed by the following formula: , N=1,2,…,M Where n represents the total number of samples of impedance data, Represents the actual measured impedance data, represents the impedance data of the equivalent circuit fit, is the Nth equivalent model in the integer-order model library. The evaluation index of the Nth model in the integer-order model library can be obtained as follows: .
[0082] Then, the second preset condition may be that the first evaluation index is less than a preset threshold, that is, RMSE is less than a preset threshold. When the EIS fitting error of the initial equivalent circuit model satisfies the first preset condition, and the first evaluation index of the initial equivalent circuit model satisfies the second preset condition, at this time, according to each initial equivalent circuit model and its corresponding initial parameter, an equivalent circuit integer order model library and an equivalent circuit fractional order model library are constructed. If the EIS fitting error of the initial equivalent circuit model satisfies the first preset condition, but the first evaluation index of the initial equivalent circuit model does not meet the second preset condition, the initial equivalent circuit model is updated, and the update process includes: updating the second fitting parameter of the initial equivalent circuit model, increasing or decreasing the order of the RC network, etc.
[0083] In one embodiment, the method for identifying the electrochemical system of retired lithium batteries also includes: when the EIS fitting error of the initial equivalent circuit model meets the preset conditions, determining the root mean square error and the Matthews correlation coefficient when fitting each initial equivalent circuit model; and determining the comprehensive evaluation index of each initial equivalent circuit model based on the root mean square error and the Matthews correlation coefficient when fitting each initial equivalent circuit model.
[0084] The Matthews Correlation Coefficient (MCC), whose value is between 0 and 1, can be used to evaluate the goodness of model fit.
[0085] The comprehensive evaluation index can be expressed by the following formula:
[0086] The smaller S is, the better the model fitting effect is, and each model in the equivalent circuit model library is sorted according to the size of the evaluation index.
[0087] In one embodiment, the method for identifying the electrochemical system of retired lithium batteries further includes: when the ranking of equivalent circuit fractional-order models continues to be the last K within a preset time period, deleting the equivalent circuit fractional-order models with the last ranking from the equivalent circuit fractional-order model library.
[0088] For example, if a certain equivalent circuit fractional-order model is continuously ranked among the last five within a month, the equivalent circuit fractional-order model with the lowest ranking may be deleted from the equivalent circuit fractional-order model library.
[0089] Reference Figure 6 , showing a technical route flow chart of a method for identifying the electrochemical system of retired lithium batteries provided by the present invention. For retired lithium batteries with missing labels, their terminal voltage can be tested first. If the terminal voltage is too low, the battery is charged at least until the terminal voltage rise rate no longer increases significantly, and the EIS test is performed after sufficient rest. After obtaining the EIS test data, the test data can be preprocessed, and the preprocessing includes: eliminating point data with a negative imaginary part less than 0. Then, according to the EIS test data, a target equivalent circuit integer-order model is matched from the equivalent circuit integer-order model library, and a target equivalent circuit fractional-order model is matched from the equivalent circuit integer-order model library; the electrochemical system of the retired lithium battery is determined according to the target equivalent circuit integer-order model and the target equivalent circuit fractional-order model. The equivalent circuit integer-order model library and the equivalent circuit models in the equivalent circuit integer-order model library will be updated, equivalent circuit models with poor fitting effects will be deleted, and new equivalent circuit models will be added.
[0090] In one embodiment, the present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the steps of any one of the above-mentioned methods for identifying the electrochemical system of retired lithium batteries are implemented.
[0091] In one embodiment, the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by the processor, the steps of any one of the above-mentioned methods for identifying the electrochemical system of retired lithium batteries are implemented.
[0092] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.
[0093] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.
Claims
1. A method for identifying the electrochemical system of retired lithium batteries, characterized in that: include: Constructing an equivalent circuit integer-order model library and an equivalent circuit fractional-order model library; the equivalent circuit integer-order model library includes a plurality of equivalent circuit integer-order models of different orders, and the equivalent circuit fractional-order model library includes a plurality of different types of equivalent circuit fractional-order models; Obtain EIS test data of retired lithium batteries; Determining a target equivalent circuit integer-order model from the equivalent circuit integer-order model library according to the EIS test data, and determining a target equivalent circuit fractional-order model from the equivalent circuit fractional-order model library; The electrochemical system of the retired lithium battery is determined according to the target equivalent circuit integer-order model and the target equivalent circuit fractional-order model.
2. The electrochemical system identification method of retired lithium batteries according to claim 1, characterized in that: Determining a target equivalent circuit integer-order model from the equivalent circuit integer-order model library according to the EIS test data includes: Analyzing the EIS test data by a relaxation time distribution method to obtain a relaxation time distribution graph corresponding to the EIS test data; wherein the relaxation time distribution graph includes a plurality of characteristic peaks; Determine a target characteristic peak and the number of extreme value points in the target characteristic peak; wherein the target characteristic peak is the remaining characteristic peak after removing the first characteristic peak from the plurality of characteristic peaks; Determine whether the number of extreme value points is an even number or an odd number; If the number of extreme value points is an even number, the target equivalent circuit integer-order model is determined by the following formula: ,in, is the order of the integer-order model of the target equivalent circuit, N is the number of extreme points; If the number of extreme value points is an odd number, the target equivalent circuit integer-order model is determined by the following formula: 。 3. The electrochemical system identification method of retired lithium batteries according to claim 1, characterized in that: Each of the equivalent circuit fractional-order models includes initial parameters, and determining a target equivalent circuit fractional-order model from the equivalent circuit fractional-order model library includes: Inputting the EIS test data into each of the equivalent circuit fractional-order models, so that each of the equivalent circuit fractional-order models performs fitting processing according to the EIS test data; Obtaining first fitting parameters used in fitting each of the equivalent circuit fractional-order models; Determining a parameter fitting error of each of the equivalent circuit fractional-order models according to the initial parameters of each of the equivalent circuit fractional-order models and the first fitting parameters; According to the parameter fitting error, the equivalent circuit fractional-order models are sorted, and the first M equivalent circuit fractional-order models with the highest sorting order are used as target equivalent circuit fractional-order models.
4. The electrochemical system identification method of retired lithium batteries according to claim 1, characterized in that: The constructing of the equivalent circuit integer-order model library and the equivalent circuit fractional-order model library comprises: Establishing a plurality of initial equivalent circuit models; wherein the plurality of initial equivalent circuit models include a plurality of initial equivalent circuit integer-order models of different orders, and a plurality of initial equivalent circuit fractional-order models of different types; Acquire sample EIS test data, and input the sample EIS test data into each of the initial equivalent circuit models, so that each of the initial equivalent circuit models performs fitting processing according to the sample EIS test data; Obtaining second fitting parameters used in fitting each of the initial equivalent circuit models, and fitting EIS test data; Determining the EIS fitting error of each of the initial equivalent circuit models according to the sample EIS test data and each of the fitted EIS test data; When the EIS fitting error of the initial equivalent circuit model satisfies a first preset condition, using the second fitting parameters of the initial equivalent circuit model as initial parameters of the initial equivalent circuit model; According to each of the initial equivalent circuit models and the initial parameters corresponding to the initial equivalent circuit models, an equivalent circuit integer-order model library and an equivalent circuit fractional-order model library are constructed.
5. The electrochemical system identification method of retired lithium batteries according to claim 4, characterized in that: The method further comprises: When the EIS fitting error of the initial equivalent circuit model satisfies the first preset condition, obtaining a first evaluation index of the initial equivalent circuit model; Determining whether a first evaluation index of the initial equivalent circuit model meets a second preset condition; If the first evaluation index of the initial equivalent circuit model satisfies the second preset condition, constructing an equivalent circuit integer-order model library and an equivalent circuit fractional-order model library according to each of the initial equivalent circuit models and the initial parameters corresponding to the initial equivalent circuit model; If the first evaluation index of the initial equivalent circuit model does not satisfy the second preset condition, updating the initial equivalent circuit model until the first evaluation index of the initial equivalent circuit model satisfies the second preset condition; The updating process includes: adjusting at least one of a second fitting parameter of the initial equivalent circuit model, adjusting the order of the initial equivalent circuit model, and adjusting the sample EIS test data.
6. The electrochemical system identification method of retired lithium batteries according to claim 4, characterized in that: The method further comprises: When the EIS fitting error of the initial equivalent circuit model meets the preset condition, determining the root mean square error and the Matthews correlation coefficient when each of the initial equivalent circuit models is fitted; According to the root mean square error and the Matthews correlation coefficient when each of the initial equivalent circuit models is fitted, a comprehensive evaluation index of each of the initial equivalent circuit models is determined.
7. The electrochemical system identification method of retired lithium batteries according to claim 3, characterized in that: The method further comprises: When the ranking of the equivalent circuit fractional-order models continues to be the last K in a preset time period, the equivalent circuit fractional-order models ranked last are deleted from the equivalent circuit fractional-order model library.
8. The electrochemical system identification method of retired lithium batteries according to claim 1, characterized in that: Before obtaining the EIS test data of the retired lithium battery, the method further includes: Measuring the terminal voltage of the retired lithium battery; If the terminal voltage is less than a preset voltage threshold, charging the retired lithium battery; When the terminal voltage rising speed of the retired lithium battery is less than a preset speed threshold, charging is stopped.
9. An electronic device, characterized in that: comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the method for identifying the electrochemical system of retired lithium batteries as described in any one of claims 1 to 8.
10. A computer-readable storage medium, characterized in that: Used to store computer-readable programs, which, when executed by a processor, can implement the steps of the electrochemical system identification method for retired lithium batteries as described in any one of claims 1 to 8.