Method, medium and equipment for detecting internal structure fault of lead-acid battery

Abstract

The invention relates to a lead-acid battery internal structure fault detection method, a medium and equipment. The method comprises the following steps: simulating actual fault categories of electrode pairs in an electrolyte through a cyclic aging experiment to obtain impedance data of the electrode pairs under each specification under the corresponding actual fault category and an actual classification result; taking the impedance data and the actual classification result as a data set, and dividing the data set into a training set and a test set; and inputting the training set into a kirschner-extreme gradient enhancement algorithm, randomly screening impedance characteristics from the impedance data through a kirschner optimization algorithm, and then inputting the impedance data corresponding to the impedance characteristics into the extreme gradient enhancement algorithm for iterative training. The actual fault type in the lead-acid battery can be determined through the impedance data of the lead-acid battery, the lead-acid battery can be repaired, the working condition environment of the lead-acid battery can be optimized, the service life of the lead-acid battery can be prolonged, and the economic benefit of the whole life cycle of the lead-acid battery can be improved.

Description

Technical Field

[0001] This application relates to the field of battery testing technology, and in particular to a method, medium and equipment for detecting internal structural faults in lead-acid batteries. Background Technology

[0002] Lead-acid batteries are currently widely used in substations, base stations, and vehicle starting applications. Current research cannot determine the causes of changes in the health status of lead-acid batteries, identify internal structural faults, or further repair or optimize battery operation, thus accelerating battery aging and wasting the remaining value of the battery.

[0003] Therefore, it is necessary to provide a new technical solution to improve one or more of the problems existing in the above solutions.

[0004] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this application, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this application is to provide a method, medium, and device for detecting internal structural faults in lead-acid batteries, thereby overcoming, to at least some extent, one or more problems caused by limitations and defects in related technologies.

[0006] According to a first aspect of the embodiments of this application, a method for detecting internal structural faults in a lead-acid battery is provided, the method comprising:

[0007] The positive and negative plates of lead-acid batteries of the same specifications were used to form an electrode pair, which were then placed in an electrolyte with initial electrolyte height and initial electrolyte concentration and left for 12 hours.

[0008] The actual fault categories of the electrode pairs in the electrolyte are simulated by cyclic aging experiments to obtain impedance data of the electrode pairs under the corresponding actual fault categories for each specification, as well as the actual classification results; wherein, the actual fault categories include electrolyte concentration increase, electrolyte loss and electrode plate corrosion, and the actual classification results include the number of each actual fault category;

[0009] The impedance data and the actual classification results are used as a dataset, and the dataset is divided into a training set and a test set;

[0010] The training set is input into the kingfisher-extreme gradient enhancement algorithm. The kingfisher optimization algorithm randomly selects impedance features from the impedance data. Then, the impedance data corresponding to the impedance features is input into the extreme gradient enhancement algorithm for iterative training to obtain the trained extreme gradient enhancement algorithm and the optimal combination of impedance features.

[0011] Based on the optimal impedance feature combination, corresponding impedance data is selected from the test set and input into the trained extreme gradient enhancement algorithm to output the final prediction classification result.

[0012] In the embodiments of this application, before the step of simulating the actual fault categories of the electrode pairs in the electrolyte through cyclic aging experiments to obtain impedance data of the electrode pairs under the corresponding actual fault categories for each specification, and the actual classification results, the following steps are included:

[0013] Under constant temperature conditions of 25°C, the electrode pair in the electrolyte is charged and discharged using a current of 1A to determine the initial total battery capacity of the electrode pair.

[0014] In the embodiments of this application, the step of simulating the actual fault categories of the electrode pairs in the electrolyte through cyclic aging experiments to obtain impedance data of the electrode pairs under the corresponding actual fault categories for each specification, and the actual classification results, includes:

[0015] The electrode pair was subjected to a cyclic aging test using a charge-discharge method under a constant temperature of 25℃. After 10 cycles, the impedance data of the electrode pair was measured at SOC = 0.2, 0.4, 0.6, 0.8 and 1, respectively. The electrolyte concentration and electrolyte height of the electrode pair at SOC = 1 were taken as the initial state of the electrode pair; where SOC represents the battery capacity.

[0016] Impedance data were measured at each SOC for random combinations of electrolyte concentration and electrolyte height; wherein the electrolyte concentration included 21.7%, 30.4%, and 43.7%, and the electrolyte height included 7 cm, 8 cm, 9 cm, and 10 cm.

[0017] After the measurement is completed, the electrode pair is restored to its initial state of SOC=1;

[0018] Under the condition that the electrolyte height and SOC remain unchanged, the impedance data measured by changing the electrolyte concentration is regarded as Class 1 fault data; wherein, the Class 1 fault data corresponds to the impedance data when the electrolyte concentration increases;

[0019] Under the condition that the electrolyte concentration and the electrolyte height remain unchanged, the impedance data measured by changing the SOC is regarded as two types of fault data; wherein, the two types of fault data correspond to the impedance data of the plate corrosion.

[0020] Under the condition that the electrolyte concentration and SOC remain unchanged, the impedance data measured by changing the electrolyte height is regarded as three types of fault data; wherein, the three types of fault data correspond to the impedance data of electrolyte loss;

[0021] Repeat the aging cycle until the current total battery capacity of the electrode pair is less than 0.8 of the initial total battery capacity.

[0022] In the embodiments of this application, the step of inputting the training set into the kingfisher-extreme gradient enhancement algorithm, randomly selecting impedance features from the impedance data using the kingfisher optimization algorithm, and then inputting the impedance data corresponding to the impedance features into the extreme gradient enhancement algorithm for iterative training to obtain the trained extreme gradient enhancement algorithm and the optimal combination of impedance features includes:

[0023] Initialize the parameters of the extreme gradient boosting algorithm, which include the maximum depth of the tree, learning rate, weights, sampling ratio of each decision tree, column ratio of each decision tree, and number of parallel trees; wherein the maximum depth of the tree is 8, the learning rate is 0.1, the weights are 1, the sampling ratio of each decision tree is 0.9, the column ratio of each decision tree is 1, and the number of parallel trees is 1.

[0024] Initialize the parameters of the kingfisher optimization algorithm, which include the maximum number of iterations, the population size, and the crest angle; wherein the maximum number of iterations is 10, the population size is 8, and the crest angle is a random number between 0 and 2π.

[0025] The kingfisher optimization algorithm is used to randomly generate a binary array with the same dimension as the impedance data in the training set, containing only 1s and 0s. The feature positions with a value of 1 are retained, and the feature positions are used as impedance features.

[0026] The impedance data corresponding to the impedance feature is used as population particles to input the extreme gradient enhancement algorithm for iterative training, and the predicted classification result is output. The kingfisher optimization algorithm calculates the loss function based on the predicted classification result and the actual classification result.

[0027] The loss function is used as the fitness of each particle. Then, the positions of all particles are adjusted according to the particle with the smallest fitness among all particles until the number of iterations meets the maximum number of iterations. Training is then stopped, and the trained extreme gradient enhancement algorithm and the optimal impedance feature combination are obtained.

[0028] In the embodiments of this application, the step of using the impedance data corresponding to the impedance feature as population particles to input the extreme gradient enhancement algorithm for iterative training and outputting the predicted classification result, and the step of the kingfisher optimization algorithm calculating the loss function based on the predicted classification result and the actual classification result, includes:

[0029] The classification process of the extreme gradient boosting algorithm is as follows:

[0030]

[0031] Where, This represents the classification prediction results of the first t decision trees for particle i. f represents the classification result predicted by the first t-1 decision trees for particle i. t (x i Let x represent the t-th decision tree. i This represents the impedance data corresponding to the characteristic location where the value is 1;

[0032] The objective function of the t-th decision tree is:

[0033]

[0034] In the formula, Obj (t) (θ) represents the objective function of the t-th decision tree. Represents the loss function. Ω(f) represents the actual classification result of the t-th decision tree for particle i. t (x i )) represents the regularization term, and N represents the total number of particles;

[0035]

[0036] In the formula, num represents the total number.

[0037] In the embodiments of this application, the step of using the loss function as the fitness of each particle, and then adjusting the positions of all particles according to the particle with the lowest fitness among all particles until the number of iterations meets the maximum number of iterations, stopping training, and obtaining the trained extreme gradient enhancement algorithm and the optimal impedance feature combination includes:

[0038] The particle position update formula is as follows:

[0039]

[0040] In the formula, t′ represents the number of iterations t′, t′+1 represents the number of iterations t′+1, and X i (t′) represents the position of the i-th particle in the t′ iteration number, X j (t′) represents the position of the j-th particle in the t′ iteration number, X i (t′+1) represents the position of the i-th particle in the t′+1 iteration, α=2×R-1, R is a random number that satisfies a normal distribution, and T represents the habitat strategy parameter.

[0041] In the embodiments of this application, the step of using the loss function as the fitness of each particle, and then adjusting the positions of all particles according to the particle with the lowest fitness among all particles until the number of iterations meets the maximum number of iterations, stopping training, and obtaining the trained extreme gradient enhancement algorithm and the optimal impedance feature combination includes:

[0042] When the random number of particle states is in the range of 0.5-0.8, the expression for the habitat strategy parameter T in the particle position update formula is:

[0043]

[0044] In the formula, M represents the maximum number of iterations, M = 10, B represents a constant value, C represents the crown angle of the particle, t′ represents the number of iterations, the initial value is 1, and exp represents the expectation;

[0045] When the random number of particle states is in the range of 0-0.5, the expression for the habitat strategy parameter T in the particle position update formula is:

[0046]

[0047] In the formula, b represents the frequency of particle beating, F(i) represents the fitness of the i-th particle, and F(j) represents the fitness of the j-th particle.

[0048] When the random number of particle states is in the range of 0.8-1, the particle position update formula is:

[0049]

[0050] In the formula: X best (t′) represents the particle's optimal position, and H and o represent two parameters of hunting ability, respectively. The formulas for calculating H and o are as follows:

[0051]

[0052] In the formula, F represents the optimal fitness;

[0053] The particle position is updated again based on the influence of the environment on the particle's position; the formula for updating the particle position again is as follows:

[0054]

[0055] Where, X m X represents the position of random particle m in the population. n P' represents the position of a random particle n in the population, and P' represents the predation efficiency of the particle as a constant of 0.5.

[0056] When the number of iterations meets the maximum number of iterations of 10, the optimal position of the particle is output.

[0057] Based on the optimal position of the particle, retain the feature positions with a value of 1, and take all feature positions with a value of 1 as the optimal impedance feature combination.

[0058] In the embodiments of this application, the step of selecting corresponding impedance data from the test set according to the optimal impedance feature combination and inputting it into the trained extreme gradient enhancement algorithm to output the final prediction classification result includes:

[0059] The prediction accuracy is calculated based on the final predicted classification result and the actual classification result.

[0060] According to a second aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the method for detecting internal structural faults in lead-acid batteries as described in any of the above embodiments.

[0061] According to a third aspect of the embodiments of this application, an electronic device is provided, comprising:

[0062] processor; and

[0063] Memory for storing the executable instructions of the processor;

[0064] The processor is configured to execute the steps of the lead-acid battery internal structure fault detection method described in any of the above embodiments by executing the executable instructions.

[0065] The technical solutions provided by the embodiments of this application may include the following beneficial effects:

[0066] In one embodiment of this application, the above method simulates the actual fault categories of electrode pairs after disassembly of a lead-acid battery through a cyclic aging experiment. Impedance data and actual classification results of the electrode pairs under the corresponding actual fault categories are obtained, clarifying the correspondence between the impedance data of the lead-acid battery and the actual fault categories within the battery. The impedance data and actual classification results are used as a dataset, which is divided into a training set and a test set. The training set is input into the kingfisher-extreme gradient enhancement algorithm. The kingfisher optimization algorithm filters impedance features from the impedance data, and then the impedance data corresponding to the impedance features is input into the extreme gradient enhancement algorithm for iterative training to obtain a trained extreme gradient enhancement algorithm and the optimal combination of impedance features. Based on the optimal combination of impedance features, corresponding impedance data is selected from the test set and input into the trained extreme gradient enhancement algorithm to output the final predicted classification result. This application can determine the actual fault categories within a lead-acid battery using its impedance data, which not only helps repair lead-acid batteries and optimize their operating environment, extending their service life, but also improves the economic benefits throughout the entire life cycle of the lead-acid battery.

[0067] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0068] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0069] Figure 1 This schematically illustrates a flowchart of the steps of a method for detecting internal structural faults in a lead-acid battery according to an exemplary embodiment of this application.

[0070] Figure 2 This schematic diagram illustrates the electrode pair in the electrolyte in an exemplary embodiment of this application.

[0071] Figure 3 This is a schematic diagram illustrating some impedance data presented in a Nyquist plot in an exemplary embodiment of this application;

[0072] Figure 4 This schematic diagram illustrates a confusion matrix comparing predicted fault categories and actual fault categories in an exemplary embodiment of this application.

[0073] Figure 5 This illustration schematically depicts a program product according to an exemplary embodiment of this application.

[0074] Figure 6 The schematic diagram illustrates an electronic device according to an exemplary embodiment of this application. Detailed Implementation

[0075] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided to make this application more comprehensive and complete, and to fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0076] Furthermore, the accompanying drawings are merely illustrative of this application and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore, repeated descriptions of them will be omitted.

[0077] This example embodiment first provides a method for detecting internal structural faults in lead-acid batteries. (See reference...) Figure 1 As shown, the method may include steps S101 to S105.

[0078] In step S101: The positive and negative plates of lead-acid batteries of the same specifications are used to form an electrode pair, which is then placed in an electrolyte with an initial electrolyte height and initial electrolyte concentration and left for 12 hours.

[0079] Step S102: Simulate the actual fault categories of the electrode pairs in the electrolyte through cyclic aging experiments to obtain the impedance data of the electrode pairs under the corresponding actual fault categories for each specification, as well as the actual classification results; wherein, the actual fault categories include electrolyte concentration increase, electrolyte loss and plate corrosion, and the actual classification results include the number of each actual fault category.

[0080] Step S103: Use the impedance data and the actual classification results as the dataset, and divide the dataset into a training set and a test set.

[0081] Step S104: Input the training set into the kingfisher-extreme gradient enhancement algorithm, randomly select impedance features from the impedance data through the kingfisher optimization algorithm, and then input the impedance data corresponding to the impedance features into the extreme gradient enhancement algorithm for iterative training to obtain the trained extreme gradient enhancement algorithm and the optimal combination of impedance features.

[0082] Step S105: Select the corresponding impedance data from the test set according to the optimal impedance feature combination and input it into the trained extreme gradient enhancement algorithm to output the final prediction classification result.

[0083] In one embodiment of this application, the above method simulates the actual fault categories of electrode pairs after disassembly of a lead-acid battery through a cyclic aging experiment. Impedance data and actual classification results of the electrode pairs under the corresponding actual fault categories are obtained, clarifying the correspondence between the impedance data of the lead-acid battery and the actual fault categories within the battery. The impedance data and actual classification results are used as a dataset, which is divided into a training set and a test set. The training set is input into the kingfisher-extreme gradient enhancement algorithm. The kingfisher optimization algorithm randomly selects impedance features from the impedance data. Then, the impedance data corresponding to the impedance features is input into the extreme gradient enhancement algorithm for iterative training to obtain a trained extreme gradient enhancement algorithm and the optimal combination of impedance features. Based on the optimal combination of impedance features, corresponding impedance data is selected from the test set and input into the trained extreme gradient enhancement algorithm to output the final predicted classification result. This application can determine the actual fault categories within a lead-acid battery using its impedance data. This not only helps repair lead-acid batteries and optimize their operating environment, extending their service life, but also improves the economic benefits throughout the entire life cycle of the lead-acid battery.

[0084] Below, we will refer to Figures 2 to 4 The steps of the method described above in this example embodiment will be explained in more detail.

[0085] In step S101, positive and negative electrode plates of lead-acid batteries of the same specifications are combined to form an electrode pair, which is then placed in an electrolyte with initial electrolyte height and initial electrolyte concentration, and left for 12 hours.

[0086] It is understandable that "specification" refers to the battery's capacity. The positive and negative plates of a lead-acid battery can be obtained by disassembling the battery. During disassembly, lead-acid batteries of different specifications can be disassembled to obtain the positive and negative plates for each specification, thus obtaining the positive and negative plates for lead-acid batteries of the same specification. Different specifications refer to different capacities of the battery. In this application, the different specifications of lead-acid batteries include 30Ah, 45Ah, 60Ah, 90Ah, etc. Disassembling lead-acid batteries of different specifications means disassembling 30Ah, 45Ah, 60Ah, 90Ah, etc., and obtaining the positive and negative plates for each specification.

[0087] like Figure 2 As shown, positive and negative electrode pairs of each specification were assembled and placed in a transparent acrylic housing. Dilute sulfuric acid with an initial electrolyte concentration of 1:1.28 was poured into the housing to a height of 10 cm and allowed to stand for 12 hours to ensure sufficient contact between the electrolyte and the electrode pairs. The assembled electrode pairs were then connected to a battery tester and placed in a constant temperature chamber set at 25°C for 1 hour. Figure 2In the process, the Pb negative electrode (positive plate) and the PbO2 positive electrode (positive plate) form an electrode pair, and the scale is used to measure the height of the electrolyte.

[0088] In step S102, the actual fault categories of the electrode pairs in the electrolyte are simulated through a cyclic aging experiment to obtain the impedance data of the electrode pairs under the corresponding actual fault categories for each specification, as well as the actual classification results; wherein, the actual fault categories include electrolyte concentration increase, electrolyte loss and electrode plate corrosion, and the actual classification results include the number of each actual fault category.

[0089] In one embodiment, prior to the step of simulating the actual fault categories of electrode pairs in the electrolyte through cyclic aging experiments to obtain impedance data of electrode pairs under the corresponding actual fault categories for each specification, and the actual classification results, the following steps are included:

[0090] Under constant temperature conditions of 25°C, the electrode pair in the electrolyte was charged and discharged using a current of 1A to determine the initial capacity of the electrode pair.

[0091] Under normal conditions (25℃ constant temperature), the electrode pair in the electrolyte is charged and discharged using a 1A current to determine the State of Charge (SOC) of the electrode pair at different battery capacities, and the impedance data of the electrode pair is measured at SOC = 0.2, 0.4, 0.6, 0.8, and 1. The SOC measurement method uses the ampere-hour integration method. The electrode pair is charged with a 1A current until the cutoff voltage of 2.4V, and then discharged with a 1A current until the cutoff voltage of 1.8V. The current during discharge is multiplied by the discharge time to obtain the current total battery capacity C. max Repeat the above charge-discharge process three times, and take the average discharge capacity of the three cycles as the current total battery capacity C. max Unaged C max This is the initial total battery capacity. The battery charge of the electrode pair is expressed through the current total battery capacity C. max The current battery capacity, etc., can be calculated using the following formula:

[0092]

[0093] In the formula, SOC initial The current represents the current level of the battery, I represents the current during charging or discharging, z represents the charging time, and C represents the current. max This indicates the current total battery capacity.

[0094] In one embodiment, the step of simulating the actual fault categories of electrode pairs in the electrolyte through cyclic aging experiments to obtain impedance data of electrode pairs under the corresponding actual fault categories for each specification, as well as the actual classification results, includes:

[0095] The electrode pair was subjected to a cyclic aging test using a charge-discharge method under a constant temperature of 25℃. After 10 cycles, the impedance data of the electrode pair was measured at SOC = 0.2, 0.4, 0.6, 0.8 and 1 respectively. The electrolyte concentration and electrolyte height of the electrode pair at SOC = 1 were taken as the initial state of the electrode pair. Here, SOC represents the battery capacity.

[0096] Impedance data were measured at each SOC for random combinations of electrolyte concentration and electrolyte height; the electrolyte concentrations included 21.7%, 30.4%, and 43.7%, and the electrolyte heights included 7 cm, 8 cm, 9 cm, and 10 cm.

[0097] After the measurement is completed, restore the electrode pair to its initial state of SOC=1;

[0098] Under the condition that the electrolyte level and SOC remain constant, the impedance data measured by changing the electrolyte concentration is regarded as Class 1 fault data; among them, Class 1 fault data corresponds to the impedance data when the electrolyte concentration increases.

[0099] Under the condition that the electrolyte concentration and electrolyte height remain constant, the impedance data measured by changing SOC are regarded as two types of fault data; among them, the two types of fault data correspond to the impedance data of plate corrosion.

[0100] Under the condition that the electrolyte concentration and SOC remain constant, the impedance data of changing the electrolyte height measurement are regarded as three types of fault data; among them, the three types of fault data correspond to the impedance data of electrolyte loss.

[0101] Repeated cyclic aging until the current total battery capacity C of the electrode pair is reached. max Less than 0.8 of the battery's initial total capacity.

[0102] Understandably, by conducting cyclic aging tests on electrode pairs using charge-discharge methods, impedance data for each specification of electrode pair under the corresponding actual fault category, as well as the actual classification results, can be obtained.

[0103] Type 1 faults correspond to an increase in electrolyte concentration, Type 2 faults correspond to plate corrosion, and Type 3 faults correspond to electrolyte loss.

[0104] Figure 3The impedance data obtained from some measurements are displayed, presented as a Nyquist plot. The horizontal axis represents the real part of the impedance, and the vertical axis represents the negative of the imaginary part. Each data point represents the impedance data of the electrode pair in different states. The curves from left to right represent the impedance data during the battery aging process, and the impedance data correspond to the corresponding actual fault categories, including electrolyte loss, plate corrosion, and electrolyte concentration increase. The impedance data of the electrode pair was measured using an electrochemical workstation with a frequency sweep measurement method, covering a frequency range of 0.1Hz-10kHz, measuring 61 points, and using an excitation current of 0.1A. The electrode pair was allowed to stand for 30 minutes before each measurement. Figure 3 In the diagram, Class 1 represents Class 1 faults, Class 2 represents Class 2 faults, and Class 3 represents Class 3 faults.

[0105] In step S103, the impedance data and the actual classification results are used as the dataset, and the dataset is divided into a training set and a test set.

[0106] Understandably, this application randomly splits the dataset, using 70% as the training set and 30% as the test set. The training set facilitates subsequent training of the extreme gradient boosting algorithm, while the test set facilitates the validation of the trained extreme gradient boosting algorithm.

[0107] In step S104, the training set is input into the kingfisher-extreme gradient enhancement algorithm. The kingfisher optimization algorithm randomly selects impedance features from the impedance data, and then the impedance data corresponding to the impedance features is input into the extreme gradient enhancement algorithm for iterative training to obtain the trained extreme gradient enhancement algorithm and the optimal combination of impedance features.

[0108] In one embodiment, the step of inputting the training set into the kingfisher-extreme gradient boosting algorithm, randomly selecting impedance features from the impedance data using the kingfisher optimization algorithm, and then inputting the impedance data corresponding to the impedance features into the extreme gradient boosting algorithm for iterative training to obtain the trained extreme gradient boosting algorithm and the optimal combination of impedance features includes:

[0109] Initialize the parameters of the extreme gradient boosting algorithm. The parameters of the extreme gradient boosting algorithm include the maximum tree depth, learning rate, weights, sampling ratio of each decision tree, column ratio of each decision tree, and number of parallel processes. Among them, the maximum tree depth is 8, the learning rate is 0.1, the weights are 1, the sampling ratio of each decision tree is 0.9, the column ratio of each decision tree is 1, and the number of parallel processes is 1.

[0110] Initialize the parameters of the kingfisher optimization algorithm, which include the maximum number of iterations, the population size, and the crest angle; where the maximum number of iterations is 10, the population size is 8, and the crest angle is a random number between 0 and 2π.

[0111] The kingfisher optimization algorithm is used to randomly generate a binary array with the same dimension as the impedance data in the training set, containing only 1s and 0s. The feature positions with a value of 1 are retained and used as impedance features.

[0112] The impedance data corresponding to the impedance features are used as population particles to perform iterative training of the extreme gradient enhancement algorithm, and the predicted classification results are output. The kingfisher optimization algorithm calculates the loss function based on the predicted classification results and the actual classification results.

[0113] The loss function is used as the fitness of each particle. Then, the positions of all particles are adjusted according to the particle with the smallest fitness among all particles until the number of iterations meets the maximum number of iterations. Training is then stopped, and the trained extreme gradient boosting algorithm and the optimal impedance feature combination are obtained.

[0114] Understandably, the kingfisher optimization algorithm randomly generates a binary array with the same dimension as the impedance data in the training set, containing only 1s and 0s. Non-feature positions with a value of 0 are discarded, meaning the impedance data corresponding to these non-feature positions is discarded. Only feature positions with a value of 1 are retained, meaning the impedance data corresponding to these feature positions is retained. These feature positions with a value of 1 are used as impedance features. Using impedance features facilitates the subsequent selection of corresponding impedance data from the training set, effectively reducing the amount of impedance data during training.

[0115] Furthermore, the impedance data corresponding to the impedance features is used as population particles to input the extreme gradient enhancement algorithm for iterative training, outputting the predicted classification result. The step of the kingfisher optimization algorithm calculating the loss function based on the predicted classification result and the actual classification result includes:

[0116] The classification process of the extreme gradient boosting algorithm is as follows:

[0117]

[0118] In the formula, This represents the classification prediction results of the first t decision trees for particle i. f represents the classification result predicted by the first t-1 decision trees for particle i. t (x i Let x represent the t-th decision tree. i This represents the impedance data corresponding to the characteristic location where the value is 1;

[0119] The objective function of the t-th decision tree is:

[0120]

[0121] In the formula, Obj (t) (θ) represents the objective function of the t-th decision tree. Represents the loss function. Ω(f) represents the actual classification result of the t-th decision tree for particle i. t (x i )) represents the regularization term, and N represents the total number of particles;

[0122]

[0123] In the formula, num represents the total number.

[0124] It is understandable that classification can be achieved through formula (1) in the classification process of the extreme gradient boosting algorithm, that is, the output of the predicted classification result. In the process of iterative training of the extreme gradient boosting algorithm in this application, the objective function is iteratively calculated. The objective function can be calculated through formula (2). In the process of iterative training of the extreme gradient boosting algorithm in this application, the objective function is mainly calculated iteratively. In the process of iteratively calculating the objective function, the loss function is mainly calculated. The loss function can be calculated through formula (3). The loss function facilitates the subsequent adjustment of particle positions.

[0125] Furthermore, the loss function is used as the fitness of each particle. Then, the positions of all particles are adjusted based on the particle with the lowest fitness until the maximum number of iterations is reached, at which point training stops. The steps to obtain the trained extreme gradient boosting algorithm and the optimal impedance feature combination include:

[0126] The particle position update formula is:

[0127]

[0128] In the formula, t′ represents the number of iterations t′, t′+1 represents the number of iterations t′+1, and X i (t′) represents the position of the i-th particle in the t′ iteration number, X j (t′) represents the position of the j-th particle in the t′ iteration number, X i (t′+1) represents the position of the i-th particle in the t′+1 iteration, α=2×R-1, R is a random number that satisfies a normal distribution, and T represents the habitat strategy parameter.

[0129] It is understandable that when adjusting the positions of all particles based on the particle with the lowest fitness among all particles, the particle positions can be adjusted according to formula (4).

[0130] Furthermore, the loss function is used as the fitness of each particle. Then, the positions of all particles are adjusted based on the particle with the lowest fitness until the maximum number of iterations is reached, at which point training stops. The steps to obtain the trained extreme gradient boosting algorithm and the optimal impedance feature combination include:

[0131] When the random number of particle states is in the range of 0.5-0.8, the expression for the habitat strategy parameter T in the particle position update formula is:

[0132]

[0133] In the formula, M represents the maximum number of iterations, M = 10, B represents a constant value, C represents the crown angle of the particle, t′ represents the number of iterations, the initial value is 1, and exp represents the expectation;

[0134] When the random number of particle states is in the range of 0-0.5, the expression for parameter T in the particle position update formula is:

[0135]

[0136] In the formula, b represents the frequency of particle beating, F(i) represents the fitness of the i-th particle, and F(j) represents the fitness of the j-th particle.

[0137] When the random number of particle states is in the range of 0.8-1, the particle position update formula is:

[0138]

[0139] In the formula: X best (t′) represents the particle's optimal position, and H and o represent two parameters of hunting ability, respectively. The formulas for calculating H and o are as follows:

[0140]

[0141] In the formula, F represents the optimal fitness;

[0142] The particle position is updated again based on the influence of the environment on the particle's position; the formula for updating the particle position again is as follows:

[0143]

[0144] Where, X m X represents the position of random particle m in the population. n P' represents the position of a random particle n in the population, and P' represents the predation efficiency of the particle as a constant of 0.5.

[0145] When the number of iterations meets the maximum number of iterations of 10, the optimal position of the particle is output;

[0146] Based on the optimal position of the particle, retain the characteristic positions with a value of 1, and take all the characteristic positions with a value of 1 as the optimal impedance characteristic combination.

[0147] It is understandable that the random number of particle state between 0 and 1 is used to determine whether the individual kingfisher (i.e., the particle) needs to dive to hunt. When the random number of particle state is less than 0.8, it is considered that the individual kingfisher does not need to dive. In this state, the individual kingfisher also has two states: perching and hovering. These two states are random and use the same random number.

[0148] When the random number of the particle state is in the range of 0.5-0.8, the individual kingfisher is in a hovering state; when the random number of the particle state is in the range of 0-0.5, the individual kingfisher is in a roosting state; and when the random number of the particle state is in the range of 0.8-1, the individual kingfisher is in a diving and hunting state.

[0149] When the random number of the particle state is in the range of 0.5-0.8, the particle position is updated according to formula (5) and the habitat strategy parameter T in formula (4) is updated. The particle position is updated by updating the habitat strategy parameter T.

[0150] When the random number of the particle state is in the range of 0-0.5, the particle position is updated according to formula (6) and the habitat strategy parameter T in formula (4) is updated. The particle position is updated by updating the habitat strategy parameter T.

[0151] The fitness of each particle is used as the initial value, which is 1. Then the particle position is updated again according to the calculation environment, specifically according to formula (9).

[0152] When the random number range is 0.8-1, the particle position is updated according to the particle position update formula (7).

[0153] It should be noted that during the iterative training of the extreme gradient boosting algorithm, when the number of iterations reaches the maximum number of iterations of 10, the training stops and the optimal position of the particle is output. The data of the optimal position of the particle is rounded to the nearest integer, discarding non-feature positions with a value of 0 and retaining feature positions with a value of 1. All feature positions with a value of 1 retained under the optimal position of the particle are used as the best impedance feature combination, which makes it easier to select the corresponding impedance data in the future.

[0154] In step S105, the corresponding impedance data is selected from the test set according to the optimal impedance feature combination and input into the trained extreme gradient enhancement algorithm to output the final prediction classification result.

[0155] Understandably, the optimal impedance feature combination consists of all feature positions with a value of 1 retained at the particle's optimal position, while discarding non-feature positions with a value of 0, i.e., discarding the impedance data corresponding to non-feature positions with a value of 0. This effectively reduces the amount of impedance data used in the validation process by validating the trained extreme gradient boosting algorithm on a test set.

[0156] Understandably, the final prediction classification results include the number of each predicted fault category. The predicted fault categories are the same as the actual fault categories, both including electrolyte concentration increase, electrolyte loss, and plate corrosion.

[0157] In one embodiment, the step of selecting corresponding impedance data from the test set based on the optimal impedance characteristics, inputting it into a trained extreme gradient boosting algorithm, and outputting the final predicted classification result includes:

[0158] The prediction accuracy is calculated based on the final predicted classification result and the actual classification result.

[0159] Understandably, this application selects corresponding impedance data from the test set using optimal impedance features, inputs it into a trained extreme gradient enhancement algorithm, and then outputs the final predicted classification result. To verify the accuracy of its prediction of structural faults in the electrode pairs, this application calculates the prediction accuracy by comparing the final predicted classification result with the actual classification result after outputting the final predicted classification result. The final predicted classification result includes the number of each predicted fault type.

[0160] Prediction accuracy = (Number of predicted fault types / Number of corresponding actual fault categories) × 100%.

[0161] For example: when the actual fault category is Class 1, the prediction accuracy of predicting a Class 1 fault = (number of predicted Class 1 faults / number of Class 1 faults) × 100%.

[0162] When the actual fault category is Class 2, the prediction accuracy for Class 2 faults is calculated as follows: (Number of Class 2 faults / Number of Class 2 faults) × 100%.

[0163] When the actual fault category is Class 3, the prediction accuracy for Class 3 faults is calculated as follows: (Number of Class 3 faults / Number of Class 3 faults) × 100%.

[0164] like Figure 4 As shown, the horizontal axis represents the predicted fault category, and the vertical axis represents the actual fault category. In the figure, 1 on the vertical axis represents a fault of category 1, 2 on the vertical axis represents a fault of category 2, and 3 on the vertical axis represents a fault of category 3. The numbers in the squares in the figure represent the number of each predicted fault category.

[0165] Looking at the first row: 27 faults were predicted as Class 1, 0 as Class 2, and 0 as Class 3, indicating a 100% accuracy rate for predicting Class 1 faults. Looking at the second row: 1 fault was predicted as Class 1, 25 as Class 2, and 1 as Class 3, indicating a 92.7% accuracy rate for predicting Class 2 faults. Looking at the third row: 0 faults were predicted as Class 1, 0 as Class 2, and 27 as Class 3, indicating a 100% accuracy rate for predicting Class 3 faults. The verification results show that this application has a 100% accuracy rate for detecting electrolyte concentration increase (Class 1 fault), a 100% accuracy rate for detecting electrolyte loss, and a 92.7% accuracy rate for detecting plate corrosion (Class 2 fault).

[0166] In exemplary embodiments of this application, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, can implement the steps of the method for detecting internal structural faults in lead-acid batteries described in any of the above embodiments. In some possible embodiments, various aspects of the present invention can also be implemented as a program product comprising program code, which, when run on a terminal device, causes the terminal device to perform the steps of the various exemplary embodiments of the present invention described in the section on the method for detecting internal structural faults in lead-acid batteries described above.

[0167] refer to Figure 5 As shown, a program product 300 for implementing the above-described method according to an embodiment of the present invention is described. It may employ a portable compact disc read-only memory (CD-ROM) and include program code, and may run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto. In this document, the readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0168] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0169] The computer-readable storage medium may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, capable of transmitting, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0170] Program code for performing the operations of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0171] In an exemplary embodiment of this application, an electronic device is also provided, which may include a processor and a memory for storing executable instructions of the processor. The processor is configured to perform the steps of the lead-acid battery internal structure fault detection method described in any of the above embodiments by executing the executable instructions.

[0172] Those skilled in the art will understand that various aspects of the present invention can be implemented as systems, methods, or program products. Therefore, various aspects of the present invention can be specifically implemented in the following forms: entirely hardware implementations, entirely software implementations (including firmware, microcode, etc.), or implementations combining hardware and software aspects, collectively referred to herein as “circuits,” “modules,” or “systems.”

[0173] The following reference Figure 6 To describe an electronic device 600 according to this embodiment of the present invention. Figure 6 The electronic device 600 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0174] like Figure 6 As shown, the electronic device 600 is presented in the form of a general-purpose computing device. The components of the electronic device 600 may include, but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different system components (including storage unit 620 and processing unit 610), a display unit 640, etc.

[0175] The storage unit stores program code that can be executed by the processing unit 610, causing the processing unit 610 to perform the steps described in the section on the detection method for internal structural faults in lead-acid batteries according to various exemplary embodiments of the present invention. For example, the processing unit 610 can perform actions such as... Figure 1 The steps are shown in the figure.

[0176] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 6201 and / or a cache storage unit 6202, and may further include a read-only memory unit (ROM) 6203.

[0177] The storage unit 620 may also include a program / utility 6204 having a set (at least one) program module 6205, such program module 6205 including but not limited to: an operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0178] Bus 630 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the various bus structures.

[0179] Electronic device 600 can also communicate with one or more external devices 700 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 600, and / or with any device that enables electronic device 600 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 650. Furthermore, electronic device 600 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 660. Network adapter 660 can communicate with other modules of electronic device 600 via bus 630. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0180] Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this application can be embodied in the form of a software product. This software product can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, and includes several instructions to cause a computing device (such as a personal computer, server, or network device, etc.) to execute the above-described method for detecting internal structural faults in lead-acid batteries according to the embodiments of this application.

[0181] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

Claims

1. A method for detecting internal structural faults in a lead-acid battery, characterized in that, The method includes: The positive and negative plates of lead-acid batteries of the same specifications were used to form an electrode pair, which were then placed in an electrolyte with initial electrolyte height and initial electrolyte concentration and left for 12 hours. The actual fault categories of the electrode pairs in the electrolyte are simulated by cyclic aging experiments to obtain impedance data of the electrode pairs under the corresponding actual fault categories for each specification, as well as the actual classification results; wherein, the actual fault categories include electrolyte concentration increase, electrolyte loss and electrode plate corrosion, and the actual classification results include the number of each actual fault category; The impedance data and the actual classification results are used as a dataset, and the dataset is divided into a training set and a test set; The training set is input into the kingfisher-extreme gradient enhancement algorithm. The kingfisher optimization algorithm randomly selects impedance features from the impedance data. Then, the impedance data corresponding to the impedance features is input into the extreme gradient enhancement algorithm for iterative training to obtain the trained extreme gradient enhancement algorithm and the optimal combination of impedance features. Based on the optimal impedance feature combination, corresponding impedance data is selected from the test set and input into the trained extreme gradient enhancement algorithm to output the final prediction classification result.

2. The method for detecting internal structural faults in lead-acid batteries according to claim 1, characterized in that, Before the step of simulating the actual fault categories of the electrode pairs in the electrolyte through cyclic aging experiments to obtain impedance data of the electrode pairs under the corresponding actual fault categories for each specification, and the actual classification results, the following steps are included: Under constant temperature conditions of 25°C, the electrode pair in the electrolyte is charged and discharged using a current of 1A to determine the initial total battery capacity of the electrode pair.

3. The method for detecting internal structural faults in lead-acid batteries according to claim 2, characterized in that, The step of simulating the actual fault categories of the electrode pairs in the electrolyte through cyclic aging experiments to obtain impedance data of the electrode pairs under the corresponding actual fault categories for each specification, and the actual classification results, includes: The electrode pair was subjected to a cyclic aging test using a charge-discharge method under a constant temperature of 25℃. After 10 cycles, the impedance data of the electrode pair was measured at SOC = 0.2, 0.4, 0.6, 0.8 and 1, respectively. The electrolyte concentration and electrolyte height of the electrode pair at SOC = 1 were taken as the initial state of the electrode pair; where SOC represents the battery capacity. Impedance data were measured at each SOC for random combinations of electrolyte concentration and electrolyte height; wherein the electrolyte concentration included 21.7%, 30.4%, and 43.7%, and the electrolyte height included 7 cm, 8 cm, 9 cm, and 10 cm. After the measurement is completed, the electrode pair is restored to its initial state of SOC=1; Under the condition that the electrolyte height and SOC remain unchanged, the impedance data measured by changing the electrolyte concentration is regarded as Class 1 fault data; wherein, the Class 1 fault data corresponds to the impedance data when the electrolyte concentration increases; Under the condition that the electrolyte concentration and the electrolyte height remain unchanged, the impedance data measured by changing the SOC is regarded as two types of fault data; wherein, the two types of fault data correspond to the impedance data of the plate corrosion. Under the condition that the electrolyte concentration and SOC remain unchanged, the impedance data measured by changing the electrolyte height is regarded as three types of fault data; wherein, the three types of fault data correspond to the impedance data of electrolyte loss; Repeat the aging cycle until the current total battery capacity of the electrode pair is less than 0.8 of the initial total battery capacity.

4. The method for detecting internal structural faults in lead-acid batteries according to claim 1, characterized in that, The step of inputting the training set into the kingfisher-extreme gradient enhancement algorithm, randomly selecting impedance features from the impedance data using the kingfisher optimization algorithm, and then inputting the impedance data corresponding to the impedance features into the extreme gradient enhancement algorithm for iterative training to obtain the trained extreme gradient enhancement algorithm and the optimal combination of impedance features includes: Initialize the parameters of the extreme gradient boosting algorithm, which include the maximum depth of the tree, learning rate, weights, sampling ratio of each decision tree, column ratio of each decision tree, and number of parallel trees; wherein the maximum depth of the tree is 8, the learning rate is 0.1, the weights are 1, the sampling ratio of each decision tree is 0.9, the column ratio of each decision tree is 1, and the number of parallel trees is 1. Initialize the parameters of the kingfisher optimization algorithm, which include the maximum number of iterations, the population size, and the crest angle; wherein the maximum number of iterations is 10, the population size is 8, and the crest angle is a random number between 0 and 2π. The kingfisher optimization algorithm is used to randomly generate a binary array with the same dimension as the impedance data in the training set, containing only 1s and 0s. The feature positions with a value of 1 are retained, and the feature positions are used as impedance features. The impedance data corresponding to the impedance feature is used as population particles to input the extreme gradient enhancement algorithm for iterative training, and the predicted classification result is output. The kingfisher optimization algorithm calculates the loss function based on the predicted classification result and the actual classification result. The loss function is used as the fitness of each particle. Then, the positions of all particles are adjusted according to the particle with the smallest fitness among all particles until the number of iterations meets the maximum number of iterations. Training is then stopped, and the trained extreme gradient enhancement algorithm and the optimal impedance feature combination are obtained.

5. The method for detecting internal structural faults in a lead-acid battery according to claim 4, characterized in that, The step of using the impedance data corresponding to the impedance feature as population particles to input the extreme gradient enhancement algorithm for iterative training and outputting the predicted classification result, and the step of the kingfisher optimization algorithm calculating the loss function based on the predicted classification result and the actual classification result, includes: The classification process of the extreme gradient boosting algorithm is as follows: In the formula, This represents the classification prediction results of the first t decision trees for particle i. f represents the classification result predicted by the first t-1 decision trees for particle i. t (x i Let x represent the t-th decision tree. i This represents the impedance data corresponding to the characteristic location where the value is 1; The objective function of the t-th decision tree is: In the formula, Obj (t) (θ) represents the objective function of the t-th decision tree. Represents the loss function. Ω(f) represents the actual classification result of the t-th decision tree for particle i. t (x i )) represents the regularization term, and N represents the total number of particles; In the formula, num represents the total number.

6. The method for detecting internal structural faults in a lead-acid battery according to claim 5, characterized in that, The step of using the loss function as the fitness of each particle, and then adjusting the positions of all particles according to the particle with the lowest fitness, until the number of iterations meets the maximum number of iterations, stopping training, and obtaining the trained extreme gradient boosting algorithm and the optimal impedance feature combination, includes: The particle position update formula is as follows: In the formula, t′ represents the number of iterations t′, t′+1 represents the number of iterations t′+1, and X i (t′) represents the position of the i-th particle in the t′ iteration number, X j (t′) represents the position of the j-th particle in the t′ iteration number, X i (t′+1) represents the position of the i-th particle in the t′+1 iteration, α=2×R-1, R is a random number that satisfies a normal distribution, and T represents the habitat strategy parameter.

7. The method for detecting internal structural faults in a lead-acid battery according to claim 6, characterized in that, The step of using the loss function as the fitness of each particle, and then adjusting the positions of all particles according to the particle with the lowest fitness, until the number of iterations meets the maximum number of iterations, stopping training, and obtaining the trained extreme gradient boosting algorithm and the optimal impedance feature combination, includes: When the random number of particle states is in the range of 0.5-0.8, the expression for the habitat strategy parameter T in the particle position update formula is: In the formula, M represents the maximum number of iterations, M = 10, B represents a constant value, C represents the crown angle of the particle, t′ represents the number of iterations, the initial value is 1, and exp represents the expectation; When the random number of particle states is in the range of 0-0.5, the expression for parameter T in the particle position update formula is: In the formula, b represents the frequency of particle beating, F(i) represents the fitness of the i-th particle, and F(j) represents the fitness of the j-th particle. When the random number of particle states is in the range of 0.8-1, the particle position update formula is: In the formula: X best (t′) represents the particle's optimal position, and H and o represent two parameters of hunting ability, respectively. The formulas for calculating H and o are as follows: In the formula, F represents the optimal fitness; The particle position is updated again based on the influence of the environment on the particle's position; the formula for updating the particle position again is as follows: In the formula, X m X represents the position of random particle m in the population. n P' represents the position of a random particle n in the population, and P' represents the predation efficiency of the particle as a constant of 0.

5. When the number of iterations meets the maximum number of iterations of 10, the optimal position of the particle is output. Based on the optimal position of the particle, retain the feature positions with a value of 1, and take all feature positions with a value of 1 as the optimal impedance feature combination.

8. The method for detecting internal structural faults in a lead-acid battery according to claim 7, characterized in that, The step of selecting corresponding impedance data from the test set based on the optimal impedance feature combination, inputting it into the trained extreme gradient boosting algorithm, and outputting the final predicted classification result includes: The prediction accuracy is calculated based on the final predicted classification result and the actual classification result.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method for detecting internal structural faults in lead-acid batteries as described in any one of claims 1 to 8.

10. An electronic device, characterized in that, include: processor; as well as Memory for storing the executable instructions of the processor; The processor is configured to perform the steps of the method for detecting internal structural faults in a lead-acid battery according to any one of claims 1 to 8 by executing the executable instructions.

Images