Battery health degree evaluation method and system, electronic equipment and storage medium

Abstract

The invention belongs to the technical field of battery health degree evaluation, and particularly relates to a battery health degree evaluation method and system, electronic equipment and a storage medium. Based on the original sample data of the battery health degree evaluation indexes, establishing a standardized decision matrix to eliminate the dimensional difference between the indexes; constructing a correlation matrix according to the matrix, and evaluating index correlation; extracting common factors of a correlation matrix, calculating a core quantitative index, determining a main factor by taking standard reaching of an accumulated variance contribution rate as a standard, realizing data dimension reduction and retaining original index core information; and calculating a main factor score in combination with the original data and the rotated factor matrix, performing weighting according to a variance contribution rate, and performing weighted summation to obtain a health degree comprehensive score. And the health difference of the batteries can be judged in an auxiliary manner through clustering analysis, and data support is provided for reasonable configuration and fine management of the batteries. According to the method, dimension reduction weighting is carried out on various related indexes through a factor analysis method, a result is objectively deduced based on data features, subjective influences are reduced, and evaluation scientificity and reliability are improved.

Description

Technical Field

[0001] This invention belongs to the field of battery health evaluation technology, and specifically relates to a battery health evaluation method, system, electronic device, and storage medium. Background Technology

[0002] Lithium-ion batteries possess advantages such as high energy density, low self-discharge rate, no memory effect, and long cycle life, making them widely used in electric two-wheelers, electric vehicles, and energy storage. However, with increasing usage time, the capacity and power output of lithium-ion batteries inevitably degrade, affecting user experience and even posing safety hazards. Therefore, assessing the health status of lithium-ion batteries is of great practical significance and has attracted increasing attention from researchers.

[0003] Currently, reported indicators for assessing the health status of lithium batteries include a series of metrics such as discharge capacity retention, discharge energy retention, energy efficiency retention, peak power retention, internal resistance growth rate, and self-discharge growth rate. For system-level health evaluation of lithium batteries, temperature consistency and voltage consistency levels are also included. When quantifying battery health, a single indicator cannot comprehensively represent the current state of the battery, while joint scoring of multiple indicators suffers from issues such as inconsistent dimensions and potential double counting due to correlations between indicators. To address the multi-indicator evaluation problem, comprehensive evaluation methods are widely used. The core of this approach is determining the weight of each indicator, with main methods including subjective weighting (such as the analytic hierarchy process), objective weighting (such as the entropy method), and combined weighting methods. In the field of battery evaluation, relevant technologies have been reported. For example, patent application CN110927581A uses the analytic hierarchy process and entropy weight method to comprehensively evaluate the operating status indicators of energy storage batteries; patent application CN117747591A proposes a comprehensive evaluation method for fuel cells based on the AHP-EWM model; and patent application CN116258401A uses the entropy method combined with grey relational analysis to comprehensively evaluate the performance of fuel cell cells.

[0004] However, existing technologies have limitations. Subjective weighting methods rely on expert judgment, are highly subjective, and are prone to affecting the reliability of results when the number of indicators increases. Objective weighting methods (such as the entropy method) have high requirements for data quality, are easily affected by extreme values, and cannot reflect the correlation between indicators. Although the combined weighting method takes into account both subjective and objective factors, it cannot reduce the dimensionality of indicators and is difficult to simplify the evaluation process of multiple indicators. Summary of the Invention

[0005] To address the aforementioned issues, this invention employs a more objective statistical method to reduce the dimensionality of complex and intrinsically correlated evaluation indicators. This method aggregates numerous intricate indicator variables into a few independent common factors, reducing the number of variables while reflecting the intrinsic connections between them. Weights are then assigned and a comprehensive score is provided.

[0006] In a first aspect, the present invention proposes a method for evaluating battery health, comprising the following steps: A standardized decision matrix is ​​established based on the original sample data of battery health evaluation indicators; Based on the standardized decision matrix, a correlation matrix of battery health evaluation indicators is constructed to assess the correlation between various battery health evaluation indicators. Extract common factors from the correlation matrix and calculate the core quantitative indicators of the common factors. Determine the principal factors in the correlation matrix based on the core quantitative indicators. The scores of each principal factor are calculated based on the original sample data and the rotated factor matrix, and the weights of each battery health evaluation index are assigned by the variance contribution rate. Based on the weight coefficients of each battery health evaluation index, the scores of each main factor are weighted and summed to calculate the comprehensive score of battery health. Cluster analysis is performed based on the comprehensive score of battery health and the data of each principal factor. The health of the power battery is classified and profiled according to the results of the cluster analysis, and the battery health is evaluated.

[0007] Furthermore, the battery health evaluation indicators include capacity retention rate, internal resistance growth rate, self-discharge rate, voltage consistency, temperature consistency, charge / discharge efficiency, accuracy of state of charge estimation, power state, thermal diffusion rate, thermal stability, structural integrity, sealing performance, lithium plating rate, and electrolyte decomposition rate.

[0008] Furthermore, the establishment of a standardized decision matrix based on the original sample data of the battery health evaluation index includes the following steps: Obtain raw sample data of battery health evaluation indicators, and construct a decision matrix based on the raw sample data; The original sample data is preprocessed, including standardization. A standardized decision matrix is ​​constructed based on the preprocessed original sample data and the decision matrix.

[0009] Furthermore, the standardized decision matrix is ​​as follows:

[0010] In the formula, This represents the data of the m-th battery health evaluation indicator in the n-th sample after standardization; n is the total number of samples. m represents the total number of battery health evaluation indicators. .

[0011] Furthermore, the construction of a correlation matrix for battery health evaluation indicators based on the standardized decision matrix, and the assessment of the correlation between various battery health evaluation indicators, includes: A correlation matrix of the sample data for battery health evaluation indicators was constructed using the Pearson correlation coefficient. The correlation matrix was tested using the KMO test and Bartlett test to examine the correlation between the various battery health evaluation indicators. If the KMO value of the KMO test and the significance p-value of the Bartlett test both meet the preset conditions, then the battery health evaluation indicators meet the correlation requirements and are suitable for factor analysis.

[0012] Furthermore, if the KMO value of the KMO test and the significance p-value of the Bartlett test cannot simultaneously meet the preset conditions, then this set of sample data is discarded.

[0013] Furthermore, the core quantitative indicators for calculating common factors, and the determination of principal factors in the correlation matrix based on the core quantitative indicators, include: If the correlation analysis results show that the battery health evaluation indicators meet the correlation requirements, then the core quantitative indicators of each common factor in the correlation matrix are calculated. The core quantitative indicators include the eigenvalues, eigenvectors and cumulative variance contribution rates of each factor. Determine the common factors in the correlation matrix whose eigenvalues ​​are greater than 1; Determine the common factor in the correlation matrix whose eigenvalue is greater than the average of m eigenvalues, where m is the total number of battery health evaluation indicators; Extract the corresponding number of common factors whose cumulative variance contribution rate is greater than a set value; The number of common factors extracted is determined by scree plot, and the number of principal factors is determined based on the number of common factors extracted. The principal factors are determined based on the common factors determined by the eigenvalues, eigenvectors, cumulative variance contribution rate, and scree plot.

[0014] Furthermore, the scores of each principal factor are calculated based on the original sample data and the rotated factor matrix, and the weights of each battery health evaluation index are assigned using the variance contribution rate: Based on the determined principal factors, construct the initial factor loading matrix; The initial factor loading matrix is ​​rotated using orthogonal factor rotation or oblique factor rotation to obtain the rotated factor loading matrix; The factor score coefficient matrix is ​​calculated based on the rotated factor loading matrix. Using the coefficients in the factor score coefficient matrix as weights, the standardized original sample data are weighted and summed to obtain the factor score of each sample on the corresponding principal factor. The weights of the corresponding battery health evaluation indicators are determined based on the percentage of each principal factor's variance contribution rate to the cumulative variance contribution rate.

[0015] Secondly, this invention proposes a battery health evaluation system, comprising: The decision matrix building unit is used to build a standardized decision matrix based on sample data of battery health evaluation indicators. The correlation assessment unit is used to construct a correlation matrix of battery health evaluation indicators based on the standardized decision matrix, and to assess the correlation between each battery health evaluation indicator. The principal factor determination unit is used to extract common factors from the correlation matrix, calculate the core quantitative indicators of the common factors, and determine the principal factors in the correlation matrix based on the core quantitative indicators. The weight allocation unit is used to calculate the score of each principal factor based on the original sample data and the rotated factor matrix, and to allocate weights to each battery health evaluation index based on the variance contribution rate. The quantification unit is used to calculate the comprehensive score of battery health by weighting and summing the scores of each main factor based on the weight coefficients of each battery health evaluation index. The evaluation unit is used to perform cluster analysis based on the comprehensive score of the battery health and the data of each principal factor, and to classify and profile the health of the power battery according to the cluster analysis results, thereby evaluating the battery health.

[0016] Thirdly, the present invention proposes an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, which stores computer programs; The processor, when executing the program stored in the memory, implements the battery health evaluation method.

[0017] Fourthly, the present invention provides a computer-readable storage medium storing a computer program, which, when run, executes the battery health evaluation method.

[0018] The beneficial effects of this invention are: This invention proposes a method and system for evaluating the health of power batteries. Through factor analysis, it reduces the dimensionality and assigns weights to complex and intrinsically correlated evaluation indicators, providing a comprehensive score. By establishing a standardized decision matrix based on sample data of battery health evaluation indicators, it effectively eliminates the dimensional differences between different indicators, laying a unified and comparable foundation for subsequent analysis. Calculating the correlation matrix based on the standardized decision matrix and evaluating the correlation between indicators accurately captures the intrinsic relationships between each evaluation indicator, avoiding interference from redundant information in the evaluation results. By extracting common factors from the correlation matrix and selecting principal factors based on the cumulative variance contribution rate reaching a set value, it achieves data dimensionality reduction while preserving the core information of the original indicators to the maximum extent, simplifying the complexity of the evaluation method while ensuring the comprehensiveness and accuracy of the evaluation. Weighting each evaluation indicator based on the variance contribution rate and calculating the principal factor score using the rotated factor matrix, the comprehensive score is obtained through weighted summation. This ensures that the weight allocation and evaluation results are objectively derived from data characteristics, reducing the influence of subjective human factors and improving the scientific rigor and reliability of the evaluation. Finally, cluster analysis can help identify the differences between different power batteries, assist in further analysis and evaluation of the health of power batteries, and provide data support for the rational configuration and refined management of power batteries.

[0019] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures pointed out in the description, claims and drawings. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A flowchart of a power battery health evaluation method proposed in an embodiment of the present invention is shown; Figure 2 A schematic diagram of a power battery health measurement system proposed in an embodiment of the present invention is shown. Figure 3 A schematic diagram of an electronic device proposed in an embodiment of the present invention is shown. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] This invention proposes a method for evaluating the health of power batteries, such as... Figure 1 As shown, it includes the following steps: S1: Determine the battery health evaluation indicators related to the safety status of the power battery system; Specifically, in selecting evaluation indicators, as many relevant variables as possible should be collected for systematic evaluation. In some embodiments, battery health evaluation indicators include, but are not limited to, capacity retention rate, internal resistance growth rate, self-discharge rate, voltage consistency, temperature consistency, charge / discharge efficiency, accuracy of state of charge estimation, power state, thermal diffusion rate, thermal stability, structural integrity, sealing performance, lithium plating rate, and electrolyte decomposition rate.

[0024] S2: Establish a standardized decision matrix based on the original sample data of battery health evaluation indicators; In some embodiments, a standardized decision matrix is ​​established based on the original sample data of battery health evaluation indicators, including the following steps: Obtain the original sample data of the battery health evaluation index, and construct a decision matrix based on the original sample data of the battery health evaluation index;

[0025] In the formula, This represents the data for the m-th battery health evaluation indicator in the n-th sample; n is the total number of samples. m represents the total number of battery health evaluation indicators. .

[0026] The original sample data in the decision matrix are preprocessed, including standardization, and the formula for standardization is as follows:

[0027] In the formula, This refers to the standardized data of the j-th battery health evaluation index in the i-th sample. , These are the minimum and maximum raw data for the battery health evaluation index in the i-th sample, respectively.

[0028] A standardized decision matrix is ​​constructed based on the standardized data.

[0029]

[0030] In the formula, This refers to the data for the m-th battery health evaluation indicator in the n-th standardized sample, where n is the total number of samples. m represents the total number of battery health evaluation indicators. .

[0031] S3: Construct a correlation matrix for battery health evaluation indicators based on the standardized decision matrix, and evaluate the correlation between the various battery health evaluation indicators; including the following steps: A correlation matrix for the sample data of the battery health evaluation index was constructed using the Pearson correlation coefficient; the correlation matrix is ​​as follows:

[0032]

[0033] In the formula, A is the correlation matrix of battery health evaluation indicators; Let be the Pearson correlation coefficient of the m-th battery health evaluation indicator in the n-th sample, where n is the total number of samples. m represents the total number of battery health evaluation indicators. ; This refers to the data of the first battery health evaluation index in the standardized i-th sample. This refers to the data for the second battery health evaluation indicator in the standardized i-th sample.

[0034] The correlation between various battery health evaluation indicators is determined using the KMO value and Bartlett's test to assess suitability for factor analysis. Specifically, the KMO value and significance p-value of the correlation matrix are calculated. If both the KMO value and the significance p-value meet preset conditions, the battery health evaluation indicators satisfy the correlation requirement and are suitable for factor analysis. The Bartlett's test is used to examine the correlation between variables in the correlation matrix, i.e., to test whether each variable is independent. In factor analysis, if the null hypothesis is rejected, factor analysis is suitable; if the null hypothesis is not rejected, these variables may independently provide some information and are not suitable for factor analysis. Generally, a p-value of less than 0.05 is required in the Bartlett's test.

[0035] In some embodiments, the KMO statistic is taken between 0 and 1. The closer the KMO value is to 1, the stronger the correlation between variables, and the more suitable the original variables are for factor analysis. The closer the KMO value is to 0, the weaker the correlation, and the less suitable the original variables are for factor analysis. Generally, a KMO value greater than 0.6 is required.

[0036] The process of calculating the KMO value is as follows: Using elementary determinant transformations (B,E)→(E,B) -1 Solving for E yields the inverse matrix B of the correlation matrix A, where E is the identity matrix;

[0037] In the formula, B is the inverse of the correlation matrix A. Let be the Pearson correlation coefficient of the m-th battery health evaluation indicator in the n-th sample, where n is the total number of samples. m represents the total number of battery health evaluation indicators. ; It is the data in the nth row and mth column of the inverse matrix B.

[0038] Construct a diagonal matrix based on the inverse matrix;

[0039] In the formula, C is the diagonal matrix of the inverse matrix B. This represents the data in the nth row and mth column of the diagonal matrix C. , This refers to the data at the corresponding position in matrix B.

[0040] The reflection correlation matrix is ​​calculated based on the inverse matrix and the diagonal matrix; the calculation formula is as follows:

[0041] In the formula, D is the correlation matrix reflecting the image. This represents the net correlation between the i-th and j-th indicators after removing the influence of all other indicators, and is the basic data for calculating the KMO value; It is the data in the nth row and mth column of the inverse matrix B; This represents the data in the nth row and mth column of the diagonal matrix C; The elements in the correlation matrix D reflect the net correlation between the nth and mth indicators after removing the influence of all other indicators.

[0042] The KMO value of the correlation matrix is ​​calculated based on the inverse matrix and the reflection image correlation matrix, using the following formula:

[0043] In the formula, KMO is the KMO statistic. Let A be the Pearson correlation coefficient between the i-th and m-th battery health evaluation indicators in the correlation matrix A. Let A be the Pearson correlation coefficient between the nth and mth battery health evaluation indicators in the correlation matrix A, where m is the total number of battery health evaluation indicators. n is the total number of samples. ; To reflect the elements in the correlation matrix D, it reflects the net correlation between the i-th indicator and the m-th indicator after removing the influence of all other indicators.

[0044] In some embodiments, the specific steps for calculating the significance p-value using the Bartlett test are as follows: Calculate the determinant of the correlation matrix;

[0045] In the formula, det( A Let ) be the determinant of the correlation matrix A; The element in the i-th row and n-th column of the correlation matrix A is the correlation coefficient between the i-th indicator and the n-th indicator. To and Related coefficients or element combinations used to calculate determinants; The element in the i-th row and j-th column of the correlation matrix A is the correlation coefficient between the i-th indicator and the j-th indicator. For elements The cofactor, if all variables are completely uncorrelated (i.e., A is the identity matrix), then det( A If a correlation exists, det( ) = 1; A The result will be less than 1, where n is the total number of samples. m represents the total number of battery health evaluation indicators. .

[0046] Calculate the degrees of freedom of the correlation matrix A;

[0047] In the formula, df represents the degrees of freedom of the correlation matrix A, and m represents the total number of battery health evaluation indicators.

[0048] The chi-square value of the correlation matrix is ​​calculated based on the determinant of the correlation matrix and sample information, and the offset is quantified.

[0049] In the formula, X 2 Let be the chi-square value of the correlation matrix A, m be the total number of battery health evaluation indicators, and n be the total number of samples.

[0050] The p-value is calculated using the chi-square value and degrees of freedom of the correlation matrix through a chi-square distribution to determine whether the deviation is significant. The calculation formula is as follows: p=CHISQ.DIST.RT(X2 ,d f ) In the formula, p is the p-value of Bartlett's test; CHISQ.DIST.RT is the right-tail probability calculation function of the chi-square distribution, used to calculate the probability based on the chi-square value (X). 2 The p-value calculation function uses the degrees of freedom (df) to calculate the right-tail probability of the corresponding distribution; X 2 Let be the chi-square value of the correlation matrix A, and df be the degrees of freedom of the correlation matrix A.

[0051] S4: Extract common factors from the correlation matrix and calculate the core quantitative indicators of the common factors. Determine the principal factors in the correlation matrix based on these core quantitative indicators. Specifically, extract common factors (i.e., battery health evaluation indicators) from the correlation matrix based on the correlation analysis results, and calculate the core quantitative indicators (including eigenvalues, eigenvectors, and cumulative variance contribution rates) of each common factor. Determine the principal factors based on the common factors jointly determined by the eigenvalues, eigenvectors, cumulative variance contribution rates, and scree plot. This includes: If the correlation analysis results are suitable for factor analysis, then calculate the eigenvalues, eigenvectors, and cumulative variance contribution rate of the correlation matrix; In some embodiments, eigenvalues ​​are calculated using the matrix eigenvalue calculation formula, as follows:

[0052] In the formula, Let E be the eigenvalue of the correlation matrix A, E be the identity matrix, and A be the correlation matrix of the battery health evaluation index.

[0053] The eigenvector is calculated based on the eigenvalues ​​of the correlation matrix A, using the following formula:

[0054] In the formula, Eigenvalues The corresponding feature vector.

[0055] Extract common factors with eigenvalues ​​greater than 1 from the correlation matrix; their explanatory power exceeds the average level of individual original variables; Extract the common factor from the correlation matrix whose eigenvalues ​​are greater than the average of m eigenvalues, where m is the total number of battery health evaluation indicators; The top q common factors with a cumulative variance contribution rate greater than a set value are extracted as candidate principal factors. The formula for calculating the cumulative variance contribution rate is as follows:

[0056] In the formula, λ represents the eigenvalue of the correlation matrix, m represents the total number of battery health evaluation indicators, and q represents the number of candidate main factors.

[0057] The number of common factors can be determined using a scree plot. Specifically, a line graph can be drawn showing the factor indices and eigenvalues. The reasonable number of common factors is indicated when the eigenvalues ​​transition from a rapid decline to a gradual flattening (similar to the turning point of a "scratch pile"). Factors after the turning point contribute very little information and can be ignored.

[0058] S5: Calculate the scores of each principal factor based on the original sample data and the rotated factor matrix, and assign weights to each battery health evaluation index based on the variance contribution rate. In some embodiments, the following steps are specifically included: Based on the principal factors selected from the original sample data, an initial factor loading matrix is ​​constructed. This initial factor loading matrix is ​​then rotated using either orthogonal or oblique factor rotation to obtain the rotated factor loading matrix. For example, orthogonal factor rotation can be achieved by selecting an angle... To bring the coordinate axes closer to the clusters formed by the points, the rotated factor loading matrix can be obtained through image measurement or by calculating using the following formula:

[0059] In the formula, , This is the rotated factor loading matrix. Let T be the initial factor loading matrix, and T be the rotation matrix. In orthogonal factor rotation, The angle is the rotation angle.

[0060] The factor component score coefficients are obtained by rotating the component matrix of the common factors according to the maximum variance method; the purpose of the maximum variance method is to find the rotational load that maximizes the variance of the square of the loads in each column of the loading matrix. Based on the factor component score coefficients obtained after rotation, the factor scores of each indicator of power battery health are calculated.

[0061] S6: Based on the weighting coefficients of each battery health evaluation indicator, the scores of each main factor are weighted and summed to calculate the comprehensive score of battery health; the formula for calculating the comprehensive score is as follows:

[0062] In the formula, S is the overall score of battery health, and q is the number of determined main factors. The weight of the k-th principal factor. This is the score of the k-th principal factor.

[0063] S7: Based on the comprehensive score of the battery health and the data of each principal factor, perform cluster analysis, classify and profile the health of the power battery according to the cluster analysis results, and evaluate the battery health.

[0064] In some embodiments, K-means clustering is used for analysis. The same cluster has high similarity and different clusters have low similarity to classify the power batteries and create a classification profile. This allows for further analysis and evaluation of the power battery health, providing data support for the rational configuration and refined management of power batteries.

[0065] Based on the same inventive concept, this invention proposes a battery health evaluation system, such as... Figure 2 As shown, it includes: The decision matrix building unit is used to build a standardized decision matrix based on the original sample data of battery health evaluation indicators. The correlation assessment unit is used to construct a correlation matrix of battery health evaluation indicators based on the standardized decision matrix, and to assess the correlation between each battery health evaluation indicator. The principal factor determination unit is used to extract common factors from the correlation matrix, calculate the core quantitative indicators of the common factors, and determine the principal factors in the correlation matrix based on the core quantitative indicators. The weighting unit is used to calculate the scores of each principal factor based on the original sample data and the rotated factor matrix, and to assign weights to each battery health evaluation index based on the variance contribution rate. The quantification unit is used to calculate the comprehensive score of battery health by weighting and summing the scores of each main factor based on the weight coefficients of each battery health evaluation index. The evaluation unit is used to perform cluster analysis based on the comprehensive score of the battery health and the data of each principal factor, and to classify and profile the health of the power battery according to the cluster analysis results, thereby evaluating the battery health.

[0066] Another exemplary embodiment of the present invention provides an electronic device. For example... Figure 3 As shown, the electronic device includes at least one processor 301, at least one communication interface 302, at least one memory 303, and at least one communication bus 304; wherein the processor 301, communication interface 302, and memory 303 communicate with each other through the communication bus 304. Memory 303 stores computer programs; The processor 301 is used to execute the program stored in the memory 303 to implement the battery health evaluation method.

[0067] Another exemplary embodiment of the present invention provides a computer-readable storage medium storing a computer program, which, when executed, performs the battery health evaluation method.

[0068] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A battery health degree evaluation method characterized by comprising: The method comprises the following steps: a standardized decision matrix is established based on original sample data of battery health evaluation indexes; a correlation matrix of the battery health evaluation indexes is constructed based on the standardized decision matrix to evaluate the correlation between the battery health evaluation indexes; common factors in the correlation matrix are extracted, and a core quantitative index of the common factors is calculated to determine principal factors in the correlation matrix based on the core quantitative index; scores of the principal factors are calculated according to the original sample data and the rotated factor matrix, and the battery health evaluation indexes are weighted and distributed according to the variance contribution rate; a comprehensive score of the battery health is calculated by weighted summation of the scores of the principal factors based on the weight coefficients of the battery health evaluation indexes; cluster analysis is performed based on the comprehensive score of the battery health and the data of the principal factors, and the health of the power battery is classified and imaged according to the cluster analysis result to evaluate the battery health.

2. The battery health evaluation method according to claim 1, characterized by, The battery health evaluation indexes include capacity retention rate, internal resistance growth rate, self-discharge rate, voltage consistency, temperature consistency, charge-discharge efficiency, state of charge estimation accuracy, power state, thermal diffusion rate, thermal stability, structural integrity, sealing performance, lithium extraction rate and electrolyte decomposition rate.

3. The battery health evaluation method according to claim 1, characterized by, The establishment of the standardized decision matrix based on the original sample data of the battery health evaluation indexes comprises the following steps: original sample data of the battery health evaluation indexes are obtained, and a decision matrix is constructed based on the original sample data; the original sample data are preprocessed, and the preprocessing includes standardization processing; a standardized decision matrix is constructed based on the preprocessed original sample data and the decision matrix.

4. The battery health evaluation method according to claim 1, characterized by, The standardized decision matrix is as follows: In the formula, represents the data of the mth battery health evaluation index in the nth sample after standardization; n is the total number of samples, ; m is the total number of battery health evaluation indexes, .

5. The battery health evaluation method according to claim 1, characterized by, The construction of the correlation matrix of the battery health evaluation indexes based on the standardized decision matrix to evaluate the correlation between the battery health evaluation indexes comprises the following steps: a correlation matrix of the sample data of the battery health evaluation indexes is constructed using Pearson correlation coefficients; KMO test and Bartlett test are performed on the correlation matrix to test the correlation between the battery health evaluation indexes; if the results of the KMO value of the KMO test and the significance p value of the Bartlett test both meet the preset conditions, the battery health evaluation indexes meet the correlation requirement and are suitable for factor analysis.

6. The battery health evaluation method according to claim 5, characterized by, if the results of the KMO value of the KMO test and the significance p value of the Bartlett test cannot meet the preset conditions at the same time, the group of sample data is discarded.

7. The battery health evaluation method according to claim 1, characterized by, The calculation of the core quantitative index of the common factors and the determination of the principal factors in the correlation matrix based on the core quantitative index comprise the following steps: if the correlation analysis result is that the battery health evaluation indexes meet the correlation requirement, the core quantitative index of each common factor in the correlation matrix is calculated, and the core quantitative index includes the eigenvalue, eigenvector and cumulative variance contribution rate of each factor; common factors with eigenvalues greater than 1 in the correlation matrix are determined; common factors with eigenvalues greater than the average value of m eigenvalues in the correlation matrix are determined, m being the total number of the battery health evaluation indexes; common factors with cumulative variance contribution rates greater than a set value are extracted; The number of common factors is determined by a scree plot, and the number of principal factors is determined according to the number of extracted common factors. The principal factors are determined based on the common factors determined by the eigenvalues, eigenvectors, cumulative variance contribution rates and scree plot.

8. The battery health evaluation method according to claim 1, characterized by, The scores of the principal factors are calculated according to the original sample data and the rotated factor matrix, and the weight distribution of the battery health evaluation indexes is performed according to the variance contribution rates: Based on the determined principal factors, an initial factor loading matrix is constructed; The initial factor loading matrix is rotated by orthogonal factor rotation or oblique rotation to obtain a rotated factor loading matrix; The factor score coefficient matrix is calculated according to the rotated factor loading matrix; The factor scores of each sample on the corresponding principal factors are obtained by weighting and summing the normalized original sample data with the coefficients in the factor score coefficient matrix as weights. The weight of the corresponding battery health evaluation index is determined according to the percentage of the variance contribution rate of each principal factor in the cumulative variance contribution rate.

9. A battery state of health evaluation system characterized by comprising: It comprises: A decision matrix establishment unit for establishing a standardized decision matrix based on battery health evaluation index sample data; A correlation evaluation unit for constructing a correlation matrix of battery health evaluation indexes based on the standardized decision matrix and evaluating the correlation between each battery health evaluation index; A principal factor determination unit for extracting common factors in the correlation matrix and calculating core quantitative indexes of the common factors, and determining principal factors in the correlation matrix based on the core quantitative indexes; A weight distribution unit for calculating the scores of each principal factor according to the original sample data and the rotated factor matrix, and performing weight distribution of each battery health evaluation index according to the variance contribution rate; A quantification unit for weighting and summing the scores of each principal factor to calculate the comprehensive score of battery health based on the weight coefficients of each battery health evaluation index; An evaluation unit for performing cluster analysis based on the comprehensive score of battery health and the data of each principal factor, classifying and imaging the health of the power battery according to the cluster analysis result, and evaluating the battery health.

10. An electronic device, comprising: It comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete communication with each other through the communication bus; The memory stores a computer program; The processor is used to execute the program stored in the memory to realize the battery health evaluation method of any one of claims 1-8.

11. A computer readable storage medium storing a computer program, wherein the computer program comprises program instructions configured to cause a processor to perform the method according to any one of claims 1 to 10. The computer program is executed to perform the battery health evaluation method of any one of claims 1-8.

Images