A lithium battery SOH evaluation method and system based on the closest distance

Abstract

The present invention belongs to the technical field of electric vehicle power batteries, and specifically relates to a lithium battery SOH evaluation method and system based on the closest distance, the method comprising collecting a sample battery feature data set, determining the current total capacity and SOH value of the sample battery; performing principal component analysis on the feature data set to determine the derived features of each sample battery; classifying the sample batteries based on the K-means clustering algorithm and derived features, determining the mean of the derived features in each category after classification, and the average SOH value; obtaining the feature data set of the incremental battery, and determining the derived features of the incremental battery, calculating the Euclidean distance between the derived features of the incremental battery and the derived features of each category of sample batteries, and determining the SOH value of the incremental battery according to the Euclidean distance. The SOH estimation method proposed in the present invention only needs to collect the required historical data, and the SOH value can be calculated without the battery being left to stand, which greatly improves the real-time performance of the SOH estimation.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric vehicle power batteries, and in particular relates to a lithium battery SOH evaluation method and system based on the closest distance. Background Art

[0002] In the actual engineering application of SOH estimation of new energy vehicle power batteries, the current commonly used method is to use the static voltage of the vehicle before charging and the static voltage after charging, use the battery OCV-SOC data to calibrate the SOC, and then divide the charged capacity calculated by the ampere-hour integration method by the SOC change during the charging process to calculate the battery SOH.

[0003] However, the above method still has the following disadvantages when it is implemented: SOH cannot be calculated in real time, because when using the OCV-SOC data table to calibrate SOC, the battery needs to be left at rest for a sufficiently long time, such as one to two hours, and it is very difficult for the user to achieve the calibration conditions and obtain the voltage data before and after the battery is left at rest during actual use of the vehicle. Summary of the invention

[0004] The purpose of the present invention is to provide a lithium battery SOH evaluation method and system based on the closest distance to solve the problems raised in the background technology.

[0005] The present invention achieves the above-mentioned purpose through the following technical solutions:

[0006] In a first aspect, the present invention proposes a lithium battery SOH evaluation method based on the closest distance, the method comprising:

[0007] S1. Collect characteristic data sets including charge and discharge times in historical data of sample batteries, and determine the current total capacity and SOH value of the sample batteries in combination with the total capacity of the sample batteries at the factory;

[0008] S2, performing principal component analysis on the feature data set to obtain principal component coefficients, and determining the derived features of each sample battery according to the principal component coefficients;

[0009] S3, classifying the sample batteries based on the K-means clustering algorithm and the derived features, and determining the mean of the derived features in each category after classification, and the average SOH value of the sample batteries in each category;

[0010] S4. Obtain the feature data set of the incremental battery, determine the derived features of the incremental battery, calculate the Euclidean distance between the derived features of the incremental battery and the derived features of each category of sample batteries, and determine the SOH value of the incremental battery according to the Euclidean distance.

[0011] Furthermore, in step S1, the current total capacity and SOH value of the sample battery are determined based on the following formula:

[0012] ;

[0013] in, is the total capacity of the battery in its current state, is the total capacity of the battery when it leaves the factory. is the charge and discharge capacity of the battery cell over a period of time. For charging terminal , The charging start time .

[0014] Furthermore, the step S2 includes

[0015] S2.1, set the feature data set to the original data matrix :

[0016] ;

[0017] in, is the number of batteries, is the number of features, For the The first Features, ,The 10 features are low-temperature charging times, low-temperature discharging times, high-temperature charging times, high-temperature discharging times, high-current charging times, high-current discharging times, over-discharging times, over-charging times, low-current discharging times, and low-current charging times;

[0018] S2.2. Standardize the original data matrix so that the mean of each feature is 0 and the variance is 1:

[0019] ;

[0020] in: is the mean vector of each feature, is the standard deviation vector of each feature;

[0021] The standardized data matrix Z is:

[0022] ;

[0023] S2.3. Calculate the covariance matrix As follows:

[0024] ;

[0025] in, is the number of batteries;

[0026] S2.4. Calculate eigenvalues ​​and eigenvectors:

[0027] Covariance matrix Perform eigenvalue decomposition and obtain the eigenvalue and the eigenvector , the calculated eigenvalues ​​and corresponding eigenvectors are: Eigenvalues: ; Eigenvector: ;

[0028] ;

[0029] S2.5. Select the first two eigenvectors with the largest eigenvalues ​​as the principal component coefficients to form the principal component matrix V:

[0030] ;

[0031] S2.6. Multiply the standardized data matrix Z by the principal component matrix V to obtain the principal component score matrix Y:

[0032] ;

[0033] in: For the The first principal component score of the battery, For the The second principal component score of each battery;

[0034] S2.7. Use the first principal component score and the second principal component score as the derived features.

[0035] Further, step S3 is specifically as follows: clustering the sample batteries according to the derived features based on the K-means clustering algorithm, dividing the principal component score matrix Y into k categories, calculating the mean of the two principal component scores in each category, and the average SOH of all sample batteries in each category: let k=2, let the first two samples be included in the first category, then the mean of the two derived features is:

[0036] ;

[0037] The second category contains the remaining samples, and the means of the two derived features are:

[0038] ;

[0039] Calculating the average SOH of all batteries in each category includes:

[0040] The average SOH for the first category is:

[0041] ;

[0042] The average SOH for the second category is:

[0043] .

[0044] Furthermore, in step S4, the Euclidean distance between the incremental battery derived feature and the derived feature of each category sample battery is calculated, and the SOH value of the incremental battery is determined according to the Euclidean distance, including:

[0045] Calculate incremental battery Euclidean distance to the first category:

[0046] ;

[0047] Calculate incremental battery Euclidean distance to the second category:

[0048] ;

[0049] Select the category with the smallest distance: If , then the SOH of the incremental battery is , otherwise the SOH of the incremental battery is .

[0050] In a second aspect, the present invention proposes a lithium battery SOH evaluation system based on the closest distance, which is applied to execute any of the lithium battery SOH evaluation methods described in the above claims. The system includes

[0051] The acquisition module is used to collect characteristic data sets including charge and discharge times in the historical data of the sample battery, and determine the current total capacity and SOH value of the sample battery in combination with the total capacity of the sample battery at the factory;

[0052] A first analysis module, configured to perform principal component analysis on the feature data set to obtain principal component coefficients, and determine the derived features of each sample battery according to the principal component coefficients;

[0053] A second analysis module is used to classify the sample batteries based on the K-means clustering algorithm and the derived features, and determine the mean of the derived features in each category after classification, and the average SOH value of the sample batteries in each category;

[0054] The judgment module is used to obtain the feature data set of the incremental battery, determine the derived features of the incremental battery, calculate the Euclidean distance between the derived features of the incremental battery and the derived features of each category of sample batteries, and determine the SOH value of the incremental battery according to the Euclidean distance.

[0055] In a third aspect, the present invention proposes an electronic device, comprising: one or more processors; and a memory associated with the one or more processors, wherein the memory is used to store program instructions, and when the program instructions are read and executed by the one or more processors, the steps of any of the methods described above are executed.

[0056] In a fourth aspect, the present invention proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods described above.

[0057] The beneficial effects of the present invention are:

[0058] The SOH estimation method proposed in the present invention only needs to collect the required historical data, and the SOH value can be calculated without the battery being left stationary, which greatly improves the real-time performance of SOH estimation and enables users to obtain the battery health status at any time during vehicle use; mature algorithms such as principal component analysis and K-means clustering are used in the implementation algorithm, which is easy to implement and promote. SOH evaluation can be completed through a simple data collection and processing process without the need for complex equipment and operations.

[0059] The present invention can accurately reflect the health status of the battery by analyzing historical data and principal component characteristics. The K-means clustering method can effectively classify and identify the health status of the battery, ensuring the accuracy of the estimation result. Moreover, the method is applicable to various types of electric vehicle power batteries, and only needs to adjust the appropriate parameters for different batteries, which has a wide range of application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 Schematic diagram of a flow chart of a lithium battery SOH evaluation method based on the closest distance in an embodiment of the present invention.

[0061] Figure 2 It is another flow chart of a lithium battery SOH evaluation method based on the closest distance in an embodiment of the present invention.

[0062] Figure 3 4 is a system block diagram of a lithium battery SOH evaluation system based on the closest distance in an embodiment of the present invention. DETAILED DESCRIPTION

[0063] The present application is further described in detail below in conjunction with the accompanying drawings. It is necessary to point out here that the following specific implementation methods are only used to further illustrate the present application and cannot be understood as limiting the scope of protection of the present application. Technical personnel in this field can make some non-essential improvements and adjustments to the present application based on the above application content. Example

[0064] like Figure 1-2 As shown, this embodiment proposes a lithium battery SOH evaluation method based on the closest distance, and the method includes the following steps:

[0065] S1. Collect characteristic data sets including charge and discharge times in historical data of sample batteries, and determine the current total capacity and SOH value of the sample batteries in combination with the total capacity of the sample batteries at the factory;

[0066] S2, performing principal component analysis on the feature data set to obtain principal component coefficients, and determining the derived characteristics of each sample battery according to the principal component coefficients;

[0067] S3, classifying the sample batteries based on the K-means clustering algorithm and the derived features, determining the mean of the derived features in each category after classification, and the average SOH value of the sample batteries in each category;

[0068] S4. Obtain a feature data set of the incremental battery, determine the derived features of the incremental battery, calculate the Euclidean distance between the derived features of the incremental battery and the derived features of each category of sample batteries, and determine the SOH value of the incremental battery based on the Euclidean distance.

[0069] In the scenario of this embodiment, the characteristic data set of charge and discharge times in the sample battery history data includes low-temperature charge times, low-temperature discharge times, high-temperature charge times, high-temperature discharge times, high-current charge times, high-current discharge times, over-discharge times, over-charge times, low-current discharge times, and low-current charge times.

[0070] Further preferably, in step S1, before determining the current total capacity and SOH value of the sample battery, it also includes obtaining an OCV-SOC table, obtaining the static voltage before charging and the static voltage after charging, and using the battery OCV-SOC data to obtain the SOC before charging and the SOC after charging, so as to calculate the change in SOC. Then, the charged capacity calculated by the ampere-hour integration method is divided by the SOC change in the charging process, and the total capacity of the battery when it leaves the factory can be used to calculate the SOH of the battery. It is specifically calculated based on the following formula:

[0071] ;

[0072] in, is the total capacity of the battery in its current state, is the total capacity of the battery when it leaves the factory. is the charge and discharge capacity of the battery cell over a period of time. is the SOC at the charging end, is the SOC at the start of charging.

[0073] Further preferably, step S2 includes calculating the derived features of the two batteries: performing principal component analysis on the data in S1, extracting the main information in the features, obtaining the principal component coefficients, and calculating the first principal component score and the second principal component score of each battery, that is, each battery generates two derived features. The specific steps are as follows:

[0074] S2.1. Set the feature data set to the original data matrix for:

[0075] ;

[0076] in, is the number of batteries, is the number of features, For the The first Features, , the 10 features are low-temperature charging times, low-temperature discharging times, high-temperature charging times, high-temperature discharging times, high-current charging times, high-current discharging times, over-discharging times, over-charging times, low-current discharging times, and low-current charging times;

[0077] S2.2. Standardize the original data matrix so that the mean of each feature is 0 and the variance is 1:

[0078] ;

[0079] in: is the mean vector of each feature, is the standard deviation vector of each feature;

[0080] The standardized data matrix Z is:

[0081] ;

[0082] S2.3. Calculate the covariance matrix As follows:

[0083] ;

[0084] in, is the number of batteries;

[0085] S2.4. Calculate eigenvalues ​​and eigenvectors:

[0086] Covariance matrix Perform eigenvalue decomposition and obtain the eigenvalue and the eigenvector , the calculated eigenvalues ​​and corresponding eigenvectors are: Eigenvalues: ; Eigenvector: ;

[0087] ;

[0088] S2.5. Select the first two eigenvectors with the largest eigenvalues ​​as the principal component coefficients to form the principal component matrix V:

[0089] ;

[0090] S2.6. Multiply the standardized data matrix Z by the principal component matrix V to obtain the principal component score matrix Y:

[0091] ;

[0092] in: For the The first principal component score of the battery, For the The second principal component score of each battery;

[0093] S2.7. Use the first principal component score and the second principal component score as derived features.

[0094] Further preferably, step S3 clusters the batteries using the two derived features, and calculates the mean of the two derived features for each category and the average SOH of all batteries in each category.

[0095] The steps of cluster analysis are as follows:

[0096] Use the K-means clustering algorithm to cluster the batteries according to the principal component scores. The steps of the K-means algorithm are as follows:

[0097] (1) Select the number of clusters k: Select an appropriate number of clusters based on the characteristics of the data.

[0098] (2) Initialize cluster centers: Randomly select k data points as initial cluster centers.

[0099] (3) Assigning data points: Calculate the distance between each data point and all cluster centers, and assign the data point to the nearest cluster center.

[0100] (4) Update cluster centers: Recalculate the center of each cluster, that is, the mean of all data points in the cluster.

[0101] (5) Repeat steps 3 and 4 until the cluster center no longer changes or the preset number of iterations is reached.

[0102] The means of the two derived features for each category were calculated as follows: For each cluster, the means of the two derived features (i.e., principal component scores) were calculated.

[0103] Calculate the average SOH of all batteries in each category: For each category, calculate the average SOH of all batteries in that category.

[0104] In this embodiment, step S3 is specifically as follows: clustering the sample batteries according to the derived features based on the K-means clustering algorithm, dividing the principal component score matrix Y into k categories, calculating the mean of the two principal component scores in each category, and the average SOH of all sample batteries in each category: let k=2, let the first two samples be included in the first category, then the mean of the two derived features is:

[0105] ;

[0106] The second category contains the remaining samples, and the means of the two derived features are:

[0107] ;

[0108] Calculating the average SOH of all batteries in each category includes:

[0109] The average SOH for the first category is:

[0110] ;

[0111] The average SOH for the second category is:

[0112] .

[0113] Further preferably, in step S4, the Euclidean distance between the derived features of the incremental battery (battery without SOH value) (derived features obtained by the above step S2) and the derived features of each category of sample battery is calculated, and the SOH value of the incremental battery is determined according to the Euclidean distance, including:

[0114] Calculate incremental battery Euclidean distance to the first category:

[0115] ;

[0116] Calculate incremental battery Euclidean distance to the second category:

[0117] ;

[0118] Select the category with the smallest distance: If , then the SOH of the incremental battery is , otherwise the SOH of the incremental battery is .

[0119] According to an embodiment of the present invention, SOH estimation in the prior art requires the battery to be stationary. The evaluation method proposed in this embodiment only needs to collect the required historical data, which greatly improves the real-time performance of SOH evaluation and enables users to obtain the battery health status at any time during the use of the vehicle.

[0120] Combination Figure 3 As shown, based on the same inventive concept, this embodiment also proposes a lithium battery SOH evaluation system based on the closest distance, which is applied to execute the above lithium battery SOH evaluation method, and the system includes:

[0121] The acquisition module is used to collect characteristic data sets including charge and discharge times in the historical data of the sample battery, and determine the current total capacity and SOH value of the sample battery in combination with the total capacity of the sample battery at the factory;

[0122] The first analysis module is used to perform principal component analysis on the feature data set to obtain principal component coefficients, and determine the derived characteristics of each sample battery according to the principal component coefficients;

[0123] The second analysis module is used to classify the sample batteries based on the K-means clustering algorithm and the derived features, and determine the mean of the derived features in each category after classification, and the average SOH value of the sample batteries in each category;

[0124] The judgment module is used to obtain the feature data set of the incremental battery, determine the derivative features of the incremental battery, calculate the Euclidean distance between the derivative features of the incremental battery and the derivative features of each category of sample batteries, and determine the SOH value of the incremental battery based on the Euclidean distance.

[0125] This embodiment proposes an electronic device, including: one or more processors; and a memory associated with the one or more processors, the memory is used to store program instructions, and when the program instructions are read and executed by the one or more processors, the steps of the above method are executed.

[0126] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps of the above method are implemented.

[0127] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0128] In addition, each functional module in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0129] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in the embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0130] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features can be replaced by equivalents. However, these modifications or replacements 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 application.

Claims

1. A lithium battery SOH evaluation method based on the closest distance, characterized in that: The method comprises: S1. Collect characteristic data sets including charge and discharge times in historical data of sample batteries, and determine the current total capacity and SOH value of the sample batteries in combination with the total capacity of the sample batteries at the factory; S2, performing principal component analysis on the feature data set to obtain principal component coefficients, and determining the derived features of each sample battery according to the principal component coefficients; S3, classifying the sample batteries based on the K-means clustering algorithm and the derived features, and determining the mean of the derived features in each category after classification, and the average SOH value of the sample batteries in each category; S4, obtaining the feature data set of the incremental battery, determining the derived features of the incremental battery, calculating the Euclidean distance between the derived features of the incremental battery and the derived features of each category of sample batteries, and determining the SOH value of the incremental battery according to the Euclidean distance; In step S1, the current total capacity and SOH value of the sample battery are determined based on the following formula: ； in, is the total capacity of the battery in its current state, is the total capacity of the battery when it leaves the factory. is the charge and discharge capacity of the battery cell over a period of time. For charging terminal , The charging start time ; Step S2 includes: S2.1, set the feature data set to the original data matrix for: ; in, is the number of batteries, is the number of features, For the The first Features, ,The 10 features are low-temperature charging times, low-temperature discharging times, high-temperature charging times, high-temperature discharging times, high-current charging times, high-current discharging times, over-discharging times, over-charging times, low-current discharging times, and low-current charging times; S2.

2. Standardize the original data matrix so that the mean of each feature is 0 and the variance is 1: ; in: is the mean vector of each feature, is the standard deviation vector of each feature; The standardized data matrix Z is: ; S2.

3. Calculate the covariance matrix As follows: ; in, is the number of batteries; S2.

4. Calculate eigenvalues ​​and eigenvectors: Covariance matrix Perform eigenvalue decomposition and obtain the eigenvalue and the eigenvector , the calculated eigenvalues ​​and corresponding eigenvectors are: Eigenvalues: ; Eigenvector: ; ; S2.

5. Select the first two eigenvectors with the largest eigenvalues ​​as the principal component coefficients to form the principal component matrix V: ; S2.

6. Multiply the standardized data matrix Z by the principal component matrix V to obtain the principal component score matrix Y: ; in: For the The first principal component score of the battery, For the S2.7, using the first principal component score and the second principal component score as the derived features; Step S3 is specifically as follows: clustering the sample batteries according to the derived features based on the K-means clustering algorithm, dividing the principal component score matrix Y into k categories, calculating the mean of the two principal component scores in each category, and the average SOH of all sample batteries in each category: let k=2, let the first two samples be included in the first category, then the mean of the two derived features is: ; The second category contains the remaining samples, and the means of the two derived features are: ; Calculating the average SOH of all batteries in each category includes: The average SOH for the first category is: ; The average SOH for the second category is: ; In step S4, the Euclidean distance between the incremental battery derived feature and the derived feature of each category sample battery is calculated, and the SOH value of the incremental battery is determined according to the Euclidean distance, including: Calculate incremental battery Euclidean distance to the first category: ; Calculate incremental battery Euclidean distance to the second category: ; Select the category with the smallest distance: If , then the incremental battery for , otherwise the incremental battery for .

2. A lithium battery SOH evaluation system based on the nearest distance, characterized in that: Applied to executing the lithium battery SOH evaluation method according to claim 1, the system comprises: The acquisition module is used to collect characteristic data sets including charge and discharge times in the historical data of the sample battery, and determine the current total capacity and SOH value of the sample battery in combination with the total capacity of the sample battery at the factory; A first analysis module, configured to perform principal component analysis on the feature data set to obtain principal component coefficients, and determine the derived features of each sample battery according to the principal component coefficients; A second analysis module is used to classify the sample batteries based on the K-means clustering algorithm and the derived features, and determine the mean of the derived features in each category after classification, and the average SOH value of the sample batteries in each category; The judgment module is used to obtain the feature data set of the incremental battery, determine the derived features of the incremental battery, calculate the Euclidean distance between the derived features of the incremental battery and the derived features of each category of sample batteries, and determine the SOH value of the incremental battery according to the Euclidean distance.

3. An electronic device, characterized in that: include: one or more processors; and a memory associated with the one or more processors, the memory being used to store program instructions, wherein when the program instructions are read and executed by the one or more processors, the steps of the method according to claim 1 are executed.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in claim 1 are implemented.

Images