Battery consistency evaluation method, device and equipment and storage medium

Abstract

The invention provides a battery consistency evaluation method and device, equipment and a storage medium. The method comprises the steps of obtaining real-time operation data of a battery module and storing the real-time operation data in a time sequence database; extracting evaluation data from the time sequence database; for each key performance parameter, performing feature extraction on the evaluation data to obtain a curve similarity index, an intra-group variable coefficient index and an information entropy index; and determining the consistency score of the battery module according to the curve similarity index, the intra-group variation coefficient index and the information entropy index corresponding to the multiple key performance parameters. According to the method, multiple key performance parameters are comprehensively considered, and the consistency of the battery modules is evaluated from different angles by adopting the curve similarity index, the intra-group variation coefficient index and the information entropy index, so that the one-sidedness of single-index evaluation is avoided, the actual consistency condition of the battery modules can be reflected more comprehensively and accurately, and the evaluation accuracy of the battery modules is improved. And the accuracy and reliability of evaluation are improved.

Description

Technical Field

[0001] The present application relates to the technical field of battery detection, and in particular, to a method, device, equipment, and storage medium for evaluating battery consistency. Background Art

[0002] In a battery management system, it is crucial to comprehensively and accurately evaluate the characteristics of each battery in a battery pack. Each unit in the battery pack has its unique performance, and the differences between these units may significantly affect the overall performance, safety, and service life of the battery pack. The consistency score within the battery pack can be used to measure the degree of difference between individual battery units. By monitoring the consistency score, potential problems can be detected in a timely manner, and corresponding measures can be taken, such as adjusting the charge and discharge strategy, performing battery balancing, or replacing defective battery units when necessary. This not only helps to extend the service life of the battery pack but also improves the safety and reliability of the system.

[0003] Currently, when evaluating battery consistency, one way is to evaluate the battery state by selecting appropriate characteristic values, and another way is to predict the battery life based on parameters such as current, voltage, and battery capacity. Analyzing the multi-dimensional characteristics of battery data is an effective means of evaluating the battery consistency state, where the multi-dimensional characteristics refer to the key electrical parameters related to the battery pack consistency. Researchers usually calculate these multi-dimensional characteristics through mathematical means such as weighting to evaluate the battery state.

[0004] However, the characteristic statistical values used in current research are usually relatively rough statistics such as the maximum value, minimum value, difference, and average value, which cannot significantly highlight the changes in the battery pack state. At the same time, most research only focuses on the time-series changes of a single battery, and rarely involves the consistency problem of multiple batteries within the group at different levels, that is, whether their performances are consistent at the same moment. Summary of the Invention

[0005] Embodiments of the present application provide a method, device, equipment, and storage medium for evaluating battery consistency to solve the problem that the characteristic statistical values used in the prior art for evaluating battery consistency are too rough to effectively highlight the changes in the battery pack state.

[0006] In a first aspect, embodiments of the present application provide a method for evaluating battery consistency, the method comprising:

[0007] Obtain the real-time operation data of the battery module and save it to the time-series database; wherein, the real-time operation data is obtained by respectively monitoring multiple key performance parameters corresponding to each battery unit in the battery module;

[0008] When the battery module meets the triggering condition for battery consistency evaluation, extract the evaluation data for consistency evaluation of the battery module from the time series database;

[0009] For each of the multiple key performance parameters, extract features from the evaluation data to obtain a curve similarity index, an intra-group coefficient of variation index, and an information entropy index corresponding to the key performance parameter; wherein, the curve similarity index is used to characterize the degree of difference between the battery cells in the battery module, the intra-group coefficient of variation index is used to characterize the intra-group variation degree of the battery module, and the information entropy index is used to characterize the fluctuation degree of the battery cells in the battery module;

[0010] Determine the consistency score of the battery module according to the curve similarity index, the intra-group coefficient of variation index, and the information entropy index respectively corresponding to the multiple key performance parameters.

[0011] In a second aspect, an embodiment of the present application provides a battery consistency evaluation device, and the device includes:

[0012] A data acquisition module, configured to acquire real-time operation data of the battery module and save it to the time series database; wherein, the real-time operation data is obtained by respectively monitoring multiple key performance parameters corresponding to each battery cell in the battery module;

[0013] A data extraction module, configured to extract the evaluation data for consistency evaluation of the battery module from the time series database when the battery module meets the triggering condition for battery consistency evaluation;

[0014] A feature extraction module, configured to extract features from the evaluation data for each of the multiple key performance parameters to obtain a curve similarity index, an intra-group coefficient of variation index, and an information entropy index corresponding to the key performance parameter; wherein, the curve similarity index is used to characterize the degree of difference between the battery cells in the battery module, the intra-group coefficient of variation index is used to characterize the intra-group variation degree of the battery module, and the information entropy index is used to characterize the fluctuation degree of the battery cells in the battery module;

[0015] A consistency score determination module, configured to determine the consistency score of the battery module according to the curve similarity index, the intra-group coefficient of variation index, and the information entropy index respectively corresponding to the multiple key performance parameters.

[0016] In a third aspect, an embodiment of the present application further provides an electronic device, which includes a processor, a memory, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, the steps of the above battery consistency evaluation method are implemented.

[0017] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above battery consistency evaluation method are implemented.

[0018] The embodiments of the present application at least include the following technical effects:

[0019] The technical solution of the embodiment of the present application obtains real-time operation data of the battery module and timely conducts battery consistency evaluation when the triggering condition for consistency evaluation is met. During the evaluation process, multiple key performance parameters are comprehensively considered, and the curve similarity index, intra-group coefficient of variation index, and information entropy index are used to evaluate the consistency of the battery module from different angles, avoiding the one-sidedness of single-index evaluation, and being able to more comprehensively and accurately reflect the actual consistency of the battery module, improving the accuracy and reliability of the evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0021] Figure 1 is a schematic flowchart of the battery consistency evaluation method provided by the embodiment of the present application;

[0022] Figure 2 is a schematic structural diagram of the battery consistency evaluation device provided by the embodiment of the present application;

[0023] Figure 3 is a block diagram of the electronic device provided by the embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0025] The terms "first", "second", etc. in the description and claims of this application are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of this application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are usually of the same category, and do not limit the number of objects. For example, the first object can be one or more. In addition, the term "and / or" in the description and claims is used to describe the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. In the embodiments of this application, the term "multiple" refers to two or more, and other quantifiers are similar.

[0026] As Figure 1 shown, an embodiment of this application provides a method for evaluating battery consistency, and the method includes:

[0027] Step 101, obtain the real-time operation data of the battery module and save it to the time-series database; wherein, the real-time operation data is obtained by respectively monitoring multiple key performance parameters corresponding to each battery cell in the battery module.

[0028] In the embodiments of this application, multiple key performance parameters of each battery cell in the battery module are monitored in real time through sensors to obtain real-time operation data. Therefore, the real-time operation data includes the parameter values corresponding to multiple key performance parameters of each battery cell in the battery module.

[0029] Specifically, the multiple key performance parameters may include at least one of battery key performance parameters such as battery voltage, battery temperature, State of Charge (SOC), State of Health (SOH), State of Energy (SOE), and internal resistance.

[0030] The time-series database can efficiently process data arranged in chronological order. In the embodiments of this application, after obtaining the real-time operation data of the battery module, these data are saved to the time-series database, and database technology is used to store the data in an orderly manner for extraction and analysis at any time, providing a data basis for subsequent battery consistency scoring.

[0031] Step 102, when the battery module meets the trigger condition for battery consistency evaluation, extract the evaluation data for evaluating the consistency of the battery module from the time-series database.

[0032] Specifically, the triggering conditions for battery consistency evaluation can be preset. The triggering conditions can include reaching a preset period or receiving a target triggering event. Among them, the preset period can be that the operation time of the battery module reaches a certain duration, the number of charge and discharge cycles reaches a certain quantity, etc., and the target triggering event can include an operation of applying for battery consistency evaluation proposed by the user. Exemplarily, the triggering condition for battery consistency evaluation can be set to reach the preset period, or the triggering condition for battery consistency evaluation can be set to receive the target triggering event. It is also possible to set both reaching the preset period and receiving the target triggering event as the triggering conditions for battery consistency evaluation. At this time, when the preset period is not reached, the consistency evaluation of the battery can be triggered by the target triggering event.

[0033] In the embodiment of the present application, when the preset period is reached or the target triggering event is received, it is determined that the battery module meets the triggering conditions for battery consistency evaluation.

[0034] After determining that the triggering conditions for consistency evaluation are met, the evaluation data for performing consistency evaluation on the battery module is extracted from the time series database. By setting the triggering conditions for battery consistency evaluation in the present application, frequent battery evaluation operations can be avoided, and at the same time, it is ensured that the consistency evaluation operation of the battery module can be performed at key nodes.

[0035] Step 103: For each of the multiple key performance parameters, feature extraction is performed on the evaluation data to obtain a curve similarity index, an intra-group coefficient of variation index, and an information entropy index corresponding to the key performance parameter; wherein, the curve similarity index is used to characterize the degree of difference between the battery cells in the battery module, the intra-group coefficient of variation index is used to characterize the intra-group variation degree of the battery module, and the information entropy index is used to characterize the fluctuation degree of the battery cells in the battery module.

[0036] After obtaining the evaluation data, in order to obtain key indicators that can accurately evaluate the consistency of the battery module, it is necessary to further perform feature extraction on the evaluation data for multiple key performance parameters.

[0037] In the embodiment of the present application, the key indicators include a curve similarity index, an intra-group coefficient of variation index, and an information entropy index. Moreover, each key performance parameter corresponds to a set of key indicators.

[0038] Specifically, the curve similarity index can characterize the degree of difference in key performance parameters between each battery cell in the battery module. The smaller the curve similarity index, the smaller the degree of difference in key performance parameters between each battery cell. The larger the curve similarity index, the greater the degree of difference in key performance parameters between each battery cell. The intra-group variation coefficient index can characterize the intra-group variation degree of the battery module. The larger the intra-group variation coefficient, the greater the intra-group variation degree of the battery module, that is, the greater the degree of dispersion. The smaller the intra-group variation coefficient, the smaller the intra-group variation degree of the battery module, that is, the better the consistency. The information entropy index can characterize the degree of fluctuation of each battery cell in the battery module. The larger the information entropy index, the greater the degree of fluctuation of each battery cell in the battery module in key performance parameters. The smaller the information entropy index, the smaller the degree of fluctuation of each battery cell in the battery module in key performance parameters.

[0039] Step 104: Determine the consistency score of the battery module according to the curve similarity index, the intra-group variation coefficient index, and the information entropy index corresponding to the multiple key performance parameters.

[0040] Through step 103, the curve similarity index, the intra-group variation coefficient index and the information entropy index corresponding to the multiple key performance parameters can be obtained. By comprehensively analyzing the above indicators, the consistency score of the battery module can be obtained, so that the consistency of each battery cell in the battery module can be evaluated more comprehensively and accurately, avoiding the one-sidedness of a single indicator evaluation.

[0041] It should be noted that these multiple indicators not only evaluate the data consistency at a single time point, but also take into account the changing trend of the time series, incorporate the variation in both time and space dimensions, and provide a more comprehensive perspective, so that the consistency score can reflect the performance of the module at different event points and spatial locations, and can better capture possible systematic problems or uneven spatial distribution of the battery module, thereby more accurately evaluating the overall performance of the battery module.

[0042] In the embodiment of the present application, real-time operating data of the battery module is obtained, and the battery consistency evaluation is performed in a timely manner when the triggering conditions of the consistency evaluation are met. During the evaluation process, multiple key performance parameters are comprehensively considered, and the curve similarity index, the intra-group variation coefficient index and the information entropy index are used to evaluate the consistency of the battery module from different angles, thereby avoiding the one-sidedness of a single indicator evaluation, being able to more comprehensively and accurately reflect the actual consistency of the battery module, and improving the accuracy and reliability of the evaluation.

[0043] Further, considering that the 104 industrial protocol used when collecting the real-time operation data of the battery module may cause inconsistent data frequencies of each data point saved in the time series database. For example, for the data point monitoring the voltage in the battery module, it may be updated frequently when the charge and discharge state of the battery changes; while for the data point monitoring the temperature, due to the relatively slow temperature change, the update frequency is low. This inconsistency will bring difficulties to subsequent data processing and analysis. In an optional embodiment of the present application, extracting the evaluation data for evaluating the consistency of the battery module from the time series database includes:

[0044] Extracting the target real-time operation data of the battery module within a target time period from the time series database; wherein, the end time of the target time period is the time when the battery module meets the trigger condition for battery consistency evaluation, and the target time period corresponds to a preset duration;

[0045] Cleaning, filtering, and reorganizing the target real-time operation data according to a preset data frequency to obtain the evaluation data with the same data frequency.

[0046] Specifically, when the battery module meets the trigger condition for battery consistency evaluation, it is necessary to extract the evaluation data for evaluating the consistency of the battery module from the time series database. When extracting the evaluation data, it is necessary to first determine the target time period, that is, to determine a time interval, and the evaluation data is the data within this time interval. The end time of the target time period can be the time when the battery module meets the trigger condition for consistency evaluation, and the duration between the start time and the end time of the target time period is the preset duration. In this way, both the current state of the battery module can be concerned about, and the recent battery operation conditions can be combined to achieve a comprehensive analysis of the battery module. After determining the target time period, extract the data corresponding to the target time period from the time series database to obtain the target real-time operation data.

[0047] It should be noted that the setting of the preset duration needs to comprehensively consider the application scenario and data characteristics of the battery module. In an electric vehicle, since the battery module is used frequently and the working conditions are complex, the preset duration may be set to 1 to 2 days to obtain enough data under different working conditions. In some relatively stable energy storage application scenarios, the preset duration can be set to about one week. Specifically, the preset duration can be set by configuring parameters. After determining the end time, calculate the start time of the target time period according to the preset duration.

[0048] Considering that the real-time operation data of the battery module may be interfered by various factors during the acquisition and transmission processes, resulting in problems such as data noise, outliers, or missing values. To avoid these interferences from affecting the accuracy of the consistency assessment, it is necessary to clean and filter the extracted target real-time operation data, remove these interfering data, and retain the real and effective data. Exemplarily, a filtering algorithm can be used for the data cleaning and filtering operation.

[0049] In addition, the data collected by different sensors and the data at different times may have inconsistent data frequencies. To facilitate subsequent feature extraction and consistency assessment, it is necessary to reorganize the data after cleaning and filtering according to a preset data frequency. Through data reorganization, all evaluation data have the same data frequency, providing a unified basis for subsequent analysis.

[0050] Specifically, when performing data reorganization, first determine the preset data frequency, for example, collect data every 5s. Then, according to the preset data frequency, perform data reorganization. For data with a collection frequency higher than the preset frequency, perform downsampling processing. Based on the preset data frequency, determine the time points. When there is data at a time point, directly use this data as the data at this time point after reorganization. When there is no data at a time point, you can find the data at the previous collection moment forward or backward and assign this data to this time point. For example, for a voltage time point, there is data of 3.5V at the 10th second, no data at the 5th second, and by looking forward, it is found that the data at the 12th second is 3.4V. Then, assign 3.4V to the 15th second as the data after reorganization.

[0051] After data reorganization, determine the reorganized data as the evaluation data.

[0052] In the above implementation scheme of the present application, through data reorganization, the evaluation data extracted from the time series database have the same data frequency, providing a reliable data basis for subsequent feature extraction and consistency assessment, and thus improving the accuracy of battery consistency assessment.

[0053] Specifically, in the energy storage system, there is a clear hierarchical structure, from the bottom-level battery modules, to the intermediate-level energy storage units, then to the upper-level energy storage system, and finally to the top-level energy storage power station. Each sub-level is the basic unit that constitutes the parent level, and they are interrelated and affect the overall performance of the parent level. The consistency of the battery modules is directly related to the performance of the energy storage units, and the comprehensive performance of multiple energy storage units determines the stability of the energy storage system, ultimately affecting the operation efficiency and reliability of the energy storage power station. To further evaluate the battery consistency of a higher-level structure, in an optional embodiment of the present application, after determining the consistency score of the battery module, the method further includes:

[0054] Obtain the consistency score and hierarchical weight coefficient corresponding to each sub - level under the parent level;

[0055] Determine the consistency score of the parent level through weighted operation according to the consistency score and hierarchical weight coefficient corresponding to each sub - level under the parent level;

[0056] Among them, the energy storage unit is the parent level of the battery module, the energy storage system is the parent level of the energy storage unit, and the energy storage power station is the parent level of the energy storage system.

[0057] Specifically, first, it is necessary to obtain the consistency score corresponding to each sub - level under the parent level. In addition, based on the different contributions of each sub - level to the consistency of the parent level, it is also necessary to obtain the hierarchical weight coefficient corresponding to each sub - level respectively. The hierarchical weight coefficient reflects the relative importance of the sub - level in the parent level. For example, in an energy storage unit, different battery modules may have different degrees of influence on the consistency of the energy storage unit due to factors such as capacity and usage frequency, so different weights will be assigned. Through weighted operation, the consistency scores of the sub - levels are comprehensively calculated according to their weights, and a score reflecting the overall consistency of the parent level can be obtained. The consistency score of the parent level obtained in this way can not only evaluate the overall consistency level, but also identify the key sub - levels that have a greater impact on the overall performance.

[0058] Through the above - mentioned method for determining the consistency score of the parent level, based on the hierarchical relationship, the consistency scores of each level in the energy storage system can be determined.

[0059] Exemplarily, the weighted average algorithm is as follows:

[0060]

[0061] Among them, avg is the weighted average value, that is, the consistency score of the parent level, x i is the consistency score of the i - th sub - level, w i / ∑w i is the hierarchical weight coefficient of the i - th sub - level. In this example, w i is the reciprocal of x i Therefore, the lower the consistency score of the sub - level, the greater the hierarchical weight coefficient corresponding to this sub - level. That is to say, the sub - levels with low consistency scores will account for a larger proportion when calculating the consistency score of the parent level and have a greater impact on the parent - level score, so as to highlight the impact of the sub - levels with poor performance on the whole, and thus more accurately reflect the consistency of the parent level.

[0062] In addition, through step-by-step analysis, the battery module with abnormal performance can be quickly located, and repair measures can be taken in a timely manner, such as charge equalization operation. In this way, not only the accuracy of fault detection is improved, but also the reliability and maintenance management ability of the system are enhanced, making the monitoring of the battery pack more efficient and intelligent.

[0063] In the above implementation scheme of this application, by comprehensively evaluating the consistency of the battery module based on multiple key performance parameters and creating a hierarchical consistency scoring system based on the consistency scores of the battery modules, multi-level consistency evaluation from the module level to the energy storage power station level is achieved. It can comprehensively evaluate the consistency of the energy storage system from multiple levels, making the evaluation of the overall consistency of the energy storage system more accurate and comprehensive. It can also quickly locate the specific levels and sub-levels with consistency problems in the energy storage system, providing strong support for fault diagnosis and early warning. When it is found that the consistency score of the energy storage power station decreases, it is possible to quickly trace back to the abnormal energy storage system, energy storage unit and battery module, take maintenance measures in advance, avoid the expansion of faults, and improve the reliability and stability of the energy storage system.

[0064] Next, how to extract features from the evaluation data to obtain the curve similarity index, within-group coefficient of variation index, and information entropy index will be introduced respectively.

[0065] In an optional embodiment of this application, feature extraction is performed on the evaluation data to obtain the curve similarity index corresponding to the key performance parameter, including:

[0066] Extract the parameter values of the key performance parameter corresponding to each battery cell in the battery module from the evaluation data to obtain a first parameter value sequence corresponding to each battery cell;

[0067] According to the first parameter value sequence corresponding to each battery cell in the battery module, a reference sequence is determined, where the parameter value at the Nth position in the reference sequence is the mean value of the parameter values at the Nth position in the first parameter value sequences of each battery cell in the battery module, and N is a positive integer;

[0068] For each battery cell, calculate the similarity between the first parameter value sequence corresponding to the battery cell and the reference sequence;

[0069] Determine the battery cell with the lowest similarity as the target battery cell;

[0070] Perform numerical conversion on the similarity corresponding to the target battery cell to obtain the curve similarity index.

[0071] When determining the curve similarity index, first, parameter values of key performance parameters corresponding to each battery cell in the battery module are extracted from the evaluation data to obtain a first parameter value sequence corresponding to each battery cell. Then, taking the mean value of the parameter values at the corresponding positions in the first parameter value sequences of each battery cell as the reference sequence, a reference standard representing the overall average level of the battery module is established to measure the difference between each battery cell and the overall average level. Next, the similarity between the first parameter value sequence of each battery cell and the reference sequence is calculated to obtain the degree of closeness of each battery cell to the overall average performance. Among them, the higher the similarity, the smaller the performance difference between the battery cell and other battery cells; conversely, the greater the difference. After obtaining the similarity of each battery cell, the battery cell with the lowest similarity is determined as the target battery cell, and this target battery cell has the largest difference from the overall average performance and can best reflect the inconsistency between battery cells. Finally, a numerical conversion is performed on the similarity of the target battery cell to obtain the curve similarity index, so as to quantify the similarity result into a specific index that can be used to evaluate the consistency of the battery module.

[0072] In the embodiment of the present application, during the operation of the battery module, the evaluation data includes measurement values of key performance parameters (such as voltage, current, temperature, etc.) of each battery cell at different time points. From these data, the key performance parameter values of each battery cell are extracted in chronological order to form a first parameter value sequence.

[0073] In the embodiment of the present application, according to the first parameter value sequence corresponding to each battery cell in the battery module, a reference sequence is determined, where the parameter value at the Nth position in the reference sequence is the mean value of the parameter values at the Nth position in the first parameter value sequences of each battery cell in the battery module. Exemplarily, the battery module includes 3 battery cells. For the key performance parameter of voltage, the first parameter value sequence of each battery cell is extracted from the evaluation data. Among them, the first parameter value sequence of battery cell 1 is [3.5, 3.6, 3.4, 3.5, 3.6], the first parameter value sequence of battery cell 2 is [3.4, 3.5, 3.3, 3.4, 3.5], and the first parameter value sequence of battery cell 3 is [3.8, 3.7, 3.6, 3.7, 3.8]. The mean value of the parameter values at the Nth position in the first parameter value sequences of each battery cell is calculated respectively to obtain the reference sequence, and the reference sequence is [3.57, 3.6, 3.43, 3.53, 3.63]. By creating the reference sequence, each first parameter value sequence can be compared with the reference sequence. Compared with pairwise comparison of each first parameter value sequence, the number of comparisons can be reduced and the calculation consumption can be reduced.

[0074] When calculating the similarity between the first parameter value sequence of each battery cell and the reference sequence, the cosine similarity between the first parameter value sequence and the reference sequence can be calculated. Cosine Similarity is a measure used to quantify the similarity between two non-zero vectors. The calculation formula is as follows:

[0075] Where, is the similarity between the first parameter value sequence and the reference sequence, A i is the first parameter value sequence of the i-th battery cell, is the reference sequence.

[0076] It should be noted that the value range of cosine similarity is between [-1, 1]. The closer the value is to 1, the more similar the two vectors are; the closer the value is to -1, the less similar they are; and a value of 0 indicates that the two vectors are orthogonal (independent of each other).

[0077] In the embodiments of the present application, the battery cell with the smallest similarity is used as the target battery cell because it usually represents the individual with the largest difference from the average behavior. This feature is highly sensitive to inconsistent battery cells and can ensure the sensitivity of the system to these inconsistencies. In a battery management system, inconsistent battery cells may lead to the instability of the entire system or pose safety hazards. By identifying and paying attention to these least similar battery cells, the problem can be quickly located to a specific battery module, and corresponding countermeasures can be taken in a timely manner.

[0078] Since the value of cosine similarity is between -1 and 1, for the needs of subsequent analysis and threshold setting, the similarity corresponding to the target battery cell is numerically transformed to obtain the curve similarity index. The specific formula is as follows:

[0079] Score = 2 - (1 + similarity)

[0080] Where, Score is the curve similarity index, and similarity is the similarity of the target battery cell. (1 + similarity) is between 0 and 2, and the higher the similarity, the larger the value. By subtracting (1 + similarity) from the maximum value 2, it is inversely proportional to the consistency score. This processing method helps to more clearly define the level of consistency score. Through numerical transformation, the closer the curve similarity index is to 0, the more similar it is; on the contrary, the larger the curve similarity index, the greater the difference.

[0081] In the above implementation scheme of the present application, by comparing the first parameter value sequence of each battery cell with the reference sequence, the battery cell with the largest performance difference is found, and the curve similarity index is determined based on this battery cell, providing a strong basis for evaluating the consistency of the battery module.

[0082] In an alternative embodiment of the present application, feature extraction is performed on the evaluation data to obtain the within-group coefficient of variation index corresponding to the key performance parameter, including:

[0083] Extract each parameter value of the key performance parameter corresponding to each time point from the evaluation data to obtain a second parameter value sequence corresponding to each time point;

[0084] For each time point, calculate the mean of each parameter value in the second parameter value sequence corresponding to the time point and the extreme value standard deviation corresponding to the second parameter value sequence; wherein, the extreme value standard deviation is obtained by taking the square root of the ratio of the square of the difference between the maximum parameter value and the minimum parameter value in the second parameter value sequence to the number of battery cells in the battery module;

[0085] Calculate the ratio of the extreme value standard deviation to the mean to obtain the coefficient of variation corresponding to the time point;

[0086] Determine the mean of the coefficients of variation corresponding to each time point in the evaluation data as the within-group coefficient of variation index.

[0087] Specifically, the evaluation data records the change of each key performance parameter of each battery cell in the battery module over time. When determining the within-group coefficient of variation index, first extract each parameter value of the key performance parameter corresponding to each time point from the review data, that is, the key performance parameter values of all battery cells at this time point, to form a second parameter value sequence. Exemplarily, the battery module includes N battery cells. At the Mth time point, the second parameter value sequence can be expressed as group={X M1 , X M2 , X M3 ,..., X MN}. Among them, each time point corresponds to a second parameter value sequence, and the number of elements included in the second parameter value sequence is equal to the number of battery cells included in the battery module. The second parameter value sequence reflects the distribution of the according performance parameters of each battery cell in the battery module at the same moment.

[0088] Then, calculate the mean and extreme value standard deviation of the second parameter value sequence corresponding to each time point. Among them, the average level of the key performance parameter of the battery module at this moment can be obtained by calculating the mean. The extreme value standard deviation takes into account the maximum parameter value and the minimum parameter value in the second parameter value sequence, can highlight the degree of data dispersion, and reflect the amplitude of the performance difference between battery cells.

[0089] The mean calculation formula is as follows:

[0090]

[0091] Among them, is the mean value, X i is the i-th parameter value in the second parameter value sequence, and N is the number of battery cells in the battery module.

[0092] The calculation formula of the extreme value standard deviation is as follows:

[0093]

[0094] Among them, Y is the extreme value standard deviation, X max is the maximum parameter value in the second parameter value sequence, X min is the difference between the minimum parameter value in the second parameter value sequence, and N is the number of battery cells in the battery module.

[0095] It should be noted that the traditional standard deviation calculation method is based on the deviation of all data points relative to the average value, considering the difference between each data point and the average value, and comprehensively calculating the standard deviation. When facing extreme values or mutation points in the data, this method may be masked by the overall smoothing effect, especially when these mutation points are not part of the mainstream trend. Since the maximum and minimum values usually represent extreme situations or outliers in the data, in the embodiments of the present application, the extreme value standard deviation adopted uses the maximum and minimum values as the measurement criteria, is more sensitive to single or a small number of mutation points, so as to more significantly highlight the variability within the group, and thus can avoid the possible defects of traditional statistical methods. Especially in some cases, the traditional method may mask the true existence of mutation points. By directly focusing on the extreme values in the data, the present application can more clearly identify abnormal behaviors and extreme changes within the group.

[0096] After obtaining the mean value and the extreme value standard deviation Y, it is necessary to perform normalization processing, calculate the ratio of the extreme value standard deviation to the mean value, obtain the coefficient of variation corresponding to the time point, and this coefficient of variation is a dimensionless eigenvalue, so that the variation degrees between different battery modules can be directly compared, and it also helps to ensure the comparability of the eigenvalues between different battery modules or time periods, and more accurately reflects the consistency and variation of each battery module.

[0097] The coefficient of variation in the embodiments of the present application can eliminate the influence of the mean value size on the measurement of the dispersion degree, and make the variation degrees between different time points or different key performance parameters comparable. Among them, each time point corresponds to a coefficient of variation.

[0098] The calculation formula of the coefficient of variation is as follows:

[0099]

[0100] Among them, CV is the coefficient of variation.

[0101] Finally, considering that in practical applications, some time steps may have accidental high coefficients of variation due to specific operating conditions or environmental changes, after obtaining the coefficients of variation corresponding to each time point, it is necessary to average the coefficients of variation to smooth the influence of these accidental factors, so as to more accurately reflect the overall consistency level of the battery pack. That is, taking the values of the coefficients of variation at each time point as the within-group coefficient of variation index, comprehensively considering the variation of the battery module during the entire evaluation time period, a quantitative index that can overall reflect the degree of within-group variation is obtained, thus better representing the actual performance of the battery module rather than being dominated by individual abnormal data. The calculation formula of the within-group coefficient of variation index is as follows:

[0102]

[0103] Among them, is the within-group coefficient of variation index, CV i is the coefficient of variation at the i-th time point, and M is the number of time points.

[0104] In the above implementation scheme of the present application, by quantifying the within-group variation degree of the battery module during the corresponding time period of the evaluation data, a within-group coefficient of variation index that can characterize the difference in key performance parameters between battery cells is obtained, providing a strong basis for evaluating the consistency of the battery module.

[0105] In an optional embodiment of the present application, feature extraction is performed on the evaluation data to obtain an information entropy index corresponding to the key performance parameters, including:

[0106] Extract the parameter values of the key performance parameters corresponding to each battery cell in the battery module from the evaluation data to obtain a first parameter value sequence corresponding to each battery cell;

[0107] For each battery cell, according to the first parameter value sequence corresponding to the battery cell, determine a difference sequence, where the value at the M-th position in the difference sequence is the difference between the parameter value at the M-th position and the parameter value at the M-1-th position in the first parameter value sequence;

[0108] Determine multiple numerical intervals according to the difference sequence;

[0109] Determine the probability corresponding to each numerical interval according to the number of differences in the difference sequence and the number of differences included in each of the multiple data intervals;

[0110] Calculate the information entropy corresponding to the battery cell according to the probabilities corresponding to the multiple numerical intervals;

[0111] Determine the mean of the information entropy corresponding to each battery cell in the battery module as the information entropy index.

[0112] Specifically, in the process of data analysis, extracting the features of time series is crucial for understanding the behavior and changes of data. The information entropy algorithm mainly realizes the extraction of time features from battery data and focuses on the data change rate. When determining the information entropy index, first extract the key performance parameter values of each battery cell from the evaluation data to form a first parameter value sequence, and obtain the change trajectory of the key performance parameters of each battery cell over time, laying a foundation for subsequent analysis of its fluctuation characteristics.

[0113] Considering that the information entropy of the sequence difference can more accurately reflect the data change rate and fluctuation, while the information entropy of the overall curve may be masked by the long-term trend of the data, making it difficult to detect local abnormal fluctuations. In the embodiments of the present application, choosing to focus on the information difference of the sequence difference rather than the information entropy of the overall curve can reduce the influence of the long-term trend, make the entropy value more concentrated in reflecting short-term fluctuations and noises, help highlight the true change characteristics in the data, and thus improve the sensitivity and accuracy of the consistency evaluation.

[0114] Specifically, based on the fact that the change amount between adjacent parameter values can directly reflect the fluctuation of the parameter, and the size and direction of the difference reflect the rate and trend of the change of the key performance parameter, calculate the difference between adjacent parameter values of the first parameter value sequence to obtain a difference sequence, that is, for each battery cell, determine the difference sequence according to the first parameter value sequence corresponding to the battery cell. The value at the Mth position in the difference sequence is the difference between the parameter value at the Mth position and the parameter value at the M - 1th position in the first parameter value sequence.

[0115] After obtaining the difference sequence, in order to discretize the difference for facilitating the statistics of the occurrence frequencies of different fluctuation degrees, it is necessary to determine multiple numerical intervals. Then calculate the probability based on the total number of differences included in the difference sequence and the number of differences in each interval, and obtain the probability corresponding to each numerical interval. This probability represents the possibility that the difference falls into different intervals and provides necessary parameters for information entropy calculation. The probability calculation formula for the ith numerical interval is as follows:

[0116]

[0117] where, p i is the probability corresponding to the ith numerical interval, h i is the number of differences in the ith numerical interval, and n is the total number of differences included in the difference sequence.

[0118] Finally, the information entropy is calculated using the probabilities of each interval. Based on the principle of measuring uncertainty in information theory, the information entropy reflects the degree of uncertainty in the fluctuations of the key performance parameters of the battery cell. The more irregular the fluctuations, the greater the information entropy. The calculation formula for the information entropy is as follows:

[0119] En = -∑p i ·log2(p i )

[0120] where En is the information entropy, and p i is the probability corresponding to the i-th numerical interval. This information entropy is a non-negative number. The greater the fluctuation of the difference sequence, the greater the information entropy.

[0121] The information entropy is a characteristic of the change of a single battery cell over a period of time. In a battery module, there are usually multiple battery cells. Therefore, the information entropy of a single battery cell cannot comprehensively represent the performance of the entire battery module. To more accurately reflect the overall consistency of the battery module, calculate the information entropy of all battery cells during this time period and select their average value as the information entropy index. This comprehensively considers the fluctuations of each battery cell in the battery module, helps to balance the influence of each battery cell, and provides a more comprehensive evaluation of the consistency of the battery module. By using the average information entropy as a characteristic quantity, potential problems within the battery pack can be more effectively identified and analyzed, thereby improving the reliability and performance of the overall system.

[0122] In the above implementation of this application, by accurately quantifying the degree of fluctuation of the key performance parameters of each battery cell in the battery module and presenting it intuitively in numerical form, it provides a strong basis for evaluating the consistency of the battery module.

[0123] In an optional embodiment of this application, according to the curve similarity index, the within-group coefficient of variation index, and the information entropy index corresponding to the multiple key performance parameters, determine the consistency score of the battery module, including:

[0124] Determine the upper limit of the score of the battery module according to the curve similarity index and the within-group coefficient of variation index corresponding to the multiple key performance parameters;

[0125] Determine the deduction value of the battery module through weighted operation according to the information entropy index and the entropy characteristic weight coefficient corresponding to the multiple key performance parameters;

[0126] Determine the consistency score of the battery module according to the upper limit of the score and the deduction value.

[0127] Specifically, in the embodiments of the present application, the curve similarity index reflects the degree of similarity between the key performance parameter curves of each battery cell and the overall reference sequence. The higher the similarity, the closer the performance curves of the battery cells are, and the better the consistency of the battery module. The within-group coefficient of variation index measures the degree of dispersion of the parameters between battery cells at the same time point by calculating the ratio of the standard deviation of the extreme values of the key performance parameters to the mean at each time point. The smaller the coefficient of variation, the smaller the performance difference between the battery cells and the higher the consistency. Selecting the curve similarity index and the within-group coefficient of variation index to delimit the information score, that is, determining the upper limit of the score of the battery module, utilizes the positive or negative influence relationship of these two indexes on the consistency of the battery module, converts it into an initial upper limit of the score, and provides a basic range for the subsequent scoring. This upper limit of the score reflects the influence of the overall stability and uniformity of the performance of the battery module on the score, in order to highlight these key features for mutation names and ensure the comprehensiveness and reliability of the scoring.

[0128] Then, considering that the information entropy index measures the data fluctuation and uncertainty of the battery module, introducing the information entropy index as a penalty term in the scoring can effectively identify and handle the noise and abnormal fluctuations in the data. The larger the information entropy index, the more complex and unstable the parameter changes, and the worse the consistency of the battery module may be. When the change trend of the time series is clear and stable, the value of the information entropy index is close to 0, while when there are large fluctuations or noises in the sequence, the value of the information entropy index will increase. Exemplarily, taking voltage as an example, during the process of calibrating the consistency, it is expected that the voltage will rise during charging, fall during discharging, and remain unchanged during energy storage over time. These behaviors reflect the expected performance of the battery in different operating stages. However, in actual situations, the voltage may be affected by noise, environmental changes, or faults, resulting in fluctuations. In the consistency scoring, the information entropy index as a penalty term can handle and control these fluctuations, balancing the difference between the actual fluctuations of the voltage and the ideal behavior. Through the penalty based on the information entropy index, the battery cells showing large inconsistencies or abnormal fluctuations can be effectively scored, thereby prompting the system to identify and solve potential problems.

[0129] In the embodiments of the present application, the deduction score is determined through the information entropy index, that is, a penalty is imposed on the upper limit of the score. Specifically, the information entropy index corresponding to each of multiple key performance parameters is weighted with a preset entropy feature weight coefficient to obtain the deduction score of the battery module. Among them, the entropy feature weight coefficient can be set according to the actual data distribution. Exemplarily, for the key performance parameter of temperature, it usually remains relatively stable, while for the key performance parameter of voltage, obvious fluctuations will occur. Therefore, when setting the entropy feature weight coefficient, it is necessary to increase the entropy feature weight coefficient corresponding to temperature and correspondingly decrease the entropy feature weight coefficient corresponding to voltage. Thus, the stability of temperature and the fluctuation of voltage can be more accurately reflected, thereby improving the accuracy of the overall score and the sensitivity of the system to abnormal situations.

[0130] Finally, the initial upper limit of the score is adjusted according to the magnitude of the information entropy, that is, the consistency score is obtained by subtracting the deduction score from the upper limit of the score, which comprehensively considers various factors and more comprehensively evaluates the consistency of the battery module. Exemplarily, the calculation formula for the consistency score is as follows:

[0131] consistency score =S - α i En i

[0132] Wherein, consistency score is the consistency score, S is the upper limit of the score, En i is the information entropy index of the i-th key performance parameter, and α i is the entropy feature weight coefficient of the i-th key performance parameter.

[0133] In the above implementation scheme of the present application, different characteristic indexes of multiple key performance parameters are comprehensively considered when determining the consistency score, avoiding the limitations of single-index evaluation, and being able to comprehensively reflect the consistency status of the battery module from multiple perspectives, making the evaluation result more accurate and reliable.

[0134] In an optional embodiment of the present application, determining the upper limit of the score of the battery module according to the curve similarity index and the within-group coefficient of variation index corresponding to each of the multiple key performance parameters includes:

[0135] Obtaining the preset threshold values of the curve similarity index and the within-group coefficient of variation index corresponding to each of the multiple key performance parameters;

[0136] Obtaining the first quantity of the curve similarity index greater than the curve similarity index threshold value among the multiple key performance parameters;

[0137] Obtaining the second quantity of the within-group coefficient of variation index greater than the within-group coefficient of variation index threshold value among the multiple key performance parameters;

[0138] Deduct the first quantity of preset scores and the second quantity of preset scores from the total consistency score of the battery module to obtain the upper limit of the score of the battery module.

[0139] Specifically, based on the design requirements of the battery module, industry standards, and a large amount of experimental data, etc., the threshold of the curve similarity index and the threshold of the within-group coefficient of variation index for each key performance parameter are preset as the benchmark for judging whether the consistency of the key performance parameter is good. Among them, the threshold of the curve similarity index and the threshold of the within-group coefficient of variation index represent the standards that each index should reach under the ideal or acceptable consistency level. By comparing the actually measured index values with the thresholds, it can be judged whether the consistency performance of the battery module on each key performance parameter is good.

[0140] Count the first quantity of the curve similarity index greater than the curve similarity index threshold and the second quantity of the within-group coefficient of variation index greater than the within-group coefficient of variation index threshold. The first quantity reflects the number of parameters with poor performance of the curve similarity index among the key performance parameters; the second quantity reflects the number of parameters with poor performance of the within-group coefficient of variation index among the key performance parameters.

[0141] Deduct the first quantity of preset scores and the second quantity of preset scores from the total consistency score of the battery module to determine the upper limit of the score. The preset score here can be determined according to factors such as the importance of each key performance parameter to the overall performance of the battery module. Considering the consistency performance of multiple key performance parameters comprehensively, a reasonable upper limit of the score is finally determined. This upper limit reflects the highest possible score based on the current consistency situation of the key performance parameters. The upper limit of the score calculated in this way can prevent the abnormal influence of a single feature on the overall score and can also avoid ignoring other important features due to the abnormality of a single index.

[0142] Exemplarily, the consistency score of the battery module can adopt a hundred-mark system, that is, the total consistency score is 100 points, and the preset score can be set to 20 points. That is to say, among multiple key performance parameters, for each occurrence of the curve similarity index greater than the curve similarity index threshold, 20 points are deducted, and for each occurrence of the within-group coefficient of variation index greater than the within-group coefficient of variation index threshold, 20 points are also deducted.

[0143] In the above implementation scheme of the present application, by comparing the actual curve similarity index and the within-group coefficient of variation index with the preset thresholds, the number of parameters that do not meet the requirements is obtained, and then the corresponding scores are deducted from the total score, accurately reflecting the performance consistency performance of the battery module and providing an accurate quantitative basis for evaluating its overall quality.

[0144] In an optional embodiment of the present application, after determining the consistency score of the battery module, the method further includes:

[0145] Determine whether the consistency score is lower than the warning score threshold;

[0146] When the consistency score is lower than the warning score threshold, send a warning signal to the battery management system corresponding to the battery module, so that the battery management system turns on the balancing circuit to balance the power of the battery module.

[0147] In the embodiment of the present application, by obtaining the real-time operation data of the battery module and performing battery consistency evaluation in a timely manner when the trigger condition for consistency evaluation is met, multiple key performance parameters are comprehensively considered during the evaluation process, and curve similarity index, within-group coefficient of variation index, and information entropy index are used to evaluate the consistency of the battery module from different perspectives, obtaining a consistency score that can comprehensively and accurately reflect the actual consistency of the battery module. Among them, the lower the consistency score of the battery module, the greater the difference in the key performance parameters of each battery unit in the battery module, that is, the worse the consistency between the battery units.

[0148] Specifically, a warning score threshold for determining whether there is a problem with the consistency of the battery module can be set in advance. This warning score threshold represents the lower limit of the normal acceptable range of the battery module performance. Optionally, this warning score threshold can be determined through a large number of experiments, theoretical analysis, and practical application experience based on the theoretical model and the specific operating environment, equipment performance, and actual operating status of the battery module. The monitoring system of the battery module can continuously monitor the consistency score of the battery module and compare it with the warning score threshold. Once it is detected that the consistency score of the battery module is lower than the warning score threshold, the warning mechanism will be automatically triggered. Based on the main function of the battery management system is to ensure the safe and efficient operation of the battery module, this warning mechanism can be to send a warning signal to the battery management system corresponding to the battery module. After receiving the warning signal, the battery management system immediately starts the internal balancing circuit to balance the power of the battery module. Specifically, for the battery with a higher power, the balancing circuit will consume its excess power, while for the battery with a lower power, it will be appropriately supplemented, so that the power of each battery in the entire battery module tends to be consistent, improving the performance consistency of the battery module.

[0149] In the above embodiments of the present application, it is determined whether the consistency score of the battery module is lower than the warning score threshold, and when the consistency score is lower than the warning score threshold, it is linked with the battery management system to timely turn on the equalization circuit to equalize the power, reduce the power difference between the individual battery cells in the battery module, improve the overall performance consistency of the battery module, so as to be able to timely detect the early performance problems of the battery module, avoid serious failures caused by the accumulation of problems, and improve the reliability and safety of the battery system.

[0150] Further, the warning score threshold can be determined in advance by using a machine learning algorithm. In an optional embodiment of the present application, the method further includes:

[0151] Collect the historical operation data and battery operation status labels respectively corresponding to a plurality of battery modules, and determine the consistency score of the battery module based on the historical operation data;

[0152] Use a machine learning algorithm to train the consistency scores and battery operation status labels respectively corresponding to the plurality of battery modules to obtain a warning score threshold.

[0153] In the embodiments of the present application, in order to comprehensively and accurately reflect the operation of the battery module, the historical operation data of a plurality of battery modules under different times, different usage environments and different working conditions can be collected, and the consistency score of the battery module can be determined based on the above battery module consistency scoring algorithm. At the same time, when collecting the historical operation data, the battery operation status labels also need to be collected. Exemplarily, the battery operation status label can include digital labels 0 and 1, where 0 represents a bad battery and 1 represents a good battery; the battery operation status label can also include text labels, such as normal, slightly abnormal, and severely abnormal.

[0154] After obtaining the consistency scores and battery operating status labels corresponding to multiple battery modules respectively, machine learning algorithms (such as support vector machines, random forests, or neural networks) can be used for training. That is, the consistency scores corresponding to multiple battery modules are used as input features, and the corresponding battery operating status labels are used as target outputs, and input into the selected machine learning algorithm for training. During the training process, the algorithm will measure the difference between the model prediction result and the actual label according to a certain loss function, and continuously adjust the parameters of the model through an optimization algorithm (such as stochastic gradient descent method) to make the value of the loss function gradually decrease. The training process usually needs to divide the dataset into a training set, a validation set, and a test set. The training set is used for learning the model parameters, the validation set is used to monitor the performance of the model during the training process to prevent overfitting, and when the performance of the model on the validation set no longer improves, the training is stopped. Finally, the test set is used to evaluate the trained model to determine the generalization ability of the model. After multiple rounds of training and optimization, the model gradually learns the relationship between the consistency score and the battery operating status, so as to accurately predict the battery operating status according to the new consistency score, and at the same time determine the warning score threshold, that is, when the consistency score is lower than this threshold, the model predicts that the battery operating status is abnormal.

[0155] In the above implementation of the present application, through machine learning training based on a large amount of actual data, a warning score threshold that can accurately reflect the actual operating status of the battery module is obtained, improving the accuracy of the warning.

[0156] The battery consistency evaluation method provided by the embodiments of the present application is introduced above. Next, the battery consistency evaluation device provided by the embodiments of the present application will be introduced with reference to the accompanying drawings.

[0157] As Figure 2 shown, the embodiments of the present application also provide a battery consistency evaluation device, and the device includes:

[0158] A data acquisition module 201, configured to acquire real-time operation data of the battery module and store it in a time-series database; wherein, the real-time operation data is obtained by respectively monitoring multiple key performance parameters corresponding to each battery cell in the battery module;

[0159] A data extraction module 202, configured to extract evaluation data for performing consistency evaluation on the battery module from the time-series database when the battery module meets the trigger condition for battery consistency evaluation;

[0160] A feature extraction module 203, which is used to extract features from the evaluation data for each of the multiple key performance parameters, so as to obtain a curve similarity index, an intra-group coefficient of variation index, and an information entropy index corresponding to the key performance parameter; wherein, the curve similarity index is used to characterize the degree of difference between the battery cells in the battery module, the intra-group coefficient of variation index is used to characterize the intra-group variation degree of the battery module, and the information entropy index is used to characterize the fluctuation degree of the battery cells in the battery module.

[0161] A consistency score determination module 204, which is used to determine the consistency score of the battery module according to the curve similarity index, the intra-group coefficient of variation index, and the information entropy index respectively corresponding to the multiple key performance parameters.

[0162] Optionally, the data extraction module includes:

[0163] A data extraction sub-module, which is used to extract the target real-time operation data of the battery module within a target time period from the time series database; wherein, the end time corresponding to the target time period is the time when the battery module meets the trigger condition for battery consistency evaluation, and the target time period corresponds to a preset duration.

[0164] A data processing sub-module, which is used to clean, filter, and reorganize the target real-time operation data according to a preset data frequency, so as to obtain the evaluation data with the same data frequency.

[0165] Optionally, after determining the consistency score of the battery module, the device further includes:

[0166] An acquisition module, which is used to acquire the consistency score and the hierarchical weight coefficient corresponding to each sub-hierarchy under the parent hierarchy.

[0167] A first determination module, which is used to determine the consistency score of the parent hierarchy through weighted operation according to the consistency score and the hierarchical weight coefficient corresponding to each sub-hierarchy under the parent hierarchy.

[0168] Wherein, the energy storage unit is the parent hierarchy of the battery module, the energy storage system is the parent hierarchy of the energy storage unit, and the energy storage power station is the parent hierarchy of the energy storage system.

[0169] Optionally, the feature extraction module includes:

[0170] A first extraction sub-module, which is used to extract each parameter value of the key performance parameter corresponding to each battery cell in the battery module from the evaluation data, so as to obtain a first parameter value sequence corresponding to each battery cell.

[0171] The first determination sub-module is used to determine a reference sequence according to the first parameter value sequences corresponding to each battery cell in the battery module. Wherein, the parameter value at the Nth position in the reference sequence is the mean value of the parameter values at the Nth position in the first parameter value sequences of the battery cells in the battery module, and N is a positive integer;

[0172] The first calculation sub-module is used to calculate the similarity between the first parameter value sequence corresponding to each battery cell and the reference sequence;

[0173] The second determination sub-module is used to determine the battery cell with the lowest similarity as the target battery cell;

[0174] The first processing sub-module is used to perform numerical conversion on the similarity corresponding to the target battery cell to obtain the curve similarity index.

[0175] Optionally, the feature extraction module includes:

[0176] The second extraction sub-module is used to extract the parameter values of each key performance parameter corresponding to each time point from the evaluation data to obtain a second parameter value sequence corresponding to each time point;

[0177] The second calculation sub-module is used to calculate the mean value of the parameter values in the second parameter value sequence corresponding to each time point and the extreme value standard deviation corresponding to the second parameter value sequence; wherein, the extreme value standard deviation is obtained by taking the square root of the ratio of the square of the difference between the maximum parameter value and the minimum parameter value in the second parameter value sequence to the number of battery cells in the battery module;

[0178] The third calculation sub-module is used to calculate the ratio of the extreme value standard deviation to the mean value to obtain the coefficient of variation corresponding to the time point;

[0179] The third determination sub-module is used to determine the mean value of the coefficients of variation corresponding to each time point in the evaluation data as the within-group coefficient of variation index.

[0180] Optionally, the feature extraction module includes:

[0181] The third extraction sub-module is used to extract the parameter values of each key performance parameter corresponding to each battery cell in the battery module from the evaluation data to obtain a first parameter value sequence corresponding to each battery cell;

[0182] The fourth determination sub-module is used to determine a difference sequence for each battery cell according to the first parameter value sequence corresponding to the battery cell. Wherein, the value at the Mth position in the difference sequence is the difference between the parameter value at the Mth position and the parameter value at the M-1th position in the first parameter value sequence;

[0183] A fifth determination sub-module, configured to determine a plurality of numerical intervals according to the difference sequence;

[0184] A sixth determination sub-module, configured to determine the probability corresponding to each numerical interval according to the number of differences in the difference sequence and the number of differences included in each of the plurality of data intervals;

[0185] A fourth calculation sub-module, configured to calculate the information entropy corresponding to the battery cell according to the probabilities corresponding to the plurality of numerical intervals respectively;

[0186] A seventh determination sub-module, configured to determine the mean value of the information entropies corresponding to the battery cells in the battery module as the information entropy index.

[0187] Optionally, the consistency score determination module includes:

[0188] A score upper limit determination sub-module, configured to determine the score upper limit of the battery module according to the curve similarity index and the within-group coefficient of variation index corresponding to the plurality of key performance parameters respectively;

[0189] A deduction value determination sub-module, configured to determine the deduction value of the battery module through weighted operation according to the information entropy index and the entropy feature weight coefficient corresponding to the plurality of key performance parameters respectively;

[0190] A consistency score determination sub-module, configured to determine the consistency score of the battery module according to the score upper limit and the deduction value.

[0191] Optionally, the score upper limit determination sub-module includes:

[0192] A first acquisition unit, configured to acquire the curve similarity index threshold and the within-group coefficient of variation index threshold respectively corresponding to the plurality of key performance parameters set in advance;

[0193] A second acquisition unit, configured to acquire the first number of curve similarity indexes greater than the curve similarity index threshold among the plurality of key performance parameters;

[0194] A third acquisition unit, configured to acquire the second number of within-group coefficient of variation indexes greater than the within-group coefficient of variation index threshold among the plurality of key performance parameters;

[0195] A determination unit, configured to deduct the first number of preset scores and the second number of preset scores from the total score value of the consistency score of the battery module to obtain the score upper limit of the battery module.

[0196] Optionally, after determining the consistency score of the battery module, the device further includes:

[0197] A judgment module, configured to judge whether the consistency score is lower than a warning score threshold;

[0198] A sending module, configured to send a warning signal to the battery management system corresponding to the battery module when the consistency score is lower than the warning score threshold, so that the battery management system turns on an equalization circuit to perform power equalization on the battery module.

[0199] Optionally, the device further includes:

[0200] A collection module, configured to collect historical operation data and battery operation status labels respectively corresponding to a plurality of battery modules, and determine the consistency score of the battery module based on the historical operation data;

[0201] A training module, configured to use a machine learning algorithm to train the consistency scores and battery operation status labels respectively corresponding to the plurality of battery modules to obtain a warning score threshold.

[0202] Optionally, the triggering conditions for the battery consistency evaluation include that a preset period arrives, or a target triggering event is received; the device further includes:

[0203] A second determination module, configured to determine that the battery module meets the triggering conditions for the battery consistency evaluation when the preset period arrives or the target triggering event is received.

[0204] The battery consistency evaluation device provided in this application obtains the real-time operation data of the battery module, and timely performs battery consistency evaluation when the triggering conditions for the consistency evaluation are met. During the evaluation process, multiple key performance parameters are comprehensively considered, and curve similarity index, intra-group coefficient of variation index and information entropy index are used to evaluate the consistency of the battery module from different angles, avoiding the one-sidedness of single-index evaluation, and being able to more comprehensively and accurately reflect the actual consistency of the battery module, improving the accuracy and reliability of the evaluation.

[0205] An embodiment of this application further provides an electronic device, including: a processor, a memory, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, it implements each process of the above-mentioned battery consistency evaluation method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0206] For example, Figure 3 The schematic physical structure diagram of an electronic device is shown. As Figure 3As shown in the figure, the electronic device may include: a processor 310, a communications interface 320, a memory 330, and a communication bus 340. Among them, the processor 310, the communications interface 320, and the memory 330 complete communication with each other through the communication bus 340. The processor 310 may call the logical instructions in the memory 330. The processor 310 is configured to perform the following steps: obtaining real-time operation data of the battery module and storing it in the time series database; wherein, the real-time operation data is obtained by respectively monitoring multiple key performance parameters corresponding to each battery cell in the battery module; when the battery module meets the trigger condition for battery consistency evaluation, extracting evaluation data for consistency evaluation of the battery module from the time series database; for each key performance parameter among the multiple key performance parameters, performing feature extraction on the evaluation data to obtain a curve similarity index, an intra-group coefficient of variation index, and an information entropy index corresponding to the key performance parameter; wherein, the curve similarity index is used to characterize the degree of difference between the battery cells in the battery module, the intra-group coefficient of variation index is used to characterize the intra-group variation degree of the battery module, and the information entropy index is used to characterize the fluctuation degree of the battery cells in the battery module; determining a consistency score of the battery module according to the curve similarity index, the intra-group coefficient of variation index, and the information entropy index respectively corresponding to the multiple key performance parameters. The processor 310 may also execute other solutions in the embodiments of the present application, which will not be further elaborated here.

[0207] In addition, when the logical instructions in the above-mentioned memory 330 can be implemented in the form of software functional units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application.

[0208] The embodiments of the present application also provide a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it realizes each process of the above-mentioned embodiments of the battery consistency evaluation method and can achieve the same technical effects. To avoid repetition, details are not described here again. Among them, the computer-readable storage medium includes, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc, etc.

[0209] It should be noted that in this text, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, article or device including that element.

[0210] From the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc) and includes several instructions for causing a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in the various embodiments of the present application.

[0211] The embodiments of the present application have been described above with reference to the accompanying drawings. However, the present application is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present application, those of ordinary skill in the art can also make many forms without departing from the purpose of the present application and the scope protected by the claims, and all of them belong to the protection scope of the present application.

[0212] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in the embodiments of the present application can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraint conditions of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0213] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be repeated herein.

[0214] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections between each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0215] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0216] In addition, the functional units in the various embodiments of the present application can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0217] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical discs that can store program codes.

[0218] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present application, and all should be covered by the protection scope of the present application.

Claims

1. A method for evaluating battery consistency, characterized in that The method includes: Obtaining real-time operation data of the battery module and storing it in the time-series database; wherein, the real-time operation data is obtained by respectively monitoring multiple key performance parameters corresponding to each battery cell in the battery module; When the battery module meets the trigger condition for battery consistency evaluation, extracting evaluation data for performing consistency evaluation on the battery module from the time-series database; For each of the multiple key performance parameters, performing feature extraction on the evaluation data to obtain a curve similarity index, an intra-group coefficient of variation index, and an information entropy index corresponding to the key performance parameter; wherein, the curve similarity index is used to characterize the difference degree between battery cells in the battery module, the intra-group coefficient of variation index is used to characterize the intra-group variation degree of the battery module, and the information entropy index is used to characterize the fluctuation degree of battery cells in the battery module; Determining the consistency score of the battery module according to the curve similarity index, the intra-group coefficient of variation index, and the information entropy index respectively corresponding to the multiple key performance parameters.

2. The battery consistency evaluation method according to claim 1, wherein After determining the consistency score of the battery module, the method further includes: Judging whether the consistency score is lower than the warning score threshold; When the consistency score is lower than the warning score threshold, sending a warning signal to the battery management system corresponding to the battery module, so that the battery management system turns on the balancing circuit to perform power balancing on the battery module.

3. The battery consistency evaluation method according to claim 2, wherein The method further includes: Collecting historical operation data and battery operation status labels respectively corresponding to multiple battery modules, and determining the consistency score of the battery module based on the historical operation data; Training the consistency scores and battery operation status labels respectively corresponding to the multiple battery modules by using a machine learning algorithm to obtain the warning score threshold.

4. The battery consistency evaluation method according to claim 1, wherein Extracting the evaluation data for performing consistency evaluation on the battery module from the time-series database includes: Extracting the target real-time operation data of the battery module within a target time period from the time-series database; wherein, the end moment corresponding to the target time period is the moment when the battery module meets the trigger condition for battery consistency evaluation, and the target time period corresponds to a preset duration; Cleaning, filtering, and reorganizing the target real-time operation data according to a preset data frequency to obtain the evaluation data with the same data frequency.

5. The battery consistency evaluation method according to claim 1, wherein After determining the consistency score of the battery module, the method further includes: Obtaining the consistency score and the hierarchical weight coefficient corresponding to each sub-level under the parent level; Determining the consistency score of the parent level through weighted operation according to the consistency score and the hierarchical weight coefficient corresponding to each sub-level under the parent level; Wherein, the energy storage unit is the parent level of the battery module, the energy storage system is the parent level of the energy storage unit, and the energy storage power station is the parent level of the energy storage system.

6. The battery consistency evaluation method according to claim 1, wherein Performing feature extraction on the evaluation data to obtain the curve similarity index corresponding to the key performance parameter, including: Extract the parameter values of the key performance parameters corresponding to each battery cell in the battery module from the evaluation data to obtain a first parameter value sequence corresponding to each battery cell; Determine a reference sequence according to the first parameter value sequences corresponding to each battery cell in the battery module, where the parameter value at the Nth position in the reference sequence is the mean of the parameter values at the Nth position in the first parameter value sequences of the battery cells in the battery module, and N is a positive integer; For each battery cell, calculate the similarity between the first parameter value sequence corresponding to the battery cell and the reference sequence; Determine the battery cell with the lowest similarity as the target battery cell; Perform numerical conversion on the similarity corresponding to the target battery cell to obtain the curve similarity index; 7. The battery consistency evaluation method according to claim 1, wherein Extract features from the evaluation data to obtain the within-group coefficient of variation index corresponding to the key performance parameters, including: Extract the parameter values of the key performance parameters corresponding to each time point from the evaluation data to obtain a second parameter value sequence corresponding to each time point; For each time point, calculate the mean of the parameter values in the second parameter value sequence corresponding to the time point and the extreme value standard deviation corresponding to the second parameter value sequence; where the extreme value standard deviation is obtained by taking the square root of the ratio of the square of the difference between the maximum parameter value and the minimum parameter value in the second parameter value sequence to the number of battery cells in the battery module; Calculate the ratio of the extreme value standard deviation to the mean to obtain the coefficient of variation corresponding to the time point; Determine the mean of the coefficients of variation corresponding to each time point in the evaluation data as the within-group coefficient of variation index; 8. The battery consistency evaluation method according to claim 1, characterized in that Extract features from the evaluation data to obtain the information entropy index corresponding to the key performance parameters, including: Extract the parameter values of the key performance parameters corresponding to each battery cell in the battery module from the evaluation data to obtain a first parameter value sequence corresponding to each battery cell; For each battery cell, determine a difference sequence according to the first parameter value sequence corresponding to the battery cell, where the value at the Mth position in the difference sequence is the difference between the parameter value at the Mth position and the parameter value at the M - 1th position in the first parameter value sequence; Determine multiple numerical intervals according to the difference sequence; Determine the probability corresponding to each numerical interval according to the number of differences in the difference sequence and the number of differences included in each of the multiple data intervals; Calculate the information entropy corresponding to the battery cell according to the probabilities corresponding to the multiple numerical intervals; Determine the mean of the information entropies corresponding to each battery cell in the battery module as the information entropy index; 9. The battery consistency evaluation method according to claim 1, wherein Determine the consistency score of the battery module according to the curve similarity index, within-group coefficient of variation index, and information entropy index corresponding to the multiple key performance parameters, including: Determine the upper limit of the score of the battery module according to the curve similarity index and within-group coefficient of variation index corresponding to the multiple key performance parameters; Based on the information entropy index and entropy feature weight coefficient corresponding to each of the multiple key performance parameters, determine the deduction score of the battery module through weighted operation; Based on the score upper limit and the deduction score, determine the consistency score of the battery module.

10. The battery consistency evaluation method according to claim 9, wherein Determine the score upper limit of the battery module according to the curve similarity index and the within-group coefficient of variation index corresponding to each of the multiple key performance parameters, including: Obtain the pre-set threshold values of the curve similarity index and the within-group coefficient of variation index corresponding to each of the multiple key performance parameters; Obtain the first quantity of the curve similarity index greater than the curve similarity index threshold value among the multiple key performance parameters; Obtain the second quantity of the within-group coefficient of variation index greater than the within-group coefficient of variation index threshold value among the multiple key performance parameters; Deduct the first quantity of pre-set scores and the second quantity of pre-set scores from the total score value of the battery module's consistency score to obtain the score upper limit of the battery module.

11. The battery consistency evaluation method according to claim 1, characterized in that The triggering conditions for the battery consistency evaluation include that a preset period arrives, or a target trigger event is received; The method further includes: When the preset period arrives or a target trigger event is received, determine that the battery module meets the triggering conditions for the battery consistency evaluation.

12. A battery consistency evaluation device, characterized in that, The device includes: A data acquisition module, configured to acquire the real-time operation data of the battery module and save it to the time series database; wherein, the real-time operation data is obtained by respectively monitoring multiple key performance parameters corresponding to each battery cell in the battery module; A data extraction module, configured to extract evaluation data for consistency evaluation of the battery module from the time series database when the battery module meets the triggering conditions for the battery consistency evaluation; A feature extraction module, configured to perform feature extraction on the evaluation data for each of the multiple key performance parameters to obtain the curve similarity index, the within-group coefficient of variation index, and the information entropy index corresponding to the key performance parameter; wherein, the curve similarity index is used to characterize the difference degree between battery cells in the battery module, the within-group coefficient of variation index is used to characterize the within-group variation degree of the battery module, and the information entropy index is used to characterize the fluctuation degree of battery cells in the battery module; A consistency score determination module, configured to determine the consistency score of the battery module according to the curve similarity index, the within-group coefficient of variation index, and the information entropy index corresponding to each of the multiple key performance parameters.

13. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, the steps of the battery consistency evaluation method according to any one of claims 1 to 11 are implemented.

14. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium. When the computer program is executed by the processor, the steps of the battery consistency evaluation method according to any one of claims 1 to 11 are implemented.

Images