Abnormal battery cell identification method and device, computer equipment and readable storage medium

Abstract

The invention discloses an abnormal battery cell identification method and device, computer equipment and a readable storage medium. The method comprises the following steps: acquiring to-be-identified physical parameters of a to-be-identified battery cell at specified time; performing clustering processing on the to-be-identified physical parameters to obtain a plurality of clustered clusters; obtaining clustering index data corresponding to the corresponding clustering cluster based on the physical parameters in any clustering cluster; determining a target clustering cluster in the plurality of clustering clusters according to a comparison result between the clustering index data corresponding to the any clustering cluster and preset index data; and determining abnormal physical parameters according to the physical parameters in the target cluster, and determining an abnormal battery cell set according to the battery cells corresponding to the abnormal physical parameters. Therefore, the abnormal battery cells with different physical states from other battery cells in the same battery cluster can be more accurately identified, and the operation and maintenance efficiency is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage batteries, and in particular to an abnormal battery cell identification method, device, computer equipment and readable storage medium. Background Art

[0002] As the world pays more and more attention to the application of green energy, energy storage systems are increasingly used as an important supporting facility in addition to wind power and photovoltaic power generation. The stability of energy storage systems is crucial to maintaining the stability and reliability of power systems. Therefore, the operation and maintenance management of energy storage systems plays an important role.

[0003] In the related art, the voltage difference between the highest voltage and the lowest voltage of the cells in each battery cluster in the energy storage system is monitored to identify abnormal cells, thereby achieving operation and maintenance management. However, the operation and maintenance efficiency in the related art needs to be improved. Summary of the invention

[0004] The present invention aims to solve one of the technical problems in the related art to at least a certain extent. To this end, the present invention proposes an abnormal battery cell identification method, device, computer equipment and readable storage medium, which can accurately and quickly identify abnormal battery cells, and identify battery cells with different degrees of degradation before the abnormal battery cells reach the conditions for triggering the shutdown of the energy storage system, thereby saving operation and maintenance costs and improving operation and maintenance efficiency.

[0005] To achieve the above-mentioned purpose, a first embodiment of the present invention proposes a method for identifying abnormal battery cells, the method comprising: obtaining physical parameters to be identified of the battery cells to be identified at a specified time; wherein the battery cells to be identified are located in the same battery cluster in the energy storage system; clustering the physical parameters to be identified to obtain a number of clustering sub-clusters after clustering; obtaining clustering index data corresponding to the corresponding clustering sub-clusters based on the physical parameters in any clustering sub-clusters; determining a target clustering sub-clusters among the several clustering sub-clusters based on a comparison result between the clustering index data corresponding to any clustering sub-clusters and preset index data; determining abnormal physical parameters based on the physical parameters in the target clustering sub-clusters, and determining an abnormal battery cell set based on the battery cells corresponding to the abnormal physical parameters.

[0006] According to one embodiment of the present invention, determining a target cluster among the several clusters based on a comparison result between the clustering indicator data corresponding to any cluster and the preset indicator data includes: if the absolute value of the difference between the clustering indicator data corresponding to any cluster and the preset indicator data is greater than or equal to a preset difference threshold, determining the corresponding cluster as the target cluster.

[0007] According to one embodiment of the present invention, the number of physical parameters in the target clustering sub-clustering is at least one; determining the abnormal physical parameter based on the physical parameter in the target clustering sub-clustering includes: if the deviation data between any physical parameter in the target clustering sub-clustering and the preset physical parameter is greater than or equal to a preset deviation threshold, obtaining the abnormal physical parameter according to the corresponding physical parameter.

[0008] According to one embodiment of the present invention, the designated time includes designated time periods under multiple dates; the abnormal physical parameters include abnormal parameters of designated time periods under the multiple dates; determining the abnormal battery cell set according to the battery cells corresponding to the abnormal physical parameters includes: obtaining the frequency of occurrence of the corresponding battery cells in the multiple dates according to the battery cells corresponding to the abnormal parameters on each date; if the frequency of occurrence of any battery cell in the multiple dates is greater than or equal to a preset number threshold, determining the corresponding battery cell as an abnormal battery cell; wherein the abnormal battery cell is included in the abnormal battery cell set.

[0009] According to an embodiment of the present invention, the designated time includes a designated discharge period; and the physical parameter to be identified includes voltage variation data of the battery cell to be identified during the designated discharge period.

[0010] According to an embodiment of the present invention, the designated time includes a designated discharge end time; and the physical parameter to be identified includes power data of the battery cell to be identified at the designated discharge end time.

[0011] According to one embodiment of the present invention, clustering the physical parameters to be identified includes: clustering the physical parameters to be identified using a profile clustering method.

[0012] To achieve the above-mentioned purpose, the second embodiment of the present invention proposes an abnormal battery cell identification device, which includes: a physical parameter acquisition module, used to obtain the physical parameters to be identified of the battery cell to be identified at a specified time; wherein the battery cell to be identified is located in the same battery cluster in the energy storage system; a clustering acquisition module, used to cluster the physical parameters to be identified to obtain a number of clustering clusters after clustering; an indicator data acquisition module, used to obtain clustering indicator data corresponding to the corresponding clustering cluster based on the physical parameters in any clustering cluster; a target cluster determination module, used to determine a target clustering cluster among the several clustering clusters according to a comparison result between the clustering indicator data corresponding to any clustering cluster and preset indicator data; an abnormal battery cell determination module, used to determine abnormal physical parameters according to the physical parameters in the target clustering cluster, and determine an abnormal battery cell set according to the battery cells corresponding to the abnormal physical parameters.

[0013] To achieve the above-mentioned purpose, a third aspect of the present invention proposes a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the abnormal battery cell identification method described in any of the aforementioned embodiments are implemented.

[0014] To achieve the above objectives, a fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the abnormal battery cell identification method described in any of the aforementioned embodiments.

[0015] According to the multiple embodiments provided by the present invention, cluster analysis can more accurately identify abnormal cells with different physical states from other cells in the same battery cluster, better adapt to the performance differences between cells, and more accurately and quickly identify abnormal cells, reducing false positives and missed positives. At the same time, this method can identify cells with different degrees of deterioration before the abnormal cells reach the conditions that trigger the shutdown of the energy storage system, so that operation and maintenance personnel can take preventive measures in advance, and can formulate targeted maintenance plans according to actual conditions, thereby saving operation and maintenance costs and improving operation and maintenance efficiency.

[0016] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1a The present invention is a flowchart of a method for identifying abnormal battery cells according to an embodiment of the present invention.

[0018] Figure 1b A diagram showing voltage changes during the charge and discharge process of a battery cell according to an embodiment of the present specification.

[0019] Figure 2 The present invention is a schematic diagram of a process for determining an abnormal battery cell set according to an embodiment of the present specification.

[0020] Figure 3a A schematic diagram of an operation and maintenance process provided according to an implementation method of this specification.

[0021] Figure 3b Another operation and maintenance process diagram provided according to an implementation method of this specification.

[0022] Figure 4 The present invention is a structural block diagram of an abnormal battery cell identification device provided according to an embodiment of the present specification. DETAILED DESCRIPTION

[0023] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0024] As the world pays more and more attention to the application of green energy, energy storage power stations are increasingly used as an important supporting facility in addition to wind power and photovoltaic power generation. In 2023 alone, a certain country added 26.7GWh (Gigawatt-hour) of energy storage capacity, that is, the total energy storage capacity of the energy storage system increased by 26.7GWh, and the cumulative energy storage capacity reached 86.5GWh. The stability of the energy storage system plays an important role in maintaining the stability and reliability of the power system.

[0025] At present, the energy storage system is mainly composed of lithium iron phosphate cells. Lithium iron phosphate cells have the characteristic that the performance of the cells will continue to deteriorate as the number of charge and discharge times increases. The most intuitive effect of cell degradation is that the voltage difference between the cells continues to increase. In the relevant technology, the basis of the energy storage system control logic is to monitor the voltage difference between the highest voltage and the lowest voltage of the cells in each battery cluster. When it is detected that the voltage difference between the cells in a battery cluster reaches a certain threshold at a certain moment, the energy storage system will take shutdown measures. For example, due to degradation, the voltage of a cell in a battery cluster drops faster than the voltage of other cells in the cluster during the discharge process, resulting in the lowest voltage of the cell during the discharge process being lower than the voltage of other cells at the same time, and triggering the voltage difference limit with the highest voltage of other cells, thereby causing a shutdown. At this time, other cells still have a higher voltage, and some of the electricity in the cells has not been discharged, which reduces the discharge of the energy storage system and the charging and discharging efficiency, resulting in a decrease in the revenue of the power station.

[0026] On the other hand, there are generally hundreds of cells in the same battery cluster. According to the above control logic, when the control module of the battery cluster detects that the voltage difference between the cells in the cluster reaches the set limit value, it will take shutdown measures. If there are multiple cells with different degrees of degradation in the same cluster, after a cell triggers the shutdown condition and is replaced by an organization, another cell may be triggered to shut down again for the same reason after a period of time (such as a week or half a month), so that the cells need to be replaced frequently, resulting in unnecessary waste of manpower, material resources and operation and maintenance time, and low operation and maintenance efficiency. At the same time, frequent replacement of cells in a short period of time requires operation and maintenance personnel to frequently travel back and forth on site, resulting in a great waste of operation and maintenance costs such as operation and maintenance manpower hours and round-trip vehicle distance. In addition, frequent shutdowns of power stations will also lead to a decrease in power station revenue, and each replacement of the plug box requires the power station to be shut down, resulting in the inability of the power station to charge and discharge normally, which further affects the power station revenue.

[0027] In order to improve the operation and maintenance efficiency of the energy storage system and realize the accurate and rapid identification of the cells with different degrees of degradation in the energy storage system, it is necessary to propose an abnormal cell identification method, device, computer equipment and readable storage medium. The clustering result is obtained by clustering the physical parameters of the cells to be identified in the same battery cluster, wherein the clustering result includes several clustering sub-clusters. According to the physical parameters in any clustering sub-clusters obtained after the clustering process, the clustering index data corresponding to the corresponding clustering sub-clusters is obtained to represent the overall characteristics of the physical parameters in the corresponding clustering sub-clusters. The clustering index data corresponding to any clustering sub-clusters is compared with the preset index data to determine whether the corresponding clustering sub-clusters are the target clustering sub-clusters that meet the screening conditions according to the comparison results. The abnormal physical parameters are determined according to the physical parameters in the target clustering sub-clusters determined in the several clustering sub-clusters, so as to obtain an abnormal cell set including at least one abnormal cell according to the cells corresponding to the abnormal physical parameters.

[0028] Therefore, through cluster analysis, abnormal cells with different physical states from other cells in the same battery cluster can be more accurately identified, and the performance differences between cells can be better adapted to identify abnormal cells more accurately and quickly, reducing false positives and missed positives. At the same time, this method can identify cells with different degrees of degradation before the abnormal cells reach the conditions that trigger the shutdown of the energy storage system, so that operation and maintenance personnel can take preventive measures in advance and formulate targeted maintenance plans according to actual conditions, thereby saving operation and maintenance costs and improving operation and maintenance efficiency.

[0029] This specification provides a method for identifying abnormal cells. Figure 1a As shown, the abnormal battery cell identification method may include the following steps.

[0030] S110, obtaining physical parameters of the battery cells to be identified at a specified time; wherein the battery cells to be identified are located in the same battery cluster in the energy storage system.

[0031] S120, clustering the physical parameters to be identified to obtain a number of clusters.

[0032] S130: Obtain clustering index data corresponding to a corresponding clustering based on the physical parameters in any clustering.

[0033] S140: Determine a target clustering among several clusterings according to a comparison result between clustering indicator data corresponding to any clustering and preset indicator data.

[0034] S150, determining abnormal physical parameters according to the physical parameters in the target clustering, and determining an abnormal battery cell set according to the battery cells corresponding to the abnormal physical parameters.

[0035] The specified time may include a specified time period or a specified time point.

[0036] The physical parameter to be identified (physical parameter) may include at least one of the voltage, current, internal resistance, capacitance, etc. of the battery cell. It is understandable that for each battery cell, the collected physical parameters of the battery cell may be labeled using the unique number of the battery cell.

[0037] If a battery cluster includes multiple battery cells, the number of battery cells to be identified is multiple.

[0038] The clustering index data can be used to represent the overall characteristics of the physical parameters in the corresponding clustering, such as consistency, distribution, etc.

[0039] The battery cell corresponding to the abnormal physical parameter is a degraded abnormal battery cell. The abnormal battery cell set includes at least one degraded abnormal battery cell. If the abnormal battery cell set includes multiple abnormal batteries, the degree of degradation of each abnormal battery cell may be different.

[0040] The abnormal cell collection is initialized to an empty collection.

[0041] Specifically, for the same battery cluster in the energy storage system, the physical parameters of multiple cells to be identified in the cluster at a specified time are collected as the physical parameters to be identified. The collected physical parameters to be identified are clustered to divide them into several cluster sub-clusters, and the physical parameters in each sub-cluster have similar characteristics.

[0042] For each cluster, calculate its corresponding cluster index data. These indicators can be any of the statistics such as mean, standard deviation, maximum and minimum difference, etc., to describe the overall characteristics of the battery cells in the cluster. Compare the cluster index data of each cluster with the preset index data respectively to obtain the comparison results corresponding to each cluster. If the cluster index data of a cluster deviates significantly from the preset index data, the cluster can be determined as a target cluster, which contains the physical parameters of abnormal batteries. Thereby, it is possible to determine the target cluster among several clusters.

[0043] Abnormal physical parameters are obtained according to the physical parameters in the target clustering, and the battery cells corresponding to the abnormal physical parameters can be determined as abnormal battery cells, so as to put the abnormal battery cells into the abnormal battery cell set to obtain the final abnormal battery cell set.

[0044] In some embodiments, according to the determined target clustering, all physical parameters in the clustering may be taken as abnormal physical parameters.

[0045] The preset indicator data may be set based on historical data or specification data provided by the manufacturer, or may be obtained by statistics based on the collected physical parameters to be identified.

[0046] Exemplarily, each battery cluster in the energy storage system contains 240 battery cells. For any battery cluster, the 240 battery cells it contains are the battery cells to be identified. The charging period and discharging period of each day in each quarter of the energy storage system can be used as the designated time, and the voltage data of the 240 battery cells in each battery cluster are regularly exported from the energy storage system each quarter as the physical parameters to be identified. It can be understood that the voltage data of each battery cell during the charging period and the discharging period of each day can be presented in the form of a curve (or a broken line), so it can also be called voltage curve data, and each voltage curve data corresponds to a battery cell number. Reference Figure 1b The figure shows the voltage variation of 240 cells in a battery cluster during the charge and discharge process on a certain date. The horizontal axis is time and the vertical axis is voltage value. The time period from 13:00 to 23:59 is the discharge period. At the end of this period, there are several curve segments with significantly lower voltage (corresponding to the framed part in the figure). The voltage curve data containing these curve segments is the voltage data that may be abnormal and needs to be identified through clustering.

[0047] For the same battery cluster, the voltage curve data of the charging and discharging periods of each quarter of the 240 cells in the cluster are clustered to obtain multiple cluster sub-clusters, each of which contains voltage curve data with similar characteristics. For a certain sub-cluster, the voltage curve data in the cluster can be averaged to obtain the voltage curve mean data of the cluster as the cluster index data corresponding to the cluster, or the voltage curve data of the center point of the cluster can be used as the cluster index data corresponding to the cluster. In this way, the cluster index data corresponding to each cluster sub-cluster can be obtained.

[0048] According to the obtained voltage curve data of the battery cells in the battery cluster, the corresponding average data can be statistically obtained as the preset index data, which is used to describe the standard physical parameter index of the battery cells in the battery cluster. The cluster index data corresponding to each cluster sub-cluster is compared with the preset index data to find the cluster index data that obviously deviates from the preset index data, and the cluster sub-cluster corresponding to the corresponding cluster index data is determined as the target cluster sub-cluster.

[0049] The physical parameters in the target clustering are taken as abnormal physical parameters, and the cells corresponding to the abnormal physical parameters are determined as abnormal cells to obtain an abnormal cell set.

[0050] For example, the discharge period of each day in each quarter of the energy storage system can also be used as the designated time, and the voltage data of the discharge period of each battery cell in each battery cluster can be regularly exported from the energy storage system each quarter as the physical parameter to be identified. The description of clustering processing and determining the target clustering sub-cluster and abnormal battery cell set is similar to that in the above example, and will not be repeated here.

[0051] Exemplarily, the voltage data of each battery cell in each battery cluster can also be exported from the energy storage system at regular intervals every day. For example, the voltage data of each battery cell in each battery cluster can be exported from the energy storage system every 10 minutes every day as the physical parameter to be identified. Alternatively, the voltage data of each battery cell on that day or the voltage data during the discharge period of that day can be exported from the energy storage system at the end of each discharge as the physical parameter to be identified.

[0052] It should be noted that in the above embodiment, any one of the internal resistance data and the power data of each battery cell in each battery cluster, or at least two of the voltage data, internal resistance data, and power data of each battery cell, etc., may be periodically derived from the energy storage system as the physical parameter to be identified. At least one of the voltage change data, internal resistance change data, and power change data of each battery cell may also be periodically derived from the energy storage system as the physical parameter to be identified.

[0053] At the same time, for a certain clustering sub-cluster, mean square error calculation or other calculations can also be performed on the physical parameters in the sub-cluster to obtain the sub-clustering index data corresponding to the cluster. The description of the specific process is similar to the process description in the above example and will not be repeated here.

[0054] In other embodiments, the physical parameters in the target clustering may be further analyzed to find physical parameters that are significantly different from the physical parameters of normal cells, or physical parameters that are more abnormal, and determine them as abnormal physical parameters.

[0055] It should be noted that the setting of the specified time can be determined according to the actual application scenario or application requirements, and is not specifically limited in this specification.

[0056] In the above implementation, cluster analysis can be used to more accurately identify abnormal cells with different physical states from other cells in the same battery cluster. Compared with the traditional voltage difference threshold comparison method, the abnormal cell identification method provided in this specification can process multiple physical characteristics of the cell at the same time, and can better adapt to the performance differences between cells, so as to more accurately and quickly identify abnormal cells and reduce false alarms and missed reports. At the same time, this method can identify cells with different degrees of deterioration before the abnormal cells reach the conditions that trigger the shutdown of the energy storage system, so that operation and maintenance personnel can take preventive measures in advance to avoid potential safety risks and economic losses. After identifying the abnormal cells, a targeted maintenance plan can be formulated according to the actual situation, thereby saving costs and improving operation and maintenance efficiency. By timely discovering abnormal cells and performing targeted maintenance, it helps to maintain the health of the entire energy storage system, slow down the overall aging rate, and thus extend the service life of the system.

[0057] In some embodiments, determining a target cluster among several clusters based on a comparison result between cluster indicator data corresponding to any cluster and preset indicator data may include: if the absolute value of the difference between the cluster indicator data corresponding to any cluster and the preset indicator data is greater than or equal to a preset difference threshold, determining the corresponding cluster as the target cluster.

[0058] Specifically, for a certain cluster, the cluster index data corresponding to the cluster is calculated according to the physical parameters in the cluster. The cluster index data of the cluster is compared with the preset index data, and the difference between the two is calculated, and the absolute value is taken. The absolute value of the difference between the cluster index data of the cluster and the preset index data is further compared with the preset difference threshold. If the absolute value of the difference is greater than or equal to the preset difference threshold, the cluster is determined as the target cluster, indicating that the physical parameters in the cluster have obvious abnormalities, and the battery cells corresponding to these physical parameters may have obvious abnormalities. In this way, it can be judged whether each cluster is a target cluster that meets the screening conditions.

[0059] Exemplarily, the physical parameters to be identified include the voltage curve data of the cells to be identified on the same date, and the minimum extreme voltage and the extreme moment corresponding to the minimum extreme voltage can be determined according to the extreme voltage of the discharge period in each voltage curve data. The voltage data of all cells to be identified in the same battery cluster at the same moment are averaged to obtain the preset indicator data, or the maximum voltage data among these voltage data is used as the preset indicator data.

[0060] For a certain cluster, the average voltage curve data of the cluster is obtained by averaging the voltage curve data in the cluster, and the voltage at the extreme value moment in the average voltage curve data can be used as the cluster index data corresponding to the cluster. The difference between the voltage and the preset index data is calculated and the absolute value is taken. Assuming that the preset difference threshold is 200mV, if the absolute value is greater than or equal to 200mV, the cluster is determined as the target cluster.

[0061] Exemplarily, the physical parameters to be identified include the voltage drop rate value of the battery cell to be identified in the same discharge period, and the preset indicator data is obtained by averaging the voltage drop rate values ​​of all the battery cells to be identified in the same battery cluster in the same period, or the minimum voltage drop rate value among these voltage drop rate values ​​is used as the preset indicator data.

[0062] For a certain cluster, the average voltage drop rate of the cluster is obtained by averaging the voltage drop rate values ​​in the cluster, which is used as the cluster indicator data corresponding to the cluster. The difference between the average voltage drop rate and the preset indicator data is calculated and the absolute value is taken. Assuming that the preset difference threshold is 40%, if the absolute value is greater than or equal to 40%, the cluster is determined as the target cluster.

[0063] It should be noted that the preset difference threshold can be determined according to actual application requirements or prior experience, and is not specifically limited in this specification.

[0064] In some embodiments, the number of physical parameters in the target clustering sub-cluster is at least one; determining the abnormal physical parameter based on the physical parameter in the target clustering sub-cluster may include: if the deviation data between any physical parameter in the target clustering sub-cluster and the preset physical parameter is greater than or equal to a preset deviation threshold, obtaining the abnormal physical parameter based on the corresponding physical parameter.

[0065] The deviation data may be used to measure the difference or gap between the corresponding physical parameter and the preset physical parameter.

[0066] Specifically, for each physical parameter in the target clustering, it is compared with the preset normal physical parameter, and the deviation data between each physical parameter and the preset physical parameter is calculated respectively. For a certain physical parameter, if the deviation data between the physical parameter and the preset physical parameter is greater than or equal to the preset deviation threshold, the physical parameter can be determined as an abnormal physical parameter, indicating that the battery cell corresponding to the physical parameter may be abnormal.

[0067] Exemplarily, for any physical parameter, it is assumed that it includes voltage curve data. Based on the voltage curve data of the battery cells in the same battery cluster, the voltage curve data closest to the normal battery cell voltage curve can be used as the preset physical parameter. For each physical parameter in the target clustering sub-cluster, any one of the calculation methods such as Euclidean distance, mean square error, function fitting, etc. can be used to calculate the deviation data between the corresponding physical parameter and the preset physical parameter, and the deviation data corresponding to each physical parameter is compared with the preset deviation threshold, so as to determine the abnormal physical parameter in the target clustering sub-cluster.

[0068] Exemplarily, for any physical parameter, it is assumed that it includes voltage change rate data and charge change rate data. According to the voltage change rate data and charge change rate data of each battery cell in the same battery cluster, the voltage change rate and charge change rate that are closest to the normal battery cell can be screened out, or the lowest voltage change rate and charge change rate can be screened out as the preset physical parameter. For each physical parameter in the target clustering sub-cluster, any one of the calculation methods such as Euclidean distance or mean square error can be used to calculate the voltage deviation data between the voltage change rate data in the corresponding physical parameter and the voltage change rate in the preset physical parameter, and the charge deviation data between the charge change rate data in the corresponding physical parameter and the charge change rate in the preset physical parameter.

[0069] The voltage deviation data corresponding to each physical parameter in the target clustering sub-cluster is compared with the preset voltage deviation threshold, and the power deviation data corresponding to each physical parameter is compared with the preset power deviation threshold, so as to determine the abnormal physical parameter in the target clustering sub-cluster. Alternatively, for any physical parameter in the target clustering sub-cluster, weighted operations such as weighted operations can be performed according to the corresponding voltage deviation data and power deviation data to obtain the deviation data corresponding to the physical parameter, and the deviation data is compared with the preset deviation threshold to determine the abnormal physical parameter in the target clustering sub-cluster.

[0070] It should be noted that the preset physical parameters can be obtained by statistics based on the obtained physical parameters to be identified, or can be set based on historical data, manufacturer specifications or experience values. The deviation data can be a difference, a percentage difference or other forms of measurement. In this specification, the preset physical parameters and the preset indicator data can be the same or different. The preset deviation threshold value can be determined based on actual application requirements or prior experience, and is not specifically limited in this specification.

[0071] In the above implementation, by setting a reasonable preset deviation threshold, the abnormal physical parameters with more obvious abnormal performance in the target clustering can be further identified to reduce false positives and false negatives. At the same time, by adopting flexible threshold setting and dynamic adjustment mechanism, it can adapt to different types of battery cells and application scenarios, and has strong versatility and scalability.

[0072] In some embodiments, the specified time includes a specified time period on a plurality of dates; and the abnormal physical parameter includes an abnormal parameter in a specified time period on a plurality of dates. Figure 2 As shown, determining an abnormal cell set according to cells corresponding to abnormal physical parameters may include the following steps.

[0073] S210, obtaining the frequency of occurrence of the corresponding battery cells on multiple dates according to the battery cells corresponding to the abnormal parameters on each date.

[0074] S220: If the frequency of any battery cell appearing on multiple dates is greater than or equal to a preset number threshold, determine the corresponding battery cell as an abnormal battery cell; wherein the abnormal battery cell set includes the abnormal battery cell.

[0075] Specifically, according to the respective specified time periods under multiple dates, the physical parameters of the corresponding cells to be identified are collected respectively to obtain the physical parameters to be identified. Accordingly, the abnormal physical parameters in the target clustering sub-cluster determined by the above method may include the abnormal parameters of the respective specified time periods under the corresponding dates.

[0076] For each date, the corresponding battery cell can be determined based on any abnormal parameter on that date, so that the battery cells that are abnormal on that date can be recorded. Based on the abnormal parameters on multiple dates, it is possible to count which battery cells are abnormal on the corresponding dates, and count the frequency of the corresponding battery cells appearing on these dates. For example, if a battery cell has abnormal parameters on 5 different dates, the frequency of the battery cell appearing is 5.

[0077] For a certain battery cell among these battery cells, the frequency of occurrence of the battery cell is compared with a preset number threshold. If the frequency of occurrence of the battery cell is greater than or equal to the preset number threshold, it is considered that the battery cell frequently has abnormalities, and the battery cell is determined as an abnormal battery cell and can be placed in an abnormal battery cell set (or abnormal set list).

[0078] It should be noted that the preset number threshold may need to be determined according to the specific battery cell type, application scenario, historical data, etc., and is not specifically limited in this specification.

[0079] In the above implementation, by considering the physical parameters of the battery cells on multiple dates, the accidental impact of the data at a single time point can be reduced, and by counting the abnormal frequency of the battery cells on multiple dates, it is possible to more accurately distinguish between the battery cells that occasionally show abnormalities and the battery cells that continuously show abnormalities, thereby reducing false positives and missed positives. Compared with simple single detection, this method can more accurately identify those battery cells that continuously show abnormal behavior, so as to achieve better identification results, thereby helping to improve operation and maintenance efficiency.

[0080] In some embodiments, the specified time includes a specified discharge period; and the physical parameter to be identified includes voltage change data of the battery cell to be identified during the specified discharge period.

[0081] The voltage change data is the voltage change rate.

[0082] Specifically, the discharge period of the energy storage system can be specified as a specified time. For the same battery cluster in the energy storage system, the voltage data of each battery cell to be identified in the battery cluster during the specified discharge period can be collected, and the voltage change data of the corresponding battery cell during the specified discharge period can be statistically obtained to obtain the physical parameters to be identified.

[0083] In some embodiments, the designated time may include designated discharge time periods under multiple dates.

[0084] Exemplarily, the discharge period of each day in a quarter of the energy storage system is taken as the designated time, and the voltage data of each battery cell in each battery cluster is regularly exported from the energy storage system every quarter, and the voltage change rate of each battery cell in the discharge period of each day in the corresponding quarter is calculated to obtain the physical parameter to be identified.

[0085] It should be noted that by analyzing the voltage change rate of each battery cell in the corresponding discharge period on multiple dates, the degradation trend of each battery cell can also be analyzed in stages to evaluate the degree of degradation of each battery cell, so that operation and maintenance personnel can specify the operation and maintenance plan of the energy storage system in a targeted manner.

[0086] In other embodiments, the specified time may include a specified discharge period under the current date.

[0087] Exemplarily, after the energy storage system finishes discharging on the current date, the discharge period of the energy storage system on the current date can be used as the designated time to collect voltage data of each battery cell during the period, and the corresponding voltage change rate is statistically obtained to obtain the physical parameter to be identified.

[0088] For example, during the discharge operation of the energy storage system on the current date, the voltage data of each battery cell in the energy storage system can be collected at intervals, and the corresponding voltage change rate can be obtained by statistics. Then, the specified time is a plurality of preset intervals during the discharge process on the current date.

[0089] Further, illustratively, reference is made to Figure 3a As shown, in the related technology, the processing flow of realizing operation and maintenance based on the energy storage system (also called energy storage energy management system EMS, Energy Management System) control program mainly includes: first, the EMS collects the voltage information of each battery cell during the charging and discharging process of the energy storage power station, and then the EMS control program judges the voltage difference of the battery cells in the same cluster, and when it is judged that the voltage difference between the battery cells in the same cluster reaches 400mV, an alarm is issued to shut down, and finally the operation and maintenance personnel go to the site to replace the deteriorated battery cells.

[0090] This manual adds big data analysis and early warning setting logic based on the above energy storage system control program. Figure 3b As shown, in Figure 3a On the basis of the operation and maintenance process shown, the energy storage cell operation and maintenance method provided in this specification can collect and analyze the voltage change rate differences between each cell in the same battery cluster from the background regularly (for example, one month or one quarter, etc.) to confirm the deteriorated cell. On the other hand, a big data analysis module for comparing the voltage change rate differences between each cell in the same battery cluster is added to the energy storage system control program at regular intervals (for example, every 5 minutes or 10 minutes, etc.). When the difference in the voltage difference change rate between cells in the same cluster reaches the threshold, a voltage change rate difference list will be issued separately, that is, a list of abnormal voltage change rates will be issued separately to display the voltage change rate with a large deviation from the minimum voltage change rate in the same cluster and the corresponding cell number. These cells have a certain degree of deterioration, but have not yet reached the conditions for shutting down the energy storage power station. Therefore, cells with more serious deterioration can be distinguished in advance, and cells with different degrees of deterioration can be displayed. The operation and maintenance personnel can choose to replace the cells in the deterioration list at one time to reduce the number of visits to the site, improve the operation and maintenance efficiency, reduce labor costs, and at the same time reduce the number of power outages of the entire station and reduce the loss of the power station.

[0091] By adding a cell big data analysis alarm setting on the basis of the existing energy storage system, the energy storage system can constantly compare the big data of the voltages of all cells, and constantly alarm to display the cells that are seriously deteriorated. The pressure difference between the cells can remind the operation and maintenance personnel before it reaches the system alarm shutdown pressure difference threshold, so that the operation and maintenance personnel can replace them at one time in a unified manner, reducing the number of system shutdowns due to pressure difference problems, and at the same time reducing the situation where the system suddenly shuts down due to sudden degradation of the cells during a longer interval of regular data analysis.

[0092] In some embodiments, the designated time includes a designated discharge end time; and the physical parameter to be identified includes power data of the battery cell to be identified at the designated discharge end time.

[0093] The designated discharge end time is the end time of the designated discharge period.

[0094] The power data is also called capacity data or remaining power data, which can be calculated through the cut-off voltage of the battery cell.

[0095] Specifically, the discharge end time of the energy storage system can be designated as a designated time, and for the same battery cluster in the energy storage system, the power data of each battery cell to be identified in the battery cluster at the designated discharge end time can be collected to obtain the physical parameters to be identified.

[0096] It should be noted that the designated time may include designated discharge end times on multiple dates, or may only include designated discharge end times on the current date.

[0097] Furthermore, if the specified time includes specified discharge end times on multiple dates, then based on the power data of each battery cell to be identified in the same battery cluster at the specified discharge end times on the multiple dates, the power change data (power change rate) of the corresponding battery cell on the multiple dates can also be calculated to analyze the degradation change trend of each battery cell, and the physical parameters to be identified can also include these power change data.

[0098] Regarding the description of obtaining the physical parameter to be identified based on the power data in this embodiment, please refer to the description of obtaining the physical parameter to be identified based on the voltage change data in this specification, and the details will not be repeated here.

[0099] It should be noted that the specified time may include a specified discharge period and the end time of the specified discharge period, and the physical parameters to be identified may include the voltage change data of the battery cell to be identified during the specified discharge period, and the power data of the battery cell to be identified at the end time of the specified discharge period.

[0100] In some implementations, clustering the physical parameters to be identified may include: clustering the physical parameters to be identified using a profile clustering method.

[0101] Among them, the silhouette clustering method can also be called a clustering method based on silhouette coefficient.

[0102] Specifically, any one of the algorithms such as K-means (K-means clustering algorithm), DBSCAN (Density-Based Spatial Clustering of Applications with Noise), hierarchical clustering, etc. can be used to perform preliminary clustering on the physical parameters to be identified, and obtain preliminary clustering clusters. For any preliminary clustering cluster, the silhouette coefficient of each physical parameter can be calculated respectively, and the average silhouette coefficient of the cluster can be calculated as a measure of the overall clustering quality. According to the size of the average silhouette coefficient, the quality of the current clustering result is evaluated. A higher average silhouette coefficient indicates a better clustering effect, while a lower average silhouette coefficient indicates a poor clustering effect.

[0103] By adjusting the parameters of the clustering algorithm (such as the number of clusters k in K-means, the neighborhood radius and the minimum number of samples in DBSCAN, etc.), re-clustering and recalculating the average silhouette coefficient, the parameter setting that maximizes the average silhouette coefficient is selected as the optimal configuration.

[0104] In the above implementation, the clustering method based on the silhouette coefficient adjusts different clustering parameters to perform clustering operations respectively, and obtains the corresponding optimal clustering results through the silhouette coefficient. When the intra-class distance of the clustering cluster is less than the inter-class distance, the clustering result is more compact and the contour is more obvious. Compared with the clustering clustering divided according to a fixed number of clusters, the physical parameters are usually only divided into two clusters, and it is difficult to distinguish abnormal parameters. The use of the silhouette coefficient method to evaluate the selected new number of clusters can better divide the differences in variable points (physical parameters) under different numbers of iterations. The monomer cells with high consistency of physical parameters have similar characteristics and will be divided into the same cluster; while other variable points with poor consistency (abnormal monomer cells) will be divided into other clusters. In this way, abnormal monomer clusters and normal monomer clusters can be quickly distinguished, thereby realizing the rapid identification of early monomer abnormalities.

[0105] The embodiment of this specification also provides an abnormal battery cell identification device, referring to Figure 4 As shown, the abnormal cell identification device 400 may include: a physical parameter acquisition module 410 , a clustering acquisition module 420 , an indicator data acquisition module 430 , a target clustering determination module 440 , and an abnormal cell determination module 450 .

[0106] The physical parameter acquisition module 410 is used to acquire the physical parameters of the battery cells to be identified at a specified time; wherein the battery cells to be identified are located in the same battery cluster in the energy storage system.

[0107] The clustering sub-clustering acquisition module 420 is used to perform clustering processing on the physical parameters to be identified to obtain a plurality of clusters after clustering.

[0108] The index data acquisition module 430 is used to obtain cluster index data corresponding to any cluster based on the physical parameters in the cluster.

[0109] The target cluster determination module 440 is used to determine a target cluster among a plurality of clusters according to a comparison result between cluster indicator data corresponding to any cluster and preset indicator data.

[0110] The abnormal cell determination module 450 is used to determine abnormal physical parameters according to the physical parameters in the target clustering, and determine an abnormal cell set according to the cells corresponding to the abnormal physical parameters.

[0111] For the specific limitations of the abnormal battery cell identification device, please refer to the limitations of the abnormal battery cell identification method above, which will not be repeated here. Each module in the above-mentioned abnormal battery cell identification device can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0112] The embodiments of this specification also provide a computer device, which may include a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the abnormal battery cell identification method in any of the aforementioned embodiments are implemented.

[0113] The embodiments of the present specification also provide a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the abnormal battery cell identification method in any of the aforementioned embodiments are implemented.

[0114] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0115] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0116] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0117] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0118] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0119] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A method for identifying abnormal battery cells, characterized in that: The method comprises: Obtaining physical parameters of the battery cell to be identified at a specified time; wherein the battery cell to be identified is located in the same battery cluster in the energy storage system; Performing clustering processing on the physical parameters to be identified to obtain a plurality of clustering sub-clusters after clustering; Based on the physical parameters in any clustering sub-cluster, the clustering index data corresponding to the corresponding clustering sub-cluster is obtained; Determining a target clustering sub-cluster among the plurality of clustering sub-clusters according to a comparison result between the clustering index data corresponding to any clustering sub-clusters and the preset index data; An abnormal physical parameter is determined according to the physical parameter in the target clustering, and an abnormal battery cell set is determined according to the battery cells corresponding to the abnormal physical parameter.

2. The method according to claim 1, characterized in that: The determining a target clustering from the plurality of clusterings according to a comparison result between the clustering indicator data corresponding to any clustering and the preset indicator data comprises: If the absolute value of the difference between the clustering indicator data corresponding to any clustering sub-cluster and the preset indicator data is greater than or equal to a preset difference threshold, the corresponding clustering sub-cluster is determined as the target clustering sub-cluster.

3. The method according to claim 1, characterized in that The number of the physical parameters in the target clustering sub-cluster is at least one; and determining the abnormal physical parameter according to the physical parameter in the target clustering sub-cluster includes: If the deviation data between any physical parameter in the target clustering and the preset physical parameter is greater than or equal to a preset deviation threshold, the abnormal physical parameter is obtained according to the corresponding physical parameter.

4. The method according to claim 1, characterized in that: The specified time includes a specified time period of each of the multiple dates; the abnormal physical parameters include abnormal parameters of the specified time period of each of the multiple dates; and determining the abnormal battery cell set according to the battery cells corresponding to the abnormal physical parameters includes: According to the cells corresponding to the abnormal parameters on each date, the frequency of occurrence of the corresponding cells on the multiple dates is obtained; If the frequency of any battery cell appearing in the multiple dates is greater than or equal to a preset number threshold, the corresponding battery cell is determined as an abnormal battery cell; wherein the abnormal battery cell set includes the abnormal battery cell.

5. The method according to claim 1, characterized in that The specified time includes a specified discharge period; the physical parameter to be identified includes voltage change data of the battery cell to be identified during the specified discharge period.

6. The method according to claim 1, characterized in that The designated time includes a designated discharge end time; the physical parameter to be identified includes the power data of the battery cell to be identified at the designated discharge end time.

7. The method according to any one of claims 1 to 6, characterized in that: The clustering process of the physical parameters to be identified includes: The physical parameters to be identified are clustered using a contour clustering method.

8. An abnormal battery cell identification device, characterized in that: The device comprises: A physical parameter acquisition module, used to acquire the physical parameters of the battery cell to be identified at a specified time; wherein the battery cell to be identified is located in the same battery cluster in the energy storage system; A clustering sub-clustering acquisition module is used to perform clustering processing on the physical parameters to be identified to obtain a plurality of clustering sub-clusterings after clustering; An index data acquisition module, used to obtain cluster index data corresponding to a corresponding cluster based on the physical parameters in any cluster; A target cluster determination module, configured to determine a target cluster among the plurality of clusters according to a comparison result between cluster indicator data corresponding to any cluster and preset indicator data; The abnormal cell determination module is used to determine abnormal physical parameters according to the physical parameters in the target clustering, and determine an abnormal cell set according to the cells corresponding to the abnormal physical parameters.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images