Battery cell state evaluation method, device, equipment and medium

Abstract

The invention discloses a battery cell state evaluation method, device and equipment and a medium, and the method comprises the steps: obtaining historical operation data of an energy storage power station, and screening out the historical operation data of an abnormal battery cell; calculating feature data vectors of the abnormal battery cells according to the historical operation data of the abnormal battery cells, and forming a comprehensive feature vector matrix of the abnormal battery cells; according to the comprehensive feature vector matrix, establishing an energy storage cell evaluation model; and through the energy storage cell evaluation model, predicting to obtain a state evaluation result of the to-be-tested cell. According to the method, feature processing and state evaluation can be performed on the battery cell data in a targeted manner, so that on one hand, the accuracy of abnormal battery cell identification is improved, and on the other hand, related technicians can be guided to execute corresponding operation and maintenance decisions without opening the energy storage equipment and stopping the energy storage equipment; therefore, the operation and maintenance inspection efficiency is improved, the fault-free operation time of the energy storage power station is prolonged, and the cell state monitoring cost is reduced.

Description

Technical Field

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

[0002] As one of the important components of the new energy system, electrochemical energy storage technology has developed rapidly in recent years. Among them, lithium-ion batteries have become an ideal choice for portable terminal devices, transportation, and large-scale energy storage systems due to their high energy density, low self-discharge rate, long cycle life, fast charging and discharging, and low environmental pollution. In actual applications, lithium-ion batteries may experience abnormal conditions and performance degradation, which not only brings safety hazards such as battery expansion, shell rupture, abnormal temperature rise, but also may cause fires and explosions. In addition, different battery operating conditions will also cause differential degradation of the performance of lithium-ion batteries, which may cause the consistency of the batteries in the energy storage power station to deteriorate, the charging and discharging capacity to fail to meet the standards, and the daily operating efficiency cannot meet the operational needs. To deal with these problems, conventional solutions include strengthening manual inspections, unpacking visual inspections, pre-buried gas and sound and light signal characteristic sensors for battery appearance inspections or unpacking inspections.

[0003] However, the existing solutions have the following problems: On the one hand, for large-scale energy storage power stations, the efficiency of analyzing and troubleshooting a large number of battery cells one by one, whether it is manual inspection and unpacking inspection, or uploading characteristic signal reminders through sensors, is relatively low, and the hardware cost, manual operation and maintenance cost and technical difficulty are high. On the other hand, adding equipment and manual unpacking will destroy the original structure of the battery, and the battery warranty and subsequent use will also have different degrees of impact due to different construction integration processes. Summary of the invention

[0004] In view of the above problems, the present application is proposed to provide a battery cell status assessment method, device, equipment and medium, so as to achieve the technical effects of improving the accuracy of abnormal battery cell identification, improving operation and maintenance inspection efficiency, and reducing the cost of battery cell status monitoring.

[0005] According to a first aspect of the present application, a method for evaluating a battery cell state is provided, the method comprising:

[0006] Obtain historical operation data of energy storage power stations and filter out abnormal battery cell historical operation data;

[0007] Calculate the characteristic data vector of the abnormal cell according to the historical operation data of the abnormal cell, and form a comprehensive characteristic vector matrix of the abnormal cell;

[0008] Establishing an energy storage cell evaluation model according to the comprehensive eigenvector matrix;

[0009] The energy storage cell evaluation model is used to predict the state evaluation result of the cell to be tested.

[0010] Optionally, according to the historical operation data of the abnormal battery cell, a plurality of characteristic data vectors of the abnormal battery cell are calculated and a comprehensive characteristic vector matrix of the abnormal battery cell is formed, including:

[0011] According to the historical operation data of the abnormal battery cell, the abnormal operation data of the abnormal battery cell during the complete charge and discharge process within the evaluation cycle is screened out;

[0012] According to the abnormal operation data in the complete charging and discharging process, a plurality of characteristic data vectors of the abnormal battery cell are calculated, and a comprehensive characteristic vector matrix between the classification results of the abnormal battery cell and the characteristic data vector is formed.

[0013] Optionally, the characteristic data vector of the abnormal cell includes at least one of the following:

[0014] Voltage-related characteristic data vector, current-related characteristic data vector, temperature-related characteristic data vector, internal resistance-related characteristic data vector, range-related characteristic data vector, charge and discharge depth-related characteristic data vector, and historical data longitudinal comparison ranking-related characteristic data vector.

[0015] Optionally, screening out abnormal battery cell historical operation data includes:

[0016] According to the historical operation data of the energy storage power station, the available capacity and / or the deformation rate of the battery cells of all the battery cells in the energy storage power station are obtained respectively;

[0017] According to the available capacity of each battery cell and / or the battery cell deformation rate, the classification results of abnormal batteries are screened and obtained.

[0018] Optionally, the classification result of the abnormal battery cell includes at least one of the following: a seriously abnormal battery cell, a generally abnormal battery cell, and a slightly abnormal battery cell.

[0019] Optionally, establishing an energy storage cell evaluation model according to the comprehensive eigenvector matrix includes:

[0020] The gradient boosting decision tree algorithm is used to establish an energy storage cell evaluation model;

[0021] The model is obtained through machine learning training using multiple sets of training data, the input of the model is the feature data vector corresponding to each abnormal battery cell in the comprehensive feature vector matrix, and the output of the model is the classification result of the abnormal battery cell.

[0022] Optionally, before calculating the multiple characteristic data vectors of the abnormal battery cell, the method further includes:

[0023] Pre-process the historical operation data of the screened abnormal cells;

[0024] The preprocessing includes filtering and cleaning the historical operation data that exceeds the rated operation range, and performing null value interpolation on the filtered and cleaned historical operation data.

[0025] According to a second aspect of the present application, a battery cell status evaluation device is provided, the device comprising:

[0026] A data screening unit is used to obtain the historical operation data of the energy storage power station and screen out the historical operation data of abnormal cells;

[0027] A feature vector calculation unit, used to calculate the feature data vector of the abnormal battery cell according to the historical operation data of the abnormal battery cell, and form a comprehensive feature vector matrix of the abnormal battery cell;

[0028] A model building unit, used to build an energy storage cell evaluation model according to the comprehensive eigenvector matrix;

[0029] The battery cell evaluation unit is used to predict the state evaluation result of the battery cell to be tested through the energy storage battery cell evaluation model.

[0030] According to a third aspect of the present application, an electronic device is provided, comprising: a processor; and a memory arranged to store computer executable instructions, wherein when the executable instructions are executed, the processor executes the evaluation method as described in any one of the first aspects above.

[0031] According to a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores one or more programs, and when the one or more programs are executed by a processor, the evaluation method as described in any one of the first aspects above is implemented.

[0032] As can be seen from the above, at least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects: first, the historical operation data of the energy storage power station is obtained, and the historical operation data of the abnormal battery cells is screened out; second, according to the historical operation data of the abnormal battery cells, the characteristic data vector of the abnormal battery cells is calculated, and a comprehensive characteristic vector matrix of the abnormal battery cells is formed; third, according to the comprehensive characteristic vector matrix, an energy storage battery cell evaluation model is established; finally, the state evaluation result of the battery cell to be tested is predicted through the energy storage battery cell evaluation model. Through the battery cell state evaluation method of the embodiment of the present application, on the one hand, it is possible to perform targeted feature processing on the battery cell data, more accurately reflect the state change process inside the battery cell, and improve the accuracy of abnormal battery cell identification; on the other hand, there is no need to unpack and shut down the energy storage equipment, which improves the efficiency of operation and maintenance inspection, and also improves the trouble-free operation time and economic benefits of the energy storage power station; at the same time, it is possible to judge battery abnormalities in a non-invasive manner, without the need to add additional sensors, without destroying the original structure of the battery, and reducing the cost of battery cell state monitoring.

[0033] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present application. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0035] Figure 1 A schematic diagram of a flow chart of a method for evaluating a battery cell state in one embodiment of the present application;

[0036] Figure 2 It is an abnormal battery cell status analysis curve of a user-side energy storage power station in one embodiment of the present application;

[0037] Figure 3 This is a schematic diagram of the structure of a battery cell status assessment device in one embodiment of the present application;

[0038] Figure 4 This is a schematic diagram of the structure of an electronic device in one embodiment of the present application. DETAILED DESCRIPTION

[0039] The exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0040] The technical concept of the present application is that, first, by taking advantage of the massive data of the energy storage cloud platform, based on the real-time online data of the battery, combined with feature engineering analysis technology, feature vectors of multiple secondary processing dimensions are constructed and associated with historical battery abnormal data, thereby building a machine learning model; second, the online real-time data of the energy storage system is input into the constructed model to analyze the battery operating status and abnormal trends, and realize remote monitoring of the battery status, which not only avoids structural changes to the battery, but can also guarantee the battery operation quality to the greatest extent, while improving the operation and maintenance efficiency and reducing the operation and maintenance costs.

[0041] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0042] In one embodiment of the present application, Figure 1 As shown, a flow chart of a method for evaluating a battery cell state is provided, and the method for evaluating a battery cell state includes at least the following steps S110 to S140:

[0043] Step S110, obtaining historical operation data of the energy storage power station, and screening out abnormal battery cell historical operation data.

[0044] Specifically, it is necessary to obtain the historical operating data of all cells in the energy storage power station during the evaluation period, and filter out the operating data of abnormal cells from the historical operating data of all cells, and use it as the historical operating data of abnormal cells. In this implementation, the sampling device level for acquiring data is the cell, and the sampling period is preferably 1 minute; the evaluation period refers to a specific evaluation period, such as 1 day or 1 week; of course, this application does not limit the length of the evaluation period and sampling period.

[0045] Preferably, according to the battery cell operation value range given by the manufacturer, the abnormal battery cell historical operation data that is not within the range can be filtered and cleaned, and then the null values ​​can be interpolated, and the characteristic vector of the filtered interpolated data can be calculated.

[0046] Step S120, calculating the characteristic data vector of the abnormal cell according to the historical operation data of the abnormal cell, and forming a comprehensive characteristic vector matrix of the abnormal cell.

[0047] In this embodiment, it is necessary to select the historical operation data of abnormal cells or the historical operation data of abnormal cells after filtering and interpolation, and screen the above data in the complete charging and discharging process to calculate and extract the characteristic data vector of abnormal cells. At the same time, according to the multi-dimensional characteristic data vector, the characteristic data vectors of each abnormal cell are merged to form a comprehensive characteristic vector matrix of multiple abnormal cells.

[0048] Step S130: establishing an energy storage cell evaluation model according to the comprehensive eigenvector matrix.

[0049] Specifically, based on the comprehensive feature vector matrix of all abnormal cells, an algorithm model training set containing all feature data vectors is constructed. That is, the input of the model is a multi-dimensional feature data structure, and its output is the abnormal classification label of each cell.

[0050] Step S140: predicting and obtaining a state evaluation result of the battery cell to be tested by using the energy storage battery cell evaluation model.

[0051] In this embodiment, during the operation cycle to be tested of the energy storage power station, the collected online real-time data can be input into the constructed energy storage battery evaluation model, and the battery operation status and abnormal trend can be analyzed and obtained through the evaluation model, so as to classify and predict the abnormal degree of the battery cell and remotely monitor it; for example, the real-time section data related to the battery cell to be tested is processed, and multiple feature data vectors are constructed to form a comprehensive feature vector matrix, which is input into the trained model, and then the state evaluation result of the battery cell to be tested is obtained through the model output. Of course, the form of the state evaluation result of this embodiment includes but is not limited to the battery cell state classification result, state evaluation score, abnormal probability, etc., which is not limited here.

[0052] It can be seen that, through the battery cell status assessment model of the embodiment of the present application, first, it is possible to perform targeted feature processing on the battery cell data, more accurately reflect the state change process inside the battery cell, and improve the accuracy of abnormal battery cell identification; second, there is no need to unpack and shut down the energy storage equipment, which not only improves the efficiency of operation and maintenance inspections, but also increases the trouble-free operation time and economic benefits of the energy storage power station; third, it is possible to judge battery abnormalities in a non-invasive manner without adding additional sensors, which will not damage the original structure of the battery and reduce the cost of battery cell status monitoring; fourth, the battery cell assessment results obtained by the embodiment of the present application can guide relevant technical personnel to execute corresponding operation and maintenance decisions, such as timely replacement or repair of potential faulty batteries, further improving the overall performance of the energy storage system.

[0053] In some embodiments of the present application, based on the historical operation data of the abnormal battery cells, multiple feature data vectors of the abnormal battery cells are calculated and a comprehensive feature vector matrix of the abnormal battery cells is formed, including: based on the historical operation data of the abnormal battery cells, the abnormal operation data of the abnormal battery cells in the complete charge and discharge process within an evaluation period are screened out; based on the abnormal operation data in the complete charge and discharge process, multiple feature data vectors of the abnormal battery cells are calculated, and a comprehensive feature vector matrix between the classification results of the abnormal battery cells and the feature data vectors is formed.

[0054] In this embodiment, the complete charging and discharging process refers to a complete process including a pair of adjacent charging segments and discharging segments. In the evaluation cycle or operation cycle, the complete charging and discharging process may include multiple ones. Specifically, the charging segment is the process defined by the charging start signal reported by the battery management system BMS as the start and the charging termination signal as the end, and the discharging segment is the process defined by the discharge start signal reported by the battery management system BMS as the start and the discharge termination signal as the end. There is a standby process (for example, a standby process of more than 30 minutes) between the adjacent pair of charging segments and discharging segments (the charging segment is before the discharging segment), and there is also a standby process (for example, a standby process of more than 30 minutes) after the end of the discharging segment; the process that meets the above description is defined as a complete charging and discharging process; of course, the above standby process is not necessary.

[0055] Therefore, during the evaluation cycle, it is necessary to filter out the data in the complete charging and discharging process from the acquired historical operation data of the abnormal battery cells as valid data, and thereby calculate multiple feature data vectors of the abnormal battery cells, associate the multiple feature data vectors of each abnormal battery cell with its corresponding classification results, and form a comprehensive feature vector matrix of the battery.

[0056] In some embodiments, the filtering out of the abnormal battery cell historical operation data includes: obtaining the available capacity and / or battery cell deformation rate of all battery cells in the energy storage power station according to the historical operation data of the energy storage power station; and filtering out the classification results of abnormal battery cells according to the available capacity and / or battery cell deformation rate of each battery cell. Preferably, the classification results of the abnormal battery cells include at least one of the following: severely abnormal battery cells, generally abnormal battery cells, and slightly abnormal battery cells.

[0057] In one example, after obtaining the historical operation data of the energy storage power station, the cell status can be classified according to the available capacity and / or cell deformation rate in the historical operation data of all cells; for example, the corresponding data of the cells with an actual detected available capacity lower than 70% and a cell deformation rate greater than 10% are defined as severely abnormal cell data; the corresponding data of the cells with an actual detected available capacity lower than 80% but higher than 70% and no obvious deformation are defined as generally abnormal cells; the corresponding data of the cells with an actual detected available capacity lower than 85% but higher than 80% are defined as slightly abnormal cells. In this way, multiple feature data vectors of abnormal cells can be associated with their corresponding classification results, and the associated data can be used to construct a comprehensive feature vector matrix.

[0058] In some embodiments, the characteristic data vector of the abnormal battery cell includes at least one of the following: a voltage-related characteristic data vector, a current-related characteristic data vector, a temperature-related characteristic data vector, an internal resistance-related characteristic data vector, a range-related characteristic data vector, a charge and discharge depth-related characteristic data vector, and a historical data longitudinal comparison ranking-related characteristic data vector.

[0059] In one example, the voltage-related feature data vector of the abnormal battery cell includes at least the following dimensions:

[0060] During each complete charge and discharge process, the highest cell voltage V max充 ;

[0061] In each complete charge and discharge process, the minimum voltage of the battery cell in the discharge stage is: V min放 ;

[0062] In each complete charge and discharge process, the voltage change V of the battery cell from the moment charging is cut off to 30 minutes after charging is cut off r30s充 ;

[0063] In each complete charge and discharge process, the voltage change V from the moment the battery discharge is terminated to 30 minutes after the discharge is terminated r30s放 ;

[0064] During each complete charge and discharge process, the average voltage change rate per minute of the battery cluster to which the battery cell belongs during the charging process difv充 ;

[0065] During each complete charge and discharge process, the average voltage change rate per minute of the battery cluster to which the battery cell belongs during the discharge process difv放 ;

[0066] In each complete charge and discharge process, the difference between the voltage change per minute from the moment the cell is charged to 30 minutes after the charging is cut off and the average voltage change per minute of the battery cluster to which the cell belongs in the same cycle is dif (Rate V30充,Rate V30avg充 );

[0067] In each complete charge and discharge process, the difference between the voltage change per minute from the moment the cell discharge is terminated to 30 minutes after the discharge is terminated and the average voltage change per minute of the battery cluster to which the cell belongs in the same cycle is dif (Rate V30放 ,Rate V30avg放 ).

[0068] In one example, the temperature-related characteristic data vector of the abnormal battery cell includes at least the following dimensions:

[0069] During each complete charge and discharge process, the temperature change value T of the battery cell per minute in the charging stage r充 ;

[0070] During each complete charge and discharge process, the temperature change value T of the battery cell per minute in the discharge stage r放 ;

[0071] During each complete charge and discharge process, the average cell temperature T avg充 ;

[0072] During each complete charge and discharge process, the average cell temperature T avg放 ;

[0073] In one example, the internal resistance-related characteristic data vector of the abnormal battery cell includes at least the following dimensions:

[0074] In each complete charging and discharging process, the maximum equivalent internal resistance R of the battery cell in the charging stage max充 ;

[0075] In each complete charge and discharge process, the maximum equivalent internal resistance R of the battery cell in the discharge stage max放 ;

[0076] In each complete charging and discharging process, the minimum equivalent internal resistance of the battery cell in the charging stage is R min充 ;

[0077] In each complete charge and discharge process, the minimum equivalent internal resistance of the battery cell in the discharge stage is R min放 ;

[0078] The equivalent internal resistance of this embodiment is calculated according to the ratio of the voltage change value to the current change value within a sampling period, and the sampling period is preferably 1 minute.

[0079] In one example, the current-related characteristic data vector of the abnormal cell includes at least the following dimensions:

[0080] In each complete charging and discharging process, the average current rate of the battery in the charging section is I r充 ;

[0081] In each complete charge and discharge process, the average current rate of the battery in the discharge segment is I r放 .

[0082] In one example, the extreme difference-related feature data vector of the abnormal battery cell includes at least the following dimensions:

[0083] In each complete charging and discharging process, the ratio of the cell SOC range difference to the voltage range difference in the charging section

[0084] In each complete charge and discharge process, the ratio of the cell SOC extreme difference to the voltage extreme difference in the discharge segment

[0085] In one example, the characteristic data vector related to the charge and discharge depth of the abnormal battery cell includes at least the following dimensions: the average value DOD of the charge and discharge depth of the battery cell during the last 10 complete charge and discharge processes last10avg ; The charge and discharge depth can be related to the SOC operating range set for the battery cell, and the range difference is expressed as a percentage.

[0086] In one example, the feature data vector related to the longitudinal comparison ranking of historical data of abnormal cells includes at least the following dimensions:

[0087] During each complete charge and discharge process, the battery cluster to which the battery cell belongs ranks by the highest voltage percentage of all the batteries in the charging process. maxV充 For example, the highest voltages of all cells in the battery cluster to which the cell belongs during the charging process are ranked from large to small, and the ranking of the cell in the ranking is obtained and displayed as a percentage (such as the ranking of the cell in a few percent);

[0088] During each complete charge and discharge process, the battery cluster to which the battery cell belongs ranks the percentage of the highest voltage of all the batteries in the discharge process. maxV放 , for example, ranking the highest voltages of all cells in the battery cluster to which the cell belongs during the discharge process from large to small, and obtaining the ranking of the cell in the ranking and displaying it as a percentage;

[0089] In each complete charge and discharge process, the battery cluster to which the cell belongs ranks the percentage of the lowest voltage of all cells in the charging process. minV充 For example, the lowest voltages of all cells in the battery cluster to which the cell belongs in the charging process are ranked from small to large, and the ranking of the cell in the ranking is obtained and displayed as a percentage;

[0090] In each complete charge and discharge process, the battery cluster to which the battery cell belongs ranks the percentage of the lowest voltage of all the batteries in the discharge process. minV放For example, the lowest voltages of all cells in the battery cluster to which the cell belongs in the discharge process are ranked from small to large, and the ranking of the cell in the ranking is obtained and displayed as a percentage;

[0091] In each complete charge and discharge process, the ranking percentage P of the voltage change of all cells in the battery cluster to which the cell belongs from the moment of charge cutoff to 30 minutes after charge cutoff Vrate30充 For example, from the moment of charging cutoff to 30 minutes after charging cutoff, the voltage changes of all cells in the battery cluster to which the cell belongs are sorted from large to small, and the ranking of the cell in the ranking is obtained and displayed as a percentage;

[0092] In each complete charge and discharge process, the ranking percentage P of the voltage change of all cells in the battery cluster to which the cell belongs from the moment of discharge cutoff to 30 minutes after discharge cutoff Vrate30放 For example, from the moment of discharge cut-off to 30 minutes after discharge cut-off, the voltage changes of all cells in the battery cluster to which the cell belongs are sorted from small to large, and the ranking of the cell in the ranking is obtained and displayed as a percentage;

[0093] Preferably, the feature data vector related to the vertical comparison ranking of the historical data of abnormal cells also includes: the number of times the above ranking percentages are less than 10% in all complete charging and discharging processes within the evaluation period; specifically, it may include the following dimensions:

[0094] During the evaluation period, P maxV充 Less than 10% of the time D RPmav充 ;

[0095] During the evaluation period, P maxV放 Less than 10% of the time D RPmav放 ;

[0096] During the evaluation period, P minV充 Less than 10% of the time D RPmiv充 ;

[0097] During the evaluation period, P minV放 Less than 10% of the time D RPmiv放 ;

[0098] During the evaluation period, P Vrate30充 Less than 10% of the time D RPvr30充 ;

[0099] During the evaluation period, P Vrate30放 Less than 10% of the time D RPvr30放 .

[0100] Thus, the above-mentioned multiple characteristic data vectors are merged, and then an energy storage cell evaluation model can be established according to the comprehensive characteristic vector matrix; of course, the above-mentioned meaning, dimension, quantity, etc. of the characteristic data vectors are only exemplary embodiments and cannot be understood as limitations on the present application.

[0101] In some embodiments, establishing an energy storage cell evaluation model based on the comprehensive feature vector matrix includes: using a gradient boosting decision tree algorithm to establish an energy storage cell evaluation model; wherein the model is obtained through machine learning training using multiple sets of training data, the input of the model is the feature data vector corresponding to each abnormal cell in the comprehensive feature vector matrix, and the output of the model is the classification result of the abnormal cell.

[0102] In one example, the feature data vectors calculated above are merged to form a multi-dimensional comprehensive feature vector, and the comprehensive feature vector of each abnormal battery cell is used to construct an algorithm model training set, wherein the feature data structure of the model input is defined as X, and the battery cell abnormality classification label output by the model is defined as Y, as shown below:

[0103]

[0104] Each row in the data structure X represents the comprehensive battery characteristic data corresponding to a battery cell, and the subscript c 0 Indicates the first cell, c n represents the Nth battery cell. Similarly, the abnormal classification label Y Indicates the status label corresponding to the first battery cell. Indicates the status label corresponding to the Nth battery cell. Of course, this application does not limit the number of battery cells.

[0105] Preferably, the above data is used as a training set, that is, X is used as the training input and Y is used as the training output, and then the energy storage cell evaluation model is trained in combination with the gradient boosting decision tree algorithm, and the state classification evaluation of the cell to be tested is completed through the trained model. Of course, this application does not impose specific restrictions on the algorithm for constructing the model.

[0106] In a preferred embodiment, before calculating the multiple characteristic data vectors of the abnormal battery cell, the method further includes: preprocessing the historical operation data of the screened abnormal battery cell; wherein the preprocessing includes: filtering and cleaning the historical operation data that exceeds the rated operation range, and performing null value interpolation on the filtered and cleaned historical operation data. In this way, the characteristic vector of the filtered and interpolated data can be calculated, and a comprehensive characteristic vector matrix can be constructed based on this to establish an energy storage battery cell evaluation model, which will not be described in detail here.

[0107] To facilitate understanding of the technical solution of this application, a specific application example is provided below:

[0108] Take the battery cell operation data of a user-side energy storage power station in a certain place in October 2024 as an example. Comprehensively consider the temperature, current, voltage, charge and discharge depth, state of charge (SOC), extreme difference, internal resistance, vertical comparison ranking of historical operation data and other relevant factors in the above operation data of the energy storage power station during the evaluation period, calculate the multi-dimensional feature data vector of the abnormal battery cells in the power station, and build an energy storage battery cell evaluation model for the power station; then, during real-time operation, use the model to judge the abnormal state of the battery cell to be tested. The following is an example of the specific process of extracting the feature data vector, building the model, and evaluating the output results of the power station:

[0109] According to the V max充 ,V min放 ,V r30s充 ,V r30s放 ,Rate difv充 ,Rate difv放 ,dif(Rate V30充 ,Rate V30avg充 ),dif(Rate V30放 ,Rate V30avg放 ) and other data dimensions, construct the temperature-related characteristic data vector, and the calculation result of the vector is: [3.56, 3.01, 0.13, 0.07, 1.7, 2.6, 2.4, 1.1];

[0110] According to T in the above embodiment r充 , T r放 , T avg充 , T avg放 Equal data dimensions, construct temperature-related characteristic data vector, and the calculation result of the vector is: [32.8, 34.3, 27.2, 28.3];

[0111] According to the R max充 , R max放 , R min充 , R min放 Equal data dimensions, construct the internal resistance related characteristic data vector, and the calculation result of the vector is: [0.043, 0.039, 0.017, 0.023];

[0112] According to the above embodiment I r充 ,I r放 Equal data dimensions, construct the current related characteristic data vector, and the calculation result of the vector is: [0.5, 0.53];

[0113] According to the above-mentioned embodiment Equal data dimensions, construct the range-related characteristic data vector, and the calculation result of the vector is: [35.9, 38.1];

[0114] According to the DOD in the above embodiment last10avg Construct the charge and discharge depth related characteristic data vector, and the calculation result of the vector is: [72.6];

[0115] According to the aforementioned embodiment, maxV充 , P maxV放 , P minV充 , P minV放 , P Vrate30充 , P Vrate30放 , D RPmav充 , D RPmav放 , D RPmiv充 , D RPmiv放 , D RPvr30充 , D RPvr30放 Equal dimensions, construct the historical data longitudinal comparison ranking related feature data vector, and the calculation result of the vector is: [0.0046, 0.0046, 0.1069, 0.1070, 0.0132, 0.0326, 32, 27, 20, 24, 17, 11];

[0116] The voltage-related characteristic data vector, temperature-related characteristic data vector, current-related characteristic data vector, range-related characteristic data vector, charge-discharge depth-related characteristic data vector, internal resistance-related characteristic data vector, and historical data longitudinal comparison ranking-related characteristic data vector obtained by the above calculations are combined to obtain the calculation result of the comprehensive characteristic vector matrix:

[0117] [V max充 ,V min放 ,V r30s充 ,V r30s放 ,Rate difv充 ,Rate difv放 ,dif(Rate V30充 ,Rate V30avg充 ),dif(Rate V30放 ,Rate V30avg放 ), T r充 , T r放 , T avg充 , T avg放 , R max充 , R max放 , R min充 , R min放 , I r充 , I r放 , DOD last10avg , P maxV充 , P maxV放 , PminV充 , P minV放 , P Vrate30充 , P Vrate30放 , D RPmav充 , D RPmav放 , D RPmiv充 , D RPmiv放 , D RPvr30充 , D RPvr30放 ]=[3.56,3.01,0.13,0.07,1.7,2.6,2.4,1.1,32.8,34.3,27.2,28.3,0.043,0.039,0.017,0.023,0.5,0.53,35.9,38.1,72.6,0.0046,0.0046,0.1069,0.1070,0.0132,0.0326,32,27,20,24,17,11];

[0118] The eigenvectors in the above comprehensive eigenvector matrix are input into the constructed energy storage cell evaluation model, and the evaluation and classification results of the abnormal state of the cell to be tested are output through the model, as shown in Table 1:

[0119] Table 1 Evaluation results of the battery cell status of a user-side energy storage power station in a certain place

[0120]

[0121]

[0122] Taking the A-phase 14-unit pack5, cluster 7, cell 103 and the A-phase 14-unit pack8, cluster 21, cell 109 in the evaluation results as examples for further verification, the operating data of all cells in the A-phase 14-unit during a certain charge and discharge process in the evaluation cycle were checked. It was found that the voltage operating data of the above two cells were greatly inconsistent with other cells in the same unit (average voltage of the unit cells), as shown in the following figure. Figure 2 As shown in the figure, A.BC14.P08.C21-109U represents the 21st cluster 109th cell of the A phase 14 units pack8, and A.BC14.P05.C07-103.U represents the 7th cluster 103th cell of the A phase 14 units pack5;

[0123] Combination Figure 2 It can be seen that the state analysis curves of the two cells are significantly higher than the average voltage of the cell, which means that the two cells will cause the charging to be terminated early, making it impossible for other cells to continue charging. The actual available capacity of the cell is seriously attenuated, which is 73% of the nominal capacity. According to the evaluation and classification results obtained by the model, the relevant technicians inspected the A-phase 14-cell pack5 and pack8 batteries and found that the above batteries were obviously bulging and the shell was cracked.

[0124] It can be seen that the cell status assessment results obtained by the model in this example are consistent with the inspection results, and then after the battery pack to which the cell belongs is replaced as a whole, the available capacity of the unit is restored to more than 85% of the nominal value. In addition, the model assessment results are compared with the inspection verification of the remaining cells. The results show that: through the evaluation method of the embodiment of the present application, the recall rate is 80% and the accuracy rate is 100% when troubleshooting severely abnormal cells; when troubleshooting general abnormal cells, the recall rate reaches 100% and the accuracy rate is 93%. At the same time, no slightly abnormal cells were found. It can be seen that the evaluation method of the embodiment of the present application can guide relevant technical personnel to execute subsequent operation and maintenance decisions, greatly improving the work efficiency of troubleshooting.

[0125] In some embodiments of the present application, Figure 3 As shown, a battery cell state evaluation device 300 is proposed, the device comprising:

[0126] The data screening unit 310 is used to obtain the historical operation data of the energy storage power station and screen out the abnormal battery cell historical operation data;

[0127] A feature vector calculation unit 320 is used to calculate the feature data vector of the abnormal cell according to the historical operation data of the abnormal cell, and form a comprehensive feature vector matrix of the abnormal cell;

[0128] A model building unit 330, used to build an energy storage cell evaluation model according to the comprehensive feature vector matrix;

[0129] The battery cell evaluation unit 340 is used to predict and obtain a state evaluation result of the battery cell to be tested by using the energy storage battery cell evaluation model.

[0130] It can be understood that each step of the aforementioned cell state evaluation method can be executed by the cell state evaluation device provided in this embodiment. Therefore, the relevant explanations about the cell state evaluation method are applicable to the cell state evaluation device and will not be repeated here.

[0131] In summary, the present application has achieved at least the following technical effects: on the one hand, it is possible to perform targeted feature processing on the battery cell data, more accurately reflect the state change process inside the battery cell, and improve the accuracy of abnormal battery cell identification; on the other hand, there is no need to unpack and shut down the energy storage equipment, which improves the efficiency of operation and maintenance inspections while also increasing the trouble-free operation time and economic benefits of the energy storage power station; at the same time, it is possible to determine battery abnormalities in a non-invasive manner without adding additional sensors, which will not damage the original structure of the battery and reduce the cost of monitoring the battery cell status; at the same time, the battery cell evaluation results obtained using the embodiments of the present application can guide relevant technical personnel to execute subsequent operation and maintenance decisions, further improving the operation and maintenance troubleshooting efficiency and overall performance of the energy storage system.

[0132] Figure 4 This is a schematic diagram of the structure of an electronic device in an embodiment of the present application. Figure 4 At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. Of course, the electronic device may also include hardware required for other services.

[0133] The processor, network interface and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0134] The memory is used to store the program. Specifically, the program may include a program code, and the program code includes a computer operation instruction. The memory may include a memory and a non-volatile memory, and provides instructions and data to the processor.

[0135] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a battery cell status evaluation device at the logical level. The processor executes the program stored in the memory and is specifically used to perform the following operations:

[0136] Obtain historical operation data of energy storage power stations and filter out abnormal battery cell historical operation data;

[0137] Calculate the characteristic data vector of the abnormal cell according to the historical operation data of the abnormal cell, and form a comprehensive characteristic vector matrix of the abnormal cell;

[0138] Establishing an energy storage cell evaluation model according to the comprehensive eigenvector matrix;

[0139] The energy storage cell evaluation model is used to predict the state evaluation result of the cell to be tested.

[0140] The above application Figure 1 The cell state assessment method disclosed in the illustrated embodiment can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or an instruction in software form. The above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present application can be directly embodied as being executed by a hardware decoding processor, or executed by a combination of hardware and software modules in a decoding processor. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0141] The present application also provides a computer program product, which stores one or more programs, wherein the one or more programs include instructions, which, when executed by an electronic device including multiple application programs, enable the electronic device to execute Figure 1 The cell status assessment method in the illustrated embodiment is specifically used to perform:

[0142] Obtain historical operation data of energy storage power stations and filter out abnormal battery cell historical operation data;

[0143] Calculate the characteristic data vector of the abnormal cell according to the historical operation data of the abnormal cell, and form a comprehensive characteristic vector matrix of the abnormal cell;

[0144] Establishing an energy storage cell evaluation model according to the comprehensive eigenvector matrix;

[0145] The energy storage cell evaluation model is used to predict the state evaluation result of the cell to be tested.

[0146] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0147] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0148] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0149] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0150] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0151] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0152] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0153] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0154] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0155] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for evaluating a battery cell state, characterized in that: The method comprises: Obtain historical operation data of energy storage power stations and filter out abnormal battery cell historical operation data; Calculate the characteristic data vector of the abnormal cell according to the historical operation data of the abnormal cell, and form a comprehensive characteristic vector matrix of the abnormal cell; Establishing an energy storage cell evaluation model according to the comprehensive eigenvector matrix; The energy storage cell evaluation model is used to predict the state evaluation result of the cell to be tested.

2. The evaluation method according to claim 1, characterized in that: According to the historical operation data of the abnormal battery cell, a plurality of characteristic data vectors of the abnormal battery cell are calculated and a comprehensive characteristic vector matrix of the abnormal battery cell is formed, including: According to the historical operation data of the abnormal battery cell, the abnormal operation data of the abnormal battery cell during the complete charge and discharge process within the evaluation cycle is screened out; According to the abnormal operation data in the complete charging and discharging process, a plurality of characteristic data vectors of the abnormal battery cell are calculated, and a comprehensive characteristic vector matrix between the classification results of the abnormal battery cell and the characteristic data vector is formed.

3. The evaluation method according to claim 2, characterized in that: The characteristic data vector of the abnormal cell includes at least one of the following: Voltage-related characteristic data vector, current-related characteristic data vector, temperature-related characteristic data vector, internal resistance-related characteristic data vector, range-related characteristic data vector, charge and discharge depth-related characteristic data vector, and historical data longitudinal comparison ranking-related characteristic data vector.

4. The evaluation method according to claim 2, characterized in that: The screening of abnormal battery cell historical operation data includes: According to the historical operation data of the energy storage power station, the available capacity and / or the deformation rate of the battery cells of all the battery cells in the energy storage power station are obtained respectively; According to the available capacity of each battery cell and / or the battery cell deformation rate, the classification results of abnormal batteries are screened and obtained.

5. The evaluation method according to claim 4, characterized in that: The classification result of the abnormal battery cell includes at least one of the following: a seriously abnormal battery cell, a generally abnormal battery cell, and a slightly abnormal battery cell.

6. The evaluation method according to claim 3, characterized in that: The step of establishing an energy storage cell evaluation model according to the comprehensive eigenvector matrix includes: The gradient boosting decision tree algorithm is used to establish an energy storage cell evaluation model; The model is obtained through machine learning training using multiple sets of training data, the input of the model is the feature data vector corresponding to each abnormal battery cell in the comprehensive feature vector matrix, and the output of the model is the classification result of the abnormal battery cell.

7. The evaluation method according to claim 1, characterized in that: Before calculating the plurality of characteristic data vectors of the abnormal battery cell, the method further includes: Pre-process the historical operation data of the screened abnormal cells; The preprocessing includes filtering and cleaning the historical operation data that exceeds the rated operation range, and performing null value interpolation on the filtered and cleaned historical operation data.

8. A battery cell status assessment device, characterized in that: The device comprises: A data screening unit is used to obtain the historical operation data of the energy storage power station and screen out the historical operation data of abnormal cells; A feature vector calculation unit, used to calculate the feature data vector of the abnormal battery cell according to the historical operation data of the abnormal battery cell, and form a comprehensive feature vector matrix of the abnormal battery cell; A model building unit, used to build an energy storage cell evaluation model according to the comprehensive eigenvector matrix; The battery cell evaluation unit is used to predict the state evaluation result of the battery cell to be tested through the energy storage battery cell evaluation model.

9. An electronic device, characterized in that: The electronic device comprises: a processor; and a memory arranged to store computer executable instructions, wherein when the executable instructions are executed, the processor performs the evaluation method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs, which, when executed by a processor, implement the evaluation method according to any one of claims 1 to 7.

Images