Battery pack consistency evaluation method and device and electronic equipment

Abstract

The invention discloses a battery pack consistency evaluation method and device and electronic equipment, and belongs to the technical field of lithium batteries, and the method comprises the steps: extracting a plurality of voltage characteristics of a charging and discharging voltage change curve segment of a single cell in a battery pack, and a plurality of temperature characteristics of a temperature change curve segment; adding category labels for the curve segments according to the performance of the single battery cells; training a neural network model according to the voltage change curve segments, the voltage features, the temperature change curve segments, the temperature features and the category labels to obtain a category recognition model; and identifying the category of each single cell in a target battery pack based on the category identification model to obtain the category of the single cell, and evaluating the consistency of the target battery pack according to the category of the single cell. According to the invention, through feature extraction and analysis of the voltage change curve and the temperature change curve, the type of each battery monomer is determined, and rapid and accurate battery pack consistency evaluation is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of lithium batteries, and particularly to a method, device, and electronic device for evaluating the consistency of a battery pack. Background Art

[0002] With the introduction of advanced technologies such as artificial intelligence and machine vision, the accuracy and efficiency of battery sorting will be further improved, providing strong support for the sustainable development of the battery industry. There are initial differences and parameter differences during use among battery cells, and these differences may stem from multiple links such as the design, manufacturing, storage, and use of the batteries. For example, inconsistencies in battery capacity, voltage, and internal resistance will all have a significant impact on the performance of the battery pack. The consistency of the battery pack is crucial for its service life, performance, and safety. A battery pack with precisely sorted cells can provide a more stable and reliable power output and reduce the risk of unexpected failures.

[0003] In the prior art, various parameters of single cells, such as capacity, voltage, and internal resistance, are measured and analyzed to evaluate the consistency of the battery pack. However, this evaluation method requires continuous measurement and calculation, and the evaluation result lacks accuracy due to excessive human intervention. In addition, by comparing the time length of the charging curve and the discharging curve of the battery pack to judge the consistency of the battery pack, the evaluation result also lacks accuracy.

[0004] Therefore, there is an urgent need for a method for evaluating the consistency of a battery pack that can solve the technical problem of inaccurate evaluation of the consistency of the battery pack in the prior art. Summary of the Invention

[0005] In view of this, it is necessary to provide a method, device, and electronic device for evaluating the consistency of a battery pack, which can extract and analyze the characteristics of the voltage change curve and the temperature change curve during the charging and discharging processes of each single cell in the battery pack, determine the types of each battery cell, and thereby evaluate the consistency of the battery pack, achieving a fast and accurate evaluation of the consistency of the battery pack.

[0006] To solve the above technical problems, on the one hand, the present invention provides a method for evaluating the consistency of a battery pack, including: Extracting multiple voltage characteristics of the voltage change curve segment of the single cell in the battery pack during charging and discharging, and multiple temperature characteristics of the temperature change curve segment in the corresponding time domain interval; Adding category labels to the voltage change curve segment and the temperature change curve segment according to the performance of the single cell; Training a neural network model based on the voltage change curve segment and voltage characteristics, temperature change curve segment and temperature characteristics, and category labels to obtain a category recognition model; Identify the categories of individual battery cells in the target battery pack based on the category recognition model to obtain the categories of individual battery cells, and evaluate the consistency of the target battery pack according to the categories of individual battery cells.

[0007] In a possible implementation, extract multiple voltage features of the charge and discharge voltage change curve segments of the individual battery cells in the battery pack, and multiple temperature features of the temperature change curve segments in their corresponding time domain intervals, including: Obtain the voltage change curve and temperature change curve of each individual battery cell during the constant current charge and discharge process in the battery pack; Divide the voltage change curve and temperature change curve according to the preset extraction period and extraction interval in terms of time domain features to obtain multiple voltage change curve segments and temperature change curve segments, where the time domain intervals of the voltage change curve segments and temperature change curve segments are the same; Extract multiple voltage features from the voltage change curve segments; Extract multiple temperature features from the temperature change curve segments.

[0008] In a possible implementation, the voltage features include maximum voltage, minimum voltage, average voltage, voltage difference, voltage change slope, and voltage volatility.

[0009] In a possible implementation, the temperature features include maximum temperature, minimum temperature, average temperature, temperature difference, temperature change slope, and temperature volatility.

[0010] In a possible implementation, add category labels to the voltage change curve segments and temperature change curve segments according to the performance of the individual battery cells, including: Obtain the static voltage, internal resistance, and battery capacity of the individual battery cells in the battery pack; Evaluate the performance of the individual battery cells according to the static voltage, internal resistance, and battery capacity of the individual battery cells; Determine the category of the individual battery cell according to the performance of the individual battery cell and the preset evaluation criteria; Add category labels to the voltage change curve segments and temperature change curve segments corresponding to the individual battery cells according to the category of the individual battery cell.

[0011] In a possible implementation, train a neural network model according to the voltage change curve segments and voltage features, temperature change curve segments and temperature features, and category labels to obtain a category recognition model, including: Based on the neural network, construct an initial recognition model with the voltage features of the voltage change curve of the individual battery cell charge and discharge and the temperature features of the temperature change curve as the input and the category of the individual battery cell as the output; Iteratively train the neural network model using the voltage change curve segments, voltage characteristics, temperature change curve segments, temperature characteristics, and class labels as the training dataset; Based on a preset loss function, calculate the loss function value between the output result of the initial recognition model and the class label, and optimize the initial recognition model according to the loss function value to obtain a battery cell class recognition model.

[0012] In a possible implementation manner, the initial recognition model includes an input layer, an LSTM layer for the voltage branch, an LSTM layer for the temperature branch, and a fully connected layer, including: The input layer is used to receive the voltage characteristics of the voltage change curve of the single battery cell during charge and discharge and the temperature characteristics of the temperature change curve; The LSTM layer of the voltage branch is used to extract the voltage time series relationship characteristics of the voltage characteristics; The LSTM layer of the temperature branch is used to extract the temperature time series relationship characteristics of the temperature characteristics; The fully connected layer is used to introduce a cross-modal attention mechanism, calculate the correlation between the voltage time series relationship characteristics and the temperature time series relationship characteristics, determine the association weights of the voltage branch and the temperature branch according to the correlation, and perform weighted mapping on the voltage time series relationship characteristics and the temperature time series relationship characteristics according to the association weights to obtain the category recognition result.

[0013] In a possible implementation manner, evaluating the consistency of the target battery pack according to the categories of the individual battery cells includes: Determine the categories of the individual battery cells in the target battery pack, and count the number of individual battery cells in each category; Evaluate the consistency of the target battery pack based on a preset category threshold and the number of individual battery cells in each category.

[0014] In a second aspect, the present invention also provides a battery pack consistency evaluation device, including: A feature extraction module, configured to extract multiple voltage characteristics of the voltage change curve segments of the individual battery cells in the battery pack during charge and discharge, and multiple temperature characteristics of the corresponding temperature change curve segments in the time domain interval; A label adding module, configured to add class labels to the voltage change curve segments and temperature change curve segments according to the performance of the individual battery cells; A model construction module, configured to train a neural network model according to the voltage change curve segments, voltage characteristics, temperature change curve segments, temperature characteristics, and class labels to obtain a category recognition model; A consistency evaluation module, configured to identify the categories of the individual battery cells in the target battery pack based on the category recognition model to obtain the individual battery cell categories, and evaluate the consistency of the target battery pack according to the individual battery cell categories.

[0015] In a third aspect, the present invention further provides an electronic device, including a memory and a processor. Among them, the memory is used for storing programs and data; the processor is coupled to the memory and is used for executing the programs stored in the memory to implement the battery pack consistency evaluation method as described above, and / or to implement the battery pack consistency evaluation as described above.

[0016] The beneficial effects of the present invention are as follows: First, the voltage change curves and temperature change curves during the charging and discharging processes of individual single cells in the battery pack are cut to obtain a plurality of voltage change curve segments and temperature change curve segments. Then, feature extraction is performed on the cut voltage change curve segments and temperature change curve segments to obtain a plurality of voltage features and a plurality of temperature features. The cutting of the voltage change curve and temperature change curve can improve the accuracy of reflecting the curve features of voltage and temperature while providing more data. Then, category labels are added to the voltage change curve segments and temperature change curve segments according to the actual performance of the battery cells. Then, the voltage change curve segments and voltage features, temperature change curve segments and temperature features, and category labels are used as a training data set to train a neural network model to obtain a category recognition model. Finally, according to the voltage change curves and temperature change curves of each single cell in the target battery pack, the categories of each single cell in the target battery pack are identified, and the consistency of the target battery pack is evaluated according to the categories of each single cell. The present invention extracts voltage features and temperature features from each segment of the voltage change curve and temperature change curve during the charging and discharging processes of each single cell in the battery pack, uses these voltage features and temperature features as input labels of input data, uses the performance types of single cells as output labels, iteratively trains the neural network model to obtain a category recognition model, and evaluates the consistency of the battery pack according to the types of each battery cell identified by the category recognition model, realizing fast and accurate evaluation of the battery pack consistency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those skilled in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0018] Figure 1 It is a schematic flowchart of an embodiment of the battery pack consistency evaluation method provided by the present invention; Figure 2 For the present invention Figure 1 It is a schematic flowchart of an embodiment of step S101 in the present invention; Figure 3Schematic diagram of the voltage change curve of various types of single - cell batteries provided by the present invention during charge and discharge; Figure 4 Schematic diagram of the temperature change curve of various types of single - cell batteries provided by the present invention during charge and discharge; Figure 5 Schematic diagram of a fragment of the voltage change curve of various types of single - cell batteries provided by the present invention; Figure 6 Schematic diagram of a fragment of the temperature change curve of various types of single - cell batteries provided by the present invention; Figure 7 For the present invention Figure 1 Flow schematic diagram of an embodiment of step S102 in the present invention; Figure 8 For the present invention Figure 1 Flow schematic diagram of an embodiment of step S103 in the present invention; Figure 9 For the present invention Figure 1 Flow schematic diagram of an embodiment of step S104 in the present invention; Figure 10 Schematic diagram of the structure of an embodiment of the battery pack consistency evaluation electronic device provided by the present invention; Figure 11 Schematic diagram of the structure of an embodiment of the battery pack consistency evaluation electronic equipment provided by the present invention. Detailed implementation manners

[0019] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present invention.

[0020] In the description of the embodiments of the present invention, unless otherwise specified, the meaning of "a plurality" is two or more.

[0021] In the embodiments of the present invention, the descriptions such as "first", "second", etc. are only for descriptive purposes, and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Therefore, the technical features defined with "first" and "second" may explicitly or implicitly include at least one of such features.

[0022] Reference to "embodiments" in this document means that the specific features, structures, or characteristics described in connection with the embodiments can be included in at least one embodiment of the present invention. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment each time, nor are they independent or alternative embodiments mutually exclusive of other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0023] The present invention provides a method, apparatus, and electronic device for evaluating the consistency of a battery pack, which will be described separately below.

[0024] Figure 1 FIG. is a schematic flowchart of an embodiment of the method for evaluating the consistency of a battery pack provided by the present invention. As Figure 1 shown, the method for evaluating the consistency of a battery pack includes: S101. Extract a plurality of voltage features of the charge-discharge voltage change curve segments of the individual battery cells in the battery pack, and a plurality of temperature features of the temperature change curve segments in the corresponding time domain intervals; It should be noted that a battery charge-discharge integrated machine is used to charge and discharge the battery pack; the battery management system BMS is used to charge and discharge the battery pack. The charge-discharge cabinet communicates with the battery pack to collect the voltage change data and temperature change data of each individual battery cell during the charge-discharge process, and transmits this data to the electronic terminal. Through data analysis software, the voltage change data and temperature change data are analyzed, and the voltage change curve and temperature change curve are plotted. Then, the voltage change curve and temperature change curve are segmented according to a preset extraction period and extraction interval to obtain a plurality of voltage change curve segments and temperature change curve segments. Among them, the electronic terminal includes but is not limited to smart phones, tablet computers, laptop computers, desktop computers, etc.

[0025] S102. Add category labels to the voltage change curve segments and temperature change curve segments according to the performance of the individual battery cells; It should be noted that the performance of the individual battery cell is determined according to the static voltage, static capacitance, and internal resistance of the battery cell. The higher the static voltage, the closer the static capacity is to the rated capacity, and the internal resistance is lower than the preset normal value, indicating that the performance of the individual battery cell is very good. If the static voltage gradually decreases, the static capacity is also lower than the rated capacity, and the internal resistance increases, indicating that the performance of the individual battery cell gradually decreases. Based on the preset evaluation criteria, the category of the individual battery cell is determined according to the performance of the individual battery cell. In this embodiment, the categories are divided into three types, including good, average, and poor.

[0026] S103. Train a neural network model according to the voltage change curve segments and voltage features, temperature change curve segments and temperature features, and category labels to obtain a category recognition model; It should be noted that a neural network model is constructed according to the model software on the electronic terminal, and the model is iteratively trained to obtain a final classification recognition model, which can judge the category of a single battery cell according to the voltage and temperature change curves of the single battery cell.

[0027] S104. Based on the classification recognition model, identify the categories of the single battery cells in the target battery pack to obtain the single battery cell categories, and evaluate the consistency of the target battery pack according to the single battery cell categories.

[0028] It should be noted that the target battery pack is charged and discharged, the voltage change curve and the temperature change curve are drawn according to the voltage change data and the temperature change data, the voltage change curve and the temperature change curve are used as the input of the classification recognition model to obtain the categories of the single battery cells in the target battery pack, and the consistency of the target battery pack is evaluated according to the categories of the single battery cells. For example, the performance types of the batteries are divided into three categories: very good consistency, relatively good consistency, and relatively poor consistency. When the categories of all the single battery cells in the battery pack are of the same type, it means that the consistency of the battery pack is very high. When all three types of single battery cells are included in the battery pack, it means that the consistency is relatively poor.

[0029] In this embodiment, by extracting the voltage characteristics and temperature characteristics of each segment in the voltage change curve and the temperature change curve during the charging and discharging process of each single battery cell in the battery pack, these voltage characteristics and temperature characteristics are used as the input labels of the input data, and the performance type of the single battery cell is used as the output label, and the neural network model is iteratively trained to obtain a classification recognition model. The consistency of the battery pack is evaluated according to the types of the battery cells in the battery pack identified by the classification recognition model, realizing the rapid and accurate evaluation of the consistency of the battery pack.

[0030] In some embodiments of the present invention, as Figure 2 shown, Figure 2 is a schematic flowchart of an embodiment of step S101 provided by the present invention, including: Figure 1 Including: S201. Obtain the voltage change curves and temperature change curves of each single battery cell during the constant current charging and discharging process of each single battery cell in the battery pack; It should be noted that at room temperature, the battery pack is discharged in a constant current manner until any single cell reaches the preset first cut-off voltage or the total voltage of the battery pack reaches the preset second cut-off voltage. Then, the battery pack is left standing for a preset time, and then charged in a constant current charging mode until the voltage of any single cell reaches the preset third voltage or the total voltage of the battery pack reaches the preset fourth voltage. Then, it is switched to a constant voltage charging mode and charged until the current is less than the preset first current value, and then the charging is stopped. Record the voltage change during this discharge and charging process, and draw a voltage change curve. Record the voltage change and temperature change during the discharge process, and draw a voltage change curve and a temperature change curve.

[0031] Furthermore, it should be noted that the categories of battery monomers in this embodiment are divided into three categories, including good, general, and poor. Obtain the voltage change curves and temperature change curves of these three types of single cells respectively. As Figure 3 and Figure 4 shown, Figure 3 is a schematic diagram of the voltage change curves of various types of single cells provided by the present invention during charge and discharge, Figure 4 is a schematic diagram of the temperature change curves of various types of single cells provided by the present invention during charge and discharge.

[0032] S202. Divide the voltage change curve and the temperature change curve according to the preset extraction period and extraction interval in terms of time domain characteristics, and obtain a plurality of voltage change curve segments and temperature change curve segments. Among them, the time domain intervals of the voltage change curve segments and the temperature change curve segments are the same; It should be noted that the extraction period and extraction interval in this embodiment are obtained through experimental verification, and the specific values are not limited. In this embodiment, the extraction time period is preferably set to 100 seconds, and the extraction interval is 20 seconds. The voltage change curve and the temperature change curve are segmented. At the same time, it is necessary to determine that the start and end time intervals corresponding to each voltage change curve segment have corresponding temperature change curve segments. Take these voltage change curve segments and temperature change curve segments with the same time domain intervals as a set of data. At this time, multiple sets of such voltage change curve segment and temperature change curve segment data can be used to train the model.

[0033] Specifically, as Figure 5 and Figure 6 shown, Figure 5 is a schematic diagram of the voltage change curve segments of various types of single cells provided by the present invention, Figure 6 is a schematic diagram of the temperature change curve segments of various types of single cells provided by the present invention. This set of voltage change curve segments and temperature change curve segments are composed of Figure 3 and Figure 4A set of curve segments obtained by dividing the voltage change curve and the temperature change curve in accordance with a preset extraction period and extraction interval.

[0034] S203. Extract multiple voltage features from the voltage change curve segment; Specifically, the voltage features include the maximum voltage, the minimum voltage, the average voltage, the voltage difference, the voltage change slope, and the voltage volatility; The maximum voltage is the maximum voltage value in the voltage change curve segment; The minimum voltage is the minimum voltage value in the voltage change curve segment; The average voltage is the average voltage value in the voltage change curve segment, and its calculation formula is: , where, is the average voltage, t is the time length of the voltage change curve segment, and u is the voltage value in the voltage change curve segment.

[0035] The voltage difference is the difference between the maximum voltage and the minimum voltage in the voltage change curve segment, and its calculation formula is: , where, is the voltage difference, is the maximum voltage, is the minimum voltage.

[0036] The voltage change slope is the rate of change of the voltage from the start time to the end time in the voltage change curve segment, and its calculation formula is: , where, is the voltage change slope, is the voltage value at the end time in the voltage change curve segment, is the voltage value at the start time in the voltage change curve segment, and t is the time length of the voltage change curve segment.

[0037] The voltage volatility is the ratio of the voltage standard deviation to the average voltage in the voltage change curve segment, and its calculation formula is: , , where, is the voltage volatility, is the voltage standard deviation, is the average voltage, N is the number of sampling points in the voltage change curve segment, is the voltage value at the i-th sampling point.

[0038] S204. Extract multiple temperature features from the temperature change curve segment.

[0039] Specifically, the temperature characteristics include the maximum temperature, the minimum temperature, the average temperature, the temperature difference, the temperature change slope, and the temperature volatility; The maximum temperature is the maximum temperature value in the curve segment; The minimum temperature is the minimum temperature value in the temperature change curve segment; The average temperature is the average temperature value in the temperature change curve segment, and its calculation formula is: , where, is the average temperature, t is the time length of the voltage change curve segment, and T is the temperature value in the temperature change curve segment.

[0040] The temperature difference is the difference between the maximum temperature and the minimum temperature in the temperature change curve segment, and its calculation formula is: , where, is the temperature difference, is the maximum temperature, is the minimum temperature.

[0041] The temperature change slope is the rate of change of temperature from the start time to the end time in the temperature change curve segment, and its calculation formula is: , where, is the temperature change slope, is the temperature value at the end time in the electrical temperature change curve segment, is the temperature value at the start time in the temperature change curve segment, and t is the time length of the temperature change curve segment.

[0042] The temperature volatility is the ratio of the temperature standard deviation to the average voltage in the temperature change curve segment, and its calculation formula is: , , where, is the temperature volatility, is the temperature standard deviation, is the average temperature, N is the number of sampling points in the temperature change curve segment, is the temperature value of the i-th sampling point.

[0043] This embodiment obtains multiple curve segments by segmenting the voltage change curve and temperature change curve during the charge and discharge process of the single battery cell, thereby increasing the training data of the model and extracting the voltage characteristics and temperature characteristics of each curve segment. Through multiple different characteristics, the characteristics of each curve in different charging stages can be fully reflected. Combined with the time domain characteristics, the output accuracy of the training model can be improved.

[0044] In some embodiments of the present invention, Figure 7 As shown, Figure 7 The present invention provides Figure 1 The flowchart of an embodiment of step S102 in the embodiment includes: S701, obtaining the static voltage, internal resistance and battery capacity of the battery pack single cell; Automated equipment is used to measure the voltage value, battery capacity and internal resistance of each single cell in the battery pack in a static state.

[0045] S702, evaluating the performance of the single cell according to the static voltage, internal resistance and battery capacity of the single cell; It should be noted that, in this embodiment, lithium iron phosphate batteries are taken as an example, with a nominal voltage of 2.5-3.65V. When the static voltage is within the normal range, the actual capacity is consistent with or close to the rated capacity, and the internal resistance is normal (lower than 30mΩ), the performance of the single battery is good; when the static voltage is slightly lower but still within the safe use range, the actual capacity is less than 10% of the rated capacity, and the internal resistance is slightly high (30mΩ - 50mΩ), the performance of the single battery is average; when the static voltage is significantly lower than the normal range, the actual capacity is more than 10% lower than the rated capacity, and the internal resistance is high (more than 50mΩ), the performance of the single battery is poor.

[0046] S703, determining the category of the single cell according to the performance of the single cell and a preset evaluation standard; It should be noted that the present embodiment divides the single cells into three categories, namely, category A: good, category B: average, and category C: poor. The category of the single cell is determined according to the performance of the single cell and its corresponding evaluation criteria.

[0047] S704 , adding category labels to the voltage change curve segments and the temperature change curve segments corresponding to the single battery cells according to the categories of the single battery cells.

[0048] It should be noted that after determining the category of a single cell, corresponding labels are added to all voltage change curve segments and temperature change curve segments of the single cell. When adding labels, ABC is used as the category labels, where A represents good, B represents average, and C represents poor.

[0049] In this embodiment, the type of the battery cell is determined by analyzing the performance of the battery cell, and category labels are added to the voltage change curve segment and the temperature change curve segment according to the type of the battery cell, providing an accurate basis for the training of the model.

[0050] In some embodiments of the present invention, as Figure 8 shown, Figure 8 provided by the present invention Figure 1 is a schematic flowchart of an embodiment of step S103, including: S801. Based on a neural network, taking the voltage characteristics of the voltage change curve of the single-cell battery during charge and discharge and the temperature characteristics of the temperature change curve as inputs, and the category of the single-cell battery as the output, an initial recognition model is constructed; It should be noted that in this embodiment, a time-domain neural network, such as an LSTM network, is used to analyze the long-term dependence relationship of the voltage characteristics of the voltage change curve and the temperature characteristics of the temperature change curve in the time domain, further improving the accuracy of the identification of the type of the single-cell battery and providing accurate data support for the consistency evaluation of the battery pack.

[0051] S802. Using the voltage change curve segment and voltage characteristics, temperature change curve segment and temperature characteristics, and category labels as the training data set to perform iterative training on the neural network model; It should be noted that in this embodiment, the DC internal resistance of the single-cell battery is also used as a feature of the curve segment together with the voltage change curve segment and voltage characteristics, temperature change curve segment and temperature characteristics, and category labels. These data are divided into a training data set and a test data set to train and verify the model.

[0052] S803. Based on a preset loss function, calculate the loss function value between the output result of the initial recognition model and the category label, and optimize the initial recognition model according to the loss function value to obtain a battery cell category recognition model.

[0053] It should be noted that the loss function in this embodiment is not specifically limited. Preferably, the cross-entropy loss function is selected. During the training process, the loss function value between the model recognition result and the true label is calculated for each iteration. According to the loss function value, the model is further optimized based on an optimization algorithm (such as SGD, ADAM, etc.). The optimization process includes the learning rate, adding regularization terms, etc., to improve the generalization ability and accuracy of the model. In this embodiment, by evaluating the recognition result of the test data set, its accuracy rate is calculated to be 99.0783%, which fully demonstrates the accuracy of the single-cell battery category recognition model.

[0054] In some embodiments of the present invention, the initial recognition model includes an input layer, an LSTM layer of the voltage branch, an LSTM layer of the temperature branch, and a fully connected layer, including: The input layer is used to receive the voltage characteristics of the voltage change curve and the temperature characteristics of the temperature change curve during the charge and discharge of a single cell. The LSTM layer of the voltage branch is used to extract the voltage time-series relationship characteristics of the voltage characteristics. The LSTM layer of the temperature branch is used to extract the temperature time-series relationship characteristics of the temperature characteristics. The fully connected layer is used to introduce a cross-modal attention mechanism, calculate the correlation between the voltage time-series relationship characteristics and the temperature time-series relationship characteristics, determine the association weights of the voltage branch and the temperature branch according to the correlation, and perform weighted mapping on the voltage time-series relationship characteristics and the temperature time-series relationship characteristics according to the association weights to obtain the category recognition result.

[0055] It should be noted that preferably, based on the LSTM neural network, an LSTM neural network model is constructed. This model includes two parallel branches, a voltage branch and a temperature branch. Each branch is composed of at least one LSTM layer, which is used to capture the long-term dependencies in time-series data. At the end of each branch, a fully connected layer is added to further extract high-level features. Before merging the outputs of the voltage branch and the temperature branch, a cross-modal attention mechanism is introduced. For the voltage characteristics and the temperature characteristics, their correlations are calculated respectively. According to the correlation, the association weights between the voltage and the temperature are obtained through a normalization method. The association weights are used to perform weighted fusion on the outputs of the voltage branch and the temperature branch to obtain a joint feature vector. Finally, the recognition result, that is, the category recognition result of the single cell, is generated through an activation function based on the joint feature vector.

[0056] In this embodiment, by constructing an LSTM model with a voltage branch and a temperature branch, and finally calculating and fusing the dynamic weights of the recognition results of the voltage branch and the temperature branch, the category of the single cell can be recognized from two different directions respectively, improving the accuracy of the model in category recognition.

[0057] In some embodiments of the present invention, as Figure 9 shown, Figure 9 is a schematic flowchart of an embodiment of step S104 provided by the present invention, including: Figure 1 S901. Determine the categories of the single cells in the target battery pack, and count the number of single cells in each category. S902. Evaluate the consistency of the target battery pack based on a preset category threshold and the number of single cells in each category. S902. Evaluate the consistency of the target battery pack based on a preset category threshold and the number of single cells in each category.

[0058] Specifically, the types of each single cell in the target battery pack are identified through the model. That is, the three types A, B, and C represent better performance, average performance, and poor performance respectively. The single cells are classified according to their types, and the quantity is counted. When the types of all single cells in the target battery pack are the same, it indicates that the consistency of the target battery pack is very good. When there are only two types of single cells in the target battery pack and these two types are connected (such as AB, BC), it indicates that the consistency of the target battery pack is average. When there are at least two types of single cells in the target battery pack and the types are not connected (such as AC), it indicates that the consistency of the target battery pack is poor.

[0059] To better implement the battery pack consistency evaluation method in the embodiments of the present invention, correspondingly, on the basis of the battery pack consistency evaluation method, as Figure 10 shown, the embodiments of the present invention further provide a battery pack consistency evaluation device 1000, including: A feature extraction module 1001, configured to extract a plurality of voltage features of the charge and discharge voltage change curve segments of the single cells in the battery pack, and a plurality of temperature features of the temperature change curve segments in the corresponding time domain intervals; A label adding module 1002, configured to add category labels to the voltage change curve segments and temperature change curve segments according to the performance of the single cells; A model building module 1003, configured to train a neural network model based on the voltage change curve segments and voltage features, temperature change curve segments and temperature features, and category labels to obtain a category recognition model; A consistency evaluation module 1004, configured to obtain the types of each single cell in the target battery pack based on the category recognition model according to the voltage change curves and temperature change curves of each single cell in the target battery pack, and evaluate the consistency of the target battery pack according to the types of the single cells.

[0060] The battery pack consistency evaluation device 1000 provided in the above embodiments can implement the technical solutions described in the embodiments of the above battery pack consistency evaluation method. The specific implementation principles of the above modules or units can be referred to the corresponding content in the embodiments of the above battery pack consistency evaluation method, and will not be elaborated here.

[0061] In the embodiments of the present invention, the battery pack consistency evaluation device can be an independent server, or a server network or server cluster composed of servers. For example, the battery pack consistency evaluation device described in the embodiments of the present invention includes, but is not limited to, a computer, a network host, a single network server, a set of multiple network servers, or a cloud server composed of multiple servers. Among them, the cloud server is composed of a large number of computers or network servers based on cloud computing (Cloud Computing).

[0062] AsFigure 11 As shown, the present invention also correspondingly provides a battery pack consistency evaluation electronic device 1100. The battery pack consistency evaluation electronic device 1100 includes a processor 1101, a memory 1102, and a display 1103. Figure 11 Only some components of the battery pack consistency evaluation electronic device 1100 are shown, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be alternatively implemented.

[0063] In some embodiments, the processor 1101 may be a central processing unit (CPU), a microprocessor, or other data processing chips, and is used to run the program code stored in the memory 1102 or process data, such as the battery pack consistency evaluation method in the present invention.

[0064] In some embodiments of the present invention, the processor 1101 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, the processor 1101 may be local or remote. In some embodiments, the processor 1101 may be implemented on a cloud platform. In one embodiment, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-cloud, etc., or any combination thereof.

[0065] In some embodiments, the memory 1102 may be an internal storage unit of the battery pack consistency evaluation electronic device 1100, such as the hard disk or memory of the battery pack consistency evaluation electronic device 1100. In some other embodiments, the memory 1102 may also be an external storage device of the battery pack consistency evaluation electronic device 1100, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the battery pack consistency evaluation electronic device 1100.

[0066] Furthermore, the memory 1102 may also include both the internal storage unit of the battery pack consistency evaluation electronic device 1100 and the external storage device. The memory 1102 is used to store the application software installed on the battery pack consistency evaluation electronic device 1100 and various types of data.

[0067] The display 1103 may be an LED display, a liquid crystal display, a touch liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. in some embodiments. The display 1103 is used to display information of the battery pack consistency evaluation electronic device 1100 and to display a visual user interface. Components 901-903 of the battery pack consistency evaluation electronic device 1100 communicate with each other via the system bus.

[0068] In some embodiments of the present invention, when the processor 1101 executes the battery pack consistency evaluation program in the memory 1102, the following steps can be implemented: Extract a plurality of voltage features of the charge and discharge voltage change curve segments of the individual battery cells in the battery pack, and a plurality of temperature features of the temperature change curve segments in their corresponding time domain intervals; Add class labels to the voltage change curve segments and temperature change curve segments according to the performance of the individual battery cells; Train a neural network model based on the voltage change curve segments and voltage features, temperature change curve segments and temperature features, and class labels to obtain a class recognition model; Based on the class recognition model, according to the voltage change curves and temperature change curves of the individual battery cells in the target battery pack, obtain the classes of the individual battery cells in the target battery pack, and evaluate the consistency of the target battery pack according to the classes of the individual battery cells.

[0069] It should be understood that when the processor 1101 executes the battery pack consistency evaluation program in the memory 1102, in addition to the above functions, other functions can also be implemented. For specific details, please refer to the description of the corresponding method embodiments above.

[0070] Furthermore, the embodiments of the present invention do not specifically limit the type of the battery pack consistency evaluation electronic device 1100 mentioned. The battery pack consistency evaluation electronic device 1100 may be a mobile phone, a tablet computer, a personal digital assistant (PDA), a wearable device, a laptop computer, etc., which are portable battery pack consistency evaluation electronic devices. Exemplary embodiments of the portable battery pack consistency evaluation electronic device include, but are not limited to, portable battery pack consistency evaluation electronic devices running IOS, Android, Microsoft, or other operating systems. The above-mentioned portable battery pack consistency evaluation electronic device may also be other portable battery pack consistency evaluation electronic devices. It should also be understood that in some other embodiments of the present invention, the battery pack consistency evaluation electronic device 1100 may not be a portable battery pack consistency evaluation electronic device, but a desktop computer with a touch-sensitive surface (such as a touch panel).

[0071] Those skilled in the art can understand that all or part of the processes of implementing the methods of the above embodiments can be completed by instructing relevant hardware (such as processors, controllers, etc.) through a computer program, and the computer program can be stored in a computer-readable storage medium. Among them, the computer-readable storage medium is a disk, an optical disk, a read-only memory, or a random access memory, etc.

[0072] The above has introduced in detail the battery pack consistency evaluation method, device, equipment and storage device provided by the present invention. Specific examples are used in this article to elaborate on the principles and implementation manners of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A method for evaluating the consistency of a battery pack, characterized in that, Including: Extracting multiple voltage features of the charge-discharge voltage change curve segments of the individual battery cells in the battery pack, and multiple temperature features of the temperature change curve segments in the corresponding time domain intervals; Adding category labels to the voltage change curve segments and temperature change curve segments according to the performance of the individual battery cells; Training a neural network model based on the voltage change curve segments and voltage features, temperature change curve segments and temperature features, and category labels to obtain a category recognition model; Identifying the categories of the individual battery cells in the target battery pack based on the category recognition model to obtain the categories of the individual battery cells, and evaluating the consistency of the target battery pack according to the categories of the individual battery cells.

2. The battery pack consistency evaluation method according to claim 1, wherein Extracting multiple voltage features of the charge-discharge voltage change curve segments of the individual battery cells in the battery pack, and multiple temperature features of the temperature change curve segments in the corresponding time domain intervals, including: Obtaining the voltage change curve and temperature change curve of each individual battery cell during the constant current charge and discharge process of the battery pack; Dividing the voltage change curve and temperature change curve according to the preset extraction period and extraction interval according to the time domain characteristics to obtain multiple voltage change curve segments and temperature change curve segments, where the time domain intervals of the voltage change curve segments and temperature change curve segments are the same; Extracting multiple voltage features from the voltage change curve segments; Extracting multiple temperature features from the temperature change curve segments.

3. The battery pack consistency evaluation method according to claim 2, characterized in that The voltage features include the maximum voltage, minimum voltage, average voltage, voltage difference, voltage change slope, and voltage volatility.

4. The battery pack consistency evaluation method according to claim 3, wherein The temperature features include the maximum temperature, minimum temperature, average temperature, temperature difference, temperature change slope, and temperature volatility.

5. The battery pack consistency evaluation method according to claim 1, wherein Adding category labels to the voltage change curve segments and temperature change curve segments according to the performance of the individual battery cells, including: Obtaining the static voltage, internal resistance, and battery capacity of the individual battery cells in the battery pack; Evaluating the performance of the individual battery cells according to the static voltage, internal resistance, and battery capacity of the individual battery cells; Determining the category of the individual battery cells according to the performance of the individual battery cells and the preset evaluation criteria; Adding category labels to the voltage change curve segments and temperature change curve segments corresponding to the individual battery cells according to the categories of the individual battery cells.

6. The battery pack consistency evaluation method according to claim 1, wherein Training a neural network model based on the voltage change curve segments and voltage features, temperature change curve segments and temperature features, and category labels to obtain a category recognition model, including: Based on the neural network, constructing an initial recognition model with the voltage features of the voltage change curve of the individual battery cell charge and discharge and the temperature features of the temperature change curve as the input and the category of the individual battery cell as the output; Performing iterative training on the neural network model using the voltage change curve segments and voltage features, temperature change curve segments and temperature features, and category labels as the training data set; Calculating the loss function value between the output result of the initial recognition model and the category label based on the preset loss function, and optimizing the initial recognition model according to the loss function value to obtain a battery cell category recognition model.

7. The battery pack consistency evaluation method according to claim 1, wherein The initial recognition model includes an input layer, an LSTM layer for the voltage branch, an LSTM layer for the temperature branch, and a fully connected layer, including: The input layer is used to receive the voltage characteristics of the voltage change curve and the temperature characteristics of the temperature change curve during the charge and discharge of a single cell; The LSTM layer of the voltage branch is used to extract the voltage time-series relationship characteristics of the voltage characteristics; The LSTM layer of the temperature branch is used to extract the temperature time-series relationship characteristics of the temperature characteristics; The fully connected layer is used to introduce a cross-modal attention mechanism, calculate the correlation between the voltage time-series relationship characteristics and the temperature time-series relationship characteristics, determine the association weights of the voltage branch and the temperature branch according to the correlation, and perform weighted mapping on the voltage time-series relationship characteristics and the temperature time-series relationship characteristics according to the association weights to obtain a category recognition result.

8. The battery pack consistency evaluation method according to claim 5, wherein Evaluating the consistency of the target battery pack according to the categories of the single cells includes: Determining the categories of the single cells in the target battery pack and counting the number of single cells in each category; Evaluating the consistency of the target battery pack based on a preset category threshold and the number of single cells in each category.

9. A battery pack consistency evaluation device, characterized in that Includes: A feature extraction module, which is used to extract multiple voltage characteristics of the voltage change curve segment of the single cell during charge and discharge in the battery pack, and multiple temperature characteristics of the temperature change curve segment in the corresponding time domain interval; A label adding module, which is used to add category labels to the voltage change curve segment and the temperature change curve segment according to the performance of the single cell; A model construction module, which is used to train a neural network model according to the voltage change curve segment and voltage characteristics, temperature change curve segment and temperature characteristics, and category labels to obtain a category recognition model; A consistency evaluation module, which is used to identify the categories of the single cells in the target battery pack based on the category recognition model to obtain the single cell categories, and evaluate the consistency of the target battery pack according to the single cell categories.

10. An electronic device for evaluating the consistency of a battery pack, characterized in that, Includes a processor and a memory, and a computer program is stored on the memory. When the computer program is executed by the processor, the steps of the battery pack consistency evaluation method according to any one of claims 1 to 8 are implemented.

Images