Power battery soc calibration method and device, storage medium and power battery

Abstract

This invention discloses a method, apparatus, storage medium, and power battery for State of Charge (SOC) calibration. The method includes: acquiring data on the battery pack's resting and non-resting processes; constructing and training a voltage drop prediction model based on the resting and non-resting process data; and performing SOC calibration on the battery pack using the trained voltage drop prediction model. Therefore, by constructing and training a voltage drop prediction model using the battery pack's resting and non-resting process data, and performing SOC calibration on the power battery based on the trained voltage drop prediction model, the resting time required for power battery SOC calibration can be significantly shortened using only existing vehicle data, without relying on additional battery testing experiments or fixed usage conditions or scenarios. This improves the trigger frequency and applicability of power battery SOC calibration in daily vehicle use.

Description

Technical Field

[0001] This invention relates to the field of battery management technology, and in particular to a power battery SOC calibration method, apparatus, storage medium, and power battery. Background Technology

[0002] The state of charge (SOC) of a battery is an important indicator of its remaining usable capacity. Since SOC estimation typically relies on methods such as coulomb integration, model extrapolation, or voltage mapping, it is susceptible to cumulative biases caused by factors such as current measurement deviations, model errors, battery aging, and temperature changes during long-term operation. To improve the accuracy of SOC estimation, related techniques usually employ a static calibration method to periodically correct the estimation results.

[0003] However, the problem with the related technology is that the static calibration method requires a sufficiently long static time (short static time is not enough to completely eliminate the internal chemical polarization after the battery has finished running, and does not meet the requirements of the static calibration algorithm), so the triggering frequency is low in daily vehicle use.

[0004] In addition, there are other SOC calibration improvement technologies, but these technologies usually require specific usage conditions or scenarios, such as additional calibration of the battery's OCV (Open Circuit Voltage) data or reliance on charging pile equipment, which limits their application scenarios. Summary of the Invention

[0005] This invention aims to at least partially address one of the technical problems in related technologies. Therefore, the present invention aims to provide a power battery SOC calibration method, apparatus, storage medium, and power battery to solve the problems of low triggering frequency and limited application scenarios in current SOC calibration methods during daily vehicle use.

[0006] In a first aspect, embodiments of the present invention propose a power battery SOC calibration method, the method comprising: acquiring static process data and non-static process data of the battery pack; constructing and training a voltage drop prediction model based on the static process data and non-static process data; and performing SOC calibration on the battery pack based on the trained voltage drop prediction model.

[0007] In some embodiments of the present invention, the data of the resting process and the data of the non-resting process include battery pack temperature, battery pack current and individual cell voltage data of the battery pack.

[0008] In some embodiments of the present invention, the step of constructing and training a pressure drop prediction model based on the static process data and non-static process data includes: obtaining the static historical features and static future features at each moment of the static process based on the static process data; obtaining the pre-static operating features based on the non-static process data; and constructing and training the pressure drop prediction model by using the static historical features and the pre-static operating features as input features and the static future features as output labels.

[0009] In some embodiments of the present invention, the power battery SOC calibration method further includes: establishing sample points for each moment in the resting process; constructing a sample dataset based on the sample points established for each moment in the resting process, the resting process data, and the non-resting process data; and training the voltage drop prediction model using the sample dataset to obtain the trained voltage drop prediction model.

[0010] In some embodiments of the present invention, the step of performing SOC calibration on the battery pack according to the trained voltage drop prediction model includes: obtaining the resting time before vehicle startup; if the resting time is less than a preset resting time, obtaining model input feature values ​​based on the resting and non-resting data of the battery pack before vehicle startup; inputting the model input feature values ​​into the trained voltage drop prediction model to obtain a predicted voltage drop value; obtaining the open-circuit voltage of the battery pack based on the predicted voltage drop value and the individual cell voltage before resting; and performing SOC calibration on the battery pack using an interpolation method based on the open-circuit voltage of the battery pack and the historical resting temperature.

[0011] In some embodiments of the present invention, the power battery SOC calibration method further includes: if the resting process duration is greater than or equal to the preset resting duration, then obtaining the open-circuit voltage and historical resting temperature of the battery pack based on the resting process data of the battery pack before vehicle startup; and performing SOC calibration on the battery pack using an interpolation calculation method based on the open-circuit voltage and the historical resting temperature of the battery pack.

[0012] In some embodiments of the present invention, the pressure drop prediction model is constructed and trained using a regression prediction algorithm.

[0013] Secondly, embodiments of the present invention propose a power battery SOC calibration device, the device comprising: an acquisition module for acquiring static process data and non-static process data of the battery pack; a training module for constructing and training a voltage drop prediction model based on the static process data and non-static process data; and a calibration module for performing SOC calibration on the battery pack based on the trained voltage drop prediction model.

[0014] Thirdly, embodiments of the present invention provide a computer-readable storage medium storing a computer program thereon, wherein when the computer program is executed by a processor, it implements the power battery SOC calibration method described in the first aspect embodiment.

[0015] Fourthly, embodiments of the present invention provide a power battery, the power battery including the power battery SOC calibration device as described in the second aspect embodiment.

[0016] The power battery SOC calibration method, device, storage medium, and power battery of this invention construct and train a voltage drop prediction model using data from the battery pack's resting and non-resting processes, and perform power battery SOC calibration based on the trained voltage drop prediction model. Thus, without relying on additional battery testing experiments and fixed usage conditions or scenarios, the resting time required for power battery SOC calibration can be significantly shortened using only existing vehicle data, thereby increasing the trigger frequency and applicability of power battery SOC calibration in the daily actual use of the vehicle.

[0017] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the power battery SOC calibration method according to an embodiment of the present invention. Figure 2 This is a schematic diagram illustrating the process of obtaining model input features and output labels according to an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the process of constructing and training a voltage drop prediction model according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the process for performing SOC calibration of a battery pack according to a model according to an embodiment of the present invention; Figure 5 This is a block diagram of a power battery SOC calibration device according to an embodiment of the present invention; Figure 6 This is a block diagram of a power battery according to an embodiment of the present invention. Detailed Implementation

[0019] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0020] The following description, with reference to the accompanying drawings, outlines a power battery SOC calibration method, apparatus, storage medium, and power battery according to embodiments of the present invention.

[0021] Figure 1 This is a flowchart illustrating the power battery SOC calibration method according to an embodiment of the present invention.

[0022] In some embodiments of the present invention, reference is made to Figure 1 As shown, the SOC calibration method for power batteries includes: S1, acquire data on the battery pack's resting and non-resting processes.

[0023] Specifically, the settling process data refers to the battery pack operation data that meets the requirements of the settling calibration algorithm (i.e., reaches the standard settling time). In particular, since the open-circuit voltage data of the battery pack is usually measured under the standard settling time when the battery pack leaves the factory, the specific duration of the settling process data (i.e., the preset settling time) can be the settling time spent when the open-circuit voltage data is measured when the battery pack leaves the factory. The non-settling process data refers to the operation data of the battery pack in the period adjacent to the settling time. The specific duration of the non-settling process data is determined by the method used to build and train the voltage drop prediction model, and can be adjusted according to the training effect, prediction effect and engineering practice experience of the model.

[0024] More specifically, during actual vehicle use, the battery management system acquires data on the battery pack's resting and non-resting processes. This allows for the establishment of a voltage drop prediction model based on resting process data that meets the standard resting duration and non-resting process data from the adjacent period before resting. Based on this model, SOC prediction is performed for vehicle resting scenarios (such as temporary parking or waiting at traffic lights) where the resting duration does not meet the standard. This shortens the resting time required for SOC calibration and increases the frequency of SOC calibration triggering during daily vehicle use.

[0025] For example, in some embodiments of the present invention, the data during the resting process and the data during the non-resting process include battery pack temperature, battery pack current and individual cell voltage data of the battery pack.

[0026] Specifically, in this embodiment of the present invention, the training and establishment of the voltage drop prediction model is based on a single battery cell in the battery pack as the basic unit. That is, the cell voltage refers to the voltage of a certain battery cell in the battery pack, which may be the highest cell voltage, the lowest cell voltage, or the cell voltage of a certain number, etc.

[0027] S2, based on static process data and non-static process data, construct and train a pressure drop prediction model.

[0028] Specifically, since the operating conditions of the vehicle before it is stationary may affect the elimination speed of the battery pack polarization effect during the subsequent stationary process, in this embodiment of the present invention, the operating condition data characteristics of a non-stationary process of an adjacent period before the stationary period are also considered when constructing and training the voltage drop prediction model, thereby effectively improving the adaptability and prediction accuracy of the voltage drop prediction model to different vehicle operating conditions before stationary period.

[0029] S3, perform SOC calibration on the battery pack based on the trained voltage drop prediction model.

[0030] Specifically, after the voltage drop prediction model is trained, data on the static process and non-static process that are monitored in real time during actual vehicle use and do not meet the standard can be input into the voltage drop prediction model. Based on the output of the voltage drop prediction model, the battery pack can be calibrated for SOC, thereby shortening the static time required for SOC calibration and increasing the trigger frequency of SOC calibration during daily actual vehicle use.

[0031] Furthermore, in some embodiments of the present invention, reference is made to... Figure 2 As shown, a pressure drop prediction model is constructed and trained based on data from the static and non-static processes, including: S21, Based on the static settling process data, obtain the static settling history characteristics and static settling future characteristics at each moment of the static settling process.

[0032] Specifically, in this embodiment of the invention, a set of input features is recorded for each moment of the resting process to ensure that the voltage drop prediction model can learn the data features of the battery pack resting process data for any resting time within a preset resting time range.

[0033] More specifically, the static history features include static history voltage, static history temperature, static history voltage drop, and static history duration. The static future features include static future voltage drop. In this embodiment of the invention, for each moment of the static process, the battery pack cell voltage at that moment is recorded as the input feature "static history voltage", the battery pack temperature at that moment is recorded as the input feature "static history temperature", the difference between the battery pack cell voltage at that moment and the battery pack cell voltage at the start of the static process is recorded as the input feature "static history voltage drop", the duration from that moment to the start of the static process is recorded as the feature "static history duration", and the absolute value of the difference between the battery pack cell voltage at that moment and the battery pack cell voltage at the end of the static process is recorded as the output label "static future voltage drop".

[0034] S22, based on non-static process data, obtain the operating characteristics before static settling.

[0035] Specifically, the operating characteristics before resting include the average current before resting, the standard deviation of the current before resting, the maximum current before resting, and the single-cell voltage before resting. In this embodiment of the present invention, the average value of the absolute value of the battery pack current during the non-static process is denoted as the input characteristic "average current before resting", the standard deviation of the battery pack current during the non-static process is denoted as the input characteristic "standard deviation of the current before resting", the maximum value of the absolute value of the battery pack current during the non-static process is denoted as the input characteristic "maximum current before resting", and the single-cell voltage of the battery pack at the end of the non-static process is denoted as the input characteristic "single-cell voltage before resting".

[0036] S23 uses historical features of static storage and operating features before static storage as input features, and future features of static storage as output labels to construct and train a pressure drop prediction model.

[0037] Specifically, in this embodiment of the invention, the average current before idling, the standard deviation of the current before idling, the maximum current before idling, the individual cell voltage before idling, the historical voltage before idling, the historical temperature before idling, the historical voltage drop after idling, and the historical duration after idling are used as input features, and the future voltage drop after idling is used as the output label to construct and train a voltage drop prediction model, so as to establish a mapping relationship from the average current before idling, the standard deviation of the current before idling, the maximum current before idling, the individual cell voltage before idling, the historical voltage before idling, the historical temperature after idling, the historical voltage drop after idling, and the historical duration after idling to the future voltage drop after idling.

[0038] In other words, in this embodiment of the invention, based on stationary and non-stationary process data of any stationary duration during actual vehicle use, the voltage drop prediction model is used to predict the voltage drop of the battery pack's individual cells when the vehicle continues to be stationary for a preset stationary duration. Subsequently, based on the predicted battery pack individual cell voltage drop, the battery pack individual cell voltage (i.e., open-circuit voltage) at the end of the stationary process conforming to the standard stationary duration can be calculated, and the battery pack's SOC can be calibrated using the conventional open-circuit voltage method.

[0039] Furthermore, in some embodiments of the present invention, reference is made to... Figure 3 As shown, the power battery SOC calibration method also includes: S24, establish sample points for each moment during the resting process.

[0040] Specifically, in this embodiment of the invention, sample points are established for each moment of the resting process to ensure that the voltage drop prediction model can learn the data characteristics of the battery pack resting process data for any resting time within a preset resting time range.

[0041] S25. Construct a sample dataset based on the sample points established for each moment in the resting process, the resting process data, and the non-resting process data.

[0042] Specifically, in this embodiment of the invention, the static process data (including static historical voltage, static historical temperature, static historical voltage drop, static historical duration, and static future voltage drop) at each moment during the static process and the non-static process data (including average current before static, current standard deviation before static, maximum current before static, and single-cell voltage before static) are taken as a sample data point, and all sample data points are aggregated into a sample dataset for training the voltage drop prediction model.

[0043] S26. Using the sample dataset, train the pressure drop prediction model to obtain a well-trained pressure drop prediction model.

[0044] For example, in some embodiments of the present invention, a regression prediction algorithm is used to construct and train a pressure drop prediction model.

[0045] Specifically, various regression prediction algorithms, including but not limited to MLP (Multilayer Perceptron Neural Network), SVM (Support Vector Machine), and XGBoost, can be used. In this embodiment of the invention, the MLP neural network algorithm is used to construct and train the voltage drop prediction model, which specifically includes the following four steps: 1) Data preprocessing First, all input feature data in the sample dataset are uniformly formatted, including but not limited to missing value completion and outlier removal, to improve the accuracy of the pressure drop prediction model. Then, the input feature data are normalized or standardized (such as Min-Max normalization or Z-score standardization) to ensure that features of different dimensions are comparable during model training and to avoid numerical instability during gradient descent. Finally, the dataset is divided into training, validation, and test sets according to a certain ratio (such as 7:2:1) for model training, tuning, and final evaluation.

[0046] 2) Model Structure Construction The voltage drop prediction model includes an input layer that accepts eight-dimensional battery pack operating data (including historical voltage, historical temperature, historical voltage drop, historical duration of rest, average current before rest, standard deviation of current before rest, maximum current before rest, and single-cell voltage before rest); a hidden layer, which can be set with one or more fully connected neural networks, each containing a number of neurons (e.g., 16–128), and uses activation functions such as ReLU, Sigmoid, or Tanh; an output layer that outputs a single continuous value representing the predicted future voltage drop after rest; a loss function that uses mean squared error (MSE) as the regression loss; and an optimizer that uses a gradient descent-type optimizer (such as Adam, SGD, or RMSprop) to achieve iterative parameter updates.

[0047] 3) Model Training First, the training set is input into the constructed model to obtain predicted values ​​through forward propagation, and the loss between the predicted values ​​and the actual future pressure drop under static conditions is calculated. Then, the neural network parameters are updated based on the backpropagation algorithm to gradually converge the loss function. After each iteration, the validation set is used to calculate the validation error to determine whether the model is overfitting. If the validation loss no longer decreases in several consecutive iterations, an early stopping strategy is triggered to prevent overfitting. Finally, the model parameters with the best performance on the validation set are saved as the final pressure drop prediction model.

[0048] 4) Model Evaluation First, the model is evaluated using a test set that was not used for training, and performance metrics such as MAE, MSE, and R² are calculated. Then, the model is optimized based on the evaluation metrics, including adjusting hyperparameters such as the number of neural network layers, number of neurons, activation function, and learning rate. Finally, the final model is confirmed based on the performance metrics, and the trained pressure drop prediction model is saved for real-time prediction of future pressure drop during subsequent vehicle idling.

[0049] Furthermore, in some embodiments of the present invention, reference is made to... Figure 4 As shown, the battery pack's SOC is calibrated based on the trained voltage drop prediction model, including: S31, obtain the duration of the vehicle's resting process before starting operation; Specifically, in this embodiment of the invention, the battery management system monitors the operating data of the battery pack in real time, and can obtain the resting time before the vehicle starts running through the battery management system when the vehicle starts.

[0050] S32, if the resting time is less than the preset resting time, then obtain the model input feature value based on the resting process data and non-resting process data of the battery pack before the vehicle starts running.

[0051] Specifically, if the resting period is less than the preset resting period, it indicates that the current resting period does not meet the requirements of the resting calibration algorithm, meaning that the battery pack SOC cannot be calibrated using the conventional resting calibration algorithm. Therefore, in this embodiment of the invention, it is necessary to obtain model input feature values ​​based on the battery pack resting process data and non-resting process data that do not meet the standard resting period before vehicle startup, so as to calibrate the battery pack SOC through the model.

[0052] S33: Input the model input feature values ​​into the trained pressure drop prediction model to obtain the predicted pressure drop value.

[0053] Specifically, in this embodiment of the invention, by inputting the model input feature value into the trained voltage drop prediction model, the voltage drop value that the battery pack cell voltage will generate when the vehicle continues to be stationary for a preset stationary time is predicted, and then the open circuit voltage of the battery pack is obtained based on the voltage drop value, so that the battery pack can be calibrated for SOC by the open circuit voltage method in subsequent steps.

[0054] S34: Obtain the open-circuit voltage of the battery pack based on the predicted voltage drop and the individual cell voltage before resting.

[0055] Specifically, in this embodiment of the present invention, the battery pack cell voltage at the end of the resting process that meets the standard resting time is calculated by subtracting the predicted voltage drop value from the cell voltage before resting, and the battery pack cell voltage at the end of the resting process that meets the standard resting time is used as the open circuit voltage of the battery pack.

[0056] S35 uses interpolation to calibrate the SOC of the battery pack based on its open-circuit voltage and historical resting temperature.

[0057] Specifically, when a battery pack leaves the factory, an OCV-SOC mapping curve is typically obtained through calibration. This curve reflects the SOC data of the battery pack under different temperatures and open-circuit voltages in a static state. Therefore, in this embodiment of the invention, the SOC of the battery pack can be calibrated based on its open-circuit voltage, historical static temperature, and OCV-SOC mapping curve; that is, SOC calibration of the battery pack can be performed using the open-circuit voltage method.

[0058] Furthermore, in this embodiment of the invention, an interpolation method (e.g., bilinear interpolation) is used to interpolate the curve when the accuracy of the OCV-SOC mapping curve is insufficient, thereby improving the accuracy of SOC calibration.

[0059] Furthermore, in some embodiments of the present invention, reference is made to... Figure 4 As shown, the power battery SOC calibration method also includes: S36, if the resting process duration is greater than or equal to the preset resting time, then obtain the open circuit voltage and historical resting temperature of the battery pack based on the resting process data of the battery pack before the vehicle starts running.

[0060] Specifically, if the resting period is greater than or equal to the preset resting period, it indicates that the current resting period meets the requirements of the resting calibration algorithm, meaning that the battery pack's SOC can be calibrated using a conventional resting calibration algorithm. In this case, the individual cell voltage of the battery pack at the end of the resting process is the open-circuit voltage of the battery pack.

[0061] S35 uses interpolation to calibrate the SOC of the battery pack based on its open-circuit voltage and historical resting temperature.

[0062] Specifically, if the resting period is greater than or equal to the preset resting period, the SOC of the battery pack can be directly calibrated using the open-circuit voltage method based on the open-circuit voltage and historical resting temperature of the battery pack.

[0063] In summary, by constructing and training a voltage drop prediction model using data from both the static and non-static processes of the battery pack, and then performing SOC calibration of the power battery based on the trained voltage drop prediction model, the static time required for power battery SOC calibration can be significantly shortened using only existing vehicle data, without relying on additional battery testing experiments or fixed usage conditions or scenarios. This increases the frequency and applicability of power battery SOC calibration in the daily actual use of vehicles.

[0064] The present invention also proposes a power battery SOC calibration device.

[0065] Figure 5 This is a block diagram of a power battery SOC calibration device according to an embodiment of the present invention.

[0066] In some embodiments of the present invention, reference is made to Figure 5 As shown, the power battery SOC calibration device 100 includes: an acquisition module 110, a training module 120, and a calibration module 130.

[0067] The acquisition module 110 is used to acquire data on the static and non-static processes of the battery pack; the training module 120 is used to construct and train a voltage drop prediction model based on the static and non-static process data; and the calibration module 130 is used to perform SOC calibration on the battery pack based on the trained voltage drop prediction model.

[0068] Furthermore, the training module 120 is used to obtain the historical and future characteristics of the static process at each moment based on the static process data; to obtain the running characteristics before static process based on the non-static process data; and to construct and train the voltage drop prediction model by using the historical and future characteristics of static process as input features and the future characteristics of static process as output labels.

[0069] Furthermore, the training module 120 is also used to establish sample points for each moment in the resting process; to construct a sample dataset based on the sample points established for each moment in the resting process, the resting process data, and the non-resting process data; and to train the pressure drop prediction model using the sample dataset to obtain the trained pressure drop prediction model.

[0070] Furthermore, the calibration module 130 is used to obtain the resting time before the vehicle starts running; if the resting time is less than the preset resting time, the model input feature value is obtained based on the resting and non-resting data of the battery pack before the vehicle starts running; the model input feature value is input into the trained voltage drop prediction model to obtain the predicted voltage drop value; the open circuit voltage of the battery pack is obtained based on the predicted voltage drop value and the individual cell voltage before resting; and the SOC calibration of the battery pack is performed using interpolation calculation method based on the open circuit voltage of the battery pack and the historical resting temperature.

[0071] Furthermore, the calibration module 130 is also used to obtain the open-circuit voltage and historical temperature of the battery pack based on the battery pack's static process data before the vehicle starts running, if the static process duration is greater than or equal to the preset static time; and to perform SOC calibration on the battery pack using an interpolation method based on the battery pack's open-circuit voltage and historical static temperature.

[0072] It should be understood that the specific implementation of the power battery SOC calibration device 100 of the present invention can be found in the specific implementation of the power battery SOC calibration method of the foregoing embodiments of the present invention, and will not be repeated here to reduce redundancy.

[0073] Based on the power battery SOC calibration method of the foregoing embodiments of the present invention, the present invention also proposes a computer-readable storage medium storing a power battery SOC calibration program thereon, which implements the power battery SOC calibration method of the foregoing embodiments of the present invention when executed by a processor.

[0074] It should be understood that specific embodiments of the computer-readable storage medium of the present invention can be found in the specific embodiments of the power battery SOC calibration method described in the foregoing embodiments of the present invention, and will not be repeated here to reduce redundancy.

[0075] Based on the power battery SOC calibration device 100 of the aforementioned embodiments of the present invention, the present invention also proposes a power battery 1000.

[0076] Figure 6 This is a block diagram of a power battery according to an embodiment of the present invention.

[0077] In some embodiments of the present invention, reference is made to Figure 6 As shown, the power battery 1000 includes the power battery SOC calibration device 100 described in the aforementioned embodiment of the present invention.

[0078] It should be understood that the specific implementation of the power battery 1000 of the present invention can be referred to the specific implementation of the power battery SOC calibration device 100 in the foregoing embodiments of the present invention. To reduce redundancy, it will not be described again here.

[0079] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0080] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0081] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0082] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0083] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0084] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0085] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0086] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for calibrating the state of charge (SOC) of a power battery, characterized in that, The method includes: Acquire data on the battery pack's resting and non-resting processes; Based on the static process data and non-static process data, a pressure drop prediction model is constructed and trained. The battery pack is calibrated for SOC based on the trained voltage drop prediction model.

2. The power battery SOC calibration method according to claim 1, characterized in that, The data from the static storage process and the data from the non-static storage process include battery pack temperature, battery pack current, and individual cell voltage data.

3. The power battery SOC calibration method according to claim 2, characterized in that, The step of constructing and training a pressure drop prediction model based on the static process data and the non-static process data includes: Based on the static process data, obtain the static history characteristics and static future characteristics at each moment of the static process; based on the non-static process data, obtain the running characteristics before static processing. The voltage drop prediction model is constructed and trained by using the historical features of static storage and the operating features before static storage as input features, and the future features of static storage as output labels.

4. The power battery SOC calibration method according to claim 3, characterized in that, The method further includes: Establish sample points for each moment during the settling process; A sample dataset is constructed based on the sample points established for each moment in the resting process, the resting process data, and the non-resting process data; Using the sample dataset, the pressure drop prediction model is trained to obtain the trained pressure drop prediction model.

5. The power battery SOC calibration method according to claim 3, characterized in that, The step of performing SOC calibration on the battery pack based on the trained voltage drop prediction model includes: Obtain the duration of the vehicle's resting period before starting operation; If the resting process duration is less than the preset resting time, the model input feature value is obtained based on the resting process data and non-resting process data of the battery pack before the vehicle starts running. The model input feature values ​​are input into the trained pressure drop prediction model to obtain the predicted pressure drop value; The open-circuit voltage of the battery pack is obtained based on the predicted voltage drop and the individual cell voltage before resting. Based on the open-circuit voltage of the battery pack and the historical resting temperature, the SOC calibration of the battery pack is performed using an interpolation calculation method.

6. The power battery SOC calibration method according to claim 5, characterized in that, The method further includes: If the resting process duration is greater than or equal to the preset resting time, the open-circuit voltage and historical resting temperature of the battery pack are obtained based on the resting process data of the battery pack before the vehicle starts running. Based on the open-circuit voltage of the battery pack and the historical resting temperature, the SOC calibration of the battery pack is performed using an interpolation calculation method.

7. The power battery SOC calibration method according to claim 1, characterized in that, The pressure drop prediction model is constructed and trained using a regression prediction algorithm.

8. A power battery SOC calibration device, characterized in that, The device includes: The acquisition module is used to acquire data on the battery pack's resting and non-resting processes. The training module is used to construct and train a pressure drop prediction model based on the static process data and the non-static process data. The calibration module is used to perform SOC calibration on the battery pack based on the trained voltage drop prediction model.

9. A computer-readable storage medium, characterized in that, It stores a power battery SOC calibration program, which, when executed by the processor, implements the power battery SOC calibration method as described in any one of claims 1-7.

10. A power battery, characterized in that, The power battery includes the power battery SOC calibration device as described in claim 8.

Images