Control method, device and storage medium of vehicle battery pack system

Abstract

The application relates to the technical field of battery packs, and discloses a control method and device of a vehicle battery pack system and a storage medium, which comprises the following steps: obtaining current stress characteristic data and current electrochemical characteristic data of a battery pack, and fusing the data to obtain fused characteristic data; inputting the fused characteristic data into a preset prediction model, outputting a predicted health state value and a current main attenuation factor of the battery pack through the preset prediction model; determining a current power state value of the battery pack based on the predicted health state value, current stress data and the current main attenuation factor of the battery pack, and controlling the charging and discharging power of the battery pack based on the current power state value. Therefore, the health state of the battery pack can be comprehensively predicted by combining the current stress characteristic data and the current electrochemical characteristic data. Moreover, the predicted health state value and the current main attenuation factor of the battery pack provide more accurate and forward-looking basis for subsequent power control, so that the service life of the battery pack is prolonged.

Description

Technical Field

[0001] This application relates to the field of battery pack technology, specifically to a control method, device, and storage medium for a vehicle battery pack system. Background Technology

[0002] During vehicle use, the battery pack, as one of the core components, has a State of Health (SOH) that directly affects the vehicle's performance, safety, and lifespan. SOH is an indicator that measures the ratio of the battery pack's current performance to its performance in a brand-new state. It reflects the degree of performance degradation caused by factors such as aging and wear and tear.

[0003] However, existing battery pack state of health (SOH) assessment methods often rely solely on single electrochemical characteristic parameters, such as battery capacity degradation rate and internal resistance growth, making it difficult to accurately reflect the actual health status of the battery pack. Furthermore, no strategies have been established to link battery health status prediction with battery pack system control, preventing proactive power adjustments based on future health status trends during operation. Summary of the Invention

[0004] In view of the above problems, this application provides a control method, device, and storage medium for a vehicle battery pack system. By combining current stress characteristic data and current electrochemical characteristic data, the health status of the battery pack can be predicted more comprehensively. Furthermore, predicting the health status value and the current main causes of battery pack degradation provides a more accurate and forward-looking basis for subsequent power control, thereby extending the service life of the battery pack.

[0005] The first aspect of this application provides a control method for a vehicle battery pack system, comprising: acquiring current stress characteristic data and current electrochemical characteristic data of the battery pack, and fusing the current stress characteristic data and current electrochemical characteristic data to obtain fused characteristic data; inputting the fused characteristic data into a preset prediction model to output a predicted health state value and a current attenuation cause of the battery pack through the preset prediction model; wherein, the predicted health state value is the health state value of the battery pack after a preset charge-discharge cycle; and determining the current power state value of the battery pack based on the predicted health state value, the current stress data of the battery pack, and the current attenuation cause, and controlling the charge-discharge power of the battery pack based on the current power state value.

[0006] In some specific embodiments, the step of determining the current state of power (SOP) value of the battery pack based on the predicted health status value, the current stress data of the battery pack, and the current main cause of degradation includes: determining a first coefficient and a second coefficient corresponding to the current stress data of the battery pack, and determining the weight allocation among the predicted health status value, the first coefficient, and the second coefficient according to the current main cause of degradation; wherein different main causes of degradation correspond to different weight allocations among the health status value, the first coefficient, and the second coefficient; obtaining a power correction coefficient by weighted summation of the predicted health status value, the first coefficient, and the second coefficient based on the weight allocation, and correcting the initial SOP value of the battery pack using the power correction coefficient to obtain the current SOP value; wherein the power correction coefficient is determined by the current remaining charge and temperature of the battery pack.

[0007] In some specific embodiments, the step of determining the first coefficient and the second coefficient corresponding to the current stress data of the battery pack includes: determining the stress difference between the current average stress of the battery pack and the stress threshold, and determining the first coefficient based on the stress difference; wherein, there is a preset correspondence between the stress difference and the first coefficient; determining the stress uniformity ratio between the current stress uniformity value of the battery pack and the stress uniformity threshold, and using the stress uniformity ratio as the second coefficient.

[0008] In some specific embodiments, the preset prediction model includes a first sub-model and a second sub-model; the step of inputting fused feature data into the preset prediction model to output the predicted health status value of the battery pack through the preset prediction model includes: inputting fused feature data into the first sub-model to output an initial predicted health status value through the first sub-model; inputting the initial predicted health status value and historical prediction error into the second sub-model, and correcting the initial predicted health status value based on feature importance weights to obtain the predicted health status value of the battery pack.

[0009] In some specific embodiments, the preset prediction model includes a first sub-model and a second sub-model; the step of inputting fused feature data into the preset prediction model to output the current main cause of battery pack degradation through the preset prediction model includes: inputting fused feature data into the first sub-model to output dynamic feature data of stress feature data-charge-discharge cycle and electrochemical feature data-charge-discharge cycle through the first sub-model; inputting dynamic feature data into the second sub-model to output the importance score of stress data and electrochemical data to the degradation of the battery pack's health status value through the second sub-model, so as to determine the current main cause of battery pack degradation based on the importance score.

[0010] In some specific embodiments, the step of acquiring current stress characteristic data and current electrochemical characteristic data of the battery pack, and fusing the current stress characteristic data and current electrochemical characteristic data to obtain fused characteristic data includes: acquiring any one or more of the following data during the current charge-discharge cycle: stress change rate data, stress distribution uniformity data, and stress-charge-discharge cycle correlation data of the battery pack, as current stress characteristic data of the battery pack; acquiring any one or more of the following data during the current charge-discharge cycle: individual cell voltage plateau slope, charging internal resistance growth rate, and capacity decay rate of the battery pack, as current electrochemical characteristic data of the battery pack; and fusing the current stress characteristic data and current electrochemical characteristic data through weighted fusion to obtain fused characteristic data.

[0011] In some specific embodiments, after inputting the fused feature data into a preset prediction model to output the predicted health status value of the battery pack and the current main cause of degradation through the preset prediction model, the method includes: if the difference between the predicted health status value and the current health status value of the battery pack is greater than a preset difference, then reducing the preload of the battery pack by a first preload reduction amount every preset time interval; after the number of preload reductions is greater than a preset number, if the difference is greater than the preset difference, then reducing the preload of the battery pack by a second preload reduction amount every preset time interval; wherein, the second preload reduction amount is less than the first preload reduction amount, and the preload of the battery pack after reduction is greater than or equal to a preset minimum preload.

[0012] In some specific embodiments, after determining the current power state value of the battery pack based on the predicted health state value, the current stress data of the battery pack, and the current main cause of degradation, and controlling the charging and discharging power of the battery pack based on the current power state value, the method further includes: if the current stress of the battery pack is greater than the maximum preload and less than or equal to the threshold stress, then determining a preload adjustment amount based on the difference between the current stress and the maximum preload, and adjusting the current preload of the battery pack based on the preload adjustment amount; wherein the threshold stress is greater than the maximum preload. If the current stress of the battery pack is greater than the threshold stress, then adjusting the current preload according to the preset preload adjustment amount, and limiting the current charging and discharging power of the battery pack to below a preset proportion of the current power state value, until the current stress is less than or equal to the maximum preload; wherein the preset proportion is less than 1.

[0013] A second aspect of this application provides an electronic device, comprising: a processor; and a memory for storing a computer program, wherein the computer program, when executed by the processor, implements a control method for a vehicle battery pack system as described above.

[0014] A third aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements a control method for a vehicle battery pack system as described above.

[0015] The beneficial technical effects of this application are as follows: Based on the control method, device, and storage medium for the vehicle battery pack system provided in this application, the method includes: acquiring current stress characteristic data and current electrochemical characteristic data of the battery pack, and fusing the current stress characteristic data and current electrochemical characteristic data to obtain fused characteristic data; inputting the fused characteristic data into a preset prediction model to output the predicted health state value and current degradation cause of the battery pack through the preset prediction model; wherein, the predicted health state value is the health state value of the battery pack after a preset charge-discharge cycle; based on the predicted health state value, the current stress data of the battery pack, and the current degradation cause, the current power state value of the battery pack is determined, and the charge-discharge power of the battery pack is controlled based on the current power state value. Therefore, by combining the current stress characteristic data and the current electrochemical characteristic data, the health state of the battery pack can be predicted more comprehensively. Furthermore, the predicted health state value and the current degradation cause of the battery pack provide a more accurate and forward-looking basis for subsequent power control, thereby extending the service life of the battery pack.

[0016] The above description is merely an overview of the technical solutions of the embodiments of this application. In order to better understand the technical means of the embodiments of this application and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of this application more obvious and understandable, specific implementation methods of this application are described below. Attached Figure Description

[0017] The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0018] Figure 1 This is a flowchart illustrating an embodiment of the control method for the vehicle battery pack system provided in this application;

[0019] Figure 2 This is a flowchart illustrating another embodiment of the control method for the vehicle battery pack system provided in this application;

[0020] Figure 3 This is a flowchart illustrating another embodiment of the control method for the vehicle battery pack system provided in this application;

[0021] Figure 4 This is a flowchart illustrating another embodiment of the control method for the vehicle battery pack system provided in this application;

[0022] Figure 5 This is a flowchart illustrating another embodiment of the control method for the vehicle battery pack system provided in this application;

[0023] Figure 6This is a flowchart illustrating another embodiment of the control method for the vehicle battery pack system provided in this application;

[0024] Figure 7 This is a flowchart illustrating another embodiment of the control method for the vehicle battery pack system provided in this application;

[0025] Figure 8 This is a flowchart illustrating another embodiment of the control method for the vehicle battery pack system provided in this application;

[0026] Figure 9 This is a schematic diagram of the structural framework of an embodiment of the electronic device provided in this application;

[0027] Figure 10 This is a schematic diagram of the structural framework of an embodiment of the computer-readable storage medium provided in this application. Detailed Implementation

[0028] Exemplary embodiments of the present application will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited to the embodiments set forth herein. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without inventive effort are within the scope of protection of the present application.

[0029] If the embodiments of this application involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Furthermore, if the word "and / or" appears throughout the text, it means including three parallel solutions. For example, "A and / or B" includes solution A, solution B, or a solution that simultaneously satisfies A and B. Additionally, the technical solutions of various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed in this application.

[0030] The first aspect of this application provides a control method for a vehicle battery pack system. Figure 1 This is a schematic flowchart of an embodiment of the control method for the vehicle battery pack system provided in this application. (In conjunction with...) Figure 1 This method includes the following steps:

[0031] S101: Obtain the current stress characteristic data and current electrochemical characteristic data of the battery pack, and fuse the current stress characteristic data and current electrochemical characteristic data to obtain fused characteristic data.

[0032] Specifically, high-precision pressure sensors can be used to acquire real-time mechanical stress data of the battery pack as the current stress data. It should be understood that in some applications, stress can also be referred to as expansion force, and the stress data can reflect the degree of structural deformation of the battery pack during charging and discharging. It should also be understood that the current stress data can be the stress data at the current moment, or it can be continuous monitoring values ​​within a preset time period. The preset time period can be a period from a historical moment to the current moment with a preset duration; this embodiment does not impose specific limitations.

[0033] After obtaining the current stress data, further analysis and processing are performed on the current stress data to obtain the current stress characteristic data, which to a certain extent reflects some characteristics of the stress data.

[0034] The following example illustrates how current stress data can be analyzed and processed to obtain its characteristic data: The maximum, minimum, average, and standard deviation of the current stress data can be calculated as characteristic data. These characteristic data describe the stress's features; for example, the maximum and minimum values ​​reflect the range of stress fluctuations, the average value reflects the overall stress level, and the standard deviation reflects the degree of stress dispersion. Similarly, similar analysis can be performed on current electrochemical characteristic data. For instance, the average value and trend of the slope of a single-cell voltage plateau can be calculated.

[0035] Specifically, the battery pack may include multiple cells connected in series, which are fixed between two end plates. Each cell generates stress during charging and discharging; pressure sensors can be placed between the cells to collect this stress data in real time. Through the interaction between the end plates and the cells, the battery pack maintains a stable structural state during operation, and the end plates can adjust the preload of the battery pack. The battery pack may also include an FPC (Flexible Printed Circuit) board, which can integrate various circuits, such as voltage acquisition circuitry.

[0036] After obtaining the current stress characteristic data and current electrochemical characteristic data of the battery pack, the two are fused to obtain fused characteristic data. The fusion method can employ weighted averages.

[0037] The current electrochemical characteristic data of the battery pack can be obtained through analysis of current electrochemical data, which reflects the characteristics of the current electrochemical data. This current electrochemical data can be acquired using high-precision electrochemical sensors, such as voltage sensors, current sensors, and internal resistance testers.

[0038] After obtaining the current stress characteristic data and the current electrochemical characteristic data, the current stress characteristic data and the current electrochemical characteristic data are further fused to obtain fused characteristic data. The fusion method can adopt various methods such as weighted summation and neural network fusion, and no specific limitation is made here. It should be understood that the fused characteristic data reflects the stress characteristics and electrochemical characteristics of the battery pack, and can more comprehensively reflect the overall state of the battery pack during the charging and discharging process.

[0039] After obtaining the current stress characteristic data and the current electrochemical characteristic data, the following method is used to fuse the current stress characteristic data and the current electrochemical characteristic data to obtain fused characteristic data: Based on the different degrees of influence of stress characteristics and electrochemical characteristics on the battery pack's health status, different weight values ​​are assigned to the current stress characteristic data and the current electrochemical characteristic data. For example, if stress characteristics have a greater impact on the battery pack's health status, the current stress characteristic data is assigned a larger weight; if electrochemical characteristics have a greater impact, the current electrochemical characteristic data is assigned a larger weight. Then, according to the assigned weight values, the current stress characteristic data and the current electrochemical characteristic data are weighted and summed to obtain the fused characteristic data. Alternatively, a neural network fusion method can be used to construct a neural network model containing an input layer, hidden layers, and an output layer. The current stress characteristic data and the current electrochemical characteristic data are input into the neural network model as input data. After calculation and processing by the hidden layers, the fused characteristic data is output from the output layer. The fused characteristic data obtained in this way can fully combine the information of stress characteristics and electrochemical characteristics, providing strong support for accurate prediction of the battery pack's health status. The fused feature data is specifically represented as a set of multi-dimensional vectors, which contain key information about the stress and electrochemical features after fusion processing. For example, the fused vector can be specifically represented as: fused vector F = [w1×S1, w1×S2, …, w1×Sn, w2×E1, w2×E2, …, w2×Em], where S1, S2, …, Sn represent the feature components in the current stress feature data, E1, E2, …, Em represent the feature components in the current electrochemical feature data, and w1 and w2 represent the weight values ​​corresponding to the current stress and electrochemical feature data, respectively. These weight values ​​can be adjusted and optimized according to actual needs and experimental data to achieve the best fusion effect.

[0040] S102: Input the fused feature data into the preset prediction model to output the predicted health status value of the battery pack and the current main cause of degradation through the preset prediction model; wherein, the predicted health status value is the health status value of the battery pack after a preset number of charge and discharge cycles.

[0041] After fusing the current stress characteristic data and the current electrochemical characteristic data to obtain fused characteristic data, this step further inputs the fused characteristic data into a preset prediction model. The prediction model is trained in advance with a large amount of data; for example, it can be trained using electrochemical characteristic data and stress characteristic data, and possesses the ability to accurately predict the health status of the battery pack. After inputting the fused characteristic data, the prediction model uses its complex internal algorithms and logic structure to analyze and process the input data. Through the calculations of the preset prediction model, it can output a predicted health status value for the battery pack. This predicted health status value represents the possible health status of the battery pack after a preset number of charge-discharge cycles, providing an important reference for subsequent evaluation and use of the battery pack. The preset number of cycles can be set according to actual needs, for example, 50 cycles. A charge-discharge cycle is considered when the vehicle's charge and discharge cumulatively reach the rated capacity. For example, if the battery's rated capacity is 100kWh, then a charge-discharge cycle is counted when the vehicle's charge and discharge cumulatively reach 100kWh.

[0042] On the other hand, the preset prediction model will also output the current main cause of battery pack degradation, that is, clearly point out the main factors that cause the current performance degradation of the battery pack. For example, uneven stress is the current main cause of degradation. The main cause of battery pack degradation can help to take targeted measures to improve the usage of the battery pack and extend its service life.

[0043] In some application scenarios, the predictive model structure can be as follows: The predictive model can adopt a deep neural network structure, including an input layer, multiple hidden layers, and an output layer. The input layer receives fused feature data, and the hidden layers extract and transform features from the input data through a series of nonlinear transformations to uncover deeper patterns and features in the data. For example, the hidden layers can use convolutional layers to extract local features, pooling layers for data dimensionality reduction, or fully connected layers for global feature integration. The output layer outputs the predicted health status value of the battery pack and the current main cause of degradation based on the data processed by the hidden layers. The predicted health status value can be a specific numerical value representing the health level of the battery pack after a preset charge-discharge cycle. The numerical range can be set according to actual conditions, such as between 0 and 1, with the value closer to 1 indicating a healthier battery pack. The current main cause of degradation can be represented by classification, such as uneven stress, excessive temperature, and cell aging.

[0044] It should be understood that to improve the accuracy and generalization ability of the prediction model, a large amount of historical battery pack data needs to be collected during training. This data can include stress characteristic data, electrochemical characteristic data, and corresponding actual health status values ​​and main causes of degradation. This data is divided into training, validation, and test sets. The prediction model is trained using the training set, and the model parameters are continuously adjusted to reduce the loss function value on the training set. Then, the trained model is validated using the validation set. Based on the validation results, the model is further optimized and adjusted to prevent overfitting. Finally, the optimized model is tested using the test set to evaluate its performance in real-world applications.

[0045] S103: Based on the predicted health status value, the current stress data of the battery pack, and the current main cause of degradation, determine the current power state value of the battery pack, and control the charging and discharging power of the battery pack based on the current power state value.

[0046] After obtaining the predicted health status value and the current main cause of degradation of the battery pack, this step further combines the predicted health status value, current stress data, and current main cause of degradation to accurately determine the current power state value of the battery pack. Specifically, the current power state value can be calculated using preset algorithms and logical relationships. The current power state value reflects the power level that the battery pack can safely and stably output or input under current conditions.

[0047] Once the current state of power (SOP) value is determined, the charging and discharging power of the battery pack can be precisely controlled based on this. For example, when the SOP value is high, the battery pack is allowed to charge and discharge at a higher power to meet the vehicle's driving needs or fast charging requirements; when the SOP value is low, the charging and discharging power of the battery pack is limited to prevent overheating, damage, or safety issues caused by excessive power. This charging and discharging power control method based on the current SOP value ensures that the battery pack maintains its optimal operating state under different operating conditions, improving the battery pack's efficiency and safety.

[0048] In summary, the control method for the vehicle battery pack system provided in this application, combined with current stress characteristic data and current electrochemical characteristic data, can comprehensively predict the health status of the battery pack. Furthermore, predicting the health status value and the current main causes of battery pack degradation provides a more accurate and forward-looking basis for subsequent power control, thereby extending the battery pack's service life.

[0049] Figure 2 This is a flowchart illustrating another embodiment of the control method for the vehicle battery pack system provided in this application.

[0050] Combination Figure 2In some specific embodiments, the step of determining the current state of power (SOP) of the battery pack based on the predicted health status value, the current stress data of the battery pack, and the current main cause of degradation includes:

[0051] S201: Determine the first coefficient and the second coefficient corresponding to the current stress data of the battery pack, and determine the weight allocation among the predicted health status value, the first coefficient, and the second coefficient based on the current main cause of degradation; wherein, different main causes of degradation correspond to different weight allocations among the health status value, the first coefficient, and the second coefficient.

[0052] This step first determines the corresponding first and second coefficients based on the current stress data of the battery pack. These first and second coefficients are parameters used to quantify the impact of the current stress data on the battery pack's state of power (SOP). They can be obtained through a pre-defined correspondence between stress and coefficients. For example, a correspondence between stress data and the first and second coefficients can be established based on the battery pack's performance data at different stress levels. Then, based on the real-time acquired current stress data, the corresponding first and second coefficients can be obtained through this correspondence.

[0053] After determining the first and second coefficients, the weight allocation among the predicted health state value, the first coefficient, and the second coefficient is further determined based on the current primary cause of degradation. The sum of the weights of the three coefficients is 1. Different primary causes of degradation have different degrees of impact on the battery pack's health state and power state. Therefore, this embodiment sets different weight allocation schemes for different primary causes of degradation. For example, if the current primary cause of degradation is uneven stress, the weights of the first and second coefficients can be appropriately increased to enhance the impact of stress data on the power state. If the current primary cause of degradation is electrochemical performance degradation, the weight of the predicted health state value can be appropriately increased to enhance the impact of the overall health state of the battery pack on the power state.

[0054] S202: Based on the weighted allocation, the predicted health state value, the first coefficient, and the second coefficient are weighted and summed to obtain the power correction coefficient, and the initial power state value of the battery pack is corrected by the power correction coefficient to obtain the current power state value; wherein, the initial power state value is determined by the current remaining power and temperature of the battery pack.

[0055] After obtaining the weight allocation, this step further performs a weighted summation operation on the predicted health status value, the first coefficient, and the second coefficient based on the weight allocation, thereby obtaining the power correction coefficient. The power correction coefficient is a parameter that integrates multiple factors such as the battery pack health status and the degree of influence of stress data, and can more accurately reflect the power status correction requirements of the battery pack under current conditions.

[0056] The steps for calculating the power correction coefficient by weighting and summing the predicted health status value, the first coefficient, and the second coefficient based on this weight allocation are illustrated below:

[0057] Assuming the predicted health status value is H, the first coefficient is α, and the second coefficient is β, with weights assigned according to the current primary cause of attenuation as wH, wα, and wβ (and wH + wα + wβ = 1), then the power correction coefficient K can be calculated using the following formula: K = wH × H + wα × α + wβ × β. This formula is used to weight and sum the predicted health status value, the first coefficient, and the second coefficient according to their respective weights, resulting in a comprehensive power correction coefficient.

[0058] After obtaining the power correction factor, the initial power state value (PSU) of the battery pack is corrected using this factor. The initial PSU is determined by the battery pack's current remaining charge and temperature, both of which are crucial factors affecting the pack's power output capability. For example, low remaining charge limits the pack's power output; excessively high or low temperatures also impact power efficiency and safety. In some applications, the initial PSU can be obtained by looking up a table. For instance, by consulting a Map table, assuming a temperature of 25°C and 90% remaining charge, the initial PSU is found to be 90kW. In other applications, the current PSU is obtained by correcting the initial PSU using the power correction factor. This can be achieved by multiplying the initial PSU by the power correction factor. For example, if the initial PSU is 90kW and the power correction factor is 0.9, the current PSU is 90kW multiplied by 0.9, resulting in 81kW.

[0059] Once the current state of power (SOP) value is obtained, the charging and discharging power of the battery pack can be precisely controlled based on this value. By rationally allocating the battery pack's output power according to the current SOP value, we can ensure stable and safe vehicle operation while also taking into account the future degradation trend of the battery pack.

[0060] Figure 3 This is a flowchart illustrating another embodiment of the control method for the vehicle battery pack system provided in this application.

[0061] Combination Figure 3 In some specific embodiments, the step of determining the first coefficient and the second coefficient corresponding to the current stress data of the battery pack includes:

[0062] S301: Determine the stress difference between the current average stress of the battery pack and the stress threshold, and determine the first coefficient based on the stress difference; wherein, there is a preset correspondence between the stress difference and the first coefficient.

[0063] This step specifies the method for determining the first coefficient corresponding to the current stress data of the battery pack. First, the current average stress of the battery pack can be determined. This average stress can be obtained by collecting stress data from multiple cells within the battery pack and calculating the average value. Then, the stress difference between the current average stress and a pre-set stress threshold is calculated. This stress threshold is a standard value set based on the performance and safety requirements of the battery pack.

[0064] After determining the stress difference between the current average stress and the stress threshold of the battery pack, the first coefficient is further determined based on a preset correspondence between the stress difference and the first coefficient, where the first coefficient is less than 1. For example, it can be set that when the stress difference is within a first range, the first coefficient is a first fixed value, and when the stress difference is within a second range, the first coefficient is a second fixed value. In this case, the average value of the first range can be greater than the average value of the second range, and the first fixed value is less than the second fixed value.

[0065] The step of determining the first coefficient based on the stress difference between the current average stress and the stress threshold of the battery pack is illustrated below with an example:

[0066] Assuming the stress threshold is set to 50 MPa, when the current average stress of the battery pack is 40 MPa, the stress difference is 10 MPa. If the preset correspondence stipulates that the first coefficient is 0.8 when the stress difference is in the range of 0-15 MPa, and 0.6 when the stress difference is in the range of 15-30 MPa, then since the stress difference of 10 MPa is within the 0-15 MPa range, the determined first coefficient is 0.8.

[0067] S302: Determine the stress uniformity ratio between the current stress uniformity value and the stress uniformity threshold of the battery pack, and use the stress uniformity ratio as the second coefficient.

[0068] This step specifies the method for determining the second coefficient corresponding to the current stress data of the battery pack. First, the current stress uniformity value of the battery pack can be determined. This value can be obtained by calculating the standard deviation or coefficient of variation of the stress data of each cell within the battery pack, reflecting the uniformity of stress distribution within the pack. Then, the ratio between the current stress uniformity value and a stress uniformity threshold is calculated; this ratio is used as the second coefficient. The stress uniformity threshold is preset and can serve as a standard for measuring whether the stress uniformity of the battery pack meets the required standards.

[0069] The step of determining the stress uniformity ratio between the current stress uniformity value and the stress uniformity threshold of the battery pack, and using this stress uniformity ratio as the second coefficient, is illustrated below:

[0070] Assuming the current stress uniformity value is S, calculated from the standard deviation of the stress data of each cell in the battery pack, and the preset stress uniformity threshold is S0, then the stress uniformity ratio R can be calculated using the following formula: R = S / S0. This formula compares the current stress uniformity value with the stress uniformity threshold to obtain a second coefficient reflecting the degree of stress uniformity in the battery pack.

[0071] For example, if the current stress uniformity value S is 0.7 and the stress uniformity threshold S0 is 0.8, then the stress uniformity ratio R is 0.7 divided by 0.8, which gives 0.875. This 0.875 is used as the second coefficient in subsequent calculations.

[0072] In some specific embodiments, the preset prediction model includes a first sub-model and a second sub-model. Figure 4 This is a flowchart illustrating another embodiment of the control method for the vehicle battery pack system provided in this application.

[0073] Combination Figure 4 In addition to the above, the fused feature data at this point can be data corresponding to a single time period. That is, the fused feature data is obtained by fusing stress feature data and electrochemical feature data within that time period. The steps of inputting the fused feature data into a preset prediction model to output the predicted health state value of the battery pack include:

[0074] S401: Input the fused feature data into the first sub-model to output the initial predicted health status value through the first sub-model.

[0075] In conjunction with the above embodiments, this step inputs the fused feature data into the first sub-model. Since the first sub-model has time-series processing capabilities, it can perform in-depth mining and analysis of the time-series information in the fused feature data. Specifically, the complex neural network structure in the first sub-model performs layer-by-layer feature extraction and transformation on the input fused feature data. Through the parameters accumulated during learning and training, it makes a preliminary prediction of the health status change trend of the battery pack during charging and discharging, and finally outputs an initial predicted health status value.

[0076] Specifically, the first sub-model in this embodiment can be an LSTM time series model. The LSTM time series model is a special type of recurrent neural network that effectively solves the gradient vanishing and gradient exploding problems that exist in traditional recurrent neural networks when processing long sequence data by introducing a gating mechanism. In the scenario of battery pack health status prediction, the LSTM time series model can capture the temporal dependencies in the historical data of the battery pack, thereby more accurately predicting its future health status.

[0077] The model structure and training process of the LSTM time series model in this embodiment are as follows:

[0078] In terms of model structure, LSTM time-series models typically consist of an input layer, multiple LSTM hidden layers, and an output layer. The input layer receives the input data, while the multiple LSTM hidden layers process this data layer by layer to extract the temporal features. Each LSTM hidden layer contains multiple LSTM units, and each unit controls the flow of information through input gates, forget gates, and output gates, thereby achieving accurate capture of temporal features. Finally, the output layer integrates the temporal features extracted by the LSTM hidden layers and outputs the predicted health status value of the battery pack.

[0079] In terms of the training process, the first step is to prepare a large amount of historical battery pack data as the training set. This data needs to be preprocessed, including data cleaning and normalization, to improve the training efficiency and prediction accuracy of the model. Then, the training set data is input into the LSTM time series model for training. By continuously adjusting the model's parameters, the loss function value on the training set is gradually reduced. During training, backpropagation and gradient descent optimization algorithms can be used to update the model's parameters. Simultaneously, to prevent overfitting, regularization and early stopping strategies can be employed. Finally, the trained model is validated using a validation set. Based on the validation results, the model is further optimized and adjusted until its performance on the validation set reaches a satisfactory level.

[0080] S402: Input the initial predicted health status value and historical prediction error into the second sub-model, and correct the initial predicted health status value based on the feature importance weight to obtain the predicted health status value of the battery pack.

[0081] After the first sub-model outputs the initial predicted health status value, this step further inputs the initial predicted health status value along with historical prediction errors accumulated during model training or practical application into the second sub-model. The second sub-model possesses powerful nonlinear mapping and feature selection capabilities. It can finely refine the initial predicted health status value based on the input initial predicted health status value and historical prediction errors, combined with pre-calculated feature importance weights. These feature importance weights reflect the degree of contribution of different features to the prediction result. Through these weights, the second sub-model can more accurately capture the key factors affecting the battery pack's health status, thereby outputting a more accurate predicted battery pack health status value.

[0082] Specifically, the second sub-model in this embodiment can be an XGBoost regression model. XGBoost regression is a tree-based ensemble learning model composed of multiple decision trees, each acting as a weak learner. A strong learner is built by iteratively adding new decision trees. During construction, each new tree attempts to correct the residuals predicted by all previous tree combinations, gradually improving the model's predictive ability by minimizing the loss function. Its input consists of initial predicted health status values ​​and historical prediction errors, among other feature data. At each node split, the optimal splitting feature and split point are selected based on the importance weights of the features to ensure that the resulting child nodes are as pure as possible, i.e., the prediction error of the samples in the child nodes is minimized.

[0083] In this embodiment, during the XGBoost regression model training process, a large amount of sample data, including fused feature data, initial predicted health status values, and actual health status values, is first prepared as the training set. The training set data undergoes preprocessing, such as handling missing values ​​and outliers, to ensure data quality. Then, the model parameters are initialized, and the iterative training process begins. In each iteration, the loss function value is calculated based on the current model's prediction results for the training set samples. The negative gradient for each sample is calculated using the gradient boosting algorithm and used as the target value for the next decision tree. Next, the optimal splitting features and splitting points are selected based on feature importance weights to construct a new decision tree, and the weight of this decision tree in the final prediction result is determined. The newly constructed decision tree is added to the model, updating the model's prediction results. This iterative process is repeated until a preset number of iterations is reached or the model's performance metrics (such as prediction accuracy and mean squared error on the validation set) reach a satisfactory level. During training, strategies such as regularization and cross-validation can also be used to prevent overfitting and improve the model's generalization ability.

[0084] During training, the model's inherent characteristics, such as feature gain in the XGBoost regression model, are used to calculate the importance of each feature to the prediction result, thus obtaining feature importance weights. These weights are continuously optimized as the model trains, ultimately determining a set of feature importance weight values ​​that maximize prediction accuracy. In practical applications, this pre-determined set of feature importance weights can be used to correct the initial predicted health status values, resulting in more accurate battery pack health status predictions.

[0085] It should be understood that the above correction process can effectively improve the accuracy and reliability of the prediction results, providing data support for the subsequent accurate assessment and reasonable control of the battery pack status. Furthermore, the prediction method based on the collaborative work of two models can fully leverage the advantages of different models. For example, the LSTM time series model excels at capturing long-term dependencies and dynamic trends in time series data, while the XGBoost regression model is adept at handling complex nonlinear relationships and feature interactions.

[0086] It is worth noting that the introduction of historical prediction errors in this embodiment also provides the model with an opportunity for self-correction and optimization. Furthermore, determining the feature importance weights allows analysis of the contribution of each feature to the prediction results, thereby assigning reasonable weights to different features and ultimately improving the model's prediction accuracy.

[0087] Figure 5 This is a flowchart illustrating another embodiment of the control method for the vehicle battery pack system provided in this application.

[0088] Combination Figure 5 In some specific embodiments, the preset prediction model also includes a first sub-model and a second sub-model. The step of inputting the fused feature data into the preset prediction model to output the current primary cause of battery pack degradation through the preset prediction model includes:

[0089] S501: Input the fused feature data into the first sub-model to output stress feature data - charge-discharge cycle and electrochemical feature data - dynamic feature data of charge-discharge cycle through the first sub-model.

[0090] This embodiment specifies the method for determining the current primary cause of degradation. First, the fused feature data is input into the first sub-model. Leveraging its time-series processing capabilities, the first sub-model performs in-depth analysis of the time-series information in the fused feature data, extracting dynamic feature data of stress characteristics over charge-discharge cycles and electrochemical characteristics over charge-discharge cycles. It should be understood that this dynamic feature data reflects the changes in stress and electrochemical characteristics of the battery pack over time under different charge-discharge cycles, providing crucial evidence for subsequently determining the current primary cause of degradation.

[0091] In this embodiment, the training process of the first sub-model can be as follows:

[0092] First, a large number of samples containing battery pack stress and electrochemical characteristic data under different charge-discharge cycles are collected as a training set. These samples should cover various situations of the battery pack under different usage conditions and different aging levels to ensure the generalization ability of the model. The training set data is preprocessed, such as removing noisy data and standardizing the data to make the data of different features similar in magnitude. This helps the model converge faster and better.

[0093] Then, the preprocessed data is input into the first sub-model for training. The first sub-model employs a suitable neural network structure, such as a multilayer perceptron or a more complex temporal neural network. By continuously adjusting the model parameters, the dynamic feature data output by the model is made as close as possible to the dynamic feature data in the actual samples, i.e., minimizing the loss function between the predicted and true values. During training, optimization algorithms such as mini-batch gradient descent can be used to update the model parameters, while setting appropriate hyperparameters such as the number of iterations and the learning rate to achieve better training results. After a certain number of iterations, when the loss function value of the model on the training set tends to stabilize and reach a low level, the training of the first sub-model can be considered complete.

[0094] S502: Input dynamic feature data into the second sub-model to output an importance score of the battery pack's health status value degradation based on stress data and electrochemical data, and determine the current main cause of battery pack degradation based on the importance score.

[0095] This step further inputs the dynamic characteristic data of stress feature data (charge-discharge cycle) and electrochemical feature data (charge-discharge cycle) into the second sub-model. Leveraging its powerful nonlinear processing and feature selection capabilities, the second sub-model performs detailed analysis of this dynamic characteristic data. Based on pre-set algorithms and model parameters, the second sub-model can calculate the importance scores of stress data and electrochemical data on the degradation of the battery pack's state of health. These importance scores intuitively reflect the degree to which different types of data affect the degradation of the battery pack's state of health.

[0096] After obtaining the importance scores, these scores are used to further determine the current primary cause of battery pack degradation. For example, by comparing the importance scores of stress data and electrochemical data, if the importance score of stress data is significantly higher than that of electrochemical data, then it can be determined that the current primary cause of degradation is stress-related factors, such as uneven stress or excessive stress.

[0097] In this embodiment, the training process of the second sub-model can be as follows:

[0098] First, prepare a training set containing dynamic feature data and actual battery pack health status degradation data. This training set should fully reflect the diversity of battery pack health status degradation under different combinations of stress data and electrochemical data.

[0099] The training set data is preprocessed, such as handling missing values ​​and correcting outliers, to ensure data quality. Then, the parameters of the second sub-model are initialized, and iterative training begins. In each iteration, dynamic feature data is input into the model, and the loss function is calculated by comparing the importance score output by the model with the actual health status value decay. The model parameters are adjusted using optimization algorithms, such as stochastic gradient descent, to reduce the loss function value. During training, methods such as cross-validation can be used to evaluate the model's performance and prevent overfitting.

[0100] The iterative process is repeated until the model's performance on the validation set reaches the expected standard. At this point, the training of the second sub-model is complete, enabling it to accurately output the importance scores of stress data and electrochemical data on the degradation of the battery pack's health status value.

[0101] Figure 6 This is a flowchart illustrating another embodiment of the control method for the vehicle battery pack system provided in this application.

[0102] Combination Figure 6 In some specific embodiments, the step of acquiring the current stress characteristic data and the current electrochemical characteristic data of the battery pack, and fusing the current stress characteristic data and the current electrochemical characteristic data to obtain fused characteristic data includes:

[0103] S601: Obtain one or more of the following data for the battery pack during the current charge-discharge cycle: stress change rate data, stress distribution uniformity data, and stress-charge-discharge cycle correlation data, as the current stress characteristic data of the battery pack.

[0104] This embodiment defines the current stress characteristic data, which includes any one or more of the following: rate of change data, stress distribution uniformity data, and stress-charge-discharge cycle correlation data. These data can reflect the stress condition of the battery pack during the current charge-discharge cycle from different perspectives.

[0105] For example, stress change rate data can reflect how fast the stress in the battery pack changes over time, stress distribution uniformity data can reflect the degree of balance of stress distribution in various parts of the battery pack, which is of great significance for judging whether there are problems such as local stress concentration, and stress-charge-discharge cycle correlation data can reveal the variation law of stress in different charge-discharge stages, providing a basis for analyzing the impact of stress on battery pack performance.

[0106] S602: Obtain any one or more of the following parameters during the current charge-discharge cycle: slope of the cell voltage plateau, rate of increase in internal resistance during charging, and rate of capacity decay, as current electrochemical characteristic data of the battery pack.

[0107] This embodiment defines the current stress characteristic data, which includes any one or more of the following: the slope of the cell voltage plateau, the rate of increase in charging internal resistance, and the capacity decay rate. These electrochemical characteristic data can reflect the changes in the electrochemical performance of the battery pack during the current charge-discharge cycle from different dimensions.

[0108] For example, the slope of a single-cell voltage plateau reflects the smoothness of voltage changes during charging and discharging. An abnormal slope may indicate an imbalance in certain electrochemical reactions within the battery. The rate of increase in internal resistance during charging reflects the trend of internal resistance changes during charging; excessively rapid increase in internal resistance may affect the battery's charging efficiency and energy conversion efficiency. The capacity decay rate directly reflects the degree of decline in the battery pack's ability to store charge and is one of the important indicators for assessing battery health.

[0109] S603: The current stress characteristic data and the current electrochemical characteristic data are fused together by weighted fusion to obtain fused characteristic data.

[0110] After obtaining the specific current stress and electrochemical characteristic data, this step performs a weighted fusion process on these two types of data. During the weighted fusion, appropriate weights can be assigned to different characteristic data based on their importance in reflecting the battery pack's state. For example, certain characteristic data that play a crucial role in battery pack performance evaluation, such as stress change rate data and capacity decay rate data, may be assigned higher weights.

[0111] It should be understood that this weighted fusion method can combine current stress characteristic data and current electrochemical characteristic data to form a fused characteristic data that comprehensively reflects the current state of the battery pack.

[0112] Figure 7 This is a flowchart illustrating another embodiment of the control method for the vehicle battery pack system provided in this application.

[0113] Combination Figure 7 In some specific embodiments, after the step of inputting the fused feature data into a preset prediction model to output the predicted health status value of the battery pack and the current main cause of degradation through the preset prediction model, the following steps are included:

[0114] S701: If the difference between the predicted health status value and the current health status value of the battery pack is greater than the preset difference, the preload of the battery pack is reduced by the first preload reduction amount every preset time interval.

[0115] It should be understood that if the difference between the predicted health status value and the current health status value of the battery pack is greater than the preset difference, it means that after a certain number of charge-discharge cycles, the future health status value will differ significantly from the current health status value. This implies that the battery pack may have serious potential problems, and its health status is likely to deteriorate at an accelerated pace.

[0116] Under the above circumstances, the preload of the battery pack is reduced by the first preload reduction amount every preset time interval. Reducing the preload can alleviate the stress inside the battery pack to a certain extent and reduce further damage to the battery pack structure and performance caused by stress concentration or excessive stress.

[0117] S702: After the number of times the preload is reduced exceeds a preset number, if the difference is greater than a preset difference, the preload of the battery pack is reduced by a second preload reduction amount every preset time interval; wherein, the second preload reduction amount is less than the first preload reduction amount, and the preload of the battery pack after reduction is greater than or equal to the preset minimum preload.

[0118] If the number of times the preload is reduced exceeds the preset number, and the difference is greater than the preset difference, it indicates that even if a certain degree of preload adjustment has been made, the deterioration trend of the battery pack's health status has not been effectively curbed, and the battery pack still faces significant potential risks.

[0119] In this scenario, the battery pack's preload is further reduced by a second preload reduction amount at preset intervals. This second preload reduction amount is less than the first preload reduction amount. This allows for more careful adjustment of the preload as the number of preload reductions increases, preventing excessive preload reduction from adversely affecting the battery pack's structural stability. Simultaneously, to prevent loosening or vibration of the battery pack due to excessively low preload, the preload of the battery pack must be greater than or equal to the preset minimum preload after reduction.

[0120] It should be understood that this phased preload reduction strategy allows for a more scientific and rational adjustment of the preload.

[0121] Figure 8 This is a flowchart illustrating another embodiment of the control method for the vehicle battery pack system provided in this application.

[0122] Combination Figure 8 In some specific embodiments, after determining the current state of power (SOP) of the battery pack based on the predicted health status value, the current stress data of the battery pack, and the current main cause of degradation, and controlling the charging and discharging power of the battery pack based on the current SOP, the process includes:

[0123] S801: If the current stress of the battery pack is greater than the maximum preload and less than or equal to the threshold stress, then the preload adjustment amount is determined based on the difference between the current stress and the maximum preload, so as to adjust the current preload of the battery pack based on the preload adjustment amount; wherein, the threshold stress is greater than the maximum preload.

[0124] It should be understood that if the current stress of the battery pack is greater than the maximum preload but less than or equal to the threshold stress, it indicates that the stress currently experienced by the battery pack is at a relatively high level. Although it has not yet reached the threshold stress, it has already exceeded the maximum preload set under normal operating conditions. In this situation, the excessive stress may lead to some potential risks within the battery pack, such as structural deformation and performance degradation.

[0125] To effectively address this situation, this step determines the preload adjustment based on the difference between the current stress and the maximum preload. By calculating this difference, the extent to which the current stress exceeds the normal range can be accurately determined, thus identifying the required preload adjustment. This not only ensures the structural safety and performance stability of the battery pack but also prevents unnecessary damage to the battery pack caused by over-adjusting the preload.

[0126] S802: If the current stress of the battery pack is greater than the threshold stress, the current preload is adjusted according to the preset preload adjustment amount, and the current charging and discharging power of the battery pack is limited to a preset proportion of the current power state value until the current stress is less than or equal to the maximum preload; wherein, the preset proportion is less than 1.

[0127] If the current stress of the battery pack is greater than the threshold stress, it means that the stress the battery pack is currently bearing has exceeded the safety threshold and is in a relatively dangerous state. At this time, there is a high probability that a serious problem has occurred inside the battery pack. If effective measures are not taken in time, it may lead to further damage to the battery pack or even cause a safety accident.

[0128] In this scenario, this embodiment rapidly adjusts the current preload based on a preset preload adjustment amount. By reducing the preload, it alleviates the immense stress within the battery pack, preventing further stress concentration or increase that could cause more severe damage. The preset preload adjustment amount can be a relatively large adjustment to quickly reduce the stress in the battery pack to a safe range. Simultaneously, to further ensure battery pack safety, the current charging / discharging power of the battery pack is limited to a preset percentage below the current power state value. This preset percentage is less than 1, and the specific value can be set according to actual conditions, such as 0.5 or lower, until the current stress is less than or equal to the maximum preload, ensuring the battery pack returns to a relatively safe state.

[0129] A second aspect of this application provides an electronic device, including: a processor; and a memory for storing a computer program, wherein the computer program, when executed by the processor, implements the control method of the vehicle battery pack system in any of the above embodiments.

[0130] Figure 9 This is a schematic diagram of the structural framework of an embodiment of the electronic device 600 provided in this application.

[0131] Combination Figure 9 In some specific embodiments, the electronic device 600 includes a central processing unit (CPU) 601 and a read-only memory (ROM) 602. The CPU 601 is a processor, and the ROM 602 is a memory. The CPU 601 can perform various appropriate actions and processes, such as executing the methods described in the above embodiments, based on programs stored in the ROM 602 or programs loaded from storage portion 608 into random access memory (RAM) 603. The RAM 603 also stores various programs and data required for system operation. The CPU 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0132] The following components are connected to I / O interface 605: an input section 606 including a keyboard, mouse, etc.; an output section 607 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to I / O interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 610 as needed so that computer programs read from it can be installed into storage section 608 as needed.

[0133] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 609, and / or installed from removable medium 611. When the computer program is executed by central processing unit (CPU) 601, it performs various functions defined in the system of this application.

[0134] A third aspect of this application provides a computer-readable storage medium 40, Figure 10 This is a schematic diagram of the structural framework of an embodiment of the computer-readable storage medium 40 provided in this application.

[0135] Combination Figure 10 The computer-readable storage medium 40 stores a computer program 41, which, when executed by a processor, implements the control method of the vehicle battery pack system as described in any of the above embodiments.

[0136] It should be noted that the computer-readable medium 40 shown in the embodiments of this application can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable computer program. The transmitted data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0137] In summary, the control method, device, and storage medium for the vehicle battery pack system provided in this application include: acquiring current stress characteristic data and current electrochemical characteristic data of the battery pack, and fusing the current stress characteristic data and current electrochemical characteristic data to obtain fused characteristic data; inputting the fused characteristic data into a preset prediction model to output the predicted health state value and current degradation cause of the battery pack through the preset prediction model; wherein, the predicted health state value is the health state value of the battery pack after a preset charge-discharge cycle; based on the predicted health state value, the current stress data of the battery pack, and the current degradation cause, determining the current power state value of the battery pack, and controlling the charge-discharge power of the battery pack based on the current power state value. Therefore, by combining the current stress characteristic data and the current electrochemical characteristic data, the health state of the battery pack can be predicted more comprehensively. Furthermore, the predicted health state value and the current degradation cause of the battery pack provide a more accurate and forward-looking basis for subsequent power control, thereby extending the service life of the battery pack.

[0138] The above description is merely a preferred exemplary embodiment of this application and is not intended to limit the implementation of this application. Those skilled in the art can easily make corresponding modifications or alterations based on the main concept and spirit of this application. Therefore, the scope of protection of this application should be determined by the scope of protection claimed in the claims.

Claims

1. A control method of a vehicle battery pack system, characterized by, The method comprises: obtaining current stress characteristic data and current electrochemical characteristic data of the battery pack, and fusing the current stress characteristic data and the current electrochemical characteristic data to obtain fused feature data; inputting the fused feature data into a preset prediction model to output a predicted health state value and a current attenuation main factor of the battery pack through the preset prediction model; wherein the predicted health state value is a health state value of the battery pack after a preset number of charge and discharge cycles; based on the predicted health state value, the current stress data of the battery pack and the current attenuation main factor, determining a current power state value of the battery pack, and controlling the charge and discharge power of the battery pack based on the current power state value; based on the predicted health state value, the current stress data of the battery pack and the current attenuation main factor, determining a current power state value of the battery pack, comprising: determining a first coefficient and a second coefficient corresponding to the current stress data of the battery pack, and determining a weight distribution among the predicted health state value, the first coefficient and the second coefficient according to the current attenuation main factor; wherein different attenuation main factors correspond to different weight distributions among the health state value, the first coefficient and the second coefficient; based on the weight distribution, performing weighted summation on the predicted health state value, the first coefficient and the second coefficient to obtain a power correction coefficient, and correcting an initial power state value of the battery pack through the power correction coefficient to obtain a current power state value; wherein the initial power state value is determined by the current remaining capacity and temperature of the battery pack.

2. The control method of the vehicle battery pack system according to claim 1, wherein the step of determining the first coefficient and the second coefficient corresponding to the current stress data of the battery pack comprises: determining a stress difference value between the current average stress of the battery pack and a stress threshold value, and determining the first coefficient according to the stress difference value; wherein the stress difference value and the first coefficient have a preset corresponding relationship; determining a stress uniformity ratio between the current stress uniformity value of the battery pack and a stress uniformity threshold value, and taking the stress uniformity ratio as the second coefficient.

3. The control method of the vehicle battery pack system according to claim 1, wherein the preset prediction model comprises a first sub-model and a second sub-model; the step of inputting the fused feature data into the preset prediction model to output the predicted health state value of the battery pack through the preset prediction model comprises: inputting the fused feature data into the first sub-model to output an initial predicted health state value through the first sub-model; inputting the initial predicted health state value and a historical prediction error into the second sub-model, and correcting the initial predicted health state value based on a feature importance weight to obtain the predicted health state value of the battery pack.

4. The control method of the vehicle battery pack system according to claim 1, wherein The preset prediction model comprises a first sub-model and a second sub-model; The step of inputting the fusion feature data into the preset prediction model to output the current main attenuation factor of the battery pack by the preset prediction model comprises: inputting the fusion feature data into the first sub-model to output dynamic feature data of stress feature data-charging and discharging cycle and electrochemical feature data-charging and discharging cycle by the first sub-model; inputting the dynamic feature data into the second sub-model to output an importance score of the health state value attenuation of the battery pack by the stress data and the electrochemical data by the second sub-model, and determining the current main attenuation factor of the battery pack based on the importance score.

5. The control method of the vehicle battery pack system according to claim 1, wherein The step of obtaining current stress feature data and current electrochemical feature data of the battery pack and fusing the current stress feature data and the current electrochemical feature data to obtain fusion feature data comprises: obtaining any one or more of stress change rate data, stress distribution uniformity data, and stress and charging and discharging cycle correlation data of the battery pack in the current charging and discharging cycle process as the current stress feature data of the battery pack; obtaining any one or more of monomer voltage platform slope, charging resistance growth rate, and capacity attenuation rate of the battery pack in the current charging and discharging cycle process as the current electrochemical feature data of the battery pack; fusing the current stress feature data and the current electrochemical feature data by weighted fusion to obtain fusion feature data.

6. The control method of the vehicle battery pack system according to claim 1, wherein After the step of inputting the fusion feature data into the preset prediction model to output the predicted health state value and the current main attenuation factor of the battery pack by the preset prediction model, comprising: if the difference between the predicted health state value and the current health state value of the battery pack is greater than a preset difference value, then reducing the pre-tightening force of the battery pack by a first pre-tightening force reduction amount every preset time interval; if the difference is greater than the preset difference value after the number of pre-tightening force reductions is greater than a preset number, then reducing the pre-tightening force of the battery pack by a second pre-tightening force reduction amount every preset time interval; wherein the second pre-tightening force reduction amount is less than the first pre-tightening force reduction amount, and the pre-tightening force of the battery pack after reduction is greater than or equal to a preset minimum pre-tightening force.

7. The control method of the vehicle battery pack system according to claim 1, wherein After the step of determining the current power state value of the battery pack based on the predicted health state value, the current stress data of the battery pack, and the current main attenuation factor, and controlling the charging and discharging power of the battery pack based on the current power state value, comprising: if the current stress of the battery pack is greater than the maximum pre-tightening force and less than or equal to a threshold stress, determining a pre-tightening force adjustment amount according to a difference between the current stress and the maximum pre-tightening force, to adjust the current pre-tightening force of the battery pack based on the pre-tightening force adjustment amount; wherein the threshold stress is greater than the maximum pre-tightening force; if the current stress of the battery pack is greater than the threshold stress, adjusting the current pre-tightening force according to a preset pre-tightening force adjustment amount, and limiting the current charge-discharge power of the battery pack to a preset proportion of the current power state value or less until the current stress is less than or equal to the maximum pre-tightening force; wherein the preset proportion is less than 1.

8. An electronic device, comprising: comprising: a processor; a memory for storing a computer program, the computer program being executed by the processor to implement the control method of the vehicle battery pack system according to any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The storage medium has a computer program stored therein, and the computer program is executed by the processor to implement the control method of the vehicle battery pack system according to any one of claims 1-7.

Images