Battery health state prediction method and device, terminal equipment and computer program product

Abstract

The invention relates to the technical field of batteries, in particular to a battery health state prediction method and device, terminal equipment and a computer program product. The method comprises the steps of obtaining current charging process data after vehicle charging is finished; inputting the current charging process data into a pre-trained prediction model, and performing battery health state prediction to obtain a prediction result; wherein the prediction model is obtained by taking historical charging process data of the vehicle as input and taking the battery health degree as a label for training; and outputting a prediction result. The prediction model takes historical charging process data of the vehicle as input and takes the battery health degree as a label for training, so that the accuracy and reliability of battery health state prediction are remarkably improved. According to the method, the health state of the battery can be accurately predicted in real time, the test period is short and simple, the health state of the current battery can be accurately evaluated, possible battery performance degradation is warned in advance, and the service life of the battery is effectively prolonged.

Description

Technical Field

[0001] The present invention relates to the field of battery technology, and in particular to a battery health status prediction method, device, terminal equipment and computer program product. Background Art

[0002] Existing methods for analyzing the state of health (SOH) of batteries usually use methods based on mechanism empirical models. This method predicts the health status of batteries by building models of the physical and chemical properties of batteries and simulating the chemical reactions and current flows inside the batteries. However, this method is highly complex and requires professional knowledge and technical capabilities to build and maintain the models. In addition, the method based on mechanism empirical models has a long test cycle, especially when obtaining accurate circuit models, which requires a lot of experimental data and time.

[0003] Therefore, the existing battery health status calculation method based on mechanism empirical model has the defects of high complexity and long testing cycle. Summary of the invention

[0004] In view of this, the embodiments of the present application provide a battery health status prediction method, apparatus, terminal device and computer program product, which can realize the battery health status prediction of a vehicle, and are relatively simple, have a short test cycle and high accuracy.

[0005] A first aspect of an embodiment of the present application provides a battery health status prediction method, comprising:

[0006] After the vehicle is charged, obtain the current charging process data;

[0007] Input the current charging process data into a pre-trained prediction model to predict the battery health status and obtain a prediction result; wherein the prediction model is trained with the vehicle's historical charging process data as input and the battery health status as a label;

[0008] The prediction result is output.

[0009] The prediction model of the embodiment of the present application uses the historical charging process data of the vehicle as input and is trained with the battery health (SOH) as a label. It can capture the changing rules of the battery's health status under different conditions of use, and significantly improve the accuracy and reliability of the battery health status prediction. The model trained in this way can not only learn the performance characteristics of the battery under normal working conditions, but also identify the various factors that cause the battery health to decline. In addition, the test cycle of this method is short and simple, and does not require complex experimental settings and a long data collection process. Immediately after the vehicle is charged, the data of the current charging process is obtained and input into the pre-trained prediction model, which can predict the battery health status in real time and accurately, which not only improves the timeliness of the prediction, but also can accurately evaluate the current battery health status based on the learned knowledge, and give early warning of possible battery performance degradation, thereby providing a scientific basis for battery maintenance and management, and effectively extending the battery life.

[0010] In one embodiment, the current charging process data includes at least one of voltage, current, SOC, mileage, and temperature data during the current charging process.

[0011] The current charging process data of the embodiment of the present application covers key parameters such as voltage, current, SOC (state of charge), mileage, temperature, etc. These parameters are important indicators for evaluating the health of the battery. By considering these factors, the actual working state of the battery can be more comprehensively reflected, and the accuracy and reliability of the prediction model can be improved.

[0012] In one embodiment, the historical charging process data is at least one of voltage, current, SOC, mileage, and temperature data of all charging processes in a specified time period in the past.

[0013] The historical charging process data of the embodiment of the present application selects data from a specific time period in the past, ensuring the relevance and timeliness of the data used and the current battery status, and the data covers key parameters such as voltage, current, SOC (state of charge), mileage, temperature, etc. These parameters are important indicators for evaluating the health status of the battery. By considering these factors, the actual working status of the battery can be more comprehensively reflected, and the accuracy and reliability of the prediction model can be improved.

[0014] In one embodiment, it further includes:

[0015] Based on the historical operation data of the vehicle, obtain the historical charging data of the vehicle;

[0016] Based on the historical charging data, acquiring charging process data of each charging;

[0017] Based on the charging process data of each charge, the maximum charging capacity of the battery for each charge is calculated;

[0018] The battery health for each charge is calculated based on the maximum charge capacity of the battery for each charge based on the initial capacity of the battery.

[0019] The embodiment of the present application analyzes the historical operating data of the vehicle to obtain the maximum charging capacity of each charge, and calculates the battery health on this basis as a label for training the prediction model, so that the prediction model can reflect the overall nonlinear degradation trend of the battery health status, which can improve the device perception ability of the battery management system, provide a scientific basis for early warning and preventive maintenance, and help extend the battery life and reduce maintenance costs.

[0020] In one embodiment, acquiring charging process data of each charging based on the historical charging data includes:

[0021] Filter out charging records from the historical charging data that meet the conditions that the charging current is within a range of 0.5 to 1 times the maximum rated charging current of the battery, constant current charging is adopted, and the battery temperature during charging is within a range of 20° C. to 25° C.;

[0022] The charging process data of each charging in the charging record is obtained.

[0023] The embodiment of the present application screens historical charging data based on charging record selection criteria under specific conditions (such as charging current range, constant current charging mode, and battery temperature range), thereby ensuring the quality and representativeness of the selected data, and helping to improve the accuracy of subsequent calculations of the battery's maximum charging capacity and battery health. High-quality data input improves the accuracy of the prediction model.

[0024] In one embodiment, the calculation of the maximum charging capacity of the battery for each charging based on the charging process data for each charging includes:

[0025] Based on the charging process data of each charge, determine whether the corresponding charging process is fast charging or slow charging;

[0026] If the corresponding charging process is fast charging, the maximum charging capacity of the battery is fitted through linear regression based on the corresponding charging process data, and the fitted maximum charging capacity of the battery is used as the maximum charging capacity of the battery in the corresponding charging process;

[0027] If the corresponding charging process is slow charging, the maximum charging capacity of the battery is calculated by integrating the ampere-hours based on the corresponding charging process data, and the calculated maximum charging capacity of the battery is multiplied by a preset proportional coefficient to obtain the maximum charging capacity of the battery in the corresponding charging process.

[0028] The embodiments of the present application use different calculation methods for fast charging and slow charging, which can more accurately reflect the actual capacity changes of the battery under different charging modes and improve the accuracy of the prediction. Fast charging uses linear regression fitting, while slow charging uses the ampere-hour integration method and multiplies it by a preset proportional coefficient. Linear regression can capture the complex and fluctuating data characteristics during fast charging and improve the accuracy of prediction; while ampere-hour integration is suitable for stable and linear data during slow charging, ensuring the simplicity and high accuracy of real-time monitoring. This differentiated processing method enhances the adaptability of the algorithm in different situations and improves the accuracy of calculating the maximum capacity of the battery.

[0029] In one embodiment, the prediction model is a CNN-seq2seq network model.

[0030] The embodiment of the present application uses the CNN-seq2seq network model as a prediction model, combining the advantages of the convolutional neural network (CNN) in processing local features and the ability of the sequence-to-sequence (seq2seq) model in processing long dependencies. This enables the model to effectively extract complex patterns from time series data, improving the accuracy and robustness of the prediction.

[0031] In one embodiment, the inputting the current charging process data into a pre-trained prediction model to perform battery health state prediction to obtain a prediction result includes:

[0032] Extracting time series data from the current charging process data, and inputting the time series data into a pre-trained prediction model;

[0033] Using a convolutional neural network to perform a convolution operation on the time series data to extract local features;

[0034] Using a pooling operation to reduce the dimension of the local features and extract abstract features;

[0035] The abstract features are integrated into data sequence features and input into a decoder, and decoded to obtain a prediction result, wherein the prediction result includes a battery health status.

[0036] The prediction step of the embodiment of the present application extracts time series data from the current charging process data, extracts local features using a convolutional neural network (CNN), extracts abstract features through dimensionality reduction processing through a pooling operation, and finally integrates the abstract features into data sequence features and inputs them into a decoder to generate a prediction result of the battery health state. Combining the local feature extraction capability of CNN and the time-series dependency modeling capability of the decoder, it can effectively capture the complex dynamic changes of time series data and improve the accuracy and reliability of battery health state prediction.

[0037] A second aspect of an embodiment of the present application provides a battery health status prediction device, including:

[0038] A data acquisition unit, used to acquire current charging process data after the vehicle is charged;

[0039] A prediction unit, used to input the current charging process data into a pre-trained prediction model to predict the battery health state and obtain a prediction result; wherein the prediction model is trained using the vehicle's historical charging process data as input and the battery health as a label;

[0040] An output unit is used to output the prediction result.

[0041] The third aspect of an embodiment of the present application provides a terminal device, including a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the terminal device implements the battery health status prediction method provided in the first aspect of the embodiment of the present application.

[0042] A fourth aspect of the embodiments of the present application provides a computer program product, including a computer program. When the computer program is executed, the battery health status prediction method provided in the first aspect of the embodiments of the present application is executed.

[0043] The first aspect of the embodiment of the present application provides a method for predicting the battery health status, which obtains the current charging process data after the vehicle is charged; inputs the current charging process data into a pre-trained prediction model to predict the battery health status and obtain a prediction result; wherein the prediction model is trained with the historical charging process data of the vehicle as input and the battery health as a label; and outputs the prediction result. The prediction model uses the historical charging process data of the vehicle as input and is trained with the battery health (SOH) as a label, which can capture the changing laws of the battery health status under different usage conditions, and significantly improves the accuracy and reliability of the battery health status prediction. The model trained in this way can not only learn the performance characteristics of the battery under normal working conditions, but also identify the various factors that cause the battery health to decline. In addition, the test cycle of this method is short and simple, and does not require complex experimental settings and a long data collection process. By obtaining the data of the current charging process immediately after the vehicle is charged and inputting it into the pre-trained prediction model, the battery health status can be predicted in real time and accurately. This not only improves the timeliness of the prediction, but also can accurately evaluate the current battery health status based on the learned knowledge, and provide early warning of possible battery performance degradation, thereby providing a scientific basis for battery maintenance and management and effectively extending the battery life.

[0044] It can be understood that the beneficial effects of the second to fourth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0046] Figure 1 This is a flow chart of a battery health status prediction method provided by an embodiment of the present application;

[0047] Figure 2 This is a flow chart of a battery health status prediction method provided by another embodiment of the present application;

[0048] Figure 3 This is a flow chart of a battery health status prediction method provided by another embodiment of the present application;

[0049] Figure 4 This is a flow chart of a battery health status prediction method provided by another embodiment of the present application;

[0050] Figure 5 It is a schematic diagram of the structure of the terminal device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0051] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0052] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0053] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0054] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0055] like Figure 1 As shown, the battery health status prediction method provided in the embodiment of the present application includes the following steps S101 to S103:

[0056] Step S101: After the vehicle is charged, the current charging process data is obtained.

[0057] In the application, after the vehicle is charged, the system will automatically collect various data of the current charging process, including but not limited to voltage, current, SOC (state of charge), mileage and temperature. For example, the on-board sensor can record the voltage change every minute during the charging process in real time, and record the SOC value at the beginning and end of charging, as well as the highest and lowest temperature during the entire charging process. These data can be uploaded to the cloud through the vehicle network and will be used as input for subsequent prediction models to ensure that the model can fully understand the detailed information of the current charging process.

[0058] Step S102, input the current charging process data into a pre-trained prediction model to predict the battery health status and obtain a prediction result; wherein the prediction model is trained using the vehicle's historical charging process data as input and the battery health status as a label.

[0059] In the application, the current charging process data obtained in step S101 is input into the pre-trained prediction model. The prediction model is trained based on the historical charging process data of the vehicle, where the historical data includes the voltage, current, SOC, mileage and temperature data of all charging processes in the past period of time, and the label is the calculated battery health (SOH) of each charge. For example, the charging data of the past year can be used as historical data, and the battery health after each charge can be calculated based on the historical data. The model trained with these sample data can capture the changing laws of the health status of the battery under different conditions of use, so that when faced with new charging data, the health status of the battery can be more accurately predicted. When training the prediction model, the training set and the test set can be divided into 7:3.

[0060] Step S103: output prediction results.

[0061] In the application, when the prediction model obtains the prediction result, it outputs the prediction result, so that the server can take corresponding measures such as providing maintenance suggestions based on the prediction result, such as suggesting that the user perform deep discharge during the next charge to restore battery performance, or reminding the user to contact professional maintenance personnel to check the battery as soon as possible.

[0062] The prediction model of the embodiment of the present application uses the historical charging process data of the vehicle as input and is trained with the battery health (SOH) as a label. It can capture the changing rules of the battery's health status under different conditions of use, and significantly improve the accuracy and reliability of the battery health status prediction. The model trained in this way can not only learn the performance characteristics of the battery under normal working conditions, but also identify the various factors that cause the battery health to decline. In addition, the test cycle of this method is short and simple, and does not require complex experimental settings and a long data collection process. Immediately after the vehicle is charged, the data of the current charging process is obtained and input into the pre-trained prediction model, which can predict the battery health status in real time and accurately, which not only improves the timeliness of the prediction, but also can accurately evaluate the current battery health status based on the learned knowledge, and give early warning of possible battery performance degradation, thereby providing a scientific basis for battery maintenance and management, and effectively extending the battery life.

[0063] In one embodiment, the current charging process data includes at least one of voltage, current, SOC, mileage, and temperature data during the current charging process.

[0064] The current charging process data of the embodiment of the present application covers key parameters such as voltage, current, SOC (state of charge), mileage, temperature, etc. These parameters are important indicators for evaluating the health of the battery. By considering these factors, the actual working state of the battery can be more comprehensively reflected, and the accuracy and reliability of the prediction model can be improved.

[0065] In one embodiment, the historical charging process data is at least one of voltage, current, SOC, mileage, and temperature data of all charging processes in the past specified time period.

[0066] In the application, the current charging process data covers a variety of key parameters, which are important indicators for evaluating the health of the battery. For example, voltage and current data can reflect the electrical performance of the battery during the charging process, SOC data can show the battery's state of charge, mileage data can reflect the vehicle's usage, and temperature data can provide information about the battery's working environment. By comprehensively considering these parameters, the actual working state of the battery can be more comprehensively reflected, and the accuracy and reliability of the prediction model can be improved. For example, if the temperature during the charging process is too high, it may accelerate the aging of the battery and affect its health.

[0067] The historical charging process data of the embodiment of the present application selects data from a specific time period in the past, ensuring the relevance and timeliness of the data used and the current battery status, and the data covers key parameters such as voltage, current, SOC (state of charge), mileage, temperature, etc. These parameters are important indicators for evaluating the health status of the battery. By considering these factors, the actual working status of the battery can be more comprehensively reflected, and the accuracy and reliability of the prediction model can be improved.

[0068] In one embodiment, Figure 2 As shown, the following steps S201 to S204 are also included:

[0069] Step S201: Based on the historical operation data of the vehicle, obtain the historical charging data of the vehicle.

[0070] In the application, historical operation data can be all historical operation data uploaded by the vehicle, including historical charging data. Historical charging data can be charging records and related data for each charging within a specified time period (such as the past six months, the past three months). The charging records include the charging process data for each charging.

[0071] Step S202: Based on the historical charging data, the charging process data of each charging is obtained.

[0072] In the application, detailed process data of each charge can be extracted from the historical charging data, including parameters such as voltage, current, SOC, mileage and temperature during the charging process. For example, the system can record the voltage and current changes every minute during each charging process, as well as the SOC values ​​at the beginning and end of charging. These data will be used to calculate the maximum charging capacity and health of the battery. The historical charging data contains multiple charging records, and the charging records can be filtered according to the filtering conditions to select a more representative charging process.

[0073] Step S203: Calculate the maximum charging capacity of the battery for each charging based on the charging process data for each charging.

[0074] In applications, the maximum charge capacity during slow charging can be calculated using the ampere-hour integration method, or the maximum charge capacity during fast charging can be calculated using the linear regression fitting method. These calculation methods can more accurately reflect the actual capacity changes of the battery under different charging modes.

[0075] Step S204: Calculate the battery health for each charge based on the maximum charge capacity of the battery for each charge based on the initial capacity of the battery.

[0076] In the application, the battery health (SOH) after each charge is calculated based on the initial capacity of the battery and the maximum charge capacity of each charge. For example, if the initial capacity of the battery is 100Ah and the maximum charge capacity after a certain charge is 90Ah, the battery health after this charge is 90%. In this way, the change in the health state of the battery after each charge can be quantified, providing accurate label data for model training.

[0077] The embodiment of the present application analyzes the historical operating data of the vehicle to obtain the maximum charging capacity of each charge, and calculates the battery health on this basis as a label for training the prediction model, so that the prediction model can reflect the overall nonlinear degradation trend of the battery health status, which can improve the device perception ability of the battery management system, provide a scientific basis for early warning and preventive maintenance, and help extend the battery life and reduce maintenance costs.

[0078] In one embodiment, based on historical charging data, charging process data of each charging is obtained, including:

[0079] Filter out charging records from historical charging data that meet the conditions that the charging current is within the range of 0.5 to 1 times the maximum rated charging current of the battery, constant current charging is adopted, and the battery temperature during charging is within the range of 20°C to 25°C;

[0080] Get the charging process data of each charge in the charging record.

[0081] In the application, in order to ensure the quality and representativeness of the data, the system selects charging records that meet specific conditions from the historical charging data. For example, select charging records with a charging current in the range of 0.5 to 1 times the maximum rated charging current of the battery, a constant current charging method, and a battery temperature in the range of 20°C to 25°C during charging. These conditions can exclude abnormal charging conditions and ensure the accuracy and reliability of the selected data. For example, if the charging current is too low or too high, or the charging temperature is outside the normal range, it may affect the health of the battery, so these data need to be excluded. For example, 80 representative data can be selected from 100 charging records, and detailed process data for each charge can be extracted from the selected 80 charging records, including parameters such as voltage, current, SOC, mileage, and temperature during the charging process. These data will be used for subsequent calculations and analysis to ensure the accuracy of model training.

[0082] The embodiment of the present application screens historical charging data based on charging record selection criteria under specific conditions (such as charging current range, constant current charging mode, and battery temperature range), thereby ensuring the quality and representativeness of the selected data, and helping to improve the accuracy of subsequent calculations of the battery's maximum charging capacity and battery health. High-quality data input improves the accuracy of the prediction model.

[0083] In one embodiment, Figure 3 As shown, step S203 includes the following steps S2031 to S2033:

[0084] Step S2031: Based on the charging process data of each charging, determine whether the corresponding charging process is fast charging or slow charging.

[0085] In the application, fast charging or slow charging can be judged according to the charging current. If the charging current exceeds 0.8 times the maximum rated charging current of the battery, it can be judged as fast charging; otherwise, it is slow charging. Other methods can also be used to judge fast charging or slow charging, which is not limited in this application.

[0086] Step S2032: If the corresponding charging process is fast charging, the maximum charging capacity of the battery is fitted through linear regression based on the corresponding charging process data, and the fitted maximum charging capacity of the battery is used as the maximum charging capacity of the battery in the corresponding charging process.

[0087] In the application, for the fast charging process, the linear regression fitting method is used to calculate the maximum charging capacity of the battery. For example, the voltage and current data during the charging process can be used as input to fit the maximum charging capacity of the battery through a linear regression model. Linear regression can capture the complex and fluctuating data characteristics of the fast charging process and improve the prediction accuracy.

[0088] Step S2033: If the corresponding charging process is slow charging, the maximum charging capacity of the battery is calculated by integrating the ampere-hours based on the corresponding charging process data, and the calculated maximum charging capacity of the battery is multiplied by a preset proportional coefficient to obtain the maximum charging capacity of the battery in the corresponding charging process.

[0089] In the application, for the slow charging process, the ampere-hour integration method is used to calculate the maximum charging capacity of the battery. For example, the current data during the charging process can be integrated to obtain the charging capacity of the battery. In order to improve the accuracy, the calculated charging capacity can be multiplied by a preset proportionality factor, such as 95% to 98%, to correct possible errors. The ampere-hour integration method is suitable for stable and linear data during the slow charging process, ensuring the simplicity and high accuracy of real-time monitoring.

[0090] The embodiments of the present application use different calculation methods for fast charging and slow charging, which can more accurately reflect the actual capacity changes of the battery under different charging modes and improve the accuracy of the prediction. Fast charging uses linear regression fitting, while slow charging uses the ampere-hour integration method and multiplies it by a preset proportional coefficient. Linear regression can capture the complex and fluctuating data characteristics during fast charging and improve the accuracy of prediction; while ampere-hour integration is suitable for stable and linear data during slow charging, ensuring the simplicity and high accuracy of real-time monitoring. This differentiated processing method enhances the adaptability of the algorithm in different situations and improves the accuracy of calculating the maximum capacity of the battery.

[0091] In one embodiment, the prediction model is a CNN-seq2seq network model.

[0092] The embodiment of the present application uses the CNN-seq2seq network model as a prediction model, combining the advantages of the convolutional neural network (CNN) in processing local features and the ability of the sequence-to-sequence (seq2seq) model in processing long dependencies. This enables the model to effectively extract complex patterns from time series data, improving the accuracy and robustness of the prediction.

[0093] In one embodiment, Figure 4 As shown, step S102 includes the following steps S301 to S304:

[0094] Step S301: extract time series data from current charging process data, and input the time series data into a pre-trained prediction model.

[0095] In the application, the charging process with a large range of SOC value changes can be selected in chronological order, such as a charging process with an SOC value greater than 70, and the voltage, current, SOC, mileage and temperature can be input into the model as the time series data of the model.

[0096] Step S302: Use a convolutional neural network to perform a convolution operation on the time series data to extract local features.

[0097] In the application, a convolutional neural network (CNN) is used to perform convolution operations on time series data to extract local features. For example, multiple convolutional layers and activation functions (such as ReLU) can be used to extract local change features of voltage and current during the charging process, which helps the model understand subtle changes in the charging process.

[0098] Step S303: Use pooling operation to reduce the dimension of local features and extract abstract features.

[0099] In applications, pooling operations are used to reduce the dimensionality of local features and extract abstract features. For example, you can use maximum pooling or average pooling operations to reduce the dimension of features and retain the most important information. Pooling operations help the model extract higher-level abstract features and improve the robustness of predictions.

[0100] Step S304: Integrate the abstract features into data sequence features and input them into a decoder to obtain prediction results through decoding. The prediction results include the battery health status.

[0101] In the application, the abstract features are integrated into data sequence features and input into the decoder to generate the prediction results of the battery health status. For example, a recurrent neural network such as LSTM or GRU can be used as a decoder to convert the abstract features into prediction results on the time series. The decoder can capture the long dependencies of time series data and generate accurate battery health status predictions. The final prediction result can be a specific health percentage or a set of parameters describing the battery health status.

[0102] In the application, the prediction model is retrained at set intervals to ensure that the model can make accurate predictions. For example, the prediction model is retrained every three months or six months.

[0103] The prediction step of the embodiment of the present application extracts time series data from the current charging process data, extracts local features using a convolutional neural network (CNN), extracts abstract features through dimensionality reduction processing through a pooling operation, and finally integrates the abstract features into data sequence features and inputs them into a decoder to generate a prediction result of the battery health state. Combining the local feature extraction capability of CNN and the time-series dependency modeling capability of the decoder, it can effectively capture the complex dynamic changes of time series data and improve the accuracy and reliability of battery health state prediction.

[0104] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0105] The present application also provides a battery health status prediction device for executing the steps in the above battery health status prediction method embodiment. The battery health status prediction device can be a virtual appliance in a terminal device, which is run by a processor of the terminal device, or it can be the terminal device itself.

[0106] An embodiment of the present application provides a battery health status prediction device, comprising:

[0107] A data acquisition unit, used to acquire current charging process data after the vehicle is charged;

[0108] A prediction unit, used to input the current charging process data into a pre-trained prediction model to predict the battery health state and obtain a prediction result; wherein the prediction model is trained using the vehicle's historical charging process data as input and the battery health as a label;

[0109] An output unit is used to output the prediction result.

[0110] like Figure 5 As shown, the embodiment of the present application further provides a terminal device 200, including: at least one processor 201 ( Figure 5 Only one processor is shown in the figure), a memory 202, and a computer program 203 stored in the memory 202 and executable on at least one processor 201, wherein the processor 201 implements the steps in the above-mentioned various method embodiments when executing the computer program 203.

[0111] In the application, the terminal device 200 can be a cloud server, which obtains the battery health status prediction result by inputting the data uploaded by the vehicle into the onboard prediction model, and can take corresponding measures based on the prediction result, such as providing maintenance suggestions, such as suggesting that the user perform deep discharge during the next charge to restore battery performance, or reminding the user to contact professional maintenance personnel to check the battery as soon as possible.

[0112] In the application, the terminal device may include, but is not limited to, a processor and a memory. Those skilled in the art will understand that Figure 5 It is only an example of a terminal device and does not constitute a limitation on the terminal device. The terminal device may include more or fewer components than shown in the figure, or a combination of certain components, or different components.

[0113] In applications, the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0114] In applications, the memory may be an internal storage unit of a terminal device in some embodiments, such as a hard disk or memory of the terminal device. In other embodiments, the memory may also be an external storage device of the terminal device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the terminal device. Further, the memory may also include both an internal storage unit of the terminal device and an external storage device. The memory is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as program codes of a computer program. The memory may also be used to temporarily store data that has been output or is to be output.

[0115] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0116] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0117] An embodiment of the present application provides a computer program product, including a computer program. When the computer program product runs on a terminal device, the terminal device can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0118] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device that can carry the computer program code to the device / electronic device, a recording medium, a computer memory, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), an electric carrier signal, a telecommunication signal, and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0119] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0120] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0121] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0122] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0123] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for predicting battery health status, characterized in that: include: After the vehicle is charged, obtain the current charging process data; Input the current charging process data into a pre-trained prediction model to predict the battery health status and obtain a prediction result; wherein the prediction model is trained with the vehicle's historical charging process data as input and the battery health status as a label; The prediction result is outputted.

2. The battery health status prediction method according to claim 1, characterized in that: The current charging process data includes at least one of voltage, current, SOC, mileage, and temperature data during the current charging process; The historical charging process data is at least one of voltage, current, SOC, mileage, and temperature data of all charging processes in a specified time period in the past.

3. The battery health status prediction method according to claim 1, characterized in that: Also includes: Based on the historical operation data of the vehicle, obtain the historical charging data of the vehicle; Based on the historical charging data, acquiring charging process data of each charging; Based on the charging process data of each charge, the maximum charging capacity of the battery for each charge is calculated; The battery health for each charge is calculated based on the maximum charge capacity of the battery for each charge based on the initial capacity of the battery.

4. The battery health status prediction method according to claim 3, characterized in that: The acquiring charging process data of each charging based on the historical charging data includes: Filter out charging records from the historical charging data that meet the conditions that the charging current is within a range of 0.5 times to 1 times the maximum rated charging current of the battery, constant current charging is adopted, and the battery temperature during charging is within a range of 20° C. to 25° C.; The charging process data of each charging in the charging record is obtained.

5. The battery health status prediction method according to claim 3, characterized in that: The step of calculating the maximum charging capacity of the battery for each charging based on the charging process data for each charging comprises: Based on the charging process data of each charge, determine whether the corresponding charging process is fast charging or slow charging; If the corresponding charging process is fast charging, the maximum charging capacity of the battery is fitted through linear regression based on the corresponding charging process data, and the fitted maximum charging capacity of the battery is used as the maximum charging capacity of the battery in the corresponding charging process; If the corresponding charging process is slow charging, the maximum charging capacity of the battery is calculated by integrating the ampere-hours based on the corresponding charging process data, and the calculated maximum charging capacity of the battery is multiplied by a preset proportional coefficient to obtain the maximum charging capacity of the battery in the corresponding charging process.

6. The battery health status prediction method according to claim 1, characterized in that: The prediction model is a CNN-seq2seq network model.

7. The battery health status prediction method according to claim 1, characterized in that: The step of inputting the current charging process data into a pre-trained prediction model to perform battery health state prediction and obtain a prediction result includes: Extracting time series data from the current charging process data, and inputting the time series data into a pre-trained prediction model; Using a convolutional neural network to perform a convolution operation on the time series data to extract local features; Using a pooling operation to reduce the dimension of the local features and extract abstract features; The abstract features are integrated into data sequence features and input into a decoder, and decoded to obtain a prediction result, wherein the prediction result includes a battery health status.

8. A battery health status prediction device, characterized in that: include: A data acquisition unit, used to acquire current charging process data after the vehicle is charged; A prediction unit, used to input the current charging process data into a pre-trained prediction model to predict the battery health state and obtain a prediction result; wherein the prediction model is trained using the vehicle's historical charging process data as input and the battery health as a label; An output unit is used to output the prediction result.

9. A terminal device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the terminal device implements the method as claimed in any one of claims 1 to 7.

10. A computer program product, characterized in that The invention comprises a computer program, which, when being executed, enables the method according to any one of claims 1 to 7 to be performed.

Images