Battery state of health prediction method, device, electronic equipment and readable storage medium

Abstract

The application discloses a battery health state prediction method and device, electronic equipment and a readable storage medium, and is applied to the technical field of electric vehicles. The battery health state prediction method comprises the following steps: acquiring battery data of a vehicle; performing feature extraction on the battery data to obtain a vehicle use behavior feature corresponding to the vehicle; according to the vehicle use behavior feature, a predicted vehicle use behavior feature at a next time step of the vehicle is obtained; and according to the predicted vehicle use behavior feature and a battery health state prediction model, the health degree of the battery in the vehicle is obtained. The application solves the technical problem of low prediction accuracy of the battery health state.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric vehicles, and particularly relates to a battery health state prediction method and device, electronic equipment and a readable storage medium. BACKGROUND

[0002] With the rapid development of electric vehicles, lithium batteries are widely used in electric vehicles due to their characteristics of high voltage, high specific energy and long cycle life. In order to prolong the service life of electric vehicles, ensure the safe operation of electric vehicles and improve the user experience, the battery state needs to be evaluated to optimize and adjust the energy management strategy of electric vehicles. Generally, the SOH (state of health) of the battery can be calculated to determine the battery state.

[0003] At present, the method for determining the battery state by calculating the SOH of the battery in the prior art usually predicts the SOH by using the external characteristic parameters of the battery and a neural network model to obtain the battery health state. However, this method only predicts according to the characteristics of the battery itself and does not consider other factors such as user driving behavior and user driving habits. Therefore, the prediction accuracy of the battery health state is low. SUMMARY

[0004] The main purpose of the present application is to provide a battery health state prediction method and device, electronic equipment and a readable storage medium, which aims to solve the technical problem of low prediction accuracy of the battery health state in the prior art.

[0005] To achieve the above purpose, the present application provides a battery health state prediction method applied to a battery health state prediction device, which comprises the following steps:

[0006] obtaining battery data of a vehicle;

[0007] extracting features from the battery data to obtain driving behavior features corresponding to the vehicle;

[0008] predicting the predicted driving behavior features of the vehicle at the next time step according to the driving behavior features;

[0009] predicting the health degree of the battery in the vehicle according to the predicted driving behavior features and a battery health state prediction model.

[0010] Optionally, the feature types of the driving behavior features include proportion type features and accumulation type features. The step of predicting the predicted driving behavior features of the vehicle at the next time step according to the driving behavior features comprises the following steps:

[0011] obtaining a preset linear fitting model and the vehicle usage behavior characteristics, and predicting a predicted accumulated class feature of the vehicle at a next time step according to the linear fitting model and the vehicle usage behavior characteristics;

[0012] aggregating the proportion class feature and the accumulated class feature to obtain a predicted vehicle usage behavior feature of the vehicle at the next time step.

[0013] Optionally, before the step of predicting the health degree of the battery in the vehicle according to the predicted vehicle usage behavior feature and the battery health state prediction model, the method further comprises:

[0014] obtaining a battery health state prediction model to be trained, training sample data of the battery in the vehicle, and a real health degree of the battery calculated based on the training sample data;

[0015] extracting a training vehicle usage behavior feature corresponding to the battery from the training sample data;

[0016] performing normalization processing on the training vehicle usage behavior feature to obtain a normalized vehicle usage behavior feature;

[0017] predicting a training health degree of the battery according to the normalized vehicle usage behavior feature and the battery health state prediction model to be trained;

[0018] iteratively optimizing the battery health state prediction model to be trained according to the training health degree and the real health degree to obtain the battery health state prediction model.

[0019] Optionally, the feature content of the training vehicle usage behavior feature includes user behavior characteristics and battery performance characteristics, and the step of extracting the training vehicle usage behavior feature corresponding to the battery from the training sample data comprises:

[0020] obtaining a cutoff time of battery charging data used to calculate the real health degree in the training battery data;

[0021] selecting training battery data satisfying a preset health state prediction condition before the cutoff time in the training sample data;

[0022] extracting user behavior characteristics and battery performance characteristics from the training battery data.

[0023] Optionally, the step of calculating the real health degree of the battery based on the training battery data comprises:

[0024] select battery charging data meeting preset charging working condition conditions in the training sample data, wherein the battery charging data includes a training battery current of the battery in a charging process, a first residual capacity at which the training battery starts the charging process, a second residual capacity at which the training battery ends the charging process, and a rated battery capacity of the battery;

[0025] determine a real health degree corresponding to the battery according to the training battery current, the first residual capacity, the second residual capacity, and the rated battery capacity.

[0026] Optionally, the step of selecting battery charging data meeting preset charging working condition conditions in the training sample data comprises:

[0027] select first battery data in the training sample data, wherein the first battery data is battery data with a static duration longer than a preset duration threshold after the charging process ends;

[0028] determine second battery data in the first battery data as the battery charging data, wherein the second battery data is battery data with a difference between the second residual capacity and the first residual capacity greater than a preset capacity threshold and a current at the end of the charging process less than a preset current threshold.

[0029] Optionally, the step of performing normalization processing on the training vehicle behavior features to obtain normalized vehicle behavior features comprises:

[0030] obtain preset feature thresholds corresponding to each of the training vehicle behavior features;

[0031] obtain normalized vehicle behavior features according to a ratio of each of the training vehicle behavior features to the preset feature threshold.

[0032] To achieve the above object, the application further provides a battery health state prediction device, which is applied to a battery health state prediction equipment, and comprises:

[0033] an acquisition module, configured to acquire battery data of a vehicle;

[0034] an extraction module, configured to perform feature extraction on the battery data to obtain vehicle behavior features corresponding to the vehicle;

[0035] a feature prediction module, configured to predict, according to the vehicle behavior features, predicted vehicle behavior features of the vehicle at a next time step;

[0036] a health degree prediction module, configured to predict, according to the predicted vehicle behavior features and a battery health state prediction model, a health degree of a battery in the vehicle.

[0037] Optionally, the feature type of the vehicle usage behavior feature comprises a proportion type feature and an accumulation type feature, and the feature prediction module is further configured to:

[0038] predict, according to a preset linear fitting model and the vehicle usage behavior feature, a predicted accumulation type feature of the vehicle at a next time step;

[0039] aggregate the proportion type feature and the accumulation type feature to obtain a predicted vehicle usage behavior feature of the vehicle at the next time step.

[0040] Optionally, before the step of predicting, according to the predicted vehicle usage behavior feature and a battery health state prediction model, a health degree of a battery in the vehicle, the battery health state prediction apparatus is further configured to:

[0041] obtain a battery health state prediction model to be trained, training sample data of the battery in the vehicle, and a real health degree of the battery calculated based on the training sample data;

[0042] extract a training vehicle usage behavior feature corresponding to the battery from the training sample data;

[0043] perform normalization processing on the training vehicle usage behavior feature to obtain a normalized vehicle usage behavior feature;

[0044] predict, according to the normalized vehicle usage behavior feature and the battery health state prediction model to be trained, a training health degree of the battery;

[0045] perform iterative optimization on the battery health state prediction model to be trained according to the training health degree and the real health degree to obtain the battery health state prediction model.

[0046] Optionally, the feature content of the training vehicle usage behavior feature comprises a user behavior feature and a battery performance feature, and the battery health state prediction apparatus is further configured to:

[0047] obtain a cutoff time of battery charging data used to calculate the real health degree in the training battery data;

[0048] select, from the training sample data, training battery data satisfying a preset health state prediction condition before the cutoff time;

[0049] extract a user behavior feature and a battery performance feature from the training battery data.

[0050] Optionally, the battery health state prediction apparatus is further configured to:

[0051] selecting battery charging data satisfying a preset charging condition from the training sample data, wherein the battery charging data comprises a training battery current of the battery during a charging process, a first residual capacity of the training battery at the beginning of the charging process, a second residual capacity of the training battery at the end of the charging process, and a rated battery capacity of the battery;

[0052] determining a real state of health of the battery according to the training battery current, the first residual capacity, the second residual capacity, and the rated battery capacity.

[0053] Optionally, the battery state of health prediction apparatus is further configured to:

[0054] selecting first battery data from the training sample data, wherein the first battery data is battery data with a static duration longer than a preset duration threshold after the end of the charging process;

[0055] determining second battery data in the first battery data as the battery charging data, wherein the second battery data is battery data with a difference between the second residual capacity and the first residual capacity greater than a preset capacity threshold and a current at the end of the charging process less than a preset current threshold.

[0056] Optionally, the battery state of health prediction apparatus is further configured to:

[0057] obtaining a preset feature threshold corresponding to each training vehicle behavior feature;

[0058] obtaining a normalized vehicle behavior feature according to a ratio of each training vehicle behavior feature to the preset feature threshold.

[0059] The application further provides an electronic device, comprising a memory, a processor, and a program of the battery state of health prediction method stored in the memory and executable on the processor, wherein the program of the battery state of health prediction method, when executed by the processor, can implement the steps of the battery state of health prediction method.

[0060] The application further provides a computer readable storage medium, wherein a program of a battery state of health prediction method is stored on the computer readable storage medium, and the program of the battery state of health prediction method, when executed by a processor, implements the steps of the battery state of health prediction method.

[0061] The application further provides a computer program product, comprising a computer program, wherein the computer program, when executed by a processor, implements the steps of the battery state of health prediction method.

[0062] The application provides a battery health state prediction method and device, electronic equipment and a readable storage medium. The battery data of a vehicle is obtained, the vehicle behavior feature corresponding to the vehicle is obtained by performing feature extraction on the battery data, the predicted vehicle behavior feature of the vehicle at the next time step is obtained according to the vehicle behavior feature, and the health degree of the battery in the vehicle is predicted according to the predicted vehicle behavior feature and a battery health state prediction model.

[0063] Therefore, the health degree of the battery is predicted by the vehicle behavior feature at the next time step extracted from the battery data and the battery health prediction model, so that the health degree of the battery is obtained. The battery health degree is predicted by comprehensively considering the nature of the battery itself at the next time step, the driving behavior of the user and the battery health prediction model, so that the predicted health degree is determined by multiple factors and has real-time performance, thereby improving the prediction accuracy of the battery health state. BRIEF DESCRIPTION OF DRAWINGS

[0064] The accompanying drawings incorporated in and forming a part of the specification illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the application.

[0065] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, those skilled in the art can obtain other drawings from these drawings without creative labor.

[0066] Figure 1 A flowchart of a first embodiment of the battery health state prediction method of the present application;

[0067] Figure 2 A trend graph of a vehicle behavior feature of the battery health state prediction method of the present application over time;

[0068] Figure 3 A trend graph of the real value and the predicted value of a vehicle behavior feature of the battery health state prediction method of the present application over time;

[0069] Figure 4 An example graph of the vehicle behavior feature and the predicted vehicle behavior feature of the battery health state prediction method of the present application;

[0070] Figure 5 A schematic diagram of an application scenario of the battery health state prediction method of the present application;

[0071] Figure 6 A schematic diagram of another application scenario of the battery health state prediction method of the present application;

[0072] Figure 7 Another application scenario of the battery health state prediction method of the present application is shown in the figure.

[0073] Figure 8 An example of the acquisition content of the total battery data of the battery health state prediction method of the present application is shown in the figure.

[0074] Figure 9 An example of the acquisition content of the daily battery data of the battery health state prediction method of the present application is shown in the figure.

[0075] Figure 10 An example of the content of part of the driving behavior characteristics of the battery health state prediction method of the present application is shown in the figure.

[0076] Figure 11 A device structure diagram of the hardware running environment involved in the battery health state prediction method of the present application is shown in the figure.

[0077] Figure 12 A device structure diagram of an embodiment of the battery health state prediction device of the present application is shown in the figure.

[0078] A device structure diagram of the hardware running environment involved in the battery health state prediction method of the present application is shown in the figure.

[0079] The purposes, functional features and advantages of the present application will be further described with reference to the accompanying drawings. DETAILED DESCRIPTION

[0080] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0081] The battery health state prediction method provided in the embodiments of the present application comprises:

[0082] acquiring battery data of a vehicle; extracting features of the battery data to obtain driving behavior characteristics corresponding to the vehicle; predicting predicted driving behavior characteristics of the vehicle at a next time step according to the driving behavior characteristics; and predicting a health degree of a battery in the vehicle according to the predicted driving behavior characteristics and a battery health state prediction model.

[0083] It should be noted that, in the present embodiment, in order to prolong the service life of the electric vehicle, ensure the safe operation of the electric vehicle and improve the user experience, the battery state needs to be evaluated to optimize the adjustment of the energy management strategy of the electric vehicle, and the SOH of the battery is calculated to determine the battery state. At present, the SOH is predicted by the battery external characteristic parameters and the neural network model to obtain the battery health state. However, the prediction is only based on the characteristics of the battery itself, and other factors such as user driving behavior and user driving habits are not considered. Therefore, the prediction accuracy of the battery health state is low.

[0084] In view of the above phenomenon, the battery health degree is predicted by the next time step driving behavior characteristics extracted from the battery data and the battery health prediction model, so as to obtain the battery health degree. The battery itself properties and user driving behavior of the next time step are comprehensively predicted by the battery health prediction model, so as to predict the battery health degree. The predicted health degree is determined by multiple factors, and has real-time performance, thereby improving the prediction accuracy of the battery health state.

[0085] Please refer to Figure 1 As Figure 1 shown, in the first embodiment of the battery health state prediction method of the present application, the control method of the vehicle discharge specifically includes the following steps:

[0086] Step S10, obtaining the battery data of the vehicle;

[0087] It should be noted that, in the present embodiment, the battery health state prediction method of the present application can be applied to the vehicle and the control system in the vehicle, and can also be applied to the server in communication connection with the vehicle. The communication connection can be wired connection such as interface, and can also be wireless connection such as Bluetooth and local area network, so as to realize the interactive processing of data and predict the battery health state of the battery in the vehicle. For the convenience of reading and understanding, the above vehicle is taken as the execution subject of the battery health state prediction method of the present application in the following description to specifically illustrate the present embodiment.

[0088] In the present embodiment, the battery data of the battery in the vehicle is obtained.

[0089] Step S20, extracting the driving behavior characteristics corresponding to the vehicle from the battery data to obtain the driving behavior characteristics corresponding to the vehicle;

[0090] In the present embodiment, the vehicle extracts the characteristics for representing the battery working state and user driving behavior from the battery data by a pre-set feature extractor to obtain the driving behavior characteristics corresponding to the vehicle. The driving behavior characteristics include but are not limited to driving mileage characteristics, discharge characteristics, remaining power characteristics, vehicle current characteristics, vehicle speed characteristics, vehicle operating temperature characteristics, use time characteristics, charge voltage difference characteristics and discharge voltage difference characteristics.

[0091] Step S30, predicting the predicted vehicle use behavior feature of the vehicle at the next time step according to the use behavior feature;

[0092] The use behavior feature is mapped to the predicted vehicle use behavior feature at the next time step through a preset linear fitting model.

[0093] As a feasible embodiment, in the above step S30, the feature types of the use behavior feature include the proportion type feature and the cumulative type feature, and the step of predicting the predicted vehicle use behavior feature at the next time step according to the use behavior feature includes:

[0094] Step S31, predicting the predicted cumulative type feature of the vehicle at the next time step according to the preset linear fitting model and the use behavior feature;

[0095] Step S32, aggregating the proportion type feature and the cumulative type feature to obtain the predicted vehicle use behavior feature at the next time step.

[0096] It should be noted that in this embodiment, the proportion type feature is a feature reflecting the proportion, for example, a feature for representing the proportion of parameters within a preset parameter range, and the cumulative type feature is a feature that can be accumulated over time, for example, a driving mileage feature.

[0097] In this embodiment, the cumulative type feature is mapped to the predicted cumulative type feature of the vehicle at the next time step according to the preset linear fitting model, or the calculation result of the cumulative type feature is taken as the predicted cumulative type feature of the vehicle at the next time step according to the preset linear fitting algorithm, or a time sequence cumulative type feature map of the cumulative type feature over time is constructed, and the vehicle use behavior feature value corresponding to the next time step is obtained according to the time sequence cumulative type feature map as the predicted vehicle use behavior feature, and the proportion type feature and the cumulative type feature are spliced to obtain the predicted vehicle use behavior feature at the next time step.

[0098] As an example, referring to Figure 2 , Figure 2 is a trend graph of the running total mileage of a vehicle over time, and the feature values of the running total mileage in the next three months are predicted through a preset linear fitting model or a preset linear fitting algorithm, referring to Figure 3 , Figure 3 is a trend graph of the predicted running mileage of a vehicle and the real running mileage over time. Referring to Figure 4 , Figure 4The prediction vehicle behavior characteristics of a vehicle include total mileage, full discharge cycle, discharge duration, calendar life of the vehicle, charging current 0-50A proportion, and charging end SOC proportion in 90-100. The prediction vehicle behavior characteristics include cumulative characteristics (total mileage, full discharge cycle, discharge duration, and calendar life of the vehicle) and proportion characteristics (charging current 0-50A proportion and charging end SOC proportion in 90-100). The cumulative characteristics are predicted by linear fitting, and the proportion characteristics are extended from the original characteristic values.

[0099] The cumulative characteristics are mapped into the prediction cumulative characteristics of the vehicle at the next time step by the preset linear fitting model, the prediction cumulative characteristics are obtained by linear fitting of a large amount of data, and the prediction accuracy of the prediction cumulative characteristics is improved. When the prediction cumulative characteristics of the vehicle at the next time step are calculated by the preset linear fitting algorithm, the prediction accuracy of the prediction cumulative characteristics is low due to the large limitation and low migration of the preset linear fitting algorithm. Alternatively, when the vehicle behavior characteristic values corresponding to the next time step are obtained as the prediction vehicle behavior characteristics by using the time sequence cumulative characteristic graph, the prediction accuracy of the prediction cumulative characteristics is low due to the small amount of data used to construct the time sequence cumulative characteristic graph.

[0100] In step S40, the health degree of the battery in the vehicle is predicted based on the prediction vehicle behavior characteristics and the battery health state prediction model.

[0101] In this embodiment, the prediction vehicle behavior characteristics are normalized to obtain processed prediction vehicle behavior characteristics, and the prediction vehicle behavior characteristics are mapped into the health degree of the battery in the vehicle by the battery health state prediction model.

[0102] As an example, refer to Figure 5 , Figure 5 The prediction vehicle behavior characteristics include 14 characteristic values, the battery health state prediction model includes a trained random forest model, and the health degree of the battery in the vehicle includes an output SOH result. The vehicle behavior characteristics are mapped into the health degree of the battery in the vehicle by the battery health state prediction model.

[0103] The health degree of the vehicle battery is affected by multiple factors. Multiple features are extracted from the battery data to provide more decision basis for predicting the health degree of the vehicle battery, improve the prediction accuracy of the health degree of the vehicle battery, and predict the health degree of the battery in the vehicle by the prediction vehicle behavior characteristics of the vehicle behavior characteristics and the battery health state prediction model. Real-time decision basis is provided for predicting the health degree of the vehicle battery, and therefore, the prediction accuracy of the health degree of the vehicle battery is improved.

[0104] The model algorithm of the battery health state prediction method in the above embodiment can be stored to a server in communication connection with the vehicle, Figure 6 is a schematic diagram of an application scenario of the battery health state prediction method provided by the present application. As shown in the figure, Figure 6 The application scenario can include a vehicle 01 configured with a battery, and a server 03 in communication connection with the vehicle 01 through a communication base station 02. In the Figure 2 , the vehicle 01 sends training battery data, so that the server 03 extracts driving behavior features for battery data and prediction driving behavior features, and predicts the health degree of the battery according to the prediction driving behavior features and the battery health state prediction model, to obtain the health degree of the battery in the vehicle. The storage space required by the algorithm of the battery health state prediction model is large, so as to avoid excessive storage redundancy and operation redundancy of the control system of the vehicle, and each model is preferably stored to the server.

[0105] The model algorithm of the battery health state prediction method in the above embodiment can be stored to a server in communication connection with the vehicle, Figure 7 is a schematic diagram of an application scenario of the battery health state prediction method provided by the present application. As shown in the figure, Figure 7 The application scenario can include a vehicle 01 configured with a training battery, and a control flow 04 of the battery health state prediction method stored in the vehicle 01, the control flow 04 including extracting driving behavior features from battery data, obtaining prediction driving behavior features from driving behavior features, and predicting the health degree of the battery in the vehicle according to the prediction driving behavior features.

[0106] The present application provides a battery health state prediction method, device, electronic equipment and readable storage medium, by acquiring the battery data of the vehicle; the feature extraction is carried out to the battery data, and the driving behavior feature corresponding to the vehicle is obtained; according to the driving behavior feature, the prediction driving behavior feature of the vehicle at the next time step is predicted; according to the prediction driving behavior feature and the battery health state prediction model, the health degree of the battery in the vehicle is predicted.

[0107] In this way, the battery health state prediction method of the present application predicts the health degree of the battery by extracting the driving behavior feature of the next time step from the battery data and the battery health prediction model, and obtains the health degree of the battery. The battery itself property and user driving behavior at the next time step are comprehensively predicted by the battery health prediction model, so as to predict the health degree of the battery, so that the health degree predicted by the battery is determined by multiple factors, and has real-time performance, thereby improving the prediction accuracy of the battery health state.

[0108] Further, based on the first embodiment of the battery health state prediction method of the present application, the second embodiment of the battery health state prediction method of the present application is provided.

[0109] In the embodiment, the battery health state prediction method of the application is also implemented by the vehicle as the execution subject. As a feasible embodiment, before the step of predicting the health degree of the battery in the vehicle according to the predicted vehicle behavior characteristics and the battery health state prediction model in step S40, specifically, the step can include:

[0110] Step A10, obtaining the battery health state prediction model to be trained, the training sample data of the battery in the vehicle, and the real health degree of the battery calculated based on the training sample data;

[0111] In the embodiment, the training sample data of the battery in the vehicle and the battery health state prediction model to be trained are obtained. Then, the vehicle selects the battery charging data in the ideal working condition of the actual vehicle working condition, calculates the health degree of the battery in the ideal working condition, and obtains the real health degree corresponding to the battery.

[0112] Step A20, extracting the training sample data to obtain the training vehicle behavior characteristics corresponding to the battery;

[0113] In the embodiment, the vehicle extracts the features for representing the battery working state and the user driving behavior in the training battery data by using a pre-set feature extractor, and obtains the training vehicle behavior characteristics corresponding to the training battery.

[0114] As an example, the vehicle calculates the training driving behavior data for representing the battery working state and the user driving behavior from the training sample data, and takes the features corresponding to each training driving behavior data as the training vehicle behavior characteristics corresponding to the battery.

[0115] As an example, with reference to Figure 8 , Figure 8 The vehicle type expected maximum driving range, the vehicle type expected maximum calendar life, the total driving range of the vehicle, the full discharge cycle number of the vehicle, the discharge duration of the vehicle, and the calendar life of the vehicle.

[0116] As a feasible embodiment, in the above step A20, the feature content of the training vehicle behavior characteristics includes user behavior characteristics and battery performance characteristics, and the step of extracting the training sample data to obtain the training vehicle behavior characteristics corresponding to the battery can specifically include:

[0117] Step A21, obtaining the cutoff time of the battery charging data used to calculate the real health degree in the training battery data;

[0118] In the embodiment, the cutoff time is obtained by selecting the charging end time of the battery charging data satisfying the pre-set charging condition in the training battery data.

[0119] Step A22, selecting training battery data meeting preset health state prediction conditions before the cutoff time in the training sample data;

[0120] In this embodiment, training battery data before the cutoff time in the training sample data is selected, and the vehicle is in the process of starting charging, i.e., starting the charging cycle, or not charging and the vehicle mileage gradually increases, i.e., starting the driving cycle.

[0121] As an example, refer to Figure 9 and Figure 10 , Figure 9 including the training battery data of vehicle B before the cutoff time (the total mileage, the number of full discharge cycles, the discharge duration, the calendar life of the vehicle, the proportion of the charging current 0-50A, and the proportion of the charging end SOC located in 90-100, etc.). Figure 10 including the daily training battery data of vehicle B (the cutoff mileage, the total SOC depth, and the total discharge duration).

[0122] Step A23, extracting user behavior features and battery performance features according to the training battery data.

[0123] It should be noted that in this embodiment, the preset feature extractor includes a user behavior feature extraction model and a battery performance feature extraction model.

[0124] In this embodiment, the user behavior feature extraction model is used to extract the user behavior features of the target training battery data, and the battery performance feature extraction model is used to extract the battery performance features of the target training battery data. The user behavior features and the battery performance features are spliced into the training vehicle behavior features.

[0125] Optionally, the user behavior feature extraction model is used to extract the user behavior features of the target training battery data, which can specifically include:

[0126] The user behavior feature extraction model includes a driving mileage feature extraction model, a discharge feature extraction model, a residual capacity feature extraction model, a current feature extraction model, a vehicle speed feature extraction model, and a running temperature feature extraction model. The driving mileage feature extraction model extracts the total driving mileage of the vehicle in the target training battery data to obtain a driving mileage feature. The discharge feature extraction model extracts the full discharge times and total discharge duration of the vehicle in the target training battery data to obtain a discharge feature. The residual capacity feature extraction model extracts the first proportion of the second residual capacity at the end of the charging process of the vehicle in the target training battery data, which is located in the preset charging residual capacity range, for example, the preset discharge residual capacity range can be 90-100, or 92-98, and extracts the second proportion of the third residual capacity at the end of the discharge process of the vehicle, which is located in the preset discharge residual capacity range, to obtain a residual capacity feature, for example, the preset discharge residual capacity range can be 0-15, or 5-10. The vehicle speed feature extraction model extracts the third proportion of the vehicle speed of the vehicle in the target training battery data that is greater than the preset vehicle speed threshold, to obtain a vehicle speed feature, for example, the preset vehicle speed threshold can be 120 km / h, or 130 km / h. The running temperature feature extraction model extracts the fourth proportion of the running minimum temperature of the vehicle in the target training battery data that is located in the preset running minimum temperature range, for example, the preset running minimum temperature range can be -5℃ to -25℃, or 0℃ to -15℃, and extracts the fifth proportion of the running maximum temperature of the vehicle in the target training battery data that is located in the preset running maximum temperature range, for example, the preset running maximum temperature range can be 45℃ to 50℃, or 46℃ to 49℃, to obtain a vehicle running temperature feature. The current feature extraction model extracts the sixth proportion of the charging current in the preset charging current range, for example, the preset charging current range can be 0-50A, or 0-40A, to obtain a vehicle current feature. The driving mileage feature, the discharge feature, the residual capacity feature, the vehicle current feature, the vehicle speed feature, and the vehicle running temperature feature are concatenated as user behavior features.

[0127] Optionally, the battery performance feature extraction model is used to extract the battery performance in the target training battery data to obtain a battery performance feature, which can specifically include:

[0128] The battery performance feature extraction model comprises a use time feature extraction model, a charging voltage difference feature extraction model and a discharging voltage difference feature extraction model. The use time feature is obtained by extracting the time from the factory cutoff to the real health degree of the training battery corresponding to the target training battery data through the life feature extraction model. The charging voltage difference feature is obtained by extracting the seventh proportion of the charging voltage difference in the first preset voltage difference range in the target training battery data and the eighth proportion of the charging voltage difference in the second preset voltage difference range in the target training battery data through the charging voltage difference feature extraction model. For example, the first preset voltage difference range can be 0-100 mV or 10-90 mV. The second preset voltage difference range can be 100-300 mV or 120-280 mV. The discharging voltage difference feature is obtained by extracting the ninth proportion of the discharging voltage difference in the third preset voltage difference range in the target training battery data and the tenth proportion of the discharging voltage difference in the fourth preset voltage difference range in the target training battery data through the discharging voltage difference feature extraction model. For example, the third preset voltage difference range can be 0-100 mV or 10-90 mV. The fourth preset voltage difference range can be 100-300 mV or 120-280 mV. The use time feature, the charging voltage difference feature and the discharging voltage difference feature are spliced into the battery performance feature.

[0129] In step A30, the training vehicle behavior features are normalized to obtain normalized vehicle behavior features.

[0130] In this embodiment, the cumulative features in the training vehicle behavior features are normalized to map the cumulative features to normalized cumulative features in a preset value range, and the normalized cumulative features and the proportion features are spliced into the normalized vehicle behavior features, wherein the preset value range is 0-1.

[0131] As a feasible embodiment, in step A30, the step of normalizing the training vehicle behavior features to obtain normalized vehicle behavior features can specifically comprise:

[0132] A preset feature threshold corresponding to each training vehicle behavior feature is obtained.

[0133] The normalized vehicle behavior features are obtained according to the ratio of each training vehicle behavior feature to the preset feature threshold.

[0134] It can be understood that the value range of the proportion feature is 0-1, so the proportion feature does not need to be processed.

[0135] In the embodiment, a preset feature threshold corresponding to each cumulative feature in the training vehicle behavior feature is obtained; a ratio of each cumulative feature and the preset feature threshold is taken as a normalized cumulative feature, and the normalized cumulative feature and the proportion feature are spliced into a normalized vehicle behavior feature.

[0136] Optionally, when the cumulative feature is the driving mileage feature, a preset driving mileage threshold corresponding to the driving mileage feature is obtained; a first ratio of the driving mileage feature and the preset driving mileage threshold is taken as a normalized driving mileage feature corresponding to the driving mileage feature, wherein the preset driving mileage threshold can be 700,000 km or 650,000 km; when the cumulative feature is the full discharge frequency feature, a preset full discharge frequency threshold corresponding to the full discharge frequency feature is obtained, wherein the preset full discharge frequency threshold is obtained by a second ratio of the preset driving mileage threshold and an estimated driving mileage in a preset residual power range during the discharging process, the preset residual power range can be 0-100 or 10-90, and the estimated driving mileage can be 370 km or 350 km; a fourth ratio of the full discharge frequency feature and the preset full discharge frequency threshold is taken as a normalized full discharge frequency feature corresponding to the full discharge frequency feature; when the cumulative feature is the discharging duration feature, a preset discharging duration threshold corresponding to the discharging duration feature is obtained, wherein the preset discharging duration threshold is obtained by a fourth ratio of the driving mileage threshold and an estimated average driving speed, the estimated average driving speed can be 30 km / h or 35 km / h; a fifth ratio of the discharging duration feature and the preset discharging duration threshold is taken as a normalized discharging duration feature corresponding to the discharging duration feature; when the cumulative feature is the use time feature, a preset use time threshold corresponding to the use time feature is obtained; a sixth ratio of the use time feature and the preset use time threshold is taken as a normalized use time feature corresponding to the use time feature, and the preset use time threshold can be 7,200 days or 9,000 days.

[0137] By normalizing the training vehicle behavior feature, the influence of the battery pack characteristics of different batteries on the health degree is eliminated, so that the battery health state prediction model obtained by training can be migrated between different battery packs, thereby reducing the prediction limitation of the battery health state prediction model.

[0138] In step A40, a training health degree of the battery is predicted based on the normalized vehicle behavior feature and the battery health state prediction model to be trained.

[0139] In the embodiment, the normalized vehicle behavior feature is mapped into the training health degree of the battery by the battery health state prediction model to be trained.

[0140] Step A50, according to the training health degree and the true health degree, the battery health state prediction model to be trained is iteratively optimized to obtain the battery health state prediction model.

[0141] In this embodiment, according to the difference between the training health degree and the true health degree, the model loss corresponding to the battery health prediction model to be trained is calculated, and then it is judged whether the model loss converges. If the model loss converges, the battery health prediction model to be trained is taken as the battery health prediction model. If the model loss does not converge, the gradient calculated based on the model loss is used to update the battery health prediction model to be trained by a preset model updating method, wherein the preset model updating method is gradient descent method, gradient ascent method, etc.

[0142] In the training sample data, battery health feature data is extracted, wherein the battery health feature data is used to represent the health state of the training battery; the health state of the training battery is predicted by a preset health degree prediction model and the battery health feature data, to obtain the true health degree, wherein the battery health feature data includes but is not limited to current characteristics and power characteristics.

[0143] However, the health degree of the battery is predicted by the preset health degree prediction model, and the health degree of the battery is not embodied by mapping the battery health feature data to the health degree value of the battery. The health degree of the battery is predicted by the battery health feature data and the preset health degree prediction model, which is prone to inaccurate prediction of the health state of the battery due to the selection of fewer features representing the health state of the battery, and further leads to inaccurate acquisition of the true health degree of the training battery.

[0144] As a feasible embodiment, in the above step A10, the step of calculating the true health degree of the battery based on the training battery data can specifically include:

[0145] Step A11, selecting battery charging data satisfying a preset charging condition from the training sample data, wherein the battery charging data includes training battery current of the battery during the charging process, first residual power of the battery starting the charging process, second residual power of the battery ending the charging process, and rated battery capacity of the battery.

[0146] Step A12, determining the true health degree corresponding to the battery according to the training battery current, the first residual power, the second residual power, and the rated battery capacity.

[0147] It should be noted that in the embodiment, the remaining capacity is a state of charge (SOC, also called remaining capacity). The rated battery capacity is the capacity of the battery that can be used for long-term continuous work under the rated working condition, which is determined by the battery properties of the battery. The preset charging condition is a pre-set charging condition for judging the ideal charging condition of the battery.

[0148] In the embodiment, when the battery charging data satisfying the preset charging condition is selected from the training sample data, the training battery current in the charging process, the first remaining capacity at the beginning of the charging process, the second remaining capacity at the end of the charging process, and the rated battery capacity are obtained; the difference between the second remaining capacity and the first remaining capacity is obtained, and the real health degree corresponding to the training battery is obtained according to the ratio of the training battery current to the product of the difference and the rated battery capacity.

[0149] Optionally, the real health degree corresponding to the battery is determined according to the training battery current, the first remaining capacity, the second remaining capacity, and the rated battery capacity, which can be specifically:

[0150]

[0151] Wherein, SOH is the real health degree corresponding to the battery; t β is the time when the charging process of the battery ends, t α is the time when the charging process of the battery starts; SOC α is the first remaining capacity; SOC β is the second remaining capacity; Q 额 is the rated battery capacity.

[0152] According to the battery current, the remaining capacity and the rated battery capacity, the health degree of the battery under the ideal charging condition is calculated, which realizes accurate determination of the battery health state in a quantitative way, thereby improving the determination accuracy of the battery health state, and further improving the determination accuracy of the real health degree of the battery.

[0153] As a feasible embodiment, in the above step S11, the battery charging data satisfying the preset charging condition is selected from the training sample data, which can specifically include:

[0154] Selecting the first battery data with the standing time after the end of the charging process being greater than a preset time threshold from the training sample data;

[0155] Determining the second battery data with the difference between the second remaining capacity and the first remaining capacity in the first battery data being greater than a preset capacity threshold and the current at the end of the charging process being less than a preset current threshold as the battery charging data.

[0156] It should be noted that in the present embodiment, the preset time threshold is a preset static time critical value that has less influence on the accuracy of the battery voltage value measured after the charging is completed. The preset time threshold can be 30 minutes, or 35 minutes. The preset capacity threshold is a preset critical value of the difference between the second remaining capacity and the first remaining capacity that has less influence on the accuracy of the real health degree of the training battery. The preset capacity threshold can be 50, or 55. The preset current threshold is a preset current critical value that has less influence on the remaining capacity of the training battery. The preset current threshold can be 1C, or 0.8C.

[0157] In the present embodiment, the static time of each training sample data after the charging process is completed is obtained, and the first battery data with a static time greater than the preset time threshold is selected from the training battery data. The first remaining capacity of the start charging process, the current at the end of the charging process, and the second remaining capacity at the end of the charging process of the first battery data are obtained. The second battery data with a difference between the second remaining capacity and the first remaining capacity greater than the preset capacity threshold and a current less than the preset current threshold is selected from the first battery data as the battery charging data.

[0158] It can be understood that when the current of the battery during the charging process is too large, the calculated voltage of the battery deviates from the actual voltage of the battery, resulting in inaccurate calculated remaining voltage, and further resulting in low accuracy of the calculated real health degree of the battery. When the difference between the second remaining capacity and the first remaining capacity is small, the error is large, resulting in low accuracy of the calculated real health degree of the battery. When the static time of the battery after the end of the charging process is too short, the polarization resistance of the battery is large, the calculated voltage of the battery deviates from the actual voltage of the battery, resulting in inaccurate calculated remaining voltage, and further resulting in low accuracy of the calculated real health degree of the battery.

[0159] The present embodiment selects battery charging data that meets the preset charging condition, and the preset charging condition includes conditions for restricting the difference between the second remaining capacity and the first remaining capacity of the battery during the charging process, the static time of the battery after the end of the charging process, and the current at the end of the charging process. Thus, the low accuracy of the calculated real health degree of the battery due to the large current of the battery during the charging process, and / or due to the small difference between the second remaining capacity and the first remaining capacity, and / or due to the short static time of the battery after the end of the charging process is avoided, thereby improving the determination accuracy of the real health degree of the battery.

[0160] In addition, the present application also provides a vehicle as mentioned in any of the above embodiments.

[0161] Referring toFigure 11 , Figure 11 This is a schematic diagram of the device structure of the hardware operating environment of the vehicle mentioned in the embodiments of this application.

[0162] like Figure 11 As shown, the vehicle may include: a processor 1001, such as a CPU, a communication bus 1002, a network interface 1003, and a memory 1004. The communication bus 1002 is used to establish communication between the processor 1001 and the memory 1004. The memory 1004 may be a high-speed RAM or a stable, non-volatile memory, such as a disk drive. Optionally, the memory 1004 may also be a storage device independent of the aforementioned processor 1001.

[0163] Optionally, the vehicle may also include a rectangular user interface, a network interface, a camera, RF (Radio Frequency) circuitry, sensors, audio circuitry, a WiFi module, etc. The rectangular user interface may include a display screen and an input submodule such as a keyboard. Optionally, the rectangular user interface may also include a standard wired interface or a wireless interface. The network interface may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0164] Those skilled in the art will understand that Figure 11 The structure shown does not constitute a limitation on the vehicle. Based on different design needs of actual applications, the vehicle may of course include more or fewer components than shown in different feasible implementations, or combine certain components, or have different component arrangements.

[0165] like Figure 11 As shown, the memory 1004, serving as a storage medium, may include an operating system, a network communication module, and a battery health prediction program. The operating system manages and controls programs based on vehicle hardware and software resources, supporting the operation of the battery health prediction program and other software and / or programs. The network communication module enables communication between the various components within the memory 1004, as well as communication with other hardware and software in the battery health prediction system.

[0166] exist Figure 11 In the vehicle shown, the processor 1001 is used to execute the battery health state prediction program stored in the memory 1004 to implement the steps of the battery health state prediction method described in any of the above embodiments.

[0167] The specific implementation method of this application is basically the same as the various embodiments of the above-mentioned battery health status prediction method, and will not be described again here.

[0168] Furthermore, this application also provides a battery health state prediction device. The vehicle discharge control device of this application is used to control the prediction of battery health state in the vehicle, such as... Figure 12 As shown, the battery health status prediction device of this application includes:

[0169] The acquisition module is used to acquire vehicle battery data;

[0170] The extraction module is used to extract features from the battery data to obtain the vehicle usage behavior features corresponding to the vehicle.

[0171] The feature prediction module is used to predict the vehicle's predicted vehicle behavior features in the next time step based on the vehicle behavior features.

[0172] The health prediction module is used to predict the health of the battery in the vehicle based on the predicted vehicle behavior characteristics and the battery health status prediction model.

[0173] Optionally, the feature types of the vehicle usage behavior features include percentage-based features and cumulative features, and the feature prediction module is further used for:

[0174] Based on the preset linear fitting model and the vehicle usage behavior characteristics, the predicted cumulative features of the vehicle in the next time step are predicted.

[0175] The predicted vehicle usage behavior features for the next time step are obtained by aggregating the percentage-type features and the cumulative-type features.

[0176] Optionally, before the step of predicting the health of the battery in the vehicle based on the predicted vehicle behavior characteristics and the battery health status prediction model, the battery health status prediction device is further configured to:

[0177] Obtain the battery health status prediction model to be trained, the training sample data of the battery in the vehicle, and the actual health status of the battery calculated based on the training sample data;

[0178] Feature extraction is performed on the training sample data to obtain the training vehicle behavior features corresponding to the battery;

[0179] The training vehicle behavior features are normalized to obtain normalized vehicle behavior features;

[0180] Based on the normalized vehicle behavior characteristics and the battery health status prediction model to be trained, the training health of the battery is predicted.

[0181] Based on the training health status and the actual health status, the battery health status prediction model to be trained is iteratively optimized to obtain the battery health status prediction model.

[0182] Optionally, the feature content of the training vehicle behavior characteristics includes user behavior characteristics and battery performance characteristics, and the battery health status prediction device is further used for:

[0183] The cutoff time for obtaining the battery charging data used to calculate the true health status from the training battery data;

[0184] Select training battery data that meets the preset health status prediction conditions before the cutoff time from the training sample data.

[0185] User behavior features and battery performance features are extracted based on the training battery data.

[0186] Optionally, the battery health status prediction device is further used for:

[0187] Battery charging data that meets preset charging conditions is selected from the training sample data. The battery charging data includes the training battery current during the charging process, the first remaining charge of the training battery at the start of the charging process, the second remaining charge of the training battery at the end of the charging process, and the rated battery capacity of the battery.

[0188] The actual health status of the battery is determined based on the training battery current, the first remaining charge, the second remaining charge, and the rated battery capacity.

[0189] Optionally, the battery health status prediction device is further used for:

[0190] Select first battery data from the training sample data, wherein the first battery data is battery data whose resting time after the charging process ends is greater than a preset time threshold;

[0191] The second battery data in the first battery data is determined to be the battery charging data, wherein the second battery data is the battery data in which the difference between the second remaining power and the first remaining power is greater than a preset power threshold and the current at the end of the charging process is less than a preset current threshold.

[0192] Optionally, the battery health status prediction device is further used for:

[0193] Obtain the preset feature threshold corresponding to each of the training vehicle behavior features;

[0194] Normalized vehicle behavior features are obtained based on the ratio of each training vehicle behavior feature to the preset feature threshold.

[0195] The specific implementation methods of each functional module of the vehicle discharge control device of this application are basically the same as those of the above-described vehicle discharge control method embodiments, and will not be repeated here.

[0196] This application provides a computer storage medium that stores one or more programs, which can be executed by one or more processors to implement the steps of the vehicle discharge control method described in any of the above claims.

[0197] The specific implementation of the computer storage medium in this application is basically the same as the embodiments of the above-described vehicle discharge control method, and will not be repeated here.

[0198] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described vehicle discharge control method.

[0199] The specific implementation of the computer program product in this application is basically the same as the embodiments of the above-mentioned vehicle discharge control method, and will not be repeated here.

[0200] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0201] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0202] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be an in-vehicle computer, smartphone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0203] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A battery state of health prediction method, characterized by, The battery health state prediction method comprises: acquiring battery data of a vehicle; extracting features from the battery data to obtain vehicle use behavior features corresponding to the vehicle; predicting predicted vehicle use behavior features of the vehicle at a next time step according to the vehicle use behavior features, wherein the feature types of the vehicle use behavior features include proportion type features and accumulation type features, and the step of predicting the predicted vehicle use behavior features of the vehicle at the next time step according to the vehicle use behavior features comprises: predicting predicted accumulation type features of the vehicle at the next time step according to a preset linear fitting model and the vehicle use behavior features; and aggregating the proportion type features and the accumulation type features to obtain the predicted vehicle use behavior features of the vehicle at the next time step, wherein the accumulation type features include total mileage, full discharge cycle number, discharge duration and calendar life, and the proportion type features include charging current interval proportion and charging end SOC interval proportion; predicting a health degree of a battery in the vehicle according to the predicted vehicle use behavior features and a battery health state prediction model.

2. The battery state of health prediction method of claim 1, wherein, Before the step of predicting the health degree of the battery in the vehicle according to the predicted vehicle use behavior features and the battery health state prediction model, the method further comprises: acquiring a battery health state prediction model to be trained, training sample data of the battery in the vehicle, and a real health degree of the battery calculated based on the training sample data; extracting features from the training sample data to obtain training vehicle use behavior features corresponding to the battery; performing normalization processing on the training vehicle use behavior features to obtain normalized vehicle use behavior features; predicting a training health degree of the battery according to the normalized vehicle use behavior features and the battery health state prediction model to be trained; iteratively optimizing the battery health state prediction model to be trained according to the training health degree and the real health degree to obtain the battery health state prediction model.

3. The battery state of health prediction method of claim 2, wherein, The feature content of the training vehicle use behavior features includes user behavior features and battery performance features, and the step of extracting features from the training sample data to obtain the training vehicle use behavior features corresponding to the battery comprises: acquiring a cutoff time of battery charging data used to calculate the real health degree in the training battery data; selecting training battery data satisfying a preset health state prediction condition before the cutoff time in the training sample data; extracting user behavior features and battery performance features from the training battery data.

4. The battery state of health prediction method of claim 2, wherein, The step of calculating the real health degree of the battery based on the training battery data comprises: selecting battery charging data satisfying a preset charging condition in the training sample data, wherein the battery charging data includes training battery current of the battery during a charging process, first residual capacity of the training battery at the beginning of the charging process, second residual capacity of the training battery at the end of the charging process, and rated battery capacity of the battery; According to the training battery current, the first residual power, the second residual power, and the rated battery capacity, a real health degree corresponding to the battery is determined.

5. The battery state of health prediction method of claim 4, wherein, The step of selecting battery charging data satisfying a preset charging condition from the training sample data comprises: selecting first battery data from the training sample data, wherein the first battery data is battery data with a static duration after the end of the charging process greater than a preset duration threshold; determining second battery data in the first battery data as the battery charging data, wherein the second battery data is battery data with a difference between the second residual power and the first residual power greater than a preset power threshold and a current at the end of the charging process less than a preset current threshold.

6. The battery state of health prediction method of claim 2, wherein, The step of normalizing the training vehicle behavior features to obtain normalized vehicle behavior features comprises: obtaining a preset feature threshold corresponding to each of the training vehicle behavior features; obtaining normalized vehicle behavior features according to a ratio of each of the training vehicle behavior features to the preset feature threshold.

7. A battery state of health prediction apparatus characterized by comprising: The battery health state prediction device comprises: an acquisition module configured to acquire battery data of a vehicle; an extraction module configured to perform feature extraction on the battery data to obtain vehicle behavior features corresponding to the vehicle; a feature prediction module configured to predict, according to the vehicle behavior features, predicted vehicle behavior features of the vehicle at a next time step, wherein a feature type of the vehicle behavior features comprises a proportion type feature and an accumulation type feature, and the step of predicting, according to the vehicle behavior features, predicted vehicle behavior features of the vehicle at a next time step comprises: predicting, according to a preset linear fitting model and the vehicle behavior features, predicted accumulation type features of the vehicle at a next time step; and aggregating the proportion type features and the accumulation type features to obtain predicted vehicle behavior features of the vehicle at a next time step, wherein the accumulation type features comprise total mileage, full discharge cycle number, discharge duration, and calendar life, and the proportion type features comprise a charging current interval proportion and a proportion of a charging end SOC interval; a health degree prediction module configured to predict, according to the predicted vehicle behavior features and a battery health state prediction model, a health degree of a battery in the vehicle.

8. An electronic device, comprising: The electronic device comprises: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the steps of the battery health state prediction method in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program for implementing a battery health state prediction method, and the program is executed by a processor to implement the steps of the battery health state prediction method in any one of claims 1 to 6.

Images