Battery health degree evaluation method and life prediction method based on internet-of-things charging pile

Abstract

The application relates to the technical field of charging piles, and particularly discloses a battery health degree evaluation method and a life prediction method based on an Internet of Things charging pile, wherein the evaluation method comprises the following steps: obtaining a local electric parameter change curve generated when a charging pile charges a battery; generating a charging power curve graph according to the local electric parameter change curve by using a pre-constructed charging curve model; calculating the current total capacity of the battery according to the charging power curve graph; and evaluating the battery health degree according to the current total capacity and the standard total capacity of the battery. The evaluation method is based on the local electric parameter change curve, the charging power curve graph representing the entire complete charging process of the battery is obtained by combining the charging curve model, and the current total capacity of the battery is calculated based on the charging power curve graph, so that the evaluation of the battery health degree of an electric vehicle is realized, and the evaluation method can be applied to a charging pile service platform to realize the batch, global and remote evaluation of the battery health degree of the electric vehicle.

Description

Technical Field

[0001] This application relates to the field of charging pile technology, and more specifically, to a battery health assessment method and lifespan prediction method based on Internet of Things (IoT) charging piles. Background Technology

[0002] The existing charging pile management platform is based on the Internet of Things (IoT) to realize regional detection and control and terminal power supply control. The IoT builds a communication network between charging piles, servers and users to inform users of the actual charging status of electric vehicles, and enables operators to monitor and adjust the power supply status of charging piles in real time based on the server.

[0003] Existing charging station management platforms can only monitor whether the electric vehicle battery is fully charged in order to charge the electric vehicle, but cannot monitor the battery health status. Users cannot know the battery health status except by using specialized testing equipment.

[0004] There is currently no effective technical solution to the above problems. Summary of the Invention

[0005] The purpose of this application is to provide a battery health assessment method and lifespan prediction method based on Internet of Things charging piles, which realizes the assessment of the battery health of electric vehicles based on the charging data of charging piles.

[0006] Firstly, this application provides a battery health assessment method based on IoT charging piles for assessing the battery health of electric vehicles. The assessment method includes the following steps:

[0007] Obtain the local electrical parameter change curve generated when the charging pile charges the battery;

[0008] A charging power curve is generated based on the local electrical parameter variation curve using a pre-built charging curve model.

[0009] Calculate the current total capacity of the battery based on the charging power curve;

[0010] The battery health is assessed based on the current total capacity and the battery's standard total capacity.

[0011] The battery health assessment method based on IoT charging piles in this application is based on the local electrical parameter change curve generated by the charging pile when the electric vehicle is charging. It combines the charging curve model to obtain a charging power curve that can represent the entire complete charging process of the battery, and uses this as a benchmark to calculate the current total capacity of the battery, thereby realizing the assessment of the electric vehicle battery health. The assessment method can be applied in the charging pile service platform to realize the batch, global and remote assessment of the electric vehicle battery health.

[0012] The battery health assessment method based on IoT charging piles includes a charging curve model comprising a charging stage classification model and multiple curve reconstruction models corresponding to different charging stages. The charging stage classification model is used to identify the charging stage in which the local electrical parameter change curve is located, and to place the local electrical parameter change curve into the corresponding curve reconstruction model to generate the charging power curve. The curve reconstruction model is used to generate a charging power curve containing all charging stages based on the corresponding local electrical parameter change curve.

[0013] The battery health assessment method based on IoT charging piles in this application classifies the local electrical parameter change curves based on the charging stage classification model. It uses the curve reconstruction model of the corresponding charging stage to realize the dedicated reconstruction of the charging power curve, which can avoid excessive differences in the charging power curve reconstructed by the charging curve model and ensure the accuracy of battery health assessment.

[0014] The battery health assessment method based on IoT charging piles, wherein the curve reconstruction model is a curve generator model obtained by training a generative adversarial network.

[0015] The battery health assessment method based on IoT charging piles includes three curve reconstruction models, which are used to input local electrical parameter change curves during the constant current charging stage, constant voltage charging stage, and float charging stage to generate the charging power curve.

[0016] The battery health assessment method based on IoT charging piles, wherein the step of calculating the current total capacity of the battery based on the charging power curve includes:

[0017] A fused power curve is obtained based on at least two charging power curves.

[0018] The current total capacity of the battery is calculated based on the fused power curve.

[0019] The battery health assessment method based on IoT charging piles, wherein the at least two charging power curves are generated by using different curve reconstruction models based on the local electrical parameter change curves of the corresponding charging stages.

[0020] The battery health assessment method based on IoT charging piles, wherein the standard total capacity is obtained based on user input data, or based on video recognition of electric vehicle type analysis, or is the current total capacity obtained from the first calculation of the corresponding battery.

[0021] Secondly, this application also provides a battery life prediction method based on IoT charging piles, used to predict the remaining battery life of electric vehicles. The prediction method includes the following steps:

[0022] Obtain the local electrical parameter change curve generated when the charging pile charges the battery;

[0023] A charging power curve is generated based on the local electrical parameter variation curve using a pre-built charging curve model.

[0024] Calculate the current total capacity of the battery based on the charging power curve;

[0025] The battery health is assessed based on the current total capacity and the battery's standard total capacity.

[0026] Obtain at least one historical total capacity from the historical charging database;

[0027] Battery loss rate information is obtained based on the current total capacity and the historical total capacity;

[0028] The battery life information is calculated based on the battery loss rate information and the battery health information.

[0029] The battery life prediction method based on IoT charging piles in this application can obtain the current total capacity and battery health by analyzing the local electrical parameter change curve generated by the charging pile when the electric vehicle is charging. It can also obtain battery loss rate information by combining the historical total capacity generated by historical charging behavior, so as to realize batch, global and remote prediction of electric vehicle battery life.

[0030] The battery life prediction method based on IoT charging piles, wherein the battery loss rate information is the battery loss efficiency per unit natural day, and the lifespan information is the number of natural days in which the battery health deteriorates to an unhealthy state.

[0031] This method combines the user's electric vehicle usage and charging habits to determine the electric vehicle's usable time, thus informing the user of the electric vehicle's battery lifespan from a lifespan perspective.

[0032] The battery life prediction method based on IoT charging piles, wherein the battery loss rate information is the trend curve of the battery health with respect to the number of battery charging cycles, and the lifespan information is the number of battery charging cycles required for the battery health to drop to an unhealthy state.

[0033] As can be seen from the above, this application provides a battery health assessment method and lifespan prediction method based on IoT charging piles. The battery health assessment method based on IoT charging piles is based on the local electrical parameter change curve generated by the charging pile when the electric vehicle is charging. It combines the charging curve model to obtain a charging power curve that can represent the entire complete charging process of the battery. Based on this, the current total capacity of the battery is calculated, thereby realizing the assessment of the electric vehicle battery health. The assessment method can be applied in the charging pile service platform to realize the batch, global and remote assessment of the electric vehicle battery health. Attached Figure Description

[0034] Figure 1 A flowchart of a battery health assessment method based on an IoT charging pile provided in this application embodiment.

[0035] Figure 2 The charging power curve provided for the embodiments of this application.

[0036] Figure 3 A flowchart illustrating a battery life prediction method based on IoT charging piles provided in this application embodiment. Detailed Implementation

[0037] The technical solutions of the embodiments of this 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 this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0038] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0039] Firstly, please refer to Figure 1 This application provides a battery health assessment method based on IoT charging piles for assessing the battery health of electric vehicles. The assessment method includes the following steps:

[0040] A1. Obtain the local electrical parameter change curves generated when the charging pile charges the battery;

[0041] A2. Generate a charging power curve based on the local electrical parameter variation curve using a pre-built charging curve model;

[0042] A3. Calculate the current total capacity of the battery based on the charging power curve;

[0043] A4. Assess battery health based on current total capacity and standard total capacity of the battery.

[0044] Specifically, the local electrical parameter change curve is the curve data of the electrical parameter change with respect to time. It can be the change curve of a type of electrical parameter with respect to a certain continuous time period after the charging pile is started, or it can be the change curve reconstructed from the electrical parameter data collected based on a preset collection interval after the charging pile is started. The electrical parameter is related to the charging status of the electric vehicle, so that the local electrical parameter change curve can reflect the charging progress and charging effectiveness of the electric vehicle.

[0045] More specifically, the local electrical parameter change curve is directly obtained based on the output status of the charging pile. It can be obtained based on sensors built into the charging pile or sensors installed on the charging end of the charging pile, and then uploaded to the server through the Internet of Things (IoT) communication network. The battery health assessment method based on IoT charging piles in this application embodiment is preferably a data analysis method loaded on the server, and the battery health obtained from the data analysis can be sent to the user terminal through the IoT communication network.

[0046] More specifically, such as Figure 2As shown, existing electric vehicles generally use a three-stage charger for charging. The charging process typically includes a constant current charging stage, a constant voltage charging stage, and a float charging stage. During the constant current charging stage, the charger detects the battery capacity and adaptively determines the charging current value and the switching time to constant voltage charging (simultaneously determining the voltage rise efficiency in the constant current charging stage based on the charging voltage value in the constant voltage charging stage). In the constant voltage charging stage, the charger can adaptively adjust the charging current according to the battery's state of charge and adjust the timing of entering the float charging stage based on the actual amount of charge received. During the float charging stage, the charger's indicator light typically changes from red to green, and a lower charging current is used to charge the battery, taking advantage of the remaining charge. Therefore, corresponding to different battery capacities, different electrical parameters exhibit different characteristics in different charging stages. These electrical parameters can be charging current, charging voltage, and... One or more of the charging power, i.e., the local electrical parameter change curve is a segment of the local change curve of one or more of the charging current, charging voltage, and charging power with respect to time. In conjunction with the foregoing, it can be seen that the electrical parameter data of the three charging stages of the three-stage charger affect each other. Therefore, the local electrical parameter change curve in one or more stages can serve as the basis for constructing a complete electrical parameter change curve. Among them, the information on the change of charging power with respect to time represents the amount of electricity charged into the battery. Therefore, the battery health assessment method based on IoT charging piles in this application requires obtaining a charging power curve to analyze the amount of electricity charged into the battery. Moreover, the charging power has significantly different change characteristics in the three charging stages. Therefore, the local electrical parameter change curve obtained in step A1 is preferably a local electrical parameter change curve with respect to the output power of the charging pile (corresponding to the charging power), so that step A2 can obtain a charging power curve directly related to the charging power.

[0047] More specifically, in this embodiment, the charging power curve is a curve showing the change in charging power over time as the battery charge increases from zero to full. When an electric vehicle is charging, its initial charge is generally not zero. Without the electric vehicle's internal charge detection system, the charging pile cannot directly obtain the electric vehicle's charge level, nor can it obtain relevant charge data from the charger. Even with real-time power change detection, the initial charge level of the battery cannot be obtained. The battery health assessment method based on IoT charging piles in this embodiment utilizes a pre-built charging curve model to generate a complete charging power curve based on the input local electrical parameter change curve. This complete charging power curve is used to analyze the amount of electricity consumed during the charging process from zero to full charge, thereby determining the battery's current total capacity.

[0048] More specifically, the pre-constructed charging curve model is a trained model. It can be a curve reconstruction model that generates a complete charging power curve based on the local electrical parameter change curve, or a feature correction model that extracts curve features from the local electrical parameter change curve and corrects the initial charging power curve to obtain a complete charging power curve, or a database matching model that performs feature matching based on a preset charging power curve library to retrieve a suitable charging power curve. Since the battery characteristics of different electric vehicles vary greatly and the loss patterns are very different, the charging curve model in the battery health assessment method based on IoT charging piles in this application embodiment is preferably a curve reconstruction model, so as to generate a matching charging power curve based on the curve features of the local electrical parameter change curve, ensuring the accuracy of the subsequent calculation of the current total capacity of the battery.

[0049] More specifically, the standard total capacity is the rated battery capacity of the corresponding electric vehicle battery, that is, the maximum battery capacity when no loss occurs, and the battery health is the percentage of the current total capacity to the standard total capacity.

[0050] The battery health assessment method based on IoT charging piles in this application is based on the local electrical parameter change curve generated by the charging pile when the electric vehicle is charging. It combines the charging curve model to obtain a charging power curve that can represent the entire complete charging process of the battery, and uses this as a benchmark to calculate the current total capacity of the battery, thereby realizing the assessment of the electric vehicle battery health. The assessment method can be applied in the charging pile service platform to realize the batch, global and remote assessment of the electric vehicle battery health.

[0051] In some preferred embodiments, the charging curve model includes a charging stage classification model and multiple curve reconstruction models corresponding to different charging stages. The charging stage classification model is used to identify the charging stage in which the local electrical parameter change curve is located, and to place the local electrical parameter change curve into the corresponding curve reconstruction model to generate a charging power curve. The curve reconstruction model is used to generate a charging power curve containing all charging stages based on the corresponding local electrical parameter change curve.

[0052] Specifically, based on the foregoing, the electrical parameters in the constant current charging stage, constant voltage charging stage, and float charging stage are mutually constrained and have significant numerical differences. The charging curve model, as a curve reconstruction model, may not be able to determine the stage of local electrical parameter changes. Incorrect determination can lead to significant deviations in the charging power curve, resulting in a failed battery health assessment. Therefore, the battery health assessment method based on IoT charging piles designs the charging curve model to include a charging stage classification model and multiple curve reconstruction models corresponding to different charging stages. The operation process of the charging curve model in step A2 includes:

[0053] C1. Classify the local electrical parameter change curves based on the charging stage classification model to determine the stage type of the charging stage in which the local electrical parameter change curves are located;

[0054] C2. Input the local electrical parameter change curve into the curve reconstruction model of the corresponding stage type so that the curve reconstruction model can reconstruct and generate a complete charging power curve.

[0055] It should be noted that the local electrical parameter change curve obtained in step A1 may span two charging stages. This situation will cause step C1 to fail to classify, generate classification failure information, and trigger step A1 to be re-executed.

[0056] Specifically, the battery health assessment method based on IoT charging piles in this application classifies the local electrical parameter change curves based on the charging stage classification model, so as to realize the dedicated reconstruction of the charging power curve using the curve reconstruction model of the corresponding charging stage. This can avoid excessive differences in the charging power curve reconstructed by the charging curve model and ensure the accuracy of battery health assessment.

[0057] More specifically, based on the foregoing, the changes in electrical parameters at different charging stages are significantly different. Therefore, the charging stage classification model can be trained using a general graphical classification model, that is, a classification model that classifies based on the curve features of the corresponding local electrical parameter change curves, such as HMM model, ROC curve model, SVM classification model, etc.

[0058] In some preferred embodiments, the curve reconstruction model is a curve generator model obtained by training a generative adversarial network.

[0059] Specifically, Generative Adversarial Networks (GANs) consist of a discriminator and a generator. Their training process generally includes a discriminator training process and a generator training process, which is a common training method and will not be described in detail here. In the embodiments of this application, the trained generator (i.e., the curve generator model) is a curve reconstruction model.

[0060] In some preferred embodiments, based on the foregoing, existing electric vehicle battery charging stages are three. Therefore, in the embodiments of this application, the curve reconstruction model is preferably three, which are used to insert the local electrical parameter change curves in the constant current charging stage, constant voltage charging stage and float charging stage to generate a charging power curve.

[0061] Specifically, the curve reconstruction models used for different charging stages can be trained separately. This can be done by extracting local electrical parameter change curves as input data for different charging stages based on a complete charging power curve, and using the complete charging power curve as a label for training each curve reconstruction model.

[0062] More specifically, the local electrical parameter change curves input during training can be randomly selected within a preset time range to improve the robustness, accuracy, and stability of the curve reconstruction model.

[0063] More specifically, the time length of the local electrical parameter change curve obtained in step A1 should be within the preset time range during the training process, preferably 5-15 minutes, so that the charging power curve generated by the curve reconstruction model is sufficiently accurate.

[0064] In some preferred embodiments, the step of calculating the current total capacity of the battery based on the charging power curve includes:

[0065] A31. Obtain a fused power curve based on at least two charging power curves;

[0066] A32. Calculate the current total capacity of the battery based on the fused power curve.

[0067] Specifically, in order to further improve the accuracy of calculating the current total capacity of the battery, the battery health assessment method based on IoT charging piles in this application embodiment is preferably based on obtaining at least two charging power curves in steps A1-A2, and calculating the current total capacity of the battery based on the fused power curve obtained by fusing the at least two charging power curves.

[0068] In some preferred embodiments, at least two charging power curves are generated by using different curve reconstruction models based on the local electrical parameter change curves of the corresponding charging stages.

[0069] Specifically, different charging power curves can be generated based on local electrical parameter change curves corresponding to different charging stages, or they can be generated based on local electrical parameter change curves corresponding to the same charging stage. In this embodiment, the former is preferred, and three charging power curves are preferred, so that the battery health assessment method based on IoT charging piles in this embodiment can fuse the charging power curves generated by the data from the three stages to obtain a fused power curve to more accurately calculate the current total capacity of the battery and accurately assess the battery health.

[0070] More specifically, in the embodiments of this application, the fused power curve is preferably an average curve generated by fusing the average values ​​of the charging power curves from multiple charging power curves.

[0071] In some other embodiments, step A4 includes:

[0072] A41. Obtain the average total capacity based on at least two current total capacities;

[0073] A42. Assess battery health based on the average total capacity and the standard total capacity of the battery.

[0074] Specifically, the battery health assessment method based on IoT charging piles in this embodiment is preferably based on obtaining at least two current total capacities in steps A1-A3, and assessing battery health based on the average total capacity obtained from the average of the at least two current total capacities.

[0075] More specifically, the different current total capacity can be the current total capacity calculated from the charging power curve generated based on the local electrical parameter change curves corresponding to different charging stages, or it can be the current total capacity calculated from the charging power curve generated based on the local electrical parameter change curves corresponding to the same charging stage. In the embodiments of this application, the former is preferred, and more preferably there are three current total capacities, so that the battery health assessment method based on IoT charging piles in this embodiment can integrate the current total capacity calculated from the charging power curve generated by the three stages of data to obtain the average total capacity in order to more accurately assess battery health.

[0076] In some preferred embodiments, the standard total capacity is obtained based on user input data, or based on video recognition analysis of electric vehicle type, or as the current total capacity obtained from the first calculation of the corresponding battery.

[0077] Specifically, the standard total capacity can be determined by the user through data matching based on the product model entered by the user terminal of the charging pile management platform, or by directly entering the battery specifications. Alternatively, it can be determined by using images of electric vehicles obtained from monitoring equipment near the charging pile and determining the brand and model of the electric vehicle through image analysis and matching. Or it can be determined solely by the current total capacity analyzed and recorded when charging is first performed using the charging pile management platform.

[0078] Secondly, please refer to Figure 3 Some embodiments of this application also provide a battery life prediction method based on IoT charging piles, used to predict the remaining battery life of electric vehicles. The prediction method includes the following steps:

[0079] B1. Obtain the local electrical parameter change curves generated when the charging pile charges the battery;

[0080] B2. Generate a charging power curve based on the local electrical parameter variation curve using a pre-built charging curve model;

[0081] B3. Calculate the current total capacity of the battery based on the charging power curve;

[0082] B4. Assess battery health based on current total capacity and standard total capacity of the battery;

[0083] B5. Obtain the historical total capacity based on the historical charging database;

[0084] B6. Obtain battery loss rate information based on the current total capacity and historical total capacity;

[0085] B7. Calculate and obtain battery life information based on battery loss rate information and battery health status.

[0086] Specifically, steps B1-B4 are equivalent to using the battery health assessment method based on IoT charging piles provided in the first aspect to determine the battery health; on this basis, the historical charging database in step B5 records the current total capacity of the corresponding electric vehicle battery obtained in the past using steps B1-B5, and these past current total capacities are the historical total capacities generated by the corresponding charging behavior.

[0087] More specifically, the historical total capacity represents the battery's past capacity before the current charging behavior. Therefore, step B6 can calculate the battery loss rate information based on at least one historical total capacity and the current total capacity obtained based on the analysis in steps B1-B3. The battery loss rate information can be the battery loss rate between the current battery charging behavior and the previous charging behavior (users generally choose charging piles under the same area or the same operator to charge electric vehicles), or it can be multiple battery loss rates generated by combining the current battery charging behavior and multiple past charging behaviors, or it can be a battery loss change curve composed of these multiple battery loss rates.

[0088] More specifically, battery health represents the percentage of the battery's current maximum charge to its factory-calibrated total capacity, while battery loss rate information reflects the rate or trend of battery loss. Therefore, step B7 combines both to predict and analyze the remaining usable lifespan of the electric vehicle battery. In other words, step B7 calculates the remaining lifespan of the battery before it becomes unhealthy or obsolete, based on the battery health information obtained in step B4, which represents the current state of the battery, and the rate of decrease in the battery loss rate information obtained in step B6, which represents the current changes in battery loss.

[0089] The battery life prediction method based on IoT charging piles in this application can analyze the local electrical parameter change curve generated by the charging pile when the electric vehicle is charging to obtain the current total capacity and battery health of the battery, and can combine the historical total capacity generated by historical charging behavior to obtain battery loss rate information, so as to realize batch, global and remote prediction of electric vehicle battery life.

[0090] Specifically, lifespan information is data used to reflect the remaining usability of a battery. It can be the number of times the battery can be charged, the number of complete charge and discharge cycles, or the date of use, etc.

[0091] In some preferred embodiments, the battery loss rate information is the battery loss efficiency per unit natural day, and the lifespan information is the number of natural days in which the battery health deteriorates to an unhealthy state.

[0092] Specifically, since users' electric vehicle usage behavior has a periodicity that varies with time, this implementation method is equivalent to combining users' electric vehicle usage habits and charging habits to determine the usable time of the electric vehicle, so as to inform users of the battery life information of the electric vehicle from the perspective of living time.

[0093] Specifically, the unhealthy state is determined based on a preset battery health threshold. A battery with a health level below the battery health threshold is in an unhealthy state, while a battery with a health level above or equal to the battery health threshold is in a healthy state.

[0094] More specifically, the battery health threshold can be set based on user settings, electric vehicle safety usage guidelines, or national or industry standards.

[0095] In some preferred embodiments, the battery loss rate information is a trend curve of battery health with respect to the number of battery charging cycles, and the lifespan information is the number of battery charging cycles required for the battery health to decline to an unhealthy state.

[0096] Specifically, users' electric vehicle usage behavior has a periodicity regarding the remaining battery power. For example, users generally charge the battery to full capacity when the battery power is below a certain percentage. Therefore, this implementation method is equivalent to combining the user's electric vehicle usage habits and charging habits to determine the number of times the electric vehicle can be charged, so as to inform the user of the electric vehicle battery life information from the perspective of charging behavior.

[0097] In summary, this application provides a battery health assessment method and lifespan prediction method based on IoT charging piles. The battery health assessment method based on IoT charging piles combines the local electrical parameter change curves generated by the charging pile during electric vehicle charging with a charging curve model to obtain a charging power curve that represents the entire complete charging process of the battery. Based on this, the current total capacity of the battery is calculated, thereby realizing the assessment of the electric vehicle battery health. The assessment method can be applied in a charging pile service platform to achieve batch, global, and remote assessment of electric vehicle battery health.

[0098] In the embodiments provided in this application, it should be understood that the disclosed methods can be implemented in other ways.

[0099] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0100] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A battery health assessment method based on IoT charging piles, used to assess the battery health of electric vehicles, characterized in that, The evaluation method includes the following steps: Obtain the local electrical parameter change curve generated when the charging pile charges the battery; A charging power curve is generated based on the local electrical parameter variation curve using a pre-built charging curve model. Calculate the current total capacity of the battery based on the charging power curve; The battery health is assessed based on the current total capacity and the battery's standard total capacity. The charging curve model includes a charging stage classification model and multiple curve reconstruction models corresponding to different charging stages. The charging stage classification model is used to identify the charging stage in which the local electrical parameter change curve is located, and to place the local electrical parameter change curve into the corresponding curve reconstruction model to generate the charging power curve. The curve reconstruction model is used to generate a charging power curve that includes all charging stages based on the corresponding local electrical parameter change curve.

2. The battery health assessment method based on IoT charging piles according to claim 1, characterized in that, The curve reconstruction model is a curve generator model obtained by training a generative adversarial network.

3. The battery health assessment method based on IoT charging piles according to claim 1, characterized in that, The curve reconstruction model consists of three parts, which are used to input the local electrical parameter change curves during the constant current charging stage, constant voltage charging stage, and float charging stage to generate the charging power curve.

4. The battery health assessment method based on IoT charging piles according to claim 1, characterized in that, The step of calculating the current total capacity of the battery based on the charging power curve includes: A fused power curve is obtained based on at least two charging power curves. The current total capacity of the battery is calculated based on the fused power curve.

5. The battery health assessment method based on IoT charging piles according to claim 4, characterized in that, The at least two charging power curves are generated by using different curve reconstruction models based on the local electrical parameter change curves of the corresponding charging stages.

6. The battery health assessment method based on IoT charging piles according to claim 1, characterized in that, The standard total capacity is obtained based on user input data, or based on video recognition analysis of electric vehicle type, or is the current total capacity obtained from the first calculation of the corresponding battery.

7. A battery life prediction method based on IoT charging piles, used to predict the remaining battery life of electric vehicles, characterized in that, The prediction method includes the following steps: Obtain the local electrical parameter change curve generated when the charging pile charges the battery; A charging power curve is generated based on the local electrical parameter variation curve using a pre-built charging curve model. Calculate the current total capacity of the battery based on the charging power curve; The battery health is assessed based on the current total capacity and the battery's standard total capacity. Obtain at least one historical total capacity from the historical charging database; Battery loss rate information is obtained based on the current total capacity and the historical total capacity; The battery's lifespan information is calculated based on the battery loss rate information and the battery health information. The charging curve model includes a charging stage classification model and multiple curve reconstruction models corresponding to different charging stages. The charging stage classification model is used to identify the charging stage in which the local electrical parameter change curve is located, and to place the local electrical parameter change curve into the corresponding curve reconstruction model to generate the charging power curve. The curve reconstruction model is used to generate a charging power curve that includes all charging stages based on the corresponding local electrical parameter change curve.

8. The battery life prediction method based on IoT charging piles according to claim 7, characterized in that, The battery loss rate information is the battery loss efficiency per unit of natural day, and the lifespan information is the number of natural days in which the battery health deteriorates to an unhealthy state.

9. The battery life prediction method based on IoT charging piles according to claim 7, characterized in that, The battery loss rate information is the trend curve of the battery health with respect to the number of battery charging cycles, and the lifespan information is the number of battery charging cycles required for the battery health to drop to an unhealthy state.

Images