Method and system for determining service life of battery based on charging data

Abstract

The invention provides a method and system for determining battery life based on charging data, and belongs to the technical field of battery life prediction. The method comprises the following steps: S100, when ith charging is finished, if the charging proportion reaches a first preset proportion value, recording an ith charging data set and an ith endurance value; s200, when the j-th charging is finished, if the charging proportion reaches a first preset proportion value, recording a j-th charging data set and a j-th endurance value; s300, when the difference value between the ith endurance value and the jth endurance value and the sub-stage overlap ratio are both larger than a preset threshold value, entering the step S400; and S400, uploading the ith charging data set, the jth charging data set and the interval driving data to a cloud server to determine the service life of the battery. According to the method, the charging time node and the driving node which may influence the service life of the battery are determined firstly, so that the collection and transmission of related charging data for battery prediction are more targeted, and the operation pressure of a server model is relieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of battery life prediction, and in particular, relates to a method and system for determining battery life based on charging data. Background Art

[0002] Accurate prediction of the life of power batteries, especially lithium batteries, is crucial for the normal operation of electrical equipment (such as new energy vehicles). However, due to the nonlinearity of the battery capacity degradation process and the uncertainty of operating conditions, accurate prediction of battery life faces challenges. Since the operating conditions of power batteries on different new energy vehicles are highly complex, their degradation rates vary greatly with temperature, driving habits and charging habits. This also leads to a significant increase in the uncertainty of vehicle range (an indicator closely related to battery life). Drivers' mileage anxiety cannot be eliminated, which will further hinder the development of the new energy vehicle industry.

[0003] With the development of technologies such as machine learning, artificial intelligence, and deep learning, the prior art has reported a variety of battery life (remaining endurance of power batteries) prediction methods and has been successfully applied. At the same time, based on the support of the Internet of Things / Internet of Vehicles, new energy vehicles are gradually becoming intelligent and cloud-based. Users can collect and upload data on the vehicle side to the cloud backend server at any time to detect the current battery health status for battery life detection, and can also go to a professional testing agency for life testing. For example, Chinese invention patent application CN202410034809 proposes a method for predicting the remaining endurance of a power battery based on a time series large model algorithm. The model is established with vehicle life cycle data. Through the combination of data features of the static, driving, and charging processes, the vehicle's power battery can be more comprehensively characterized under various operating conditions. The decline in life.

[0004] As the number of new energy vehicles continues to grow and battery life detection methods become more and more convenient, users are paying more and more attention to the health status of batteries. When the remaining power indicators displayed by the vehicle system (including cruising range, charging ratio, power consumption data, charging time, etc.) change, users may become psychologically anxious and frequently monitor the battery health status. That is, a large number of vehicle terminal users frequently upload local data to the cloud with the intention of performing life prediction. This is even more true in charging scenarios under different hardware conditions, which puts great pressure on the operation of the cloud backend server model of related car dealers.

[0005] However, in fact, battery life (full-charge endurance) is greatly affected by operating conditions, driving speed, temperature, charging habits, and charging environment. Not every change in the remaining power indicators (including cruising range, charging ratio, power consumption data, charging time, etc.) displayed by the car system means a change in the battery health status. Most of the time, the car system is still in a normal range of changes, and users do not need to upload data or frequently request to go to the store for inspection. In addition, some new energy vehicles also support battery replacement (i.e. battery replacement). Even under the same operating conditions, the battery before and after replacement will have normal differences in the displayed cruising range. If users frequently upload relevant data that is not related to life to the large model, based on the influence of iteration errors, it is also possible that the training accuracy of the large model will be reduced, and the accuracy of the output results will also be affected.

[0006] Therefore, how to identify the charging time nodes and driving nodes that are more likely to affect the battery life, so as to make the collection and transmission of relevant charging data used for battery prediction more targeted, thereby reducing the operating pressure of the server model, is one of the technical issues that need to be considered when determining the battery life based on charging data. Summary of the invention

[0007] In response to the above technical problems, the present invention proposes a method and system for determining battery life based on charging data. By first determining the charging time nodes and driving nodes that may affect the battery life, the collection and transmission of relevant charging data used for battery prediction are more targeted, thereby reducing the operating pressure of the server model.

[0008] In a first aspect of the present invention, a method for determining battery life based on charging data is provided, wherein the method is applied to a vehicle system.

[0009] The method comprises the following steps:

[0010] S100: At the end of the i-th charging, if the charging ratio displayed by the vehicle system reaches a first preset ratio value, then recording the i-th charging data set and the i-th endurance value displayed by the vehicle system at the end of the i-th charging;

[0011] S200: At the end of the jth charging, if the charging ratio displayed by the vehicle system reaches the first preset ratio value, record the jth charging data set and the jth endurance value displayed by the vehicle system at the end of the jth charging;

[0012] S300: Calculating the difference between the i-th endurance value and the j-th endurance value and the sub-phase overlap between the i-th charging data set and the j-th charging data set;

[0013] When the difference is greater than a first preset threshold and the sub-phase overlap between the i-th charging data set and the j-th charging data set is greater than a second preset threshold, proceeding to step S400;

[0014] When the difference is smaller than the first preset threshold, or the overlap degree is smaller than the second preset threshold, the process returns to step S100.

[0015] S400: uploading the i-th charging data set, the j-th charging data set and the interval driving data to a cloud server through the vehicle system;

[0016] The cloud server determines the battery life based on the i-th charging data set, the j-th charging data set, and the interval driving data.

[0017] The charging starting ratio of the i-th charging process and the j-th charging process is the same;

[0018] The i-th cruising range value displayed by the vehicle computer at the end of the i-th charging and the j-th cruising range value displayed by the vehicle computer at the end of the j-th charging are both determined based on the latest historical average interval driving data.

[0019] The multiple charging sub-stages included in the i-th charging process in step S100 are divided according to the following method:

[0020] Preset K split ratio points to divide the i-th charging process into K+1 charging sub-stages, where K≥2;

[0021] The multiple charging sub-stages included in the j-th charging process in step S200 are divided according to the following method:

[0022] The j-th charging process is divided into K+1 charging sub-stages according to the preset K segmentation ratio points.

[0023] The cloud server determines the battery life based on the i-th charging data set, the j-th charging data set and the interval driving data, specifically including:

[0024] Aligning the i-th charging data set and the j-th charging data set according to the K+1 charging sub-stages;

[0025] Determine the battery attenuation ratio based on the registered charging data set and the interval driving data;

[0026] The battery life is determined based on the battery attenuation ratio and the battery standard life value.

[0027] The first preset ratio value is 85%, 90%, 95% or 100%;

[0028] When the charging mode is AC charging, 95% or 100% is selected as the first preset ratio value;

[0029] When the charging mode is DC charging, 85% or 90% is selected as the first preset ratio value;

[0030] The number of preset split ratio points in the DC charging mode is greater than the number of preset split ratio points in the AC charging mode.

[0031] In a second aspect of the present invention, in order to implement the method described in the first aspect, a system for determining battery life based on charging data is proposed, and the system is applied to a vehicle-mounted terminal.

[0032] The system comprises:

[0033] A charging data storage unit, used to store a charging data set that meets the conditions; the charging data set includes two charging data;

[0034] A charging data reporting unit, used for reporting the charging data set stored in the charging data storage unit;

[0035] a battery life determination unit, configured to determine the battery life based on the charging data set reported by the charging data reporting unit;

[0036] The qualified charging data set is determined according to the following method:

[0037] When the i-th charging is completed, if the charging ratio displayed by the vehicle terminal reaches the first preset ratio value, the charging data collected during the i-th charging process is combined as the first charging data that meets the conditions;

[0038] When the jth charging is completed, if the overlap degree of the sub-stages of the i-th charging data set and the j-th charging data set is greater than the second preset threshold, and the charging ratio displayed on the vehicle terminal reaches the second preset ratio value, and the difference between the i-th endurance value displayed on the vehicle computer at the end of the i-th charging and the j-th endurance value displayed on the vehicle computer at the end of the j-th charging is greater than the first preset threshold, then the charging data collected during the j-th charging process is collected as the second charging data that meets the conditions.

[0039] The two charging data stored in the charging data storage unit are acquired in the same charging mode, and the charging mode includes an AC charging mode and a DC charging mode.

[0040] The charging starting ratio of the i-th charging process and the j-th charging process is the same;

[0041] The i-th cruising range value displayed by the vehicle computer at the end of the i-th charging and the j-th cruising range value displayed by the vehicle computer at the end of the j-th charging are both determined based on the latest historical average interval driving data.

[0042] The system further comprises: a data registration unit;

[0043] The data alignment unit performs data alignment on the charging data set reported by the charging data reporting unit and then sends the data to the battery life determination unit.

[0044] The present invention first determines the charging time nodes and driving nodes that may affect the battery life, thereby making the collection and transmission of relevant charging data for battery prediction more targeted and reducing the operating pressure of the server model.

[0045] Further advantages of the present invention will be further reflected in detail in the specific embodiments section in conjunction with the drawings of the specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 This is a schematic diagram of the main flow of a method for determining battery life based on charging data according to an embodiment of the present invention.

[0048] Figure 2 This is a schematic diagram of the display interface of a certain car system

[0049] Figure 3 This is a schematic diagram of a charging data set of multiple charging sub-stages included in a certain charging process.

[0050] Figure 4 It is a table diagram of the recorded partial i-th charging data set and j-th charging data set

[0051] Figure 5 This is a schematic diagram of the functional modules of a system for determining battery life based on charging data according to an embodiment of the present invention. DETAILED DESCRIPTION

[0052] See first Figure 1 , Figure 1 A schematic diagram showing the main flow of a method for determining battery life based on charging data according to an embodiment of the present invention.

[0053] Figure 1The method is applied to a vehicle system, especially a new energy intelligent vehicle system with pure electric endurance. The vehicle system can wirelessly communicate with a cloud backend server to perform data transmission and command interaction.

[0054] Figure 1 The method includes the following steps S100-S400, and each step is specifically implemented as follows: Figure 1 Step numbers are omitted and some descriptions are rewritten):

[0055] S100: At the end of the i-th charging, if the charging ratio displayed by the vehicle system reaches a first preset ratio value, then recording the i-th charging data set and the i-th endurance value displayed by the vehicle system at the end of the i-th charging;

[0056] S200: At the end of the jth charging, if the charging ratio displayed by the vehicle system reaches the first preset ratio value, record the jth charging data set and the jth endurance value displayed by the vehicle system at the end of the jth charging;

[0057] S300: Calculating the difference between the i-th endurance value and the j-th endurance value and the sub-phase overlap between the i-th charging data set and the j-th charging data set;

[0058] When the difference is greater than a first preset threshold and the sub-phase overlap between the i-th charging data set and the j-th charging data set is greater than a second preset threshold, proceeding to step S400;

[0059] S400: uploading the i-th charging data set, the j-th charging data set and the interval driving data to a cloud server through the vehicle system;

[0060] The cloud server determines the battery life based on the i-th charging data set, the j-th charging data set, and the interval driving data.

[0061] In specific applications, the charging data set can select different specific data types. The specific selection of which indicator data in the charging process is used as the charging data set (and then used for subsequent life prediction) is related to the type of battery life prediction model selected later.

[0062] Preferably, the i-th charging data set includes the charging voltage, charging current and charging duration of multiple charging sub-stages included in the i-th charging process; the j-th charging data set includes the charging voltage, charging current and charging duration of multiple charging sub-stages included in the j-th charging process.

[0063] Of course, the charging data set is used for subsequent battery life (range) prediction. Since different prediction models have different training data, the specific elements contained in the charging data set need to be determined based on the specific type of prediction model selected later.

[0064] In the present invention, there is no specific limitation on which battery life prediction model is used in the later stage, and this is not the focus of the present invention. Prediction models known to those skilled in the art, supported by the cloud backend server of the car dealer related to the vehicle system, and other types are included in the technical solution of the present invention. The focus of the present invention is to determine the charging time nodes and driving nodes that are most likely to affect the battery life, so that the collection and transmission of relevant charging data for battery prediction are more targeted and the operating pressure of the server model is reduced.

[0065] It is understandable that the charging data set described in the present invention may also include other driving data (such as interval driving data between two charging time points), and does not only refer to data collected during the charging process.

[0066] As an example, the trained Time-LLaMA model mentioned in the background technology of the present application is an optional model, and the required charging data set includes data such as maximum output power, vehicle theoretical endurance, maximum allowable charging power, power battery capacity, and PACK energy density;

[0067] As another example, the optional model also includes a power battery life prediction method based on capacity testing and Internet of Vehicles big data proposed in patent application CN202210147003, which requires a charging data set including historical data such as temperature, current, time, mileage, and rated capacity.

[0068] Other examples of relevant predictive models include:

[0069] (1) The Vision Transformer structure of the dual-stream framework and the efficient self-attention mechanism model: It uses a small amount of charging cycle data, including the case of using 15 charging cycle data, the model controls the prediction errors of the remaining service life and the current cycle life within 5.40% and 4.64% respectively, showing extremely high prediction accuracy. It is worth mentioning that the model can still maintain a low prediction error when facing charging strategies that do not appear in the training data set, demonstrating its strong zero-short generalization ability.

[0070] For detailed principles and model introduction, please refer to the following literature:

[0071] Liu Y, Ahmed M, Feng J, et al.Deep learning powered lifetime prediction for lithium-ion batteries based on small amounts of charging cycles[J].IEEETransactions on Transportation Electrification.

[0072] (2) Based on deep learning, neural networks, data mining and other technologies, relevant literature includes:

[0073] Xiao Haoyi, He Xiaoxia, Liang Jiajia, et al. A lithium battery life prediction method based on modal decomposition and machine learning[J]. Energy Storage Science and Technology, 2022, 11(12): 3999-4009.

[0074] Feng,X.,Zhang,Y.,Xiong,R.and Tang,A.,2023.Estimating battery state of health with 10-min relaxation voltage across various charging states ofcharge.iEnergy.

[0075] Huang Peng. Research on capacity and remaining life prediction of power lithium battery based on multi-feature extraction and data-driven[D]. Kunming University of Science and Technology, 2022.

[0076] Xie Zhize. Research on remaining life prediction method of lithium-ion batteries based on deep learning[D]. Shandong: Shandong University of Technology, 2023.

[0077] Therefore, no matter whether long-term charging data or short-term charging data, full-cycle charging data or partial-cycle charging data are obtained, the prior art has known corresponding prediction models that can be used to predict battery life.

[0078] However, it is precisely because there are so many prediction model options, coupled with the portability of data collected and uploaded by the smart car network, that users may frequently upload data or frequently request to go to the store for inspection; in addition, some new energy vehicles also support battery swapping (i.e. battery replacement), and the battery before and after replacement will have normal differences in the mileage displayed even under the same operating conditions; if users frequently upload relevant data that is not related to lifespan to the large model, due to the influence of iteration errors, it is also possible that the training accuracy of the large model will be reduced, and the accuracy of the output results will also be affected.

[0079] To this end, the technical solution of the present invention has made targeted improvements as follows.

[0080] The vehicle system of the present invention comprises a charging data storage unit, which is used to store a charging data set that meets the conditions; the charging data set includes two charging data.

[0081] Specifically, the qualified charging data set is determined according to the following method:

[0082] When the i-th charging is completed, if the charging ratio displayed by the vehicle terminal reaches the first preset ratio value, the charging voltage, charging current and charging duration of the multiple charging sub-stages included in the i-th charging process are used as the first charging data that meets the conditions;

[0083] When the jth charging is completed, if the overlap degree of the sub-stages of the i-th charging data set and the j-th charging data set is greater than the second preset threshold, and the charging ratio displayed on the vehicle terminal reaches the second preset ratio value, and the difference between the i-th endurance value displayed by the vehicle computer at the end of the i-th charging and the j-th endurance value displayed by the vehicle computer at the end of the j-th charging is greater than the first preset threshold, then the charging voltage, charging current and charging time of the multiple charging sub-stages included in the j-th charging process are used as the qualified second charging data.

[0084] The charging starting ratio of the i-th charging process and the j-th charging process is the same;

[0085] The i-th cruising range value displayed by the vehicle computer at the end of the i-th charging and the j-th cruising range value displayed by the vehicle computer at the end of the j-th charging are both determined based on the latest historical average interval driving data.

[0086] The multiple charging sub-stages included in the i-th charging process are divided according to the following method:

[0087] Preset K split ratio points to divide the i-th charging process into K+1 charging sub-stages, where K≥2;

[0088] The multiple charging sub-stages included in the j-th charging process are divided according to the following method:

[0089] The j-th charging process is divided into K+1 charging sub-stages according to the preset K segmentation ratio points.

[0090] The first preset ratio value is 85%, 90%, 95% or 100%;

[0091] When the charging mode is AC charging, 95% or 100% is selected as the first preset ratio value;

[0092] When the charging mode is DC charging, 85% or 90% is selected as the first preset ratio value;

[0093] The number of preset split ratio points in the DC charging mode is greater than the number of preset split ratio points in the AC charging mode.

[0094] The above-mentioned embodiments of the present invention define the charging time nodes and driving nodes that may affect the battery life, thereby making the collection and transmission of relevant charging data for battery prediction more targeted.

[0095] That is, in the technical solution of the present application, only the charging data sets collected from two specific charging processes and the corresponding interval driving data are collected, recorded and uploaded, and the screening of these two data is most likely to affect the battery life.

[0096] To more clearly describe how to filter out these two data, Figure 1 The various steps combined Figure 2-Figure 4 The data schematic diagram is given for detailed description.

[0097] (1) Determine the charging mode. The charging mode includes AC charging mode (commonly known as slow charging) and DC charging mode (commonly known as fast charging). Since the charging process in the slow charging state is generally smoother, its impact on battery life (reduction, loss) is lower than that of fast charging. Therefore, preferably, in the present invention, the two charging data need to be collected in the fast charging mode, that is, the i-th charging and the j-th charging are both charged in the fast charging mode;

[0098] (2) The starting point of charging and the end point of charging are the same, that is, the starting charging ratio of the i-th charging process and the j-th charging process are the same; at the end of the i-th charging, the charging ratio displayed by the vehicle system reaches a first preset ratio value; at the end of the j-th charging, the charging ratio displayed by the vehicle system also reaches the first preset ratio value;

[0099] For example, the i-th charging and the j-th charging both start from 10% and end at 90%;

[0100] (3) The difference between the i-th endurance value and the j-th endurance value is greater than a first preset threshold.

[0101] At the end of the i-th charge, the car computer displays the i-th endurance value, and at the end of the j-th charge, the car computer displays the j-th endurance value. Normally, when the difference between the i-th endurance value and the j-th endurance value is greater than a first preset threshold, the owner will think that the battery may have health problems and its life may be damaged.

[0102] In the prior art, this is the judgment standard of most car owners. However, the single endurance difference judgment does not take into account the influence of operating conditions, driving speed, temperature, charging habits, and charging environment, and the judgment accuracy is not high. Therefore, it is necessary to enter the fourth stage of judgment as follows.

[0103] (4) The sub-phase overlap between the i-th charging data set and the j-th charging data set is greater than a second preset threshold.

[0104] For example, for the i-th charging process (the process of charging from 10% to 90%), two split points can be introduced to divide it into three sub-stages, see Figure 3 Schematic diagram of charging data set of multiple charging sub-stages included in a certain charging process:

[0105] The first sub-stage i1: the sub-stage of charging from 10% to 20%;

[0106] The second sub-stage i2: a sub-stage of charging from 20% to 80%;

[0107] The third sub-stage i3: a sub-stage of charging from 90% to 90%.

[0108] It should be noted that Figure 3 The data set is only illustrative and is only used to facilitate the introduction of certain calculation processes of the embodiment, and does not represent a specific limitation on the technical solution of the present invention. For example, Figure 3 The solid line segment in the middle represents the i-th charging process, and the dotted line segment represents the j-th charging process; the relevant charging data collected on the vertical axis can be the charging current, charging voltage, etc. corresponding to different time nodes (charging ratio) during the charging process.

[0109] Of course, the j-th charging process is processed in the same manner, and three charging sub-stages j1, j2 and j3 are obtained.

[0110] Calculating the sub-stage overlap between the i-th charging data set and the j-th charging data set, specifically including:

[0111] Calculate the overlap degree ij1 between sub-stage i1 and sub-stage j1;

[0112] Calculate the overlap degree ij2 between sub-stage i2 and sub-stage j2;

[0113] Calculate the degree of overlap ij3 between sub-stage i3 and sub-stage j3.

[0114] The overlap is used to characterize the difference (similarity) between multiple different node data collected in corresponding sub-stages of different charging processes.

[0115] by Figure 4 For example, the following is an example: Figure 4It is a tabular diagram of a portion of the i-th charging data set and the j-th charging data set recorded.

[0116] exist Figure 4 In the figure, the charged capacity data collected in sub-stage i1 (the sub-stage of charging from 10% to 20%) are: 4.022, 7.793, 11.482, i.e., the data of three nodes; the charged capacity data collected in sub-stage j1 (the sub-stage of charging from 10% to 20%) are: 3.961, 8.36, 12.693, also the data of three nodes.

[0117] At this time, the overlap between sub-stages i1 and j1 is relatively high, because the differences of (4.022, 3.961), (7.793, 8.36), and (11.482, 12.693) are all within the acceptable range, which means that the data collected in these two sub-stages are considered to be highly similar.

[0118] Similarly, the overlap between sub-stages i2 and j2, and between sub-stages i3 and j3, etc. can also be calculated.

[0119] When at least one of the multiple coincidences is greater than a second preset threshold, it is considered that the sub-stage coincidence between the i-th charging data set and the j-th charging data set is greater than the second preset threshold.

[0120] It should be noted that Figure 4 The data set is only schematic and partial data, and is only used to facilitate the introduction of certain calculation processes of the embodiments. It does not represent a specific limitation on the technical solution of the present invention, nor is it all the data used in a certain embodiment of the present application.

[0121] For example, based on Figure 4 At least it can be concluded that the collected charging data set contains the following elements: charging ratio, cruising range, charging power, charged amount, and charging time.

[0122] Therefore, the above calculation of the sub-stage overlap between the i-th charging data set and the j-th charging data set can be summarized as:

[0123] Determine the number of sub-stages K+1 (the i-th charging process and the j-th charging process are both divided into K+1 sub-stages);

[0124] For each sub-stage, the charging data of multiple different nodes collected are determined.

[0125] Calculate the similarity between charging data groups of the same node in the same sub-stage of different charging processes to obtain the corresponding coincidence degree of the sub-stage;

[0126] When a specified number of overlaps among multiple overlaps of multiple sub-stages (for example, one overlap among three overlaps and two overlaps among four overlaps) are all greater than a second preset threshold, it is determined that the sub-stage overlap between the i-th charging data set and the j-th charging data set is greater than the second preset threshold.

[0127] It can be seen that in the above process of calculating the overlap, it is necessary to determine the charging data group of the same node in the same sub-stage of different charging processes.

[0128] However, in some cases, the number of nodes in the charging data group collected in the same sub-stage of different charging processes may be inconsistent.

[0129] For example, the charged capacity data collected in a certain sub-stage are: a1-10%, a2-12%, a3-18%, a4-20%, that is, the data of 4 nodes (10% / 12% / 18% / 20%); the charged capacity data collected in another identical sub-stage are: b1-10%, b2-12%, b3-15%, a4-19%, corresponding to the data of 4 nodes (10% / 12% / 15% / 19%).

[0130] For another example, the charged capacity data collected in a certain sub-stage are: a1, a2, a3, a4, that is, the data of 4 nodes; the charged capacity data collected in another same sub-stage are: b1, b2, b3, corresponding to the data of 3 nodes.

[0131] At this point, since the data is not completely corresponding, a registration operation needs to be completed.

[0132] Therefore, as a further improvement, the i-th charging data set and the j-th charging data set are aligned according to the K+1 charging sub-stages.

[0133] The registration operation can be to fit and supplement the data (for example, to fit the data of 15% / 19% to obtain the data of 18% and 20%), or to interpolate and supplement the data (for example, to interpolate and fit the data of b1, b2 and b3 to obtain the data of b4), so that the same nodes in the same sub-stages of different charging processes have corresponding charging data groups.

[0134] It can be seen that when the sub-stage overlap between the i-th charging data set and the j-th charging data set is greater than the second preset threshold, it means that the charging data of the battery in at least one stage during the charging process is the same or similar, and there is no significant difference in the battery charging and discharging process. However, if it leads to a large difference in the final displayed cruising range, it means that there may be battery structural damage caused by battery structural changes in some other charging sub-stage, and thus there is a high possibility of changes in battery life.

[0135] Therefore, the above-mentioned process of screening charging data for subsequent life prediction proposed by the present invention conforms to the actual battery structure principle and is adapted to the influencing factors of the current complex and changeable charging environment on the battery structure.

[0136] In order to further reduce the amount of data processing and storage, combined with the principles given in the above introduction, a more preferred embodiment of a method for determining battery life based on charging data of the present invention is as follows:

[0137] A method for determining battery life based on charging data, the method is applied to a vehicle system, the vehicle system is connected to a cloud server, and the cloud server provides multiple battery life prediction models; the method comprises the following steps:

[0138] S1: Determine a currently available target battery life prediction model, and determine the data type included in the charging data set based on the currently used battery life prediction model;

[0139] S2: At the end of the i-th charging, if the charging ratio displayed by the vehicle system reaches a first preset ratio value, then record the i-th charging data set and the i-th endurance value displayed by the vehicle system at the end of the i-th charging, the i-th charging process is divided into M charging sub-stages, M>2; multiple charging data are collected in each charging sub-stage;

[0140] S3: At the end of the jth charging, if the charging ratio displayed by the vehicle system reaches the first preset ratio value, record the jth endurance value displayed by the vehicle system at the end of the jth charging;

[0141] S4: Calculate the difference between the i-th cruising range value and the j-th cruising range value;

[0142] When the difference is greater than the first preset threshold, proceed to step S5; otherwise, return to step S1;

[0143] S5: Acquire a plurality of charging data collected in each of the M charging sub-stages of the j-th charging process;

[0144] S6: Calculating the sub-phase overlap between the i-th charging data set and the j-th charging data set;

[0145] When the overlap degree is greater than a second preset threshold, proceed to step S7;

[0146] Otherwise, return to step S1;

[0147] S7: uploading the i-th charging data set, the j-th charging data set and the interval driving data to the cloud server through the vehicle system;

[0148] The cloud server determines the battery life based on the i-th charging data set, the j-th charging data set and the interval driving data using the currently used target battery life prediction model.

[0149] The interval driving data includes the driving distance between the i-th charging and the j-th charging, the average driving speed, the average energy consumption (e.g. Figure 2 The specific data selected is also determined based on the target battery life prediction model currently used. The i-th endurance value displayed by the vehicle computer at the end of the i-th charging and the j-th endurance value displayed by the vehicle computer at the end of the j-th charging are both determined based on the latest historical average interval driving data.

[0150] The overlap degree in step S6 is used to characterize the similarity of data of multiple different nodes collected in corresponding sub-stages of different charging processes.

[0151] Based on the above introduction, the technical solution of the present application can also be used for the user interface prompt of the vehicle system, that is, after a certain charging, the vehicle user thinks that the mileage displayed by the vehicle system is too low, so he suspects that the battery life is reduced. In the prior art, the user may directly upload the current charging data set and other related data to the cloud backend server provided by the car dealer, and request to test the battery health status, which leads to frequent upload of test data, and the uploaded data is not targeted, which may affect the later training accuracy of the background prediction model and also bring greater data transmission pressure.

[0152] By utilizing the technical solution of the present application, when it is detected that the user clicks on the operation of uploading the current charging data, it is first determined whether the current charging data meets the upload criteria. If the upload criteria are not met, a specific reason for the judgment is given, prompting the user to upload the data only when the relevant conditions are met next time, thereby reducing the pressure on data transmission, making the collection and transmission of relevant charging data for battery prediction more targeted, and reducing the operating pressure of the server model.

[0153] Preferably, an embodiment to achieve the above technical effect is as follows:

[0154] A charging data processing method, which is applied to a vehicle-mounted terminal, comprises the following steps:

[0155] ST1: After the current charging process is completed, if it is detected that the user clicks to upload the current charging data, it is determined whether the current charging data meets the upload standard;

[0156] If satisfied, the data upload operation starts and goes to step ST2;

[0157] If not, give specific reasons for the judgment and prompt the user to upload the data next time when the relevant conditions are met; return to step ST1;

[0158] ST2: Upload the current charging data set, the saved previous charging data set and the interval driving data to the cloud server through the vehicle system;

[0159] The cloud server determines the battery life based on the current charging data set, the previously saved charging data set, and the interval driving data.

[0160] The charging data of the current time meets the upload standard, specifically, the following conditions 1 to 3 are met at the same time:

[0161] (Condition 1): The starting ratio and ending ratio of the current charge are the same as those of the previous charge, and the charging mode is the same;

[0162] (Condition 2): The difference between the battery life value displayed at the end of the previous charge and the battery life value displayed at the end of the current charge is greater than a first preset threshold;

[0163] (Condition 3): When the overlap between the sub-stage of the current charging process and the sub-stage of the previous charging process is greater than a second preset threshold.

[0164] After introducing the above method embodiment, see Figure 5 . Figure 5 A schematic diagram showing the composition of functional modules of a system for determining battery life based on charging data according to an embodiment of the present invention.

[0165] exist Figure 5 In the figure, a system for determining battery life based on charging data is shown, and the system is applied to a vehicle-mounted terminal.

[0166] The system comprises:

[0167] A charging data storage unit, used to store a charging data set that meets the conditions; the charging data set includes two charging data;

[0168] A charging data reporting unit, used for reporting the charging data set stored in the charging data storage unit;

[0169] a battery life determination unit, configured to determine the battery life based on the charging data set reported by the charging data reporting unit;

[0170] The qualified charging data set is determined according to the following method:

[0171] When the i-th charging is completed, if the charging ratio displayed by the vehicle terminal reaches the first preset ratio value, the charging voltage, charging current and charging duration of the multiple charging sub-stages included in the i-th charging process are used as the first charging data that meets the conditions;

[0172] When the jth charging is completed, if the overlap degree of the sub-stages of the i-th charging data set and the j-th charging data set is greater than the second preset threshold, and the charging ratio displayed on the vehicle terminal reaches the second preset ratio value, and the difference between the i-th endurance value displayed by the vehicle computer at the end of the i-th charging and the j-th endurance value displayed by the vehicle computer at the end of the j-th charging is greater than the first preset threshold, then the charging voltage, charging current and charging time of the multiple charging sub-stages included in the j-th charging process are used as the qualified second charging data.

[0173] The two charging data stored in the charging data storage unit are acquired in the same charging mode, and the charging mode includes an AC charging mode and a DC charging mode.

[0174] The charging starting ratio of the i-th charging process and the j-th charging process is the same;

[0175] The i-th cruising range value displayed by the vehicle computer at the end of the i-th charging and the j-th cruising range value displayed by the vehicle computer at the end of the j-th charging are both determined based on the latest historical average interval driving data.

[0176] The system further comprises: a data registration unit;

[0177] The data alignment unit performs data alignment on the charging data set reported by the charging data reporting unit and then sends the data to the battery life determination unit.

[0178] The battery life determination unit is configured with a plurality of battery life prediction models.

[0179] Preferably, the charging data storage unit is connected to the battery life determination unit. After the battery life determination unit determines the target battery life prediction model currently in use, the charging data storage unit determines the specific data element type contained in the charging data set that needs to be collected.

[0180] After the battery life determination unit replaces the currently used target battery life prediction model, feedback information is sent to the charging data storage unit so that the charging data storage unit re-determines the specific data element type contained in the charging data set that needs to be collected.

[0181] At the same time, in the specific implementation of this application, if user-related data is involved, when the embodiment of this application is applied to a specific product or technology, the user's permission or consent must be obtained, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0182] The foregoing has shown and described the method embodiments and system of the present invention, but it is understandable to those skilled in the art that various changes, modifications, substitutions and variations may be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for determining battery life based on charging data, the method being applied to a vehicle system, characterized in that: The method comprises the following steps: S100: At the end of the i-th charging, if the charging ratio displayed by the vehicle system reaches a first preset ratio value, then recording the i-th charging data set and the i-th endurance value displayed by the vehicle system at the end of the i-th charging; S200: At the end of the jth charging, if the charging ratio displayed by the vehicle system reaches the first preset ratio value, record the jth charging data set and the jth endurance value displayed by the vehicle system at the end of the jth charging; S300: Calculating the difference between the i-th endurance value and the j-th endurance value and the sub-phase overlap between the i-th charging data set and the j-th charging data set; When the difference is greater than a first preset threshold and the sub-stage overlap is greater than a second preset threshold, proceed to step S400; S400: uploading the i-th charging data set, the j-th charging data set and the interval driving data to a cloud server through the vehicle system; The cloud server determines the battery life based on the i-th charging data set, the j-th charging data set, and the interval driving data.

2. A method for determining battery life based on charging data as claimed in claim 1, characterized in that: The charging starting ratio of the i-th charging process and the j-th charging process is the same; The i-th cruising range value displayed by the vehicle computer at the end of the i-th charging and the j-th cruising range value displayed by the vehicle computer at the end of the j-th charging are both determined based on the latest historical average interval driving data.

3. A method for determining battery life based on charging data as claimed in claim 1, characterized in that: The step S300 further includes: When the difference is smaller than the first preset threshold, or the overlap degree is smaller than the second preset threshold, the process returns to step S100.

4. A method for determining battery life based on charging data as claimed in claim 1, characterized in that: The multiple charging sub-stages included in the i-th charging process in step S100 are divided according to the following method: Preset K split ratio points to divide the i-th charging process into K+1 charging sub-stages, where K≥2; The multiple charging sub-stages included in the j-th charging process in step S200 are divided according to the following method: The j-th charging process is divided into K+1 charging sub-stages according to the preset K segmentation ratio points.

5. A method for determining battery life based on charging data as claimed in claim 4, characterized in that: The cloud server determines the battery life based on the i-th charging data set, the j-th charging data set and the interval driving data, specifically including: Aligning the i-th charging data set and the j-th charging data set according to the K+1 charging sub-stages; Determine the battery attenuation ratio based on the registered charging data set and the interval driving data; The battery life is determined based on the battery attenuation ratio and the battery standard life value.

6. A method for determining battery life based on charging data as claimed in claim 4, characterized in that: The first preset ratio value is 85%, 90%, 95% or 100%; When the charging mode is AC charging, 95% or 100% is selected as the first preset ratio value; When the charging mode is DC charging, 85% or 90% is selected as the first preset ratio value; The number of preset split ratio points in the DC charging mode is greater than the number of preset split ratio points in the AC charging mode.

7. A system for determining battery life based on charging data, the system being applied to a vehicle-mounted terminal, characterized in that: The system comprises: A charging data storage unit, used to store a charging data set that meets the conditions; the charging data set includes two charging data; A charging data reporting unit, used for reporting the charging data set stored in the charging data storage unit; a battery life determination unit, configured to determine the battery life based on the charging data set reported by the charging data reporting unit; The qualified charging data set is determined according to the following method: When the i-th charging is completed, if the charging ratio displayed by the vehicle terminal reaches the first preset ratio value, the charging data collected during the i-th charging process is combined as the first charging data that meets the conditions; When the jth charging is completed, if the overlap degree of the sub-stages of the i-th charging data set and the j-th charging data set is greater than the second preset threshold, and the charging ratio displayed on the vehicle terminal reaches the second preset ratio value, and the difference between the i-th endurance value displayed on the vehicle computer at the end of the i-th charging and the j-th endurance value displayed on the vehicle computer at the end of the j-th charging is greater than the first preset threshold, then the charging data collected during the j-th charging process is collected as the second charging data that meets the conditions.

8. A system for determining battery life based on charging data as claimed in claim 7, characterized in that: The two charging data stored in the charging data storage unit are acquired in the same charging mode, and the charging mode includes an AC charging mode and a DC charging mode.

9. A system for determining battery life based on charging data as claimed in claim 7, characterized in that: The charging starting ratio of the i-th charging process and the j-th charging process is the same; The i-th cruising range value displayed by the vehicle computer at the end of the i-th charging and the j-th cruising range value displayed by the vehicle computer at the end of the j-th charging are both determined based on the latest historical average interval driving data.

10. A system for determining battery life based on charging data as claimed in claim 7, characterized in that: The system further comprises: a data registration unit; The data alignment unit performs data alignment on the charging data set reported by the charging data reporting unit and then sends the data to the battery life determination unit.

Images