Battery self-discharge rate abnormality judgment method and device based on vehicle networking data analysis

Through machine learning methods based on data analysis of Internet of Vehicles, the open circuit voltage after the battery is powered off is estimated, which solves the accuracy of battery self-discharge abnormal detection in electric vehicles, realizes real-time safety monitoring of battery packs, and improves the safety of battery packs. sex.

CN115219903BActive Publication Date: 2024-11-12CHINA FAW CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210237964.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-03-11
Publication Date
2024-11-12
Estimated Expiration
2042-03-11

AI Technical Summary

Technical Problem

The prior art is difficult to accurately detect the self-discharge abnormality of lithium-ion batteries in operating electric vehicles. Traditional methods rely on laboratory tests and cannot monitor the self-discharge status of single batteries in the battery pack in real time, resulting in an increase in safety risks.

Method used

Through a method based on the Internet of Vehicles data analysis, the machine learning model is used to estimate the open circuit voltage after the battery is powered down, and the self-discharge rate judgment is made in combination with the vehicle data to realize the early warning of abnormal self-discharge of the battery pack and improve the safety of the battery pack.

Benefits of technology

Without increasing hardware costs, real-time reminders on the risk of abnormal battery self-discharge are achieved, which significantly improves the safety of the battery pack and protects the safety of users' lives and property.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115219903B_ABST
    Figure CN115219903B_ABST
Patent Text Reader

Abstract

The present invention provides a method and device for judging the abnormality of battery self-discharge rate based on vehicle networking data analysis. The judgment method first trains an open circuit voltage model for estimating the stable state of the vehicle after powering off based on historical data, and then judges the self-discharge rate based on the open circuit voltage of the vehicle after powering on and the quasi-steady-state vehicle data read after the vehicle is powered on after standing still. The judgment device includes a receiving module, a first determination module and a first reminder module, which are respectively used to receive vehicle data reported by the vehicle, and determine the risk and level of battery pack self-discharge based on the vehicle data, and send a reminder message of battery pack self-discharge warning to the user of the vehicle and the corresponding terminal of the after-sales department if it is determined that the vehicle reported data meets the risk level of battery pack self-discharge. The present invention can realize real-time reminder of abnormal risk of battery self-discharge based on vehicle data without adding hardware, which greatly improves the safety of the battery pack and protects the life and property safety of users.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of lithium-ion batteries, and in particular relates to a method and a device for judging abnormal battery self-discharge rate based on vehicle networking data analysis. Background Art

[0002] Usually, the abnormal self-discharge of lithium-ion power batteries is caused by foreign matter inside the battery, which may cause battery spontaneous combustion accidents in severe cases. In the early days of battery production, some foreign matter was difficult to detect through methods such as voltage resistance testing. After the battery is packaged in the car, the battery management system will not actively detect the abnormal self-discharge of the battery, and the safety risk of the vehicle is increased. The traditional method is generally in the laboratory, where the lithium-ion power battery is left to stand for a period of time and the battery voltage drop is tested to analyze the self-discharge rate. However, this method cannot be used for electric vehicles in operation.

[0003] The prior art discloses a method and device for monitoring self-discharge of a power battery, a vehicle, and a storage medium. The method includes: obtaining a high-voltage power-down signal; recording the time T1 at which the high-voltage power-down signal is obtained, and recording the voltage value of each single cell; calculating the average voltage Uavgd of the single cell; obtaining a high-voltage power-on signal; recording the time T2 at which the high-voltage power-on signal is obtained, obtaining the lowest single cell voltage value Umin, and obtaining the number n of the single cell; obtaining the voltage value Un of the battery numbered n; calculating the average voltage Uavgu; calculating the self-discharge parameter K; judging whether the self-discharge parameter K meets the alarm condition, and if so, giving an alarm. The self-discharge parameter can be calculated through a normal power-down and power-on process to characterize the normal situation of self-discharge.

[0004] The prior art also discloses a method for online detection of the self-discharge state of single cells in a battery pack based on big data, taking the uploaded data of the battery pack; selecting the data of the voltage of each single cell in the battery pack when the vehicle was parked for more than T at the most recent time from the data; extracting the last voltage value uploaded before parking and the first voltage value uploaded after parking of each single cell in the data of the previous step, respectively counting them as Vx0 and Vx1, and calculating the time difference △t between the two voltage value samplings; the voltage drop value Kx of the self-discharge rate of battery No. x; then Kx=|(Vx0-Vx1) / △t|; calculating the average value K average of the self-discharge rate Kx of the single cell and its standard deviation δ, the upper limit value K threshold of the battery self-discharge=K average+Mδ; comparing the sizes of Kx and K threshold; if Kx≤K threshold, it means that the battery self-discharge rate Kx is normal; otherwise, the battery self-discharge rate is too large, and online detection and monitoring functions can be realized.

[0005] However, the existing technology for testing abnormal self-discharge of power batteries mainly relies on the laboratory to let the battery stand still and test the battery's voltage drop rate or capacity recovery rate. At present, big data analysis is performed using the Internet of Vehicles signals sent by electric vehicles in accordance with national standards. In similar schemes, most of them use the data of the previous frame of the high-voltage electrical signal to compare with the data of the next power-on frame to analyze self-discharge. However, the previous frame of the high-voltage electrical signal is not accurate due to battery polarization and voltage fluctuations during vehicle operation. To truly obtain self-discharge, it must be in an open circuit voltage state. Summary of the invention

[0006] The purpose of the present invention is to address the deficiencies of the above-mentioned prior art and provide a method for judging the abnormality of the battery self-discharge rate based on vehicle networking data analysis, which uses a machine learning method to estimate the open circuit voltage of the battery when it is powered off, and the calculation is more accurate; it also provides a device for judging the abnormality of the battery self-discharge rate based on vehicle networking data analysis to solve the problem that the existing battery system may have risks due to abnormal self-discharge rate. The present invention is based on the vehicle networking signal sent by the electric vehicle according to the national standard requirements, and after model calculation, it realizes the early warning of abnormal self-discharge of the battery pack, which greatly improves the safety of the battery pack without increasing the cost.

[0007] The objective of the present invention is achieved through the following technical solutions:

[0008] A method for judging abnormal battery self-discharge rate based on vehicle networking data analysis first trains an open circuit voltage model for estimating the stable state of a vehicle after powering off based on historical data, and then judges the self-discharge rate based on the open circuit voltage of the vehicle after powering off and the quasi-steady-state vehicle data read after powering on after standing still. Specifically, the method includes the following steps:

[0009] A. Receive vehicle data reported by the vehicle;

[0010] B. Quasi-steady state identification and selection;

[0011] C. Data screening;

[0012] D. Single cell voltage-SOC interpolation of voltage characterization quantity;

[0013] E. Calculate the extreme deltaSOC and collect training samples;

[0014] F. Summary of status before parking;

[0015] G. Machine learning regression to obtain a prediction model;

[0016] H. Quasi-steady-state data acquisition and SOC interpolation;

[0017] I. Determine the highest / lowest cell position and estimate the self-discharge rate.

[0018] Furthermore, step B specifically includes the following steps:

[0019] B1. Sort the data reported by the vehicle by time and calculate the time difference between every two frames (timediff);

[0020] B2. Extract the frame whose timediff is greater than T hours and its previous frame, wherein the frame whose timediff is greater than T hours is represented by a vehicle power-on frame, and its previous frame is represented by a vehicle power-off frame. These two frames of data appear in pairs and can be regarded as a data segment.

[0021] Furthermore, step D is specifically as follows: according to the cell voltage-SOC table of the battery characteristics, the cell voltage in each frame is interpolated with the voltage-SOC, and the interpolation method can use linear interpolation, inverse distance weighted method, spline interpolation method, Kriging method, discrete smooth interpolation and trend surface smooth interpolation method.

[0022] Furthermore, step E specifically includes the following steps:

[0023] E1. Calculate the voltage SOC extreme difference value of each frame as deltaSOC;

[0024] E2. Analyze the two indicators of deltaSOC range and deltaSOC change rate:

[0025] E21. Analyze deltaSOC range: When the deltaSOC range is less than or equal to a certain threshold value _1, enter the training sample;

[0026] E22. Calculate the deltaSOC change rate: Based on the change in mileage and the change in time, the horizontal axis is the dual coordinate axis of time / mileage, and the vertical axis is deltaSOC. Plot each deltaSOC on the graph, and obtain the difference between the two deltaSOCs, that is, the SOC extreme change. The SOC difference / interval time can be used to calculate the SOC difference change rate. Within N months, the SOC extreme change must be less than or equal to a certain threshold value_2.

[0027] Further, step G is specifically as follows: according to the classification of the vehicle state data before parking, the voltage after the vehicle is parked is used as the prediction value, and the remaining data is used as training parameters for regression training of machine learning; multiple regression methods are used for regression analysis of the five states of DC charging, AC charging, intense driving, stable driving, and parking, and the results are voted. A certain regression method with the highest accuracy rate in each vehicle state is ≥3 times, which is regarded as the optimal regression solution. The model obtained by the regression method is used as the prediction model for predicting the OCV after parking, and is trained to meet the preset model threshold value_3. The selection of the threshold is verified by using the R^2 method, and the method is:

[0028]

[0029] The numerator is the mean square error MSE, and the denominator is the variance Var. The R^2 value is between 0 and 1. The closer it is to 1, the more accurate the prediction.

[0030] Further, step H is specifically as follows: if there is a power-off static time T in the state of step E2 that is sufficient to eliminate polarization, according to the conditions of step G, regress and calculate the open circuit voltage data that eliminates polarization after power-off, and use it as the steady-state open circuit voltage after power-off, and perform SOC interpolation according to the method of step D to obtain the static predicted SOCn of each battery cell. At the same time, according to the method of step D, interpolate the voltage of each battery cell of the first frame of power-on data after power-on as the power-on SOCn of each battery cell.

[0031] Further, step I is specifically as follows: comparing the highest / lowest cell number before power-off with the highest / lowest cell number after power-on to analyze whether they are consistent, and if they are consistent, calculating the self-discharge rate:

[0032]

[0033] The monthly self-discharge rate of each battery cell can be calculated, and based on the threshold, it can be determined whether there are batteries with self-discharge rates exceeding the limit. If an over-limit battery cell is found, the user will be reminded through the APP or background to go to the store for maintenance.

[0034] A battery self-discharge rate abnormality judgment device based on vehicle networking data analysis includes a receiving module, a first determination module and a first reminder module;

[0035] The receiving module is used to receive vehicle data reported by the vehicle; the first determining module is used to determine the risk and level of battery pack self-discharge based on the vehicle data; the first reminding module is used to send a reminder message of battery pack self-discharge warning to the user of the vehicle and the corresponding terminal of the after-sales department if it is determined that the vehicle reported data meets the risk level of battery pack self-discharge.

[0036] A server, comprising:

[0037] one or more processors;

[0038] A memory for storing one or more programs;

[0039] When the one or more programs are executed by the one or more processors, the one or more processors implement the battery pack self-discharge warning method provided in the first aspect.

[0040] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the battery pack self-discharge warning method provided in the first aspect.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] Without adding hardware, the present invention can realize real-time reminder of abnormal risk of battery self-discharge according to vehicle data, greatly improving the safety of the battery pack and protecting the life and property of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 A schematic diagram of an application scenario of the method for determining abnormal battery self-discharge rate based on vehicle networking data analysis provided by the present invention;

[0045] Figure 2 A schematic flow chart of a method for determining abnormal battery self-discharge rate based on vehicle networking data analysis provided by an embodiment of the present invention;

[0046] Figure 3 A schematic diagram of the structure of a server provided in an embodiment of the present invention;

[0047] Figure 4 Schematic diagram of the calculation method of SOC difference;

[0048] Figure 5 is a schematic diagram of the calculated self-discharge rate identification result;

[0049] Figure 6 Schematic diagram of machine learning regression. DETAILED DESCRIPTION

[0050] The present invention will be further described below in conjunction with embodiments:

[0051] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0052] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0053] In a first aspect, the present invention provides a method for determining abnormal battery self-discharge rate based on vehicle networking data analysis, comprising the following steps:

[0054] 1. The server receives vehicle data reported by the vehicle;

[0055] 2. Determine the risk and level of battery self-discharge based on the vehicle data, specifically: the server has a model judgment module and a background system statistics module, and the server decodes and converts the vehicle data into readable data;

[0056] 3. If it is determined that the vehicle reported data meets the risk level of the battery self-discharge rate, a reminder message of abnormal battery self-discharge is sent to the user of the vehicle and the corresponding terminal of the after-sales department. Specifically, the model judgment module can determine whether to send a reminder message of abnormal battery self-discharge rate to the corresponding terminal of the user of the vehicle based on the vehicle data.

[0057] In a second aspect, the present invention provides a battery self-discharge rate abnormality judgment device based on vehicle networking data analysis, comprising a receiving module, a first determining module and a first reminding module;

[0058] The receiving module is used to receive vehicle data reported by the vehicle; the first determining module is used to determine the risk and level of battery pack self-discharge based on the vehicle data; the first reminding module is used to send a reminder message of battery pack self-discharge warning to the user of the vehicle and the corresponding terminal of the after-sales department if it is determined that the vehicle reported data meets the risk level of battery pack self-discharge.

[0059] In a third aspect, the present invention further provides a server, comprising: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the battery pack self-discharge warning method provided in the first aspect.

[0060] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the battery pack self-discharge warning method provided in the first aspect.

[0061] Example 1

[0062] like Figure 1As shown, the vehicle 11 will upload the vehicle data to the server 12. Exemplarily, the vehicle 11 can upload the vehicle data in accordance with the national standard GBT32960. The server 12 is provided with a model judgment module 121 and a background system statistics module 122. The server 12 decodes and converts the vehicle data and converts it into readable data. The model judgment module 121 can determine whether to send a reminder message of battery pack leakage to the terminal 13 corresponding to the user of the vehicle based on the vehicle data. The background system statistics module 122 can display vehicle data according to different dimensions. The battery self-discharge rate abnormality judgment method based on vehicle network data analysis provided in this embodiment can realize real-time early warning of battery pack leakage to users based on vehicle data.

[0063] This embodiment is applicable to the scenario of abnormal self-discharge judgment of the battery. The method can be executed by a battery management device, which can be implemented by software and / or hardware, and the abnormal self-discharge rate judgment of the battery can be integrated in a server.

[0064] like Figure 2 As shown, the method for determining abnormal battery self-discharge rate based on vehicle networking data analysis provided in this embodiment is divided into two parts:

[0065] Part 1: Train a model to estimate the open circuit voltage of the vehicle in a stable state after power is turned off based on historical data;

[0066] Part 2: Determine the self-discharge rate based on the open circuit voltage after the vehicle is powered off and the quasi-steady-state vehicle data read after being powered on after being stationary

[0067] Part 1 includes the following steps:

[0068] Step 201: Receive vehicle data reported by a vehicle.

[0069] Specifically, the vehicle in this embodiment may be a new energy vehicle, and the vehicle may upload vehicle data to the server in accordance with the provisions of the national standard GBT32960.

[0070] The vehicle data in this embodiment is the relevant information of the vehicle during driving. Optionally, the vehicle data may include but is not limited to the following items: vehicle status of the electric vehicle, vehicle mileage, vehicle current, and voltage acquisition signal of the battery pack.

[0071] Step 202: Quasi-steady state identification and selection

[0072] In order to accurately analyze the self-discharge of the battery, it is necessary to measure the open circuit voltage without current. Since electric vehicles are always running, the only way to find this state is to use the data of the vehicle after it stops and powers off, stops for a period of time, and then powers on, which can be used as quasi-static. Specifically:

[0073] 202-1: According to the data reported by the vehicle, sort by time and calculate the time difference between every two frames timediff;

[0074] 202-2: Extract the frame with timediff greater than T hours and its previous frame (two frames in total), wherein the frame with timediff greater than T hours is represented by a vehicle power-on frame, and its previous frame is represented by a vehicle power-off frame. These two frames of data appear in pairs and can be regarded as a data segment.

[0075] Step 203: Data screening.

[0076] Generally, due to abnormal data transmission of the electric vehicle and other reasons, the data in step 202 needs to be screened.

[0077] Table 1 shows the features that need to be judged for data transmission anomalies.

[0078]

[0079]

[0080] Step 204: interpolation of the cell voltage-SOC of the voltage representative quantity.

[0081] According to the single cell voltage-SOC table of battery characteristics, the voltage-SOC interpolation is performed on the single cell voltage in each frame. The interpolation method can use linear interpolation, inverse distance weighted method, spline interpolation method, Kriging method, discrete smooth interpolation, trend surface smooth interpolation and other methods.

[0082] Step 205: Calculate the range deltaSOC and collect training samples.

[0083] The voltage SOC extreme difference value (maximum cell voltage interpolation SOC-minimum cell voltage interpolation SOC) of each frame (same frame) is calculated as deltaSOC.

[0084] In order to collect training samples, two indicators, deltaSOC range and deltaSOC change rate, need to be analyzed.

[0085] 205-1deltaSOC range: When the deltaSOC range is less than or equal to a certain threshold value _1, it enters the training sample.

[0086] 205-2deltaSOC change rate:

[0087] Calculate deltaSOC according to the above method, trace back the deltaSOC of the vehicle in the previous N months, save the date, mileage, and deltaSOC, and draw a chart, for example Figure 4 .

[0088] Figure 4 As shown in the figure, the change rate of deltaSOC can be based on the change of mileage and the change of time. The horizontal axis is the dual axis of time / mileage, and the vertical axis is deltaSOC. Plot each deltaSOC on the graph, and obtain the difference between the two deltaSOCs, that is, the SOC extreme change. The SOC difference / interval time (mileage) can be used to obtain the SOC difference change rate. Within N months, the SOC extreme change must be less than or equal to a certain threshold value _2 (threshold value _2 is determined according to the battery design standard).

[0089] Step 206: Status summary before parking

[0090] For the segments that meet the status of step 205, all data of the T minutes before parking are obtained based on the characteristic requirements of the following Table 4. The characteristic parameters are described as follows:

[0091] A. No. 2, vehicle cumulative mileage represents the cumulative mileage of the vehicle so far, in units of 0.1km;

[0092] B. No. 3-4, battery temperature before / after the vehicle is parked;

[0093] C. No. 5, the time interval for parking the vehicle, in hours;

[0094] D. No. 6-7, each cell voltage before and after the vehicle is parked and other voltage representation forms, such as SOC corresponding to OCV, self-discharge is judged from the voltage before and after the vehicle is parked. At the same time, each cell voltage after the vehicle is parked is also the predicted value of machine learning;

[0095] E. No. 8, summary of the vehicle status before parking, divided into 5 states, namely: DC charging, AC charging, aggressive driving, smooth driving, and parking, allocated according to the vehicle status before parking.

[0096] F, No. 9-13, description of current value before parking;

[0097] G. Serial number 14-18, description of vehicle speed before stopping.

[0098] Table 2 Open circuit voltage prediction characteristics

[0099]

[0100] Step 207: Machine learning regression to obtain a prediction model

[0101] According to the classification of the vehicle's state data before parking, the voltage of the vehicle after parking is used as the predicted value, and the remaining data is used as training parameters for regression training of machine learning. The regression models of machine learning include but are not limited to: linear regression, logistic regression, polynomial regression, ridge regression, Lasso regression, and random forest regression. The present invention uses linear regression, ridge regression, Lasso regression, and random forest regression. Each regression method performs regression analysis on five states: DC charging, AC charging, intense driving, smooth driving, and parking, and votes on the results. A certain regression method with the highest accuracy rate ≥3 times in each vehicle state is regarded as the optimal regression solution, and the model obtained by the regression method is used as the prediction model for predicting OCV after parking. Train until the preset model threshold _3 is met. The present invention requires that the threshold is greater than 90%. And retain the regression result parameters. The selection of the threshold is based on the use of R 2 The verification method is:

[0102]

[0103] Its numerator is the mean square error MSE, and the denominator is the variance Var, then R 2 The value is between 0 and 1, and the closer it is to 1, the more accurate the prediction.

[0104] Part 2 includes the following steps:

[0105] Step 208: Quasi-steady-state data acquisition and SOC interpolation

[0106] If there is a power-off static time T (T is determined based on the battery design index) in the state 202-2 that is sufficient to eliminate polarization, according to the conditions of step 207, the open circuit voltage data for eliminating polarization after power-off is regressively calculated as the steady-state open circuit voltage after power-off, and the SOC is interpolated according to the method of step 204 to obtain the static predicted SOCn of each battery cell. At the same time, according to the method of step 204, the voltage of each battery cell of the first frame of power-on data after power-on is interpolated as the power-on SOCn of each battery cell.

[0107] Step 209: Determine the highest / lowest cell position and estimate the self-discharge rate

[0108] Determine the highest / lowest cell number before power-off and compare it with the highest / lowest cell number after power-on to analyze whether they are consistent. If they are consistent, calculate the self-discharge rate:

[0109]

[0110] The monthly self-discharge rate of each battery cell can be calculated, and based on the threshold, it can be determined whether there are batteries with self-discharge rates exceeding the limit. If an over-limit battery cell is found, the user will be reminded through the APP or background to go to the store for maintenance.

[0111] The memory 31 is a computer-readable storage medium that can be used to store software programs, computer executable programs and modules, such as program instructions and modules corresponding to the battery leakage warning method in the embodiment of the present invention. The processor 30 executes various functional applications and data processing of the server by running the software programs, instructions and modules stored in the memory 31, that is, to realize the above-mentioned battery self-discharge rate abnormality judgment method based on vehicle networking data analysis.

[0112] The memory 31 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the server, etc. In addition, the memory 31 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 31 may further include a memory remotely arranged relative to the processor 30, and these remote memories may be connected to the server via a network. Embodiments of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0113] A storage medium containing computer executable instructions, wherein the computer executable instructions are used to execute a method for determining abnormal battery self-discharge rate based on vehicle networking data analysis when executed by a computer processor, the method comprising:

[0114] Receive vehicle data reported by the vehicle;

[0115] determining a battery self-discharge rate based on the vehicle data;

[0116] If it is determined that the battery self-discharge rate exceeds the set self-discharge rate threshold, a reminder message of battery self-discharge abnormality is sent to a terminal corresponding to a user of the vehicle.

[0117] Of course, the computer executable instructions of a storage medium including computer executable instructions provided in an embodiment of the present invention are not limited to the operations of the method described above, and can also execute related operations in the battery management method provided in any embodiment of the present invention.

[0118] Through the above description of the implementation mode, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation mode. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, a read-only memory (ROM), a random access memory (RAM), a flash memory (FLASH), a hard disk or an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a vehicle, or a network device, etc.) to perform the battery self-discharge rate abnormality judgment method based on vehicle networking data analysis described in each embodiment of the present invention. It is worth noting that in the above-mentioned embodiment of the battery self-discharge rate abnormality judgment device based on vehicle networking data analysis, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be realized; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0119] Example 2

[0120] The present invention uses this method to analyze the data of a certain model of electric vehicles in April 2021, and there are 100 electric vehicles.

[0121] 1. Perform the operations of step 201, step 202, and step 203 to obtain 835 matching result fragments;

[0122] 2. Execute step 204, the interpolation method is linear interpolation;

[0123] 3. Execute step 205, where the point plotting curve of the bicycle is:

[0124] 4. Execute steps 206 and 207, use 75% of the data as the training set and 25% as the validation set, and calculate R^2 to be 96.89%, which can predict the open circuit voltage of the battery. The results are as follows:

[0125] 5. Execute step 208 to analyze the vehicle with code LFB? ? ? ? ? 23. Its monthly discharge rate is about 5%.

[0126] Through the above specific embodiments, this patent is feasible.

[0127] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for judging abnormal battery self-discharge rate based on vehicle networking data analysis, characterized in that: First, a model for estimating the open circuit voltage of the vehicle in a stable state after powering off is trained based on historical data. Then, the self-discharge rate is judged based on the open circuit voltage of the vehicle after powering off and the quasi-steady-state vehicle data read after powering on after standing still. The specific steps include: A. Receive vehicle data reported by the vehicle; B. Quasi-steady state identification and selection; C. Data screening; D. Single cell voltage-SOC interpolation of voltage characterization quantity; E. Calculate the extreme deltaSOC and collect training samples; F. Summary of status before parking; G. Machine learning regression to obtain a prediction model; H. Quasi-steady-state data acquisition and SOC interpolation; I. Determine the highest / lowest cell position and estimate the self-discharge rate; For the segments that meet the status of step E, obtain all the data of T minutes before parking as required; according to the classification of the vehicle status data before parking, use the voltage of the vehicle after parking as the prediction value, and use the remaining data as training parameters for regression training of machine learning; each regression method performs regression analysis on five states, namely DC charging, AC charging, intense driving, stable driving, and parking, and votes on the results; a regression method with the highest accuracy rate ≥3 times in each vehicle state is regarded as the optimal regression solution, and the model obtained by this regression method is used as the prediction model for predicting OCV after parking.

2. The method for determining abnormal battery self-discharge rate based on vehicle networking data analysis according to claim 1, characterized in that: Step B specifically comprises the following steps: B1. Sort the data reported by the vehicle by time and calculate the time difference between every two frames (timediff); B2. Extract the frame whose timediff is greater than T hours and its previous frame, wherein the frame whose timediff is greater than T hours is represented by a vehicle power-on frame, and its previous frame is represented by a vehicle power-off frame. These two frames of data appear in pairs and can be regarded as a data segment.

3. The method for determining abnormal battery self-discharge rate based on vehicle networking data analysis according to claim 1, characterized in that: Step D is specifically as follows: according to the cell voltage-SOC table of the battery characteristics, the cell voltage in each frame is interpolated between voltage and SOC. The interpolation method may use linear interpolation, inverse distance weighted method, spline interpolation method, Kriging method, discrete smooth interpolation and trend surface smooth interpolation method.

4. The method for determining abnormal battery self-discharge rate based on vehicle networking data analysis according to claim 1, characterized in that: Step E specifically comprises the following steps: E1. Calculate the voltage SOC extreme difference value of each frame as deltaSOC; E2. Analyze the two indicators of deltaSOC range and deltaSOC change rate: E21. Analyze deltaSOC range: When the deltaSOC range is less than or equal to a certain threshold value _1, enter the training sample; E22. Calculate the deltaSOC change rate: Based on the change in mileage and the change in time, the horizontal axis is the dual coordinate axis of time / mileage, and the vertical axis is deltaSOC. Plot each deltaSOC on the graph, and obtain the difference between the two deltaSOCs, that is, the SOC extreme change. The SOC difference / interval time can be used to calculate the SOC difference change rate. Within N months, the SOC extreme change must be less than or equal to a certain threshold value_2.

5. The method for determining abnormal battery self-discharge rate based on vehicle networking data analysis according to claim 1, characterized in that: Step G, specifically, is as follows: according to the classification of the vehicle state data before parking, the voltage of the vehicle after parking is used as the prediction value, and the remaining data is used as training parameters for regression training of machine learning; multiple regression methods are used for regression analysis of the five states of DC charging, AC charging, intense driving, stable driving, and parking, and the results are voted. A certain regression method with the highest accuracy rate in each vehicle state is ≥3 times, which is regarded as the optimal regression solution. The model obtained by the regression method is used as the prediction model for predicting the OCV after parking, and is trained to meet the preset model threshold value _3. The selection of the threshold is verified by using the R^2 method, and the method is: The numerator is the mean square error MSE, and the denominator is the variance Var. The R^2 value is between 0 and 1. The closer it is to 1, the more accurate the prediction.

6. The method for determining abnormal battery self-discharge rate based on vehicle networking data analysis according to claim 1, characterized in that: Step H, specifically: there is a power-off static time T in the state of step E2 that is sufficient to eliminate polarization. According to the conditions of step G, regress and calculate the open-circuit voltage data that eliminates polarization after power-off, and use it as the steady-state open-circuit voltage after power-off. Perform SOC interpolation according to the method of step D to obtain the static predicted SOCn of each battery cell. At the same time, according to the method of step D, interpolate the voltage of each battery cell of the first frame of power-on data after power-on as the power-on SOCn of each battery cell.

7. The method for determining abnormal battery self-discharge rate based on vehicle networking data analysis according to claim 1, characterized in that: Step I is specifically as follows: determine the highest / lowest cell number before power-off and compare it with the highest / lowest cell number after power-on to analyze whether they are consistent. If they are consistent, calculate the self-discharge rate: The monthly self-discharge rate of each battery cell can be calculated, and based on the threshold, it can be determined whether there are batteries with self-discharge rates exceeding the limit. If an over-limit battery cell is found, the user will be reminded through the APP or background to go to the store for maintenance.

8. The battery self-discharge rate abnormality judgment device according to the battery self-discharge rate abnormality judgment method based on vehicle networking data analysis according to claim 1 is characterized in that: It includes a receiving module, a first determining module and a first reminding module; The receiving module is used to receive vehicle data reported by the vehicle; the first determining module is used to determine the risk and level of battery pack self-discharge according to the vehicle data; The first reminder module is used to send a reminder message of battery pack self-discharge warning to the user of the vehicle and the corresponding terminal of the after-sales department if it is determined that the vehicle reported data meets the risk level of battery pack self-discharge.

9. A server, characterized in that: include: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement a method for determining abnormal battery self-discharge rate based on Internet of Vehicles data analysis as described in claim 1.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the program is executed by a processor, a method for determining abnormality in battery self-discharge rate based on vehicle networking data analysis as claimed in claim 1 is implemented.