A battery self-discharge fault diagnosis method and system

Abstract

The application discloses a battery self-discharge fault diagnosis method and system, and relates to the technical field of battery safety. The diagnosis method obtains actual battery data, divides the obtained actual battery data by using a preset division rule to obtain a plurality of data segments with specific characteristics, then extracts features of the divided data segments by using a preset algorithm to obtain a plurality of voltage features capable of representing battery self-discharge characteristics, and finally comprehensively evaluates each battery monomer in the battery according to the extracted voltage features to screen out battery monomers with self-discharge faults. The diagnosis method fully utilizes data generated by the battery during actual use, can accurately identify and diagnose the battery self-discharge fault in a short time, and significantly improves the safety and economy of the battery operation.

Description

Technical Field

[0001] This invention relates to the field of battery safety technology, and in particular to a method and system for diagnosing battery self-discharge faults. Background Technology

[0002] With the booming development of electrified transportation vehicles (such as electric vehicles, rail transit vehicles, electric ships, and electric aircraft) and related industries such as energy storage power stations, the safety and reliability of batteries (such as lithium-ion batteries, sodium-ion batteries, quasi-solid-state batteries, and all-solid-state batteries), as key components in these industries, are receiving increasing attention. Among these issues, abnormal self-discharge, a relatively common battery failure phenomenon, significantly impacts battery performance and lifespan. Specifically, battery self-discharge refers to the spontaneous decrease in capacity of a battery after it has been left in an open-circuit state for a period of time. When there are differences in the degree of self-discharge between individual battery cells, it leads to significant inconsistencies between the cells, which severely shorten the overall battery lifespan.

[0003] Existing diagnostic methods for battery self-discharge faults mainly focus on studying the self-discharge characteristics of batteries through experiments or modeling. Experimental methods typically require prolonged static placement and monitoring of the battery under specific experimental conditions to obtain self-discharge data. However, this method is not only time-consuming and labor-intensive, but also difficult to achieve real-time and accurate diagnosis of battery self-discharge faults in practical applications. While modeling methods can predict battery self-discharge behavior by establishing mathematical models of the battery, the accuracy and reliability of the models are easily affected by various factors, such as the battery's aging level and the usage environment. Therefore, researching a simple and rapid method for diagnosing battery self-discharge faults, capable of timely and accurately identifying and eliminating battery cells with abnormal self-discharge, is of paramount importance for improving the safety, reliability, and lifespan of battery systems in applications such as electrified vehicles and energy storage power stations. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for diagnosing battery self-discharge faults. This method and system have broad applicability and can be widely applied to various types of electrified vehicles, including electric vehicles, rail transit vehicles, electric ships, and electric aircraft, as well as energy storage power stations. This invention focuses on electric vehicles as the research object. Based on actual battery data from electric vehicles, the diagnostic method deeply mines and extracts multiple key voltage features that accurately characterize battery self-discharge properties. These extracted voltage features are then used to achieve accurate identification and diagnosis of battery self-discharge faults, effectively solving the problems of poor accuracy and insufficient reliability in existing technologies that use experimental or modeling methods to diagnose battery self-discharge faults.

[0005] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution:

[0006] This invention provides a method for diagnosing battery self-discharge faults, comprising:

[0007] Obtain actual battery data;

[0008] The acquired actual battery data is divided using preset division rules to obtain multiple data segments with specific characteristics;

[0009] The pre-defined algorithm is used to extract features from each of the divided data segments to obtain multiple voltage features that can characterize the battery's self-discharge characteristics.

[0010] The extracted voltage characteristics are used to comprehensively evaluate each individual cell in the battery in order to screen out cells with self-discharge faults.

[0011] In one embodiment of the present invention, obtaining actual battery data includes:

[0012] Acquire actual battery data generated during the actual operation of electric vehicles. The actual battery data includes diverse information about the battery under actual usage scenarios, which can reflect the comprehensive impact of user behavior, vehicle operating status and environmental factors on battery performance.

[0013] In one embodiment of the present invention, the step of dividing the acquired actual battery data using a preset division rule to obtain multiple data segments with specific characteristics includes:

[0014] Extract battery parameters from the actual battery data. The battery parameters include current, vehicle speed, vehicle charging status and state of charge. The vehicle charging status includes parking charging status, driving charging status, not charging status and charging completed status.

[0015] Based on the extracted battery parameters, the actual battery data is divided using a preset partitioning rule to obtain multiple data segments with specific characteristics.

[0016] In one embodiment of the present invention, the step of dividing the actual battery data according to the extracted battery parameters using a preset division rule to obtain multiple data segments with specific characteristics includes:

[0017] The actual battery data is divided according to the following rules: the current is 0, the vehicle speed is 0, the vehicle charging state is not charged, the state of charge is constant, and the segment time exceeds a preset time, so as to obtain multiple data segments with specific characteristics.

[0018] In one embodiment of the present invention, in the step of extracting features from each of the divided data segments using a preset algorithm to obtain multiple voltage features that can characterize the self-discharge characteristics of the battery, the voltage features include average voltage value, voltage standard deviation, and voltage change rate.

[0019] In one embodiment of the present invention, the step of using a preset algorithm to extract features from each of the divided data segments to obtain multiple voltage features that can characterize the self-discharge characteristics of the battery includes:

[0020] Based on the actual voltage values ​​of each battery cell at different time points within each data segment, the average voltage value of each battery cell within each data segment is calculated using the mean value calculation formula.

[0021] Based on the average voltage value of each battery cell in each data segment and the actual voltage value at different time points in the corresponding data segment, the standard deviation of voltage of each battery cell in each data segment is calculated using the standard deviation calculation formula.

[0022] Based on the maximum and minimum values ​​of the overall battery voltage, the maximum and minimum values ​​of the individual battery cell voltages within each data segment, and the time difference between the maximum and minimum values ​​within the corresponding data segment, the voltage change rate of the overall battery and each individual battery cell within each data segment is calculated.

[0023] In one embodiment of the present invention, the step of comprehensively evaluating each battery cell in the battery based on the extracted voltage characteristics to screen out battery cells with self-discharge faults includes:

[0024] Threshold conditions are set for each extracted voltage feature using the three-standard-deviation rule. These threshold conditions are used to determine whether the voltage features of a single battery cell are within the normal range.

[0025] Based on the extracted voltage characteristics and the set threshold conditions, each battery cell in the battery is comprehensively evaluated to screen out battery cells with self-discharge faults.

[0026] In one embodiment of the present invention, the step of setting threshold conditions for each extracted voltage feature using the three-standard-deviation rule includes:

[0027] Obtain the mean and standard deviation of the average voltage values ​​of all battery cells in each data segment, and use the three-standard-deviation rule to set threshold conditions for the average voltage values ​​of battery cells in different data segments.

[0028] The standard deviation of voltage and rate of change of voltage for each battery cell in each data segment are standardized to obtain the standardized standard deviation of voltage and rate of change of voltage. Based on the standardized standard deviation of voltage and rate of change of voltage, the threshold conditions for the standard deviation of voltage and rate of change of voltage for battery cells in different data segments are set using the three-standard deviation rule.

[0029] In one embodiment of the present invention, the step of comprehensively evaluating each battery cell in the battery based on the extracted voltage characteristics and set threshold conditions to screen out battery cells with self-discharge faults includes:

[0030] If any voltage feature extracted from any data segment does not meet the corresponding set threshold condition, the corresponding data segment is determined to be an abnormal data segment.

[0031] When the number of abnormal data segments identified for any single battery cell based on any voltage characteristic is greater than or equal to a preset number, the corresponding battery cell is determined to be a battery cell with a self-discharge fault.

[0032] Based on the same inventive concept, another embodiment of the present invention provides a battery self-discharge fault diagnosis system, wherein the system is implemented using the battery self-discharge fault diagnosis method as described in any of the above embodiments, including:

[0033] The data acquisition module is used to acquire actual battery data;

[0034] The data segmentation module is used to divide the acquired actual battery data using preset segmentation rules to obtain multiple data segments with specific characteristics.

[0035] The voltage feature extraction module is used to extract features from each of the divided data segments using a preset algorithm to obtain multiple voltage features that can characterize the battery's self-discharge characteristics.

[0036] The fault diagnosis module is used to comprehensively evaluate each battery cell in the battery based on the extracted voltage characteristics in order to screen out battery cells with self-discharge faults.

[0037] As described above, this invention provides a battery self-discharge fault diagnosis method. This method acquires actual battery data, which comprehensively covers various information about the battery in actual usage scenarios, reflecting the impact of user behavior, vehicle operating status, and environmental factors on battery performance. The acquired actual battery data is divided using preset segmentation rules to obtain multiple data segments with specific characteristics. These preset segmentation rules fully consider the inherent correlation between battery self-discharge phenomena and battery operating parameters. A preset algorithm is used to extract features from each segmented data segment to obtain multiple voltage features characterizing the battery's self-discharge properties. These voltage features accurately reflect the voltage change patterns of individual battery cells during self-discharge. Based on the extracted voltage features, each individual battery cell is comprehensively evaluated to identify those with self-discharge faults. This diagnostic method has the ability to quickly identify and accurately locate electric vehicles with self-discharge faults. Based on the extracted key voltage features, it can efficiently and accurately identify batteries with different degrees of fault. Compared to traditional methods of measuring battery self-discharge rate based on experiments or modeling, the diagnostic method described in this invention fully utilizes data generated during actual battery use, enabling the identification and diagnosis of battery self-discharge faults in a short time, thereby significantly improving the safety and economy of battery operation. Of course, any product implementing this invention does not necessarily need to achieve all the advantages described above simultaneously. Attached Figure Description

[0038] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a schematic flowchart of a battery self-discharge fault diagnosis method provided for an exemplary embodiment of this application.

[0040] Figure 2 This is a system block diagram of a battery self-discharge fault diagnosis method provided for an exemplary embodiment of this application.

[0041] Figure 3 This is a schematic diagram of a process for setting threshold conditions for voltage characteristics, provided as an exemplary embodiment of this application.

[0042] Figure 4 This is a schematic diagram of the structure of a battery self-discharge fault diagnosis system provided for another exemplary embodiment of this application. Detailed Implementation

[0043] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0044] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0045] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, publicly known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0046] Taking the electric vehicle field as an example, current research on battery self-discharge largely relies on data collected in laboratory environments. Compared to laboratory data, battery data generated during actual driving of electric vehicles can more accurately and comprehensively reflect the combined effects of user behavior (such as driving style, charging frequency, and depth of discharge), vehicle operating status (such as load, mileage, and operating conditions), and environmental factors (such as temperature, humidity, and altitude) on battery performance. Therefore, using actual battery data from electric vehicles for self-discharge fault detection is accurate and rapid, and has high application value in engineering.

[0047] To address the issues of poor accuracy and reliability in existing technologies that use experimental or modeling methods to diagnose battery self-discharge faults, this invention provides a battery self-discharge fault diagnosis method. This method is based on actual battery data from electric vehicles, deeply mines and extracts multiple key voltage features that can accurately characterize the battery's self-discharge properties, and uses the extracted voltage features to achieve accurate identification and diagnosis of battery self-discharge faults.

[0048] Please see Figure 1 As shown, in an exemplary embodiment of this application, the battery self-discharge fault diagnosis method includes the following steps:

[0049] S100: Obtain actual battery data;

[0050] S200: The acquired actual battery data is divided using preset division rules to obtain multiple data segments with specific characteristics;

[0051] S300: Use a preset algorithm to extract features from each of the divided data segments to obtain multiple voltage features that can characterize the battery's self-discharge characteristics.

[0052] S400: Based on the extracted voltage characteristics, a comprehensive evaluation of each individual cell in the battery is performed to screen out cells with self-discharge faults.

[0053] The steps of the above-mentioned battery self-discharge fault diagnosis method will be discussed in detail below.

[0054] First, please refer to Figure 1 As shown, step S100 is executed, which involves obtaining actual battery data.

[0055] In an exemplary embodiment of this application, step S100, obtaining actual battery data specifically includes: obtaining actual battery data generated by the electric vehicle during actual operation. The actual battery data includes diverse information about the battery in actual usage scenarios, which can reflect the comprehensive impact of user behavior, vehicle operating status, and environmental factors on battery performance.

[0056] It should be noted that in this embodiment, the selected research object is an electric vehicle. Therefore, the actual battery data obtained in step S100 is the real battery data generated by the electric vehicle during actual operation. However, in other embodiments, if the selected research object is changed to other electrified vehicles (such as rail transit vehicles, electric ships, electric aircraft, etc.) or energy storage power stations, the actual battery data obtained in step S100 will be adjusted accordingly to the actual battery data corresponding to the research object to ensure the applicability and accuracy of the diagnostic method.

[0057] Specifically, the actual battery data originates from real-time records of the electric vehicle during actual operation, rather than battery data from a laboratory simulation environment. While a laboratory environment allows for the study of specific battery performance through controlled variables, it cannot fully reflect the complex and varied conditions faced by the battery in real-world use. In this embodiment, the actual battery data includes diverse information about the battery in actual usage scenarios, including but not limited to user behavior habits, vehicle operating status, and environmental factors. It should be noted that user behavior habits include, but are not limited to, the user's charging frequency and driving style, which affect the battery's charging and discharging process, thereby impacting battery performance. The vehicle operating status includes the vehicle's load, mileage, speed, and road conditions. Notably, the vehicle operating status determines the battery's workload and discharge mode, significantly influencing battery wear and performance changes. Environmental factors include ambient temperature and humidity; battery performance varies significantly under different environmental conditions.

[0058] Next, please continue reading. Figure 1 As shown, step S200 is executed, which involves dividing the acquired actual battery data using a preset division rule to obtain multiple data segments with specific characteristics.

[0059] In an exemplary embodiment of this application, step S200 further includes the following steps:

[0060] S210: Extract battery parameters from the actual battery data. The battery parameters include current, vehicle speed, vehicle charging status and state of charge. The vehicle charging status includes parking charging status, driving charging status, not charging status and charging completed status.

[0061] S220: Based on the extracted battery parameters, the actual battery data is divided using a preset division rule to obtain multiple data segments with specific characteristics.

[0062] It should be noted that in step S200, the operation of extracting battery parameters and dividing the actual battery data using preset division rules aims to obtain data segments that can approximate the battery's resting state to a certain extent. When the battery is in a resting state for a long time, its voltage can be approximately equivalent to the open-circuit voltage, which is of great significance for accurately analyzing battery characteristics. Since the research object selected in this embodiment is an electric vehicle, the battery parameters extracted from the actual battery data include current, vehicle speed, vehicle charging state, and state of charge. These parameters can comprehensively reflect the battery's working state during the actual operation of the electric vehicle, providing a reliable basis for data division. However, in other embodiments, if the selected research object is changed to other electrified transportation vehicles (such as rail transit vehicles, electric ships, electric aircraft, etc.) or energy storage power stations, the battery parameters extracted in step S200 need to be adjusted accordingly to battery parameters closely related to the research object. This ensures that the actual battery data can be reasonably divided into multiple segments based on the extracted battery parameters, thereby effectively simulating the battery's resting state and laying a solid foundation for subsequent operations such as battery self-discharge fault diagnosis.

[0063] Specifically, since battery self-discharge is closely related to voltage, open-circuit voltage is commonly used to describe the battery's self-discharge rate. However, accurate measurement of the self-discharge rate generally requires complex experiments. For a more realistic and effective self-discharge scenario, please refer to [link to relevant documentation]. Figure 2 As shown, in this embodiment, battery parameters such as current, vehicle speed, vehicle charging state, and state of charge (SOC) are selected for segmentation. It should be noted that before segmentation, key battery parameters need to be extracted from actual battery data. These key battery parameters include, but are not limited to, current, vehicle speed, vehicle charging state, and SOC. The vehicle charging state describes the battery's charging status at a specific moment, including parked charging state, driving charging state, no charging state, and charging complete state. More specifically, the parked charging state indicates that the vehicle is parked and charging is in progress. The driving charging state indicates that during vehicle operation, the energy recovery system converts some kinetic energy into electrical energy to charge the battery. The no-charging state indicates that the vehicle is neither parked nor driving, and the battery is either naturally discharging or supplying power for normal vehicle operation. The charging complete state indicates that the battery is fully charged and the charging process is complete. Furthermore, the SOC represents the percentage of the battery's current remaining charge relative to its total capacity, and is an important indicator for measuring the battery's charge level.

[0064] After extracting key battery parameters, the actual battery data is divided according to a preset division rule to obtain multiple data segments with specific characteristics. In this embodiment, the actual battery data is divided according to the following rules: the current is 0, the vehicle speed is 0, the vehicle is in an uncharged state, the state of charge is constant, and the segment time exceeds a preset time, thereby obtaining multiple data segments with specific characteristics. It should be noted that when the current is 0, it indicates that the battery is neither charging nor discharging and is in a relatively stable state. When the vehicle speed is 0, it indicates that the vehicle is stationary. In addition, when the vehicle is stationary and not charging, the battery charging and discharging situation is relatively simple, and the data characteristics are more obvious. Therefore, based on the above three conditions, if the battery's state of charge remains constant for a period of time, and this period of time exceeds the preset time, the actual battery data within this period is divided into an independent data segment. It is worth noting that the preset time can be set according to actual application needs. In this embodiment, the preset time is set to 3 hours. By extracting key battery parameters and dividing the actual battery data according to the preset division rule, a data foundation is provided for subsequent battery self-discharge fault diagnosis.

[0065] Next, please continue reading. Figure 1 As shown, step S300 is executed, which involves using a preset algorithm to extract features from each of the divided data segments to obtain multiple voltage features that can characterize the battery's self-discharge characteristics.

[0066] It should be noted that you should refer to [link / reference]. Figure 2 As shown, in this embodiment, the voltage characteristics include the average voltage value, the voltage standard deviation, and the voltage change rate.

[0067] In an exemplary embodiment of this application, step S300 further includes the following steps:

[0068] S310: Based on the actual voltage values ​​of each battery cell at different time points within each data segment, the average voltage value of each battery cell within each data segment is calculated using the mean value calculation formula.

[0069] S320: Based on the average voltage value of each battery cell in each data segment and the actual voltage value at different time points in the corresponding data segment, the standard deviation of the voltage of each battery cell in each data segment is calculated using the standard deviation calculation formula.

[0070] S330: Based on the maximum and minimum values ​​of the overall battery voltage, the maximum and minimum values ​​of the individual battery cell voltages within each data segment, and the time difference between the maximum and minimum values ​​within the corresponding data segment, calculate the voltage change rate of the overall battery and the individual battery cell within each data segment.

[0071] Specifically, when left undisturbed, the voltage of a single battery cell will spontaneously decrease. However, the voltage drop of a battery cell with a self-discharge fault will be greater than that of a normal cell, and therefore its average voltage value will be lower than that of a normal battery cell. By comparing the average voltage values ​​among battery cells, it is possible to preliminarily determine which battery cells have a self-discharge fault. Based on the above data segmentation method, the average voltage value of each battery cell within each data segment can be calculated:

[0072]

[0073] in, It is the average voltage of a single battery cell. is the actual voltage of a single battery cell, and n is the number of valid data points in each data segment.

[0074] When the initial states of individual battery cells differ, using average voltage values ​​to diagnose self-discharge faults is not very effective. Because the voltage changes of self-discharge faulty cells are more severe than those of normal cells, the dispersion of their voltage changes is also greater, resulting in a larger voltage standard deviation. A more accurate self-discharge diagnosis can be achieved by comparing the voltage standard deviations between different battery cells. The voltage standard deviations of each battery cell within each data segment are as follows:

[0075]

[0076] Where S is the standard deviation of voltage.

[0077] Within the same time frame, the voltage change of the battery is closely related to the voltage change of individual battery cells; the voltage change of a self-discharge faulty cell is faster than that of a normal cell. By utilizing voltage change, the influence of the initial state of the battery cell itself is eliminated. Therefore, the self-discharge fault phenomenon reflected by the voltage change rate is more obvious and accurate than the average voltage value. The calculation formula is as follows:

[0078]

[0079] Among them, K line U represents the overall voltage change rate of the battery, while K represents the voltage change rate of a single battery cell. num It is the number of individual battery cells, δ t1 It is the time difference between the maximum and minimum values ​​of the overall battery voltage, δ. t2 It is the time difference between the maximum and minimum values ​​of the voltage of a single battery cell. V max and V min These are the maximum and minimum values ​​of the overall battery voltage, respectively. and These are the maximum and minimum values ​​of the battery cell voltage, respectively.

[0080] Finally, please continue reading Figure 1 As shown, step S400 is executed, which involves comprehensively evaluating each battery cell in the battery based on the extracted voltage characteristics in order to screen out battery cells with self-discharge faults.

[0081] In an exemplary embodiment of this application, step S400 further includes the following steps:

[0082] S410: Use the three-standard-deviation rule to set threshold conditions for each extracted voltage feature, and the threshold conditions are used to determine whether the voltage features of the battery cell are within the normal range.

[0083] S420: Based on the extracted voltage characteristics and the set threshold conditions, a comprehensive evaluation is performed on each individual cell in the battery to screen out cells with self-discharge faults.

[0084] In an exemplary embodiment of this application, step 410 further includes the following steps:

[0085] S411: Obtain the mean and standard deviation of the average voltage values ​​of all battery cells in each data segment, and use the three-standard-deviation rule to set the threshold conditions for the average voltage values ​​of battery cells in different data segments.

[0086] S412: Standardize the voltage standard deviation and voltage change rate of each battery cell in each data segment to obtain the standardized voltage standard deviation and voltage change rate. Based on the standardized voltage standard deviation and voltage change rate, use the three-standard-deviation rule to set the threshold conditions for the voltage standard deviation and voltage change rate of battery cells in different data segments.

[0087] In an exemplary embodiment of this application, step 420 further includes the following steps:

[0088] S421: When any voltage feature extracted from any data segment does not meet the corresponding set threshold condition, the corresponding data segment is determined to be an abnormal data segment.

[0089] S422: When the number of abnormal data segments determined based on any voltage characteristic of any battery cell is greater than or equal to a preset number, the corresponding battery cell is determined to be a battery cell with a self-discharge fault.

[0090] Specifically, please refer to Figure 3As shown, in this embodiment, the method for diagnosing battery self-discharge faults involves setting corresponding threshold conditions for each voltage characteristic. Most battery cells exhibit high voltage consistency, therefore the average voltage values ​​of similar battery cells conform to a normal distribution. The mean and standard deviation of the average voltage values ​​of all battery cells within each data segment are calculated, and the threshold conditions are determined using the three-standard-deviation (3σ) rule.

[0091]

[0092] in, and These are the upper and lower limits of the threshold, respectively. and These are the mean and standard deviation of the average voltage values ​​of all cells within a given data segment. If the average voltage value of a cell within a data segment exceeds a threshold, that data segment is considered an abnormal data segment.

[0093] Considering the potential differences in the initial state between different battery cells, as well as the influence of different SOCs and resting times, it is necessary to standardize the voltage standard deviation and voltage change rate, two voltage characteristics, in segments.

[0094]

[0095] in, and These are the standardized voltage standard deviation and voltage rate of change, S. ave and S std These are the mean and standard deviation of the voltage standard deviation of all individual cells within a given data segment, respectively, K. ave and K std These are the mean and standard deviation of the voltage change rate of all individual battery cells within a given data segment. In this embodiment, the standardized values ​​are... and The mean and variance are both 0 and 1, respectively. Therefore, for these two voltage characteristics, the threshold condition can be set according to the 3σ rule as follows:

[0096]

[0097] in, and These are the upper and lower limits of the threshold, respectively. When the standard deviation of the voltage of a single battery cell or the rate of change of voltage exceeds the threshold limit within a certain data segment, the data segment is determined to be an abnormal data segment.

[0098] It should be noted that in order to exclude cases where abnormal values ​​exceed the threshold range, it is necessary to set the number of abnormal data segments. In this embodiment, a battery cell with an abnormal data segment number greater than or equal to 7 segments can be diagnosed as a battery cell with a self-discharge fault.

[0099] In summary, this invention provides a battery self-discharge fault diagnosis method. This method acquires actual battery data, which comprehensively covers various information about the battery in actual usage scenarios, reflecting the impact of user behavior, vehicle operating status, and environmental factors on battery performance. The acquired actual battery data is divided using preset segmentation rules to obtain multiple data segments with specific characteristics. These preset segmentation rules fully consider the inherent correlation between battery self-discharge phenomena and battery operating parameters. A preset algorithm is used to extract features from each segmented data segment to obtain multiple voltage features characterizing the battery's self-discharge properties. These voltage features accurately reflect the voltage change patterns of individual battery cells during self-discharge. Based on the extracted voltage features, each individual battery cell is comprehensively evaluated to identify those with self-discharge faults. This diagnostic method has the ability to quickly identify and accurately locate electric vehicles with self-discharge faults, and can efficiently and accurately identify batteries with different degrees of fault based on multiple extracted key voltage features. Compared to traditional methods of measuring battery self-discharge rate based on experiments or modeling, the diagnostic method described in this invention makes full use of the data generated by the battery during actual use, and can complete the identification and diagnosis of battery self-discharge faults in a short time, thereby significantly improving the safety and economy of battery operation.

[0100] Based on the same inventive concept, please refer to Figure 4 As shown, another embodiment of the present invention also provides a battery self-discharge fault diagnosis system 100, which is implemented using the battery self-discharge fault diagnosis method as described in any of the above embodiments, including:

[0101] Data acquisition module 110 is used to acquire actual battery data;

[0102] The data segmentation module 120 is used to divide the acquired actual battery data using preset segmentation rules to obtain multiple data segments with specific characteristics.

[0103] The voltage feature extraction module 130 is used to extract features from each of the divided data segments using a preset algorithm to obtain multiple voltage features that can characterize the self-discharge characteristics of the battery.

[0104] The fault diagnosis module 140 is used to comprehensively evaluate each battery cell in the battery based on the extracted voltage characteristics in order to screen out battery cells with self-discharge faults.

[0105] It should be noted that the battery self-discharge fault diagnosis system 100 includes the battery self-discharge fault diagnosis method described in any of the above embodiments. Since the battery self-discharge fault diagnosis system 100 provided in this embodiment belongs to the same inventive concept as the battery self-discharge fault diagnosis method provided in any of the above embodiments, it has at least the same beneficial effects, and will not be elaborated further here.

[0106] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for diagnosing battery self-discharge faults, characterized in that, include: Obtain actual battery data; The acquired actual battery data is divided using preset division rules to obtain multiple data segments with specific characteristics; The pre-defined algorithm is used to extract features from each of the divided data segments to obtain multiple voltage features that can characterize the self-discharge characteristics of the battery. The voltage features include the average voltage value, voltage standard deviation, and voltage change rate. The extracted voltage characteristics are used to comprehensively evaluate each individual cell in the battery in order to screen out cells with self-discharge faults. The step of comprehensively evaluating each individual battery cell based on the extracted voltage characteristics to screen out battery cells with self-discharge faults includes: Threshold conditions are set for each extracted voltage feature using the three-standard-deviation rule. These threshold conditions are used to determine whether the voltage features of a single battery cell are within the normal range. Based on the extracted voltage characteristics and the set threshold conditions, each battery cell in the battery is comprehensively evaluated to screen out battery cells with self-discharge faults. The method of setting threshold conditions for each extracted voltage feature using the three-standard-deviation rule includes: Obtain the mean and standard deviation of the average voltage values ​​of all battery cells in each data segment, and use the three-standard-deviation rule to set threshold conditions for the average voltage values ​​of battery cells in different data segments. The standard deviation of voltage and rate of change of voltage for each battery cell in each data segment are standardized to obtain the standardized standard deviation of voltage and rate of change of voltage. Based on the standardized standard deviation of voltage and rate of change of voltage, the threshold conditions for the standard deviation of voltage and rate of change of voltage for battery cells in different data segments are set using the three-standard deviation rule.

2. The battery self-discharge fault diagnosis method according to claim 1, characterized in that, The acquisition of actual battery data includes: Acquire actual battery data generated during the actual operation of electric vehicles. The actual battery data includes diverse information about the battery under actual usage scenarios, which can reflect the comprehensive impact of user behavior, vehicle operating status and environmental factors on battery performance.

3. The battery self-discharge fault diagnosis method according to claim 1, characterized in that, The method involves dividing the acquired actual battery data using a preset partitioning rule to obtain multiple data segments with specific characteristics, including: Extract battery parameters from the actual battery data. The battery parameters include current, vehicle speed, vehicle charging status and state of charge. The vehicle charging status includes parking charging status, driving charging status, not charging status and charging completed status. Based on the extracted battery parameters, the actual battery data is divided using a preset partitioning rule to obtain multiple data segments with specific characteristics.

4. The battery self-discharge fault diagnosis method according to claim 3, characterized in that, The actual battery data is divided according to the extracted battery parameters using a preset partitioning rule to obtain multiple data segments with specific characteristics, including: The actual battery data is divided according to the following rules: the current is 0, the vehicle speed is 0, the vehicle charging state is not charged, the state of charge is constant, and the segment time exceeds a preset time, so as to obtain multiple data segments with specific characteristics.

5. The battery self-discharge fault diagnosis method according to claim 1, characterized in that, The method involves using a preset algorithm to extract features from each segment of data to obtain multiple voltage features that characterize the battery's self-discharge properties, including: Based on the actual voltage values ​​of each battery cell at different time points within each data segment, the average voltage value of each battery cell within each data segment is calculated using the mean value calculation formula. Based on the average voltage value of each battery cell in each data segment and the actual voltage value at different time points in the corresponding data segment, the standard deviation of voltage of each battery cell in each data segment is calculated using the standard deviation calculation formula. Based on the maximum and minimum values ​​of the overall battery voltage, the maximum and minimum values ​​of the individual battery cell voltages within each data segment, and the time difference between the maximum and minimum values ​​within the corresponding data segment, the voltage change rate of the overall battery and each individual battery cell within each data segment is calculated.

6. The battery self-discharge fault diagnosis method according to claim 1, characterized in that, The process involves comprehensively evaluating each individual battery cell based on extracted voltage characteristics and set threshold conditions to identify battery cells with self-discharge faults, including: If any voltage feature extracted from any data segment does not meet the corresponding set threshold condition, the corresponding data segment is determined to be an abnormal data segment. When the number of abnormal data segments identified for any single battery cell based on any voltage characteristic is greater than or equal to a preset number, the corresponding battery cell is determined to be a battery cell with a self-discharge fault.

7. A battery self-discharge fault diagnosis system, characterized in that, The system is implemented using the battery self-discharge fault diagnosis method as described in any one of claims 1 to 6, including: The data acquisition module is used to acquire actual battery data; The data segmentation module is used to divide the acquired actual battery data using preset segmentation rules to obtain multiple data segments with specific characteristics. The voltage feature extraction module is used to extract features from each of the divided data segments using a preset algorithm to obtain multiple voltage features that can characterize the battery's self-discharge characteristics. The fault diagnosis module is used to comprehensively evaluate each battery cell in the battery based on the extracted voltage characteristics in order to screen out battery cells with self-discharge faults.

Images