Battery rapid diagnosis method, storage medium and electronic device based on edge computing

Abstract

The present invention provides a rapid battery diagnosis method, storage medium, and electronic device based on edge computing. The method comprises: collecting battery information data from a battery management system at a preset frequency; convolving the battery information data using a first-order differential edge operator to obtain a convolution result; filtering out convolution results that fall outside the threshold range based on the convolution result and a threshold range; and determining whether anomalies exist in the battery information data based on the convolution results that fall outside the threshold range and the battery information data. This invention enables relatively accurate and rapid battery data processing on edge devices, enabling rapid diagnosis of battery performance and facilitating scalability.

Description

Technical Field

[0001] The present invention relates to the field of battery technology, in particular to the field of lithium battery performance analysis technology. Background Art

[0002] During normal operation of the energy storage power station, the BMS management system (Battery Management System) can monitor key parameters such as battery cell voltage, current charge and discharge current, and temperature in real time to achieve real-time protection. If you want to predict the degree of battery aging and the health of single cells, the general approach is to upload the data to the cloud platform, analyze a large amount of data on the platform, and implement predictions. Although this solution can ultimately achieve the results of data analysis, it has disadvantages such as a long collection and analysis cycle and low real-time performance. In application scenarios with more complex working conditions, such as energy storage power stations for frequency and peak regulation on the power generation side, higher requirements are placed on battery stability and safety. Therefore, a fast, relatively simple and stable method is needed to predict and diagnose the safety and health of the battery to a certain extent. Summary of the Invention

[0003] In view of the shortcomings of the prior art described above, the purpose of the present invention is to provide a battery rapid diagnosis method, storage medium and electronic device based on edge computing, which is used to solve the technical problem that the prior art cannot effectively and quickly diagnose battery performance.

[0004] To achieve the above-mentioned objectives and other related objectives, the present invention provides a battery rapid diagnosis method based on edge computing, the method comprising: collecting battery information data from a battery management system at a preset frequency; convolving the battery information data using a first-order differential edge operator to obtain a convolution result; filtering out convolution results that are not within the threshold range based on the convolution result and a threshold range; and determining whether there is an abnormality in the battery information data based on the convolution results that are not within the threshold range and the battery information data.

[0005] In one embodiment of the present invention, the battery information data includes battery cell voltage and battery charge and discharge current.

[0006] In one embodiment of the present invention, the method further includes: storing the battery cell voltages via a two-dimensional array according to the acquisition sequence, and storing the battery charge and discharge currents via a one-dimensional array according to the acquisition sequence.

[0007] In one embodiment of the present invention, one implementation method of using a first-order differential edge operator to convolve the battery information data to obtain the convolution result is: using a first-order differential edge operator to perform horizontal convolution and vertical convolution on the two-dimensional array respectively, to obtain the convolution results: horizontal gradient value and vertical gradient value.

[0008] In one embodiment of the present invention, before using a first-order differential edge operator to convolve the battery information data, it also includes: selecting a 3X3 first-order differential edge operator as an initial first-order differential edge operator; expanding the initial first-order differential edge operator to obtain the first-order differential edge operator required for convolution; horizontally expanding the initial first-order differential edge operator to an MX3-order matrix and vertically expanding it to an NX3-order matrix; wherein M is the total number of battery packs in series, and N is the total number of times the battery information data is collected.

[0009] In one embodiment of the present invention, in the process of performing horizontal convolution and vertical convolution on the two-dimensional array using the first-order differential edge operator, the first-order differential edge operator is moved with a step size of 1.

[0010] In one embodiment of the present invention, the determining whether there is an abnormality in the battery information data based on the convolution result that is not within the threshold range and the battery information data includes: if the horizontal gradient value is not within the horizontal gradient threshold range, obtaining the battery pack cell voltage data corresponding to the horizontal gradient value in the two-dimensional matrix, calculating the voltage difference value of two adjacent battery cells based on the battery pack cell voltage data, and determining whether there is an inter-group battery consistency abnormality based on the size of the voltage difference value and the voltage difference threshold; if the longitudinal gradient value is not within the longitudinal gradient threshold range, obtaining the battery pack cell voltage data corresponding to the longitudinal gradient value in the two-dimensional matrix, calculating the rate of change of the voltage difference between adjacent time points to time based on the battery pack cell voltage data, and determining whether there is a battery pack cell voltage abnormality based on the size of the change rate and the change rate threshold.

[0011] In one embodiment of the present invention, the determining whether there is an abnormality in the battery information data based on the convolution result that is not within the threshold range and the battery information data also includes: if the longitudinal gradient value is not within the longitudinal gradient threshold range, obtaining the battery pack single cell voltage data corresponding to the longitudinal gradient value in the two-dimensional matrix, calculating the rate of change of the voltage difference between adjacent time points with respect to time based on the battery pack single cell voltage data, and simultaneously obtaining the battery charge and discharge current data in the one-dimensional matrix; and determining whether there is a battery pack single cell voltage abnormality based on whether the rate of change of the voltage difference between adjacent time points with respect to time is positively correlated with the battery charge and discharge current.

[0012] To achieve the above-mentioned purpose and other related purposes, the present invention also provides a storage medium storing program instructions, which, when executed, implement the steps of the battery rapid diagnosis method based on edge computing as described above.

[0013] To achieve the above-mentioned objectives and other related objectives, the present invention also provides an electronic device, including a memory for storing a computer program; and a processor for running the computer program to implement the steps of the edge computing-based battery rapid diagnosis method as described above.

[0014] As described above, the battery rapid diagnosis method, storage medium and electronic device based on edge computing of the present invention have the following features:

[0015] Beneficial effects:

[0016] The present invention can realize relatively accurate and rapid battery data processing on edge devices, quickly diagnose battery performance, and facilitate expansion. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0018] Figure 1 Shown is a schematic diagram of the overall principle flow of a battery rapid diagnosis method based on edge computing in one embodiment of the present application;

[0019] Figure 2 Shown is a schematic diagram of the process of performing horizontal and vertical convolution on raw voltage data in a battery rapid diagnosis method based on edge computing in one embodiment of the present application;

[0020] Figure 3 Shown is a schematic diagram of a specific diagnostic process flow of a battery rapid diagnosis method based on edge computing in one embodiment of the present application;

[0021] Figure 4 Shown is a principle block diagram of an electronic device in an embodiment of the present application.

[0022] Figure 5 Shown is a schematic diagram of a network architecture of an electronic device application in one embodiment of the present application.

[0023] Component number description

[0024] 101 Electronic Equipment

[0025] 1001 Processor

[0026] 1002 Memory

[0027] Steps S100 to S400 DETAILED DESCRIPTION

[0028] The following describes the embodiments of the present invention through specific examples. 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. The 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 the following embodiments and features in the embodiments can be combined with each other unless they conflict.

[0029] The purpose of this embodiment is to provide a battery rapid diagnosis method, storage medium and electronic device based on edge computing, which is used to solve the technical problem that the existing technology cannot effectively and quickly diagnose battery performance.

[0030] In order to ensure that the battery rapid diagnosis method based on edge computing of this embodiment obtains more accurate analysis results, it is necessary to perform persistent analysis of the battery data. For edge computing devices with limited computing power and memory, the battery rapid diagnosis method based on edge computing of this embodiment adopts slicing and time-sharing to process battery data, and finally weightedly accumulates the analysis results of each time period. The following will elaborate on the principles and implementation methods of the battery rapid diagnosis method based on edge computing, storage medium and electronic device of the present invention, so that those skilled in the art can understand the battery rapid diagnosis method based on edge computing, storage medium and electronic device of the present invention without creative labor.

[0031] Example 1

[0032] This embodiment provides a battery rapid diagnosis method based on edge computing, specifically, Figure 1 As shown, the battery rapid diagnosis method based on edge computing in this embodiment includes the following steps:

[0033] Step S100, collecting battery information data from a battery management system at a preset frequency;

[0034] Step S200, convolving the battery information data using a first-order differential edge operator to obtain a convolution result;

[0035] Step S300, filtering out convolution results that are not within the threshold range based on the convolution result and the threshold range;

[0036] Step S400 : determining whether there is an abnormality in the battery information data based on the convolution result that is not within the threshold range and the battery information data.

[0037] The following describes in detail steps S100 to S400 of the edge computing-based battery rapid diagnosis method of this embodiment.

[0038] Step S100: collecting battery information data from a battery management system at a preset frequency.

[0039] In this embodiment, the battery information data includes but is not limited to battery cell voltages and battery charge and discharge currents. In this embodiment, the method further includes: storing the battery cell voltages in a two-dimensional array according to the acquisition order, and storing the battery charge and discharge currents in a one-dimensional array according to the acquisition order.

[0040] Specifically, in this embodiment, key battery information data, namely, battery cell voltages and current charge and discharge currents, is acquired from the BMS (Battery Management System) at a predetermined frequency. During the acquisition process, the battery cell voltages are stored in a two-dimensional array A[N][M], in the order in which they were acquired. The charge and discharge currents are stored in a one-dimensional array B[N], in the order in which they were acquired. Once the set matrix array is full, the next stage of data analysis is performed. For example, when N meets the set maximum value, the array is considered full, and the next stage of data analysis can be performed.

[0041] Step S200: Convolve the battery information data using a first-order differential edge operator to obtain a convolution result.

[0042] In this embodiment, the battery information data is convolved using a first-order differential edge operator to obtain a convolution result. One implementation method is to use a first-order differential edge operator to perform horizontal convolution and vertical convolution on the two-dimensional array respectively to obtain convolution results: horizontal gradient value and vertical gradient value.

[0043] In this embodiment, before convolving the battery information data using a first-order differential edge operator, the method further includes:

[0044] 1) Select a 3X3 first-order differential edge operator as the initial first-order differential edge operator.

[0045] The process of constructing a standard 3*3 first-order differential edge operator of horizontal Gx and vertical Gy is as follows:

[0046]

[0047]

[0048] 2) Expanding the initial first-order differential edge operator to obtain the first-order differential edge operator required for convolution.

[0049] 3) The initial first-order differential edge operator is expanded horizontally to an MX3-order matrix and vertically to an NX3-order matrix; wherein M is the total number of battery packs connected in series, and N is the total number of times the battery information data is collected.

[0050] Considering that the number of battery cells N and the number of acquisition times M that make up the data matrix are generated according to actual needs, the generated standard 3*3 operator is expanded horizontally to M*3 order and vertically to N*3 order to construct an operator with a larger window and increase the convolution rate.

[0051] For example, Gx expands to: Gy expands to:

[0052] It can be seen that the edge computing-based battery rapid diagnosis method of this embodiment utilizes the efficiency and directionality of the edge detection algorithm, combined with the characteristics of the matrix array composed of the original battery data, to construct suitable operators and extension forms, and quickly locate the voltage mutation at a certain moment in massive data.

[0053] In this embodiment, when the first-order differential edge operator is used to perform horizontal convolution and vertical convolution on the two-dimensional array, the first-order differential edge operator is moved with a step size of 1.

[0054] Specifically, this embodiment uses a first-order differential edge operator to perform horizontal and vertical convolution on the voltage data matrix to quickly identify outliers in a single cell at a specific moment. To ensure high throughput, the original 3x3 first-order differential edge operator is expanded horizontally to an MX3-order matrix and vertically to an NX3-order matrix. The operator is moved with a step size of 1 to ensure that gradient values ​​between adjacent battery cells and gradient values ​​between adjacent time points within a single cell are filtered out.

[0055] In this embodiment, the generated first-order differential edge operator is used to perform horizontal and vertical convolution with the original voltage data, and the horizontal and vertical voltage differential approximations can be obtained respectively. Figure 2 For example, the raw voltage data arrays from a six-cell battery pack, obtained after three acquisitions, are then convolved horizontally and vertically for the first, second, and third cells. The horizontal convolution values ​​are 3325+3327-3328*2=-1, 3326+3327-3327*2=-1, and 3320+3329-3327*2=-5, respectively. The vertical convolution values ​​are 3325+3320-3326*2=-7, 3327+3327-3327*2=0, and 3328+3329-3327*2=3. The operator is then shifted with a step size of 1 to ensure that all gradient values ​​between adjacent battery cells and the gradient values ​​between adjacent time points within a single cell are filtered out. After convolution of all data is completed, the horizontal and vertical gradient values ​​are compared with the set threshold range. If they are outside the normal range, further positioning is performed.

[0056] Step S300: Filter out convolution results that are not within the threshold range based on the convolution results and the threshold range.

[0057] That is, in this embodiment, the original voltage data is convolved to obtain the horizontal gradient value and the vertical gradient value, which are then compared with the set normal threshold range respectively. If they are not within the normal threshold range, further positioning judgment is performed.

[0058] Step S400 : determining whether there is an abnormality in the battery information data based on the convolution result that is not within the threshold range and the battery information data.

[0059] Specifically, in this embodiment, determining whether there is an abnormality in the battery information data based on the convolution result that is not within the threshold range and the battery information data includes:

[0060] 1) If the lateral gradient value is not within the lateral gradient threshold range, obtain the battery pack single cell voltage data corresponding to the lateral gradient value in the two-dimensional matrix, calculate the voltage difference value between two adjacent battery cells based on the battery pack single cell voltage data, and determine whether there is an inter-group battery consistency abnormality based on the size of the pressure difference value and the pressure difference threshold.

[0061] If the lateral gradient value is abnormal, it indicates that there is a large voltage difference between the individual cells in the collected data. The corresponding convolution data portion of the original matrix is ​​extracted, and the voltage difference value dv between two adjacent battery cells is calculated column by column. Then, the dv value is determined to be normal.

[0062] Specifically, if an abnormal transverse gradient value is detected, the corresponding data portion of the original matrix is ​​extracted, and the voltage difference dv between two adjacent battery cells is calculated column by column. The dv value is then determined to be normal. If dv is not within the normal threshold range and the number of occurrences is greater than 1, it can be determined that the consistency of the batteries in the current cell number i is abnormal.

[0063] 2) If the longitudinal gradient value is not within the longitudinal gradient threshold range, obtain the battery cell voltage data corresponding to the longitudinal gradient value in the two-dimensional matrix, calculate the rate of change of the voltage difference between adjacent time points with respect to time based on the battery cell voltage data, and determine whether there is a battery cell voltage abnormality based on the magnitude of the rate of change and the rate of change threshold.

[0064] If the longitudinal gradient value is abnormal, it means that a certain battery cell has undergone significant changes during multiple acquisitions. The corresponding convolution data portion of the original matrix is ​​extracted and the rate of change of the voltage difference at adjacent time points j-1, j, and j+1, i.e., dv / dt, is calculated row by row. The dv / dt value is then determined to be normal.

[0065] Specifically, if the longitudinal gradient value is abnormal, the corresponding data portion of the original matrix is ​​extracted and the rate of change of the voltage difference between adjacent time points j-1, j, and j+1, i.e., dv / dt, is calculated row by row. The dv / dt value is then determined to be normal. If the dv / dt value is greater than the normal threshold, it is determined that the current cell voltage is abnormal.

[0066] In addition, in this embodiment, determining whether there is an abnormality in the battery information data based on the convolution result that is not within the threshold range and the battery information data further includes:

[0067] If the longitudinal gradient value is not within the longitudinal gradient threshold range, the battery cell voltage data corresponding to the longitudinal gradient value in the two-dimensional matrix is ​​obtained. The rate of change of the voltage difference over time at adjacent time points is calculated based on the battery cell voltage data. Simultaneously, the battery charge and discharge current data in the one-dimensional matrix is ​​obtained. The presence of a battery cell voltage anomaly is determined based on whether the rate of change of the voltage difference over time at adjacent time points is positively correlated with the battery charge and discharge current. In other words, if the dv / dt ratio does not show a positive correlation with the current charge and discharge current, it can be determined that the current cell voltage is abnormal.

[0068] In order to enable those skilled in the art to further understand the principle of the battery rapid diagnosis method based on edge computing of this embodiment, the following is combined with Figure 3 The battery rapid diagnosis method based on edge computing of this embodiment is described in detail.

[0069] like Figure 3 As shown, in the S0 stage, key battery information data, namely the battery pack single cell voltage and current charge and discharge current, is obtained from the BMS (battery management system) at a certain acquisition frequency Tc.

[0070] In the S1 stage, the battery voltage data collected N times are organized into a two-dimensional matrix array A[N][M], and the current data are organized into a one-dimensional matrix array B[N].

[0071] In the S2 stage, the 3X3 first-order differential edge operator is horizontally expanded to MX3, such as Figure 2 As shown; vertical expansion is NX3, as shown Figure 3 shown.

[0072] In the S3 stage, the extended first-order differential edge operator generated in the S2 stage performs horizontal and vertical convolution on the original voltage data with a step size of 1, where the convolution value of each horizontal step is recorded as Gx[i], and the convolution value of each vertical step is recorded as Gy[j].

[0073] The S4 stage judges the convolution results of the S3 stage and filters out Gx[i] and Gy[j] that are greater than the normal threshold.

[0074] In stages S5 and S6, further analysis is performed on the abnormal Gx[i] values ​​identified in stage S4. First, the data block corresponding to Gx[i] in the original battery data matrix A[N][M] is located. The voltage differences dv1 and dv2 for two adjacent battery cells i-1, i, and i+1 are calculated column by column. If the voltage differences dv1 and dv2 are outside the normal threshold range and occur more than once, it is determined that the inter-cell consistency of the current cell i is abnormal.

[0075] In stages S7 and S8, further analysis is performed on the abnormal Gy[j] values ​​identified in stage S4. First, the data block corresponding to Gy[i] in the original battery data matrix A[N][M] is located. The rate of change of the voltage difference over time at adjacent time points j-1, j, and j+1 is calculated row by row, i.e., dv / dt, where dt = 2Tc and dv = A[i][j+1] - A[i][j-1]. Combined with the current matrix data B[N], if dv / dt is greater than the normal threshold or does not show a positive correlation with the current charge and discharge current, the current cell voltage is considered abnormal.

[0076] Therefore, the edge computing-based rapid battery diagnosis method of this embodiment can achieve relatively accurate and rapid battery data processing on edge devices, leveraging the advantages of edge computing, namely, rapid analysis of a certain amount of data in a short period of time, achieving predictive accuracy through weighted cumulative calculation results, and facilitating scalability. In terms of algorithm implementation, considering the ease of parallel computing, such as using FPGAs or GPUs to achieve parallel computing acceleration, the advantages of edge computing devices are leveraged, namely, rapid analysis of a certain amount of data in a short period of time, and accurate analysis through weighted cumulative calculation results, achieving effective analytical calculations.

[0077] The calculation process ends in stage S9, and a snapshot of the abnormal voltage cells, the corresponding raw data values, and the current data is saved. Therefore, the edge computing-based rapid battery diagnosis method of this embodiment can implement a data snapshot function, enabling comparison with cloud-based analysis data, and improving the safety of the entire power station.

[0078] In addition, in this embodiment, usually after a charge and discharge process is completed, a weight is calculated based on the cumulative number of abnormal point records of each battery, and the calculation result is uploaded to the cloud or downloaded to the BMS.

[0079] Example 2

[0080] like Figure 4As shown, this embodiment provides an electronic device 101, comprising: a processor 1001 and a memory 1002; the memory 1002 is configured to store a computer program; and the processor 1001 is configured to execute the computer program stored in the memory 1002, so that the electronic device 101 performs the steps of the edge computing-based rapid battery diagnosis method in Example 1. Since the specific implementation process of the steps of the edge computing-based rapid battery diagnosis method has been described in detail in Example 1, it will not be repeated here.

[0081] like Figure 5 As shown, the electronic device 101 of this embodiment is composed of a computing module, a communication interface module, and a storage module. The electronic device 101 is connected to the BMS (battery management system), obtains the battery pack single cell voltage and current charge and discharge current from the BMS (battery management system), and uploads the diagnosis results to the cloud platform. Therefore, the electronic device 101 of this embodiment realizes the battery data collection, analysis, calculation and storage from the energy storage power station, and can upload it to the cloud platform or send it back to the BMS (battery management system). The architecture is as follows Figure 5 As shown. The computing module is the core of the hardware unit, which encapsulates a fast processing algorithm. The data calculation and processing core of the computing module preferably uses FPGA or GPU, and data interaction uses MCU. The algorithm implemented by the computing module in this embodiment can be easily processed into multiple matrix data blocks for parallel calculation, so as to achieve a higher data processing throughput. Therefore, the electronic device 101 of this embodiment implements a fast and effective algorithm for battery pack data analysis and detection, realizes real-time and fast battery data processing and analysis on the edge device, and is easy to port to FPGA and GPU to realize parallel calculation, thereby increasing data processing throughput, reducing the burden of cloud analysis, and improving the safety of the entire power station.

[0082] The processor 1001 is a central processing unit (CPU). The memory 1002 is connected to the processor 1001 via a system bus and communicates with each other. The memory 1002 is used to store computer programs, and the processor 1001 is used to run computer programs so that the processor 1001 executes the edge computing-based rapid battery diagnosis method. The memory 1002 may include random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage.

[0083] In addition, this embodiment further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by the processor 1001, the steps of the edge computing-based rapid battery diagnosis method described in Example 1 are implemented. Example 1 has already described the edge computing-based rapid battery diagnosis method in detail and will not be repeated here.

[0084] Those skilled in the art will appreciate that all or part of the steps in the above-described method embodiments can be implemented using hardware associated with a computer program. The aforementioned computer program can be stored in a computer-readable storage medium. When executed, the program performs the steps in the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0085] In summary, the present invention enables relatively accurate and rapid battery data processing on edge devices, enabling rapid diagnosis of battery performance and facilitating scalability. Therefore, the present invention effectively overcomes the shortcomings of existing technologies and possesses high industrial application value.

[0086] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the present invention. Anyone skilled in the art may 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 persons skilled in the art without departing from the spirit and technical principles disclosed herein are intended to be covered by the claims of the present invention.

Claims

1. A rapid battery diagnosis method based on edge computing, characterized by: The method comprises: Collect battery information data from the battery management system at a preset frequency; Convolving the battery information data using a first-order differential edge operator to obtain a convolution result; Filtering out convolution results that are not within the threshold range based on the convolution result and the threshold range; It is determined whether there is an abnormality in the battery information data based on the convolution result that is not within the threshold range and the battery information data.

2. The battery rapid diagnosis method based on edge computing according to claim 1 is characterized in that: The battery information data includes battery pack single cell voltage and battery charge and discharge current.

3. The battery rapid diagnosis method based on edge computing according to claim 2, characterized in that: Also includes: The battery pack single cell voltage is stored in a two-dimensional array according to the acquisition sequence, and the battery charge and discharge current is stored in a one-dimensional array according to the acquisition sequence.

4. The battery rapid diagnosis method based on edge computing according to claim 3 is characterized in that: One implementation of using a first-order differential edge operator to convolve the battery information data to obtain a convolution result is: A first-order differential edge operator is used to perform horizontal convolution and vertical convolution on the two-dimensional array respectively to obtain convolution results: a horizontal gradient value and a vertical gradient value.

5. The battery rapid diagnosis method based on edge computing according to claim 1 or 4, characterized in that: Before convolving the battery information data using a first-order differential edge operator, the method further includes: Select the 3X3 first-order differential edge operator as the initial first-order differential edge operator; The initial first-order differential edge operator is expanded to obtain the first-order differential edge operator required for convolution; the initial first-order differential edge operator is horizontally expanded to an MX3-order matrix and vertically expanded to an NX3-order matrix; wherein M is the total number of battery packs in series, and N is the total number of times battery information data is collected.

6. The battery rapid diagnosis method based on edge computing according to claim 4 is characterized in that: In the process of performing horizontal convolution and vertical convolution on the two-dimensional array respectively using the first-order differential edge operator, when moving the first-order differential edge operator, the movement is performed with a step size of 1.

7. The battery rapid diagnosis method based on edge computing according to claim 4 is characterized in that: The determining whether there is an abnormality in the battery information data based on the convolution result that is not within the threshold range and the battery information data includes: If the transverse gradient value is not within the transverse gradient threshold range, obtaining the battery pack single cell voltage data corresponding to the transverse gradient value in the two-dimensional array, calculating the voltage difference between two adjacent battery cells based on the battery pack single cell voltage data, and determining whether there is an inter-cell battery consistency abnormality based on the difference between the voltage difference value and the voltage difference threshold; If the longitudinal gradient value is not within the longitudinal gradient threshold range, obtain the battery cell voltage data corresponding to the longitudinal gradient value in the two-dimensional array, calculate the rate of change of the voltage difference between adjacent time points with respect to time based on the battery cell voltage data, and determine whether there is a battery cell voltage abnormality based on the size of the rate of change and the rate of change threshold.

8. The battery rapid diagnosis method based on edge computing according to claim 7, characterized in that: The determining whether there is an abnormality in the battery information data based on the convolution result that is not within the threshold range and the battery information data further includes: If the longitudinal gradient value is not within the longitudinal gradient threshold range, obtaining battery cell voltage data corresponding to the longitudinal gradient value in the two-dimensional array, calculating the rate of change of voltage difference at adjacent time points relative to time based on the battery cell voltage data, and simultaneously obtaining battery charge and discharge current data in the one-dimensional array; Whether there is a battery pack single cell voltage abnormality is determined based on whether the rate of change of the voltage difference at adjacent time points with respect to time is positively correlated with the battery charge and discharge current.

9. A storage medium storing program instructions, characterized in that: When the program instructions are executed, the steps of the battery rapid diagnosis method based on edge computing as described in any one of claims 1 to 8 are implemented.

10. An electronic device, characterized in that: It includes a memory for storing a computer program; and a processor for running the computer program to implement the steps of the battery rapid diagnosis method based on edge computing as described in any one of claims 1 to 8.

Images