Battery system real-time safety early warning method, system and device and storage medium

Abstract

The invention belongs to battery safety management, and particularly relates to a battery system real-time safety early warning method, system and device and a storage medium, and the method comprises the steps: obtaining the battery characteristics of each lithium battery in a lithium battery pack; based on the battery characteristics, the time duration, the battery number and the window length, obtaining a distribution matrix; obtaining a probability distribution matrix based on the distribution matrix and probability distribution; obtaining an order reduction matrix based on the probability distribution matrix and a preset range; based on the reduced-order matrix and a preset formula, obtaining characteristic values of the lithium battery pack at all moments; and screening abnormal lithium batteries based on the characteristic values and a preset algorithm. The method has the effect of reducing the calculation amount in a mode of reducing the order of the matrix formed by the collected battery characteristics.

Description

Technical Field

[0001] The present invention relates to battery safety management, and in particular to a real-time safety early warning method, system, device and storage medium for a battery system. Background Art

[0002] Lithium-ion (Li-ion) batteries, as one of the primary energy storage devices in electric vehicles and power plants, play a key role in transportation electrification and renewable energy systems. However, the performance of lithium-ion batteries gradually changes during use. If these performance differences within a battery pack are not detected in a timely manner, they can significantly impact the safety of the entire battery pack.

[0003] In the relevant technologies, there are three main types of research on lithium battery safety, namely, based on electrochemical mechanism, based on equivalent model and inconsistency algorithm. The method based on electrochemical mechanism can provide a better description of the internal parameters of the battery. The method based on equivalent model uses a first-order or second-order model to model the circuit and then calculates its internal resistance, but it requires a static environment that meets certain SOC conditions. For actual use, there is less data that meets the conditions, and real-time calculation cannot be achieved. Research on battery consistency mainly judges the inconsistency of battery voltage and temperature.

[0004] For the above-mentioned related technologies, whether it is electrochemical mechanism, equivalent model or inconsistency algorithm, when faced with too many lithium batteries, a large amount of calculation is required to analyze the safety of each lithium battery, which wastes a lot of time. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a real-time safety warning method, system, device and storage medium for a battery system, which can reduce the amount of calculation when analyzing the safety of lithium batteries and improve the calculation speed.

[0006] A real-time safety early warning method for a battery system, comprising:

[0007] Obtain battery characteristics of each lithium battery in the lithium battery pack;

[0008] Obtaining a distribution matrix based on the battery characteristics, the time length, the number of batteries, and the window length;

[0009] Based on the distribution matrix and the probability distribution, a probability distribution matrix is ​​obtained;

[0010] Based on the probability distribution matrix and a preset range, a reduced-order matrix is ​​obtained;

[0011] Based on the reduced-order matrix and a preset formula, obtaining the eigenvalues ​​of the lithium battery pack at all times;

[0012] Based on the characteristic values ​​and the preset algorithm, abnormal lithium batteries are screened.

[0013] Optionally, obtaining a distribution matrix based on the battery characteristics, time length, number of batteries, and window length includes:

[0014] Obtaining a characteristic matrix of a lithium battery pack based on the battery characteristics, the number of batteries, and the time length;

[0015] Based on the feature matrix and the window length, the distribution matrix at a certain moment is obtained.

[0016] Optionally, obtaining a probability distribution matrix based on the distribution matrix and the probability distribution includes:

[0017] Normalizing the distribution matrix to obtain a normalized matrix;

[0018] Based on the probability distribution, a distribution range is obtained;

[0019] The probability distribution matrix is ​​obtained based on the normalized matrix and the distribution range.

[0020] Optionally, obtaining a reduced-order matrix based on the probability distribution matrix and a preset range includes:

[0021] Determining whether the probability elements in the probability distribution matrix are within a preset range;

[0022] If the probability element is within a preset range, the probability element within the preset range is eliminated to obtain the reduced-order matrix.

[0023] Optionally, based on the reduced-order matrix and a preset formula, obtaining the eigenvalues ​​of the lithium battery pack at all times includes:

[0024] Substituting each element in the reduced-order matrix into the preset formula to obtain an eigenvalue matrix;

[0025] Based on the eigenvalue matrix and the sliding window, obtaining the eigenvalues ​​of the lithium battery pack at all times;

[0026] The preset formula is:

[0027] Where X represents the set of variables xi, n is the number of variables, p(x i ) is the probability of xi appearing, and H(X) is the entropy of set X.

[0028] Optionally, obtaining the battery characteristics of each lithium battery in the lithium battery pack includes:

[0029] Get the collection period;

[0030] Based on the acquisition cycle, obtaining the initial data of the lithium battery;

[0031] The initial data is preprocessed to obtain the battery characteristics of each lithium battery in the lithium battery pack.

[0032] Optionally, the preset algorithm includes an extreme value algorithm or a DBSCAN unsupervised clustering algorithm.

[0033] A real-time safety warning system for a battery system, comprising:

[0034] An acquisition module, used to obtain battery characteristics of each lithium battery in the lithium battery pack;

[0035] A first processing module is configured to obtain a distribution matrix based on the battery characteristics, the time length, the number of batteries, and the window length;

[0036] A second processing module is configured to obtain a probability distribution matrix based on the distribution matrix and the probability distribution;

[0037] An order reduction module, configured to obtain a reduced-order matrix based on the probability distribution matrix and a preset range;

[0038] A calculation module, configured to obtain eigenvalues ​​of the lithium battery pack at all times based on the reduced-order matrix and a preset formula;

[0039] The screening module is used to screen abnormal lithium batteries based on the characteristic values ​​and preset algorithms.

[0040] A terminal device includes a memory and a processor. The memory stores a computer program that can be run on the processor. When the processor loads and executes the computer program, a real-time safety early warning method for a battery system is adopted.

[0041] A computer-readable storage medium stores a computer program. When the computer program is loaded and executed by a processor, a real-time safety early warning method for a battery system is adopted.

[0042] The present invention has the beneficial effect of dividing the collected battery characteristics at each moment into several distribution matrices according to a set window length, then filtering the data in the distribution matrix using a set probability distribution to obtain a probability distribution matrix, then removing data within a preset range from the probability distribution matrix to obtain a reduced-order matrix, and then calculating the Shannon entropy of each reduced-order matrix to obtain the eigenvalues ​​of the lithium battery, and screening out abnormal lithium batteries based on the eigenvalues. By processing the probability distribution matrix to achieve the purpose of order reduction, the amount of calculation can be reduced and the calculation speed can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A flowchart of a real-time safety early warning method for a battery system according to the present invention is shown;

[0044] Figure 2 A graph of initial data collected for characterization of the battery of the present invention;

[0045] Figure 3 This is a schematic diagram of the improvement of the initial data processing according to the present invention;

[0046] Figure 4 A schematic diagram of the calculation process of the present invention for processing initial data;

[0047] Figure 5 This is a schematic diagram of a smooth curve after the initial data is processed by the present invention. DETAILED DESCRIPTION

[0048] A real-time safety warning method for a battery system, such as Figure 1 As shown, including:

[0049] S100: Obtain battery characteristics of each lithium battery in a lithium battery pack.

[0050] Specifically, a lithium battery pack includes several lithium batteries, and battery characteristics refer to data such as the voltage or temperature of a single lithium battery at a certain moment.

[0051] S110 , obtaining a distribution matrix based on battery characteristics, time length, battery quantity, and window length.

[0052] Specifically, when collecting data from a lithium battery pack, data is collected on a timeline. The battery count refers to the number of individual lithium batteries in a lithium battery pack, the time length is the total time for data collection, and the window length is the amount of data recorded at one time. Each element in the distribution matrix represents the voltage or temperature data of a lithium battery at a specific moment in time. The number of records depends on the window length.

[0053] S120: Obtain a probability distribution matrix based on the distribution matrix and the probability distribution.

[0054] Specifically, the probability distribution is a matrix obtained by applying a 3sigma distribution to the data in the distribution matrix. Typically, the data in the distribution matrix is ​​normalized and then subjected to a 3sigma distribution to obtain the probability distribution matrix.

[0055] S130: Obtain a reduced-order matrix based on the probability distribution matrix and a preset range.

[0056] Specifically, the probability distribution matrix shows the consistency of lithium-ion batteries. If the lithium-ion batteries are highly consistent, they will all fall within a preset range, which is ±2σ. The row of lithium-ion batteries with good consistency is removed from the probability distribution matrix to achieve rank reduction, reducing the computational effort and obtaining a reduced-order matrix.

[0057] S140 : Obtaining eigenvalues ​​of the lithium battery pack at all times based on the reduced-order matrix and a preset formula.

[0058] Specifically, the preset formula is the Shannon entropy calculation formula. Each element in the reduced-order matrix is ​​substituted into the Shannon entropy calculation formula to obtain the eigenvalues ​​of each lithium battery in the lithium battery pack at all times.

[0059] S150: Screen abnormal lithium batteries based on characteristic values ​​and a preset algorithm.

[0060] Specifically, the eigenvalues ​​are filtered through the extreme value algorithm and the DBSCAN unsupervised clustering algorithm to identify abnormal lithium batteries. The data of abnormal charging cells during the period is counted, and the top n abnormality values ​​are obtained. These are identified as outliers and used as the output of the safety warning result.

[0061] In one implementation of this embodiment, step S110 obtains a distribution matrix based on battery characteristics, time length, battery quantity, and window length, including:

[0062] S200: Obtain a characteristic matrix of the lithium battery pack based on battery characteristics, battery quantity, and time length.

[0063] Specifically, the feature matrix refers to the feature matrix composed of the battery features of all lithium batteries collected within a certain period of time. The feature matrix is ​​as follows:

[0064] where x ij is the voltage, temperature and other data of the jth cell at time i. The number of cells is m, and the total recording time is n, which is the time length.

[0065] S210: Based on the feature matrix and the window length, obtain a distribution matrix at a certain moment.

[0066] Specifically, because the feature matrix contains too much data and cannot be processed all at once, we select data from the feature matrix based on the window length to obtain the distribution matrix at a certain moment. For example, if the window length is 6*6, we select data from the matrix X in 6*6 blocks to obtain several distribution matrices.

[0067] The distribution matrix is ​​as follows:

[0068]

[0069] In one implementation of this embodiment, step S120, based on the distribution matrix and the probability distribution, obtains the probability distribution matrix including:

[0070] S300 , normalizing the distribution matrix to obtain a normalized matrix.

[0071] Specifically, normalizing the distribution matrix can limit the values ​​in the distribution matrix to a certain range. The normalization process is as follows:

[0072]

[0073] Among them, X is the original data, is the mean and VAR(X) is the standard deviation.

[0074] S310: Obtain a distribution range based on the probability distribution.

[0075] S320: Obtain a probability distribution matrix based on the normalized matrix and the distribution range.

[0076] Specifically, the 3sigma distribution has a range of (-∞, -3σ), (-3σ, -2σ), (-2σ, 2σ), (2σ, 3σ), and (3σ, ∞), where σ is the standard deviation. Data within this range is retained, while outliers outside it are removed.

[0077] In one implementation of this embodiment, step S120, based on the probability distribution matrix and the preset range, obtains a reduced-order matrix including:

[0078] S400: Determine whether the probability elements in the probability distribution matrix are within a preset range.

[0079] S410: If the probability element is within a preset range, the probability element within the preset range is eliminated to obtain a reduced-order matrix.

[0080] Specifically, the preset range is ±2σ. In the battery system, most of the time, the consistency of lithium batteries is good, and the probability element will fall within the preset range. For example, assuming If the second row in is uniform, the following probability distribution matrix can be obtained:

[0081] Then, the second row with better consistency can be directly extracted. Removing the rows in the X matrix that fall within the ±2σ interval does not affect the calculation results, but can reduce the amount of calculation and increase the calculation speed. For vehicle data with a large number of lithium batteries, matrix reduction can quickly obtain results.

[0082] In one implementation of this embodiment, step S140 obtains the eigenvalues ​​of the lithium battery pack at all times based on the reduced-order matrix and a preset formula, including:

[0083] S500: Substitute each element in the reduced-order matrix into a preset formula to obtain an eigenvalue matrix.

[0084] S510 : Obtain the eigenvalues ​​of the lithium battery pack at all times based on the eigenvalue matrix and the sliding window.

[0085] The default formula is:

[0086] Where X represents the set of variables xi, n is the number of variables, p(x i ) is the probability of xi appearing, and H(X) is the entropy of set X.

[0087] Specifically, Substituting into the preset formula, we get the eigenvalue matrix:

[0088] Calculate the eigenvalue matrix of one sliding window at a time. Then combine all the sliding windows to get:

[0089] Each element in the matrix is ​​an eigenvalue of a lithium battery.

[0090] In one implementation of this embodiment, step S100, i.e., obtaining the battery characteristics of each lithium battery in the lithium battery pack, includes:

[0091] S600: Obtain a collection period.

[0092] S610: Obtain initial data of the lithium battery based on the acquisition cycle.

[0093] S620: Preprocess the initial data to obtain battery characteristics of each lithium battery in the lithium battery pack.

[0094] Specifically, the collected battery data is sent at a fixed period, and the time of the collection period can be set by yourself.

[0095] The initial data is the collected data. Due to the influence of timing, current and SOC, the initial data is step-shaped and has large fluctuations, which affects the results of feature calculation. Therefore, the initial data needs to be processed. Figure 2 The following are the initial data collected.

[0096] Since the initial data is sent periodically, discrete points are used instead of time for calculation, such as Figure 3 As shown, the initial data is divided into three segments: start and end segments, middle segment and peak segment, which are calculated as follows Figure 4 As shown. The initial data is smoothed by the improved LEAN algorithm. The preset algorithms include extreme value algorithm or DBSCAN unsupervised clustering algorithm. After the eigenvalues ​​are processed by extreme value algorithm or DBSCAN unsupervised clustering algorithm, the results are as follows Figure 5 As shown, abnormal lithium batteries are screened out, and the data of lithium batteries with abnormal charging during this period are counted to obtain the top n data of abnormal times, which are judged as abnormal values ​​and output as the result of this safety warning.

[0097] A real-time safety warning system for a battery system, characterized by comprising:

[0098] An acquisition module, used to obtain battery characteristics of each lithium battery in the lithium battery pack;

[0099] A first processing module is configured to obtain a distribution matrix based on battery characteristics, time length, number of batteries, and window length;

[0100] A second processing module is used to obtain a probability distribution matrix based on the distribution matrix and the probability distribution;

[0101] An order reduction module, used for obtaining a reduced order matrix based on the probability distribution matrix and a preset range;

[0102] A calculation module, used to obtain the eigenvalues ​​of the lithium battery pack at all times based on the reduced-order matrix and a preset formula;

[0103] The screening module is used to screen abnormal lithium batteries based on the characteristic values ​​and preset algorithms.

[0104] An embodiment of the present application also discloses a terminal device, including a memory and a processor. The memory stores a computer program that can be run on the processor. When the processor loads and executes the computer program, a real-time safety warning method for the battery system is adopted.

[0105] Among them, the terminal device can be a computer device such as a desktop computer, a laptop computer or a cloud server, and the terminal device includes but is not limited to a processor and a memory. For example, the terminal device can also include input and output devices, network access devices and buses, etc.

[0106] Among them, the processor can adopt a central processing unit (CPU). Of course, according to actual usage, other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. can also be adopted. The general-purpose processor can adopt a microprocessor or any conventional processor, etc., and this application does not impose any restrictions on this.

[0107] Among them, the memory can be an internal storage unit of the terminal device, such as the hard disk or memory of the terminal device, or it can be an external storage device of the terminal device, such as a plug-in hard disk, smart memory card (SMC), secure digital card (SD) or flash memory card (FC) equipped on the terminal device, etc., and the memory can also be a combination of the internal storage unit and the external storage device of the terminal device. The memory is used to store computer programs and other programs and data required by the terminal device. The memory can also be used to temporarily store data that has been output or is to be output. This application does not impose any restrictions on this.

[0108] Among them, through this terminal device, a real-time safety warning method for a battery system in the above embodiment is stored in the memory of the terminal device, and is loaded and executed on the processor of the terminal device for easy use.

[0109] An embodiment of the present application further discloses a computer-readable storage medium, and the computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, a real-time safety warning method for a battery system in the above embodiment is adopted.

[0110] Among them, the computer program can be stored in a computer-readable medium, the computer program includes computer program code, the computer program code can be in the form of source code, object code, executable file or certain middleware, etc. The computer-readable medium includes any entity or device that can carry computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that computer-readable medium includes but is not limited to the above-mentioned components.

[0111] Among them, through this computer-readable storage medium, a real-time safety warning method for a battery system in the above embodiment is stored in a computer-readable storage medium, and is loaded and executed on a processor to facilitate the storage and application of the above method.

[0112] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of protection of the present application is limited to these examples. In line with the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of different aspects of one or more embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.

[0113] The one or more embodiments of this application are intended to encompass all such substitutions, modifications, and variations that fall within the broad scope of this application. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of one or more embodiments of this application should be included in the scope of protection of this application.

Claims

1. A real-time safety early warning method for a battery system, characterized in that: include: Obtain battery characteristics of each lithium battery in the lithium battery pack; Obtaining a distribution matrix based on the battery characteristics, the time length, the number of batteries, and the window length; Based on the distribution matrix and the probability distribution, a probability distribution matrix is ​​obtained; Based on the probability distribution matrix and a preset range, a reduced-order matrix is ​​obtained; Based on the reduced-order matrix and a preset formula, obtaining the eigenvalues ​​of the lithium battery pack at all times; Based on the characteristic values ​​and the preset algorithm, abnormal lithium batteries are screened.

2. A battery system real-time safety early warning method according to claim 1, characterized in that: The obtaining of the distribution matrix based on the battery characteristics, time length, number of batteries, and window length includes: Obtaining a characteristic matrix of a lithium battery pack based on the battery characteristics, the number of batteries, and the time length; Based on the feature matrix and the window length, the distribution matrix at a certain moment is obtained.

3. A real-time safety early warning method for a battery system according to claim 1, characterized in that: The obtaining of a probability distribution matrix based on the distribution matrix and the probability distribution includes: Normalizing the distribution matrix to obtain a normalized matrix; Based on the probability distribution, a distribution range is obtained; The probability distribution matrix is ​​obtained based on the normalized matrix and the distribution range.

4. A battery system real-time safety early warning method according to claim 1, characterized in that: The obtaining of a reduced-order matrix based on the probability distribution matrix and a preset range includes: Determining whether the probability elements in the probability distribution matrix are within a preset range; If the probability element is within a preset range, the probability element within the preset range is eliminated to obtain the reduced-order matrix.

5. A battery system real-time safety early warning method according to claim 1, characterized in that: The method of obtaining the eigenvalues ​​of the lithium battery pack at all times based on the reduced-order matrix and the preset formula includes: Substituting each element in the reduced-order matrix into the preset formula to obtain an eigenvalue matrix; Based on the eigenvalue matrix and the sliding window, obtaining the eigenvalues ​​of the lithium battery pack at all times; The preset formula is: Where X represents the set of variables xi, n is the number of variables, p(x i ) is the probability of xi appearing, and H(X) is the entropy of set X.

6. A battery system real-time safety early warning method according to claim 1, characterized in that: The obtaining of the battery characteristics of each lithium battery in the lithium battery pack includes: Get the collection period; Based on the acquisition cycle, initial data of the lithium battery is obtained; The initial data is preprocessed to obtain the battery characteristics of each lithium battery in the lithium battery pack.

7. A real-time safety early warning method for a battery system according to claim 1, characterized in that: The preset algorithm includes an extreme value algorithm or a DBSCAN unsupervised clustering algorithm.

8. A real-time safety warning system for a battery system, characterized by: include: An acquisition module, used to obtain battery characteristics of each lithium battery in the lithium battery pack; A first processing module is configured to obtain a distribution matrix based on the battery characteristics, the time length, the number of batteries, and the window length; A second processing module is configured to obtain a probability distribution matrix based on the distribution matrix and the probability distribution; An order reduction module, configured to obtain a reduced-order matrix based on the probability distribution matrix and a preset range; A calculation module, configured to obtain eigenvalues ​​of the lithium battery pack at all times based on the reduced-order matrix and a preset formula; The screening module is used to screen abnormal lithium batteries based on the characteristic values ​​and preset algorithms.

9. A terminal device comprising a memory and a processor, characterized in that: The memory stores a computer program that can be run on the processor, and when the processor loads and executes the computer program, the method according to any one of claims 1 to 7 is adopted.

10. A computer-readable storage medium storing a computer program, wherein: When the computer program is loaded and executed by a processor, the method according to any one of claims 1 to 7 is adopted.

Images