Battery thermal runaway early warning method and system based on stress and temperature

Abstract

The invention discloses a battery thermal runaway early warning method and system based on stress and temperature, and the method comprises the steps: obtaining the temperature and stress data of each single battery in a battery system in real time, and building a time sequence with a set length; averaging and normalizing the temperature and stress data in the time sequence, and decoupling and classifying the time sequence characteristics of the temperature and stress data by using a DBSCAN (Density Based Spatial Clustering of Applications with Noise) algorithm to obtain a clustering result; performing clustering center calculation according to a clustering result, taking a clustering set with the most members as a reference object, and calculating Euclidean distances between the reference object and other clustering centers; and establishing corresponding time sequence data according to the calculated clustering center Euclidean distance, calculating a time sequence data change rate # imgabs0 # based on the clustering center Euclidean distance through a least square method, and judging whether thermal runaway early warning is needed or not according to the time sequence data change rate # imgabs1 #. According to the method, traditional real-time data noise and abnormal value interference are reduced, and early warning of the thermal runaway of the battery is timely and accurately carried out.

Description

Technical Field

[0001] The present invention belongs to the technical field of battery management, and in particular relates to a battery thermal runaway early warning method based on stress and temperature. Background Art

[0002] Lithium batteries have the advantages of high energy density and long cycle life, and are widely used in electric vehicles and energy storage. During the repeated charge and discharge process of lithium batteries, reversible lithium expansion and irreversible expansion caused by overcharge and overdischarge will occur, which will manifest as stress. It poses a threat to the health and safety of the battery, and troubles the large-scale market application of lithium-ion batteries. Therefore, the battery thermal runaway warning method based on stress and temperature is particularly important.

[0003] CN112485678A discloses a real-time status monitoring and thermal runaway alarm method for lithium batteries. By detecting the surface temperature data, deformation data, CO concentration data and VOC concentration data of the lithium battery when leakage occurs, the three typical stage characteristics of thermal runaway, namely over-temperature, over-pressure and gas leakage, are accurately reflected, and the real-time status of the lithium battery is monitored.

[0004] CN112067155A discloses a method for dynamic temperature monitoring of lithium batteries based on OFDR. The sensing optical fiber is arranged horizontally and vertically along the surface of the lithium battery to form a distributed sensing optical fiber network. The Rayleigh scattering signal in the optical fiber is collected by OFDR technology. After demodulation, the real-time temperature changes at various positions on the optical fiber are obtained, and a two-dimensional temperature distribution map of the lithium battery surface is formed to intuitively display the surface temperature conditions, thereby realizing real-time temperature monitoring.

[0005] In the above-mentioned prior art, for a battery system, how to use temperature and stress data to perform decoupling analysis in the time dimension to achieve early warning of battery thermal runaway has not been provided. Summary of the invention

[0006] The technical problem to be solved by the present invention is: based on the temperature, stress and other data collected by the battery management system, how to decouple and analyze the temperature and stress data in the time dimension to achieve data anomaly detection and early warning of battery thermal runaway.

[0007] In order to solve the above technical problems, the present invention provides a battery thermal runaway early warning method based on stress and temperature, comprising the following steps:

[0008] Step S1, acquiring the temperature and stress data of each single cell in the battery system in real time, and establishing a time series of a set length;

[0009] Step S2, performing mean processing and normalization processing on the temperature and stress data in the time series, and then using the DBSCAN algorithm to decouple and classify the time series characteristics of the temperature and stress data to obtain a clustering result;

[0010] Step S3, calculating the cluster center according to the clustering result in step S2, taking the cluster set with the most members as the reference object, and calculating the Euclidean distance between the reference object and other cluster centers;

[0011] Step S4: Establish the corresponding time series data according to the Euclidean distance of the cluster center calculated in step S3, and calculate the time series data change rate based on the Euclidean distance of the cluster center by the least squares method. , according to the rate of change of time series data Determine whether thermal runaway warning is needed. If the rate of change If it is greater than the threshold, a thermal runaway warning is issued.

[0012] In the aforementioned battery thermal runaway warning method based on stress and temperature, in step S2, the temperature and stress data in the time series are averaged, and the calculation formulas are shown in formulas (1) and (2), where is the length of the time series, Indicates the first The temperature value at the moment, represents the average temperature in this time series, Indicates the first The stress value at time, represents the average value of stress in the time series,

[0013] (1)

[0014] (2)

[0015] In the aforementioned battery thermal runaway warning method based on stress and temperature, in step S2, the calculation formulas for normalizing the data are shown in equations (3) and (4):

[0016] (3)

[0017] (4)

[0018] in, Indicates The average temperature of the battery in this time series, represents the maximum value among all battery temperature averages, Represents the minimum value among the average values ​​of all battery temperatures, Indicates The normalized average temperature of the battery; Indicates The average stress value of the battery in this time series, represents the maximum value among all battery stress averages, represents the minimum value among all battery stress averages, Indicates The normalized stress average value of the battery.

[0019] In the aforementioned stress and temperature-based battery thermal runaway early warning method, in step S2, the DBSCAN algorithm is used to perform cluster analysis on the battery, which includes the following steps:

[0020] 21) Set the scanning radius to eps;

[0021] 22) Select the mean value of the temperature time series and the mean of the stress time series As a clustering factor; calculate the Euclidean distance based on the characteristic data of each single battery, where the Battery and Battery saving European distance The calculation formula of is shown in formula (5). Represents the characteristic dimension, 0-1 corresponds to the normalized average temperature and stress;

[0022] d ij = ∑ k = 0 1 ( X i [ k ] − X j [ k ]) 2 (5);

[0023] X i [ k ] Indicates Dataset features, X i [ 0 ] Indicates The normalized average temperature of the battery, X i [ 1 ] Indicates The normalized stress average value of the battery, X j [ 0 ] Indicates The normalized average temperature of the battery, X j [ 1 ] Indicates The normalized stress average value of the battery;

[0024] 23) If the Euclidean distance is less than the scanning radius eps, the corresponding single cells are included in the same set;

[0025] 23) Determine whether the Euclidean distance between any member in the set and any member outside the set is less than the scanning radius eps. If so, include the corresponding member outside the set into the same set; the members are the normalized average stress and average temperature of the single cell;

[0026] 24) Repeat steps 21) to 23) until all batteries are sorted.

[0027] In the aforementioned battery thermal runaway warning method based on stress and temperature, in step S3, the cluster center calculation formula is as shown in formula (6), that is, the average value of all members in the set is taken for each dimensional feature,

[0028] (6)

[0029] in represents the cluster center of set A, , They are the temperature coordinate and pressure coordinate of the cluster center of set A respectively; {} represents the coordinate;

[0030] The calculation formula of the Euclidean distance between cluster centers is shown in formula (7):

[0031] d AB = ∑ k = 0 1 ( X A [ k ] − X B [ k ]) 2 (7)

[0032] in Represents the Euclidean distance between the cluster centers of set A and set B.

[0033] In the aforementioned battery thermal runaway warning method based on stress and temperature, in step S4, the Euclidean distance time series data between the cluster centers is shown in formula (8). For the cluster centers of set A and set B,

[0034] (8)

[0035] According to the Euclidean distance time series data of each cluster center, the time series data change rate is calculated by the least squares method , as shown in formula (9),

[0036] (9)

[0037] for The Euclidean distance between the cluster center of set A and the cluster center of set B at the moment, For the length of time.

[0038] In the aforementioned battery thermal runaway warning method based on stress and temperature, the threshold C is set according to the selected time interval unit.

[0039] A computer device / equipment / system comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0040] A computer-readable storage medium stores a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0041] A computer program product comprises a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0042] The beneficial effects achieved by the present invention are as follows: the method of the present invention obtains battery status characteristics by decoupling and analyzing temperature and stress data in the time dimension, thereby reducing traditional real-time data noise and outlier interference, and timely and accurately warns of battery thermal runaway. At the same time, the distance between cluster centers is calculated through DBSCAN cluster analysis, time series data of cluster center distances is established, and the least squares method is used to judge the rate of change, thereby reducing the false triggering of warnings in traditional threshold warning strategies and solving the problem that traditional methods require threshold adjustment in different scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 Flow chart of the DBSCAN clustering algorithm in Example 1 of the present invention;

[0044] Figure 2 It is a flow chart of the battery thermal runaway early warning method based on temperature and stress in Example 1 of the present invention. DETAILED DESCRIPTION

[0045] The method of the present invention is further described in detail below with reference to the accompanying drawings.

[0046] Example 1

[0047] This embodiment provides a battery thermal runaway early warning method based on stress and temperature, comprising the following steps:

[0048] Step S1, acquiring the temperature and stress data of each single cell in the battery system in real time, and establishing a time series of a set length;

[0049] Step S2, performing mean processing and normalization processing on the temperature and stress data in the time series, and then using the DBSCAN algorithm to decouple and classify the time series characteristics of the temperature and stress data to obtain a clustering result;

[0050] In step S2, the temperature and stress data in the time series are processed by averaging. The calculation formulas are shown in equations (1) and (2), where is the length of the time series, Indicates the first The temperature value at the moment, represents the average temperature in this time series, Indicates the first The stress value at time, represents the average value of stress in the time series,

[0051] (1)

[0052] (2)

[0053] In step S2, the calculation formula for normalizing the data is shown in formula (3) and formula (4):

[0054] (3)

[0055] (4)

[0056] in, Indicates The average temperature of the battery in this time series, represents the maximum value among all battery temperature averages, Represents the minimum value among the average values ​​of all battery temperatures, Indicates The normalized average temperature of the battery; Indicates The average stress value of the battery in this time series, represents the maximum value among all battery stress averages, represents the minimum value among all battery stress averages, Indicates The normalized stress average value of the battery.

[0057] In step S2, as shown in FIG1 , the process of clustering analysis of batteries using the DBSCAN algorithm includes the following steps:

[0058] 21) Set the scanning radius to eps, set it to 0.5, and the minimum number of included points to minPts, which can be 1;

[0059] 22) Calculate the Euclidean distance based on the characteristic data of each single battery, where Battery and Battery saving European distance The calculation formula of is shown in formula (5). Represents the characteristic dimension, 0-1 corresponds to the normalized average temperature and stress;

[0060] d ij = ∑ k = 0 1 ( X i [ k ] − X j [ k ]) 2 (5);

[0061] X i [ k ] Indicates Dataset features, X i [ 0 ] Indicates The normalized average temperature of the battery, X i [ 1 ] Indicates The normalized stress average value of the battery, X j [ 0 ] Indicates The normalized average temperature of the battery, X j [ 1 ] Indicates The normalized stress average value of the battery.

[0062] 23) If the Euclidean distance is less than the scanning radius eps, the corresponding single cells are included in the same set;

[0063] 23) Determine whether the Euclidean distance between any member in the set and any member outside the set is less than the scanning radius eps. If so, include the corresponding member outside the set into the same set; the members are the normalized average stress and average temperature of the single cell;

[0064] 24) Continue until all batteries have been sorted.

[0065] In step S2, the DBSCAN algorithm is used to perform cluster analysis on the battery temperature and stress data, and the mean of the temperature time series is selected. and the mean of the stress time series As clustering factors, the clustering results are shown in Table 3. The parameters extracted in this embodiment are shown in Table 1.

[0066] Table 1 Mean values ​​of battery timing series before normalization

[0067]

[0068] The parameters after normalization are shown in Table 2.

[0069] Table 2 Normalized battery time series mean

[0070] The battery data clustering results are shown in Table 3.

[0071] Table 3 Battery data clustering results

[0072]

[0073] Step S3, calculating the cluster center according to the clustering result in step S2, taking the cluster set with the most members as the reference object, and calculating the Euclidean distance between the reference object and other cluster centers.

[0074] The cluster center calculation formula is shown in formula (6), that is, the feature of each dimension takes the average value of all members in the set, where Represents the cluster center of set A, as shown in Table 4.

[0075] (6)

[0076] , They are the temperature coordinate and pressure coordinate of the cluster center of set A respectively; {} represents the coordinate;

[0077] Table 4 Cluster center coordinates

[0078]

[0079] The calculation formula of the Euclidean distance between cluster centers is shown in formula (7), where The Euclidean distance between the cluster centers of set A and set B is shown in Table 5. Since cluster A has the most members, cluster A is used as a reference to calculate the Euclidean distances between other clusters.

[0080] d AB = ∑ k = 0 1 ( X A [ k ] − X B [ k ]) 2 (7)

[0081] Table 5 Distance between cluster centers

[0082]

[0083] Step S4: Establish the corresponding time series data according to the Euclidean distance of the cluster center calculated in step S3, and calculate the time series data change rate based on the Euclidean distance of the cluster center by the least squares method. , according to the rate of change of time series data Determine whether thermal runaway warning is needed. If the rate of change If it is greater than the threshold, a thermal runaway warning is issued.

[0084] In step S4, the Euclidean distance time series data between cluster centers is shown in formula (8), For example, there are

[0085] (8)

[0086] Table 6 Time series data of distance between cluster centers

[0087] According to the Euclidean distance time series data of each cluster center, the time series data change rate is calculated by the least squares method , as shown in formula (9),

[0088] (9)

[0089] for The Euclidean distance between the cluster center of set A and the cluster center of set B at the moment, The actual value of the threshold C is related to the selected time interval unit. In this embodiment, the time interval is in seconds, and the threshold C is 0.01. Take 5, =0.01, =0.001, ≥C, so it is necessary to issue a fault warning for the batteries in set B.

[0090] Example 2

[0091] A computer device / equipment / system comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0092] Example 3

[0093] A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that the computer program / instruction implements the steps of the above method when executed by a processor.

[0094] A computer program product comprises a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0095] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0096] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0097] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0098] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A battery thermal runaway early warning method based on stress and temperature, characterized in that: The following steps are involved: Step S1, acquiring the temperature and stress data of each single cell in the battery system in real time, and establishing a time series of a set length; Step S2, performing mean processing and normalization processing on the temperature and stress data in the time series, and then using the DBSCAN algorithm to decouple and classify the time series characteristics of the temperature and stress data to obtain a clustering result; Step S3, calculating the cluster center according to the clustering result in step S2, taking the cluster set with the most members as the reference object, and calculating the Euclidean distance between the reference object and other cluster centers; Step S4: Establish the corresponding time series data according to the Euclidean distance of the cluster center calculated in step S3, and calculate the time series data change rate based on the Euclidean distance of the cluster center by the least squares method. , according to the rate of change of time series data Determine whether thermal runaway warning is needed. If the rate of change If it is greater than the threshold, a thermal runaway warning is issued.

2. The battery thermal runaway early warning method based on stress and temperature according to claim 1, characterized in that: In step S2, the temperature and stress data in the time series are processed by averaging. The calculation formulas are shown in equations (1) and (2): (1) (2) in is the length of the time series, Indicates the first The temperature value at the moment, represents the average temperature in this time series, Indicates the first The stress value at time, Represents the average value of stress in the time series.

3. A battery thermal runaway early warning method based on stress and temperature according to claim 2, characterized in that: In step S2, the calculation formula for normalizing the data is shown in formula (3) and formula (4): (3) (4) in, Indicates The average temperature of the battery in this time series, represents the maximum value among all battery temperature averages, Represents the minimum value among the average values ​​of all battery temperatures, Indicates The normalized average temperature of the battery; Indicates The average stress value of the battery in this time series, represents the maximum value among all battery stress averages, represents the minimum value among all battery stress averages, Indicates The normalized stress average value of the battery.

4. The method for early warning of battery thermal runaway based on stress and temperature according to claim 3, characterized in that: In step S2, the process of clustering the battery using the DBSCAN algorithm includes the following steps: 21) Set the scanning radius to eps; 22) Select the mean value of the temperature time series and the mean of the stress time series As a clustering factor; calculate the Euclidean distance based on the characteristic data of each single battery, where the Battery and Battery saving European distance The calculation formula of is shown in formula (5). Represents the characteristic dimension, 0-1 corresponds to the normalized average temperature and stress; (5)； Indicates Dataset features, Indicates The normalized average temperature of the battery, Indicates The normalized stress average value of the battery, Indicates The normalized average temperature of the battery, Indicates The normalized stress average value of the battery; 23) If the Euclidean distance is less than the scanning radius eps, the corresponding single cells are included in the same set; 23) Determine whether the Euclidean distance between any member in the set and any member outside the set is less than the scanning radius eps. If so, include the corresponding member outside the set into the same set; the members are the normalized average stress and average temperature of the single cell; 24) Repeat steps 21) to 23) until all batteries are sorted.

5. The method for early warning of battery thermal runaway based on stress and temperature according to claim 4, characterized in that: In step S3, the cluster center calculation formula is shown in formula (6), that is, the average value of all members in the set is taken for each dimensional feature. (6) in represents the cluster center of set A, , They are the temperature coordinate and pressure coordinate of the cluster center of set A respectively; {} represents the coordinate; The calculation formula of the Euclidean distance between cluster centers is shown in formula (7): (7) in Represents the Euclidean distance between the cluster centers of set A and set B.

6. The battery thermal runaway early warning method based on stress and temperature according to claim 1, characterized in that: In step S4, the Euclidean distance time series data between cluster centers is shown in formula (8). For cluster centers of set A and set B, we have (8)； According to the Euclidean distance time series data of each cluster center, the time series data change rate is calculated by the least squares method , as shown in formula (9), （9） for The Euclidean distance between the cluster center of set A and the cluster center of set B at the moment, For the length of time.

7. The method for early warning of battery thermal runaway based on stress and temperature according to claim 6, characterized in that: The threshold C is set according to the selected time interval unit.

8. A computer device / equipment / system comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instruction is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images