Battery cell fault diagnosis method and apparatus

Abstract

The present invention relates to the technical field of batteries. Disclosed are a battery cell fault diagnosis method and apparatus. In the present invention, on the basis of charging voltages of battery cells, a sum of MN absolute values and a Euclidean distance of a cell to be tested are calculated; on the basis of charging voltage sampling moments, a slope of a sample entropy-scale factor curve of said cell is calculated; a first Hamming proximity between the sum of MN absolute values, the Euclidean distance and the slope of the sample entropy-scale factor curve of said cell and a normal cell cluster center is calculated; a second Hamming proximity between the sum of MN absolute values, the Euclidean distance and the slope of the sample entropy-scale factor curve of said cell and a faulty cell cluster center is calculated; and if the second Hamming proximity is greater than the first Hamming proximity, it is determined that said cell is faulty.

Description

Battery cell fault diagnosis method and device

[0001] Cross-reference to related applications

[0002] Embodiments of the present application are based on and claim priority from Chinese Patent Application No. 202410809826.6 filed on June 21, 2024, the entire contents of which are incorporated herein by reference. TECHNICAL FIELD

[0003] The present application relates to the technical field of batteries, in particular to a battery cell fault diagnosis method and device. BACKGROUND

[0004] A battery generally includes a plurality of battery cells. A fault in a battery cell can cause abnormal charging and discharging of the battery, and even cause the battery to catch fire and burn. In order to ensure the safety of the battery, it is necessary to diagnose the fault of the battery cell. SUMMARY

[0005] The present application provides a battery cell fault diagnosis method and device, which solves the technical problem of how to diagnose the fault of a battery cell.

[0006] In one aspect, the present application provides the following technical solutions:

[0007] A battery cell fault diagnosis method, comprising:

[0008] obtaining charging voltages of a plurality of battery cells collected at a plurality of sampling times;

[0009] calculating the sum of the absolute values of MN and the Euclidean distance of the battery cell to be tested according to a plurality of charging voltages;

[0010] calculating the sample entropy-scale factor curve slope of the battery cell to be tested according to a plurality of sampling times;

[0011] obtaining a preset normal battery cell clustering center and a fault battery cell clustering center;

[0012] calculating a first Hamming closeness degree according to the sum of the absolute values of MN, the Euclidean distance, the sample entropy-scale factor curve slope of the battery cell to be tested, and the normal battery cell clustering center;

[0013] calculating a second Hamming closeness degree according to the sum of the absolute values of MN, the Euclidean distance, the sample entropy-scale factor curve slope of the battery cell to be tested, and the fault battery cell clustering center;

[0014] If the second Hamming closeness degree is greater than the first Hamming closeness degree, it is determined that the battery cell to be tested is faulty.

[0015] In another aspect, the present application also provides the following technical solutions:

[0016] A battery cell fault diagnosis device, comprising:

[0017] An acquisition module is configured to acquire charging voltages of a plurality of battery cells collected at a plurality of sampling moments;

[0018] A calculation module is configured to calculate a sum of absolute values of MNs and an Euclidean distance of a to-be-tested battery cell according to a plurality of charging voltages;

[0019] A sample entropy-scale factor curve slope of the to-be-tested battery cell is calculated according to a plurality of sampling moments;

[0020] The acquisition module is further configured to acquire a preset normal battery cell clustering center and a fault battery cell clustering center;

[0021] The calculation module is further configured to calculate a first Hamming closeness according to the sum of absolute values of MNs, the Euclidean distance, the sample entropy-scale factor curve slope of the to-be-tested battery cell, and the normal battery cell clustering center;

[0022] A second Hamming closeness is calculated according to the sum of absolute values of MNs, the Euclidean distance, the sample entropy-scale factor curve slope of the to-be-tested battery cell, and the fault battery cell clustering center;

[0023] A judgment module is configured to judge that the to-be-tested battery cell is faulty if the second Hamming closeness is greater than the first Hamming closeness.

[0024] In another aspect, the present application further provides the following technical solutions:

[0025] A computer device comprises a memory, a processor, and a computer program stored in the memory, and the processor executes the computer program to implement the steps of any battery cell fault diagnosis method.

[0026] In another aspect, the present application further provides the following technical solutions:

[0027] A computer readable storage medium has a computer program stored thereon, and the computer program is executed by a processor to implement the steps of any battery cell fault diagnosis method.

[0028] The one or more technical solutions provided by the present application have at least the following technical effects or advantages:

[0029] The application calculates the sum of MN absolute values and the Euclidean distance of the battery cell to be tested according to the charging voltage of the battery cell, calculates the sample entropy-scale factor curve slope of the battery cell to be tested according to the sampling time of the charging voltage, calculates the first Hamming closeness of the sum of MN absolute values, the Euclidean distance and the sample entropy-scale factor curve slope of the battery cell to be tested to the normal cell clustering center, calculates the second Hamming closeness of the sum of MN absolute values, the Euclidean distance and the sample entropy-scale factor curve slope of the battery cell to be tested to the fault cell clustering center, and judges the battery cell to be tested as fault if the second Hamming closeness is greater than the first Hamming closeness, thereby comprehensively diagnosing the battery cell fault from three dimensions. BRIEF DESCRIPTION OF DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0031] Fig. 1 is a curve diagram of the change of the charging voltage of the battery cell with time in the embodiment of the present application;

[0032] Fig. 2 is a curve diagram of the change of the MN value of the battery cell with sampling time in the embodiment of the present application;

[0033] Fig. 3 is a schematic diagram of the sum of MN absolute values of the battery cell in the embodiment of the present application;

[0034] Fig. 4 is a sample entropy-scale factor curve diagram of the battery cell in the embodiment of the present application;

[0035] Fig. 5 is a schematic diagram of the sample entropy-scale factor curve slope in the embodiment of the present application;

[0036] Fig. 6 is a schematic diagram of the Euclidean distance of the battery cell in the embodiment of the present application;

[0037] Fig. 7 is a flow chart of the battery cell fault diagnosis method in the embodiment of the present application;

[0038] Fig. 8 is a schematic diagram of the battery cell fault diagnosis device in the embodiment of the present application. Embodiment of the present application

[0039] The embodiment of the present application provides a battery cell fault diagnosis method and device, thereby solving the technical problem of how to diagnose the battery cell fault.

[0040] In order to better understand the technical solutions of the present application, the technical solutions of the present application will be described in detail in combination with the drawings in the specification and specific embodiments.

[0041] The embodiment of the application obtains the curve of the charging voltage of the six battery cells changing with time through experiments, as shown in FIG. 1, and the curve at the bottom of FIG. 1 is the charging voltage of the 5# cell, which can be seen as the charging voltage of the 5# cell is low, which is the fault cell.

[0042] The MN value curve of the six cells changing with time is obtained through experiments, as shown in FIG. 2, and the MN value curve of the 5# fault cell is at the top of FIG. 2. The sum of the absolute values of MN of the six cells is calculated according to FIG. 2, as shown in FIG. 3. It can be seen from FIG. 3 that the sum of the absolute values of MN of the 5# fault cell is obviously greater than that of the other cells, indicating that the sum of the absolute values of MN can reflect the difference of the fault cell from the normal cell, that is, the sum of the absolute values of MN can reflect the cell fault.

[0043] The sample entropy-scale factor curve of the six cells is obtained through experiments, as shown in FIG. 4, and the sample entropy-scale factor curve of the 5# fault cell is at the bottom of FIG. 4. The slope of each sample entropy-scale factor curve is obtained according to FIG. 4, as shown in FIG. 5. It can be seen that the slope of the sample entropy-scale factor curve of the 5# fault cell is obviously lower than that of the other cells, indicating that the slope of the sample entropy-scale factor curve can reflect the cell fault.

[0044] The Euclidean distance of the six cells is obtained through experiments, as shown in FIG. 6. It can be seen that the Euclidean distance of the 5# fault cell is obviously greater than that of the other cells, indicating that the Euclidean distance of the cell charging voltage curve and the average voltage curve can also reflect the cell fault.

[0045] Based on the above theory, the embodiment of the application proposes a battery cell fault diagnosis method, as shown in FIG. 7, which comprises:

[0046] Step S1, obtaining the charging voltages of the battery cells collected at multiple sampling times;

[0047] Step S2, calculating the sum of the absolute values of MN and the Euclidean distance of the to-be-tested cell according to the multiple charging voltages;

[0048] Step S3, calculating the sample entropy-scale factor curve slope of the to-be-tested cell according to the multiple sampling times;

[0049] Step S4, obtaining the preset normal cell clustering center and fault cell clustering center;

[0050] Step S5, calculating the first Hamming closeness according to the sum of the absolute values of MN, the Euclidean distance, the sample entropy-scale factor curve slope of the to-be-tested cell, and the normal cell clustering center;

[0051] Step S6, calculating the second Hamming closeness according to the sum of the absolute values of MN, the Euclidean distance, the sample entropy-scale factor curve slope of the to-be-tested cell, and the fault cell clustering center;

[0052] Step S7, if the second Hamming closeness is greater than the first Hamming closeness, it is judged that the battery cell to be measured is faulty.

[0053] In step S1, it is assumed that the charging voltages of the plurality of battery cells are collected every 1s, and a total of 20s are collected, so that the plurality of sampling times are 1, 2, 3,..., 19, and 20.

[0054] In step S2, the sum of the absolute values of MN is calculated according to the plurality of charging voltages, including: determining the maximum value of the plurality of charging voltages collected at the sampling time; calculating the average value of the plurality of charging voltages collected at the sampling time; calculating the MN value corresponding to the sampling time according to the charging voltage of the battery cell to be measured collected at the sampling time, the maximum value and the average value; and calculating the sum of the absolute values of the plurality of MN values corresponding to the plurality of sampling times to obtain the sum of the absolute values of MN. Wherein, the MN value corresponding to the sampling time is calculated according to the charging voltage of the battery cell to be measured collected at the sampling time, the maximum value and the average value, including: MN= ; MN is the MN value corresponding to a certain sampling time, U is the charging voltage of the battery cell to be measured collected at the sampling time, is the average value of all charging voltages of the battery cells collected at the sampling time, is the maximum value of all charging voltages of the battery cells collected at the sampling time. In this way, the sum of the absolute values of MN of the battery cell to be measured corresponding to a single collection time can be obtained, and each collection time corresponds to a sum of the absolute values of MN, and the sum of the absolute values of MN corresponding to all collection times can be obtained. Taking the absolute value of MN can prevent the MN value of the battery cell from jumping back and forth between positive and negative values in some special cases, so that the fault is not highlighted after taking the sum.

[0055] In step S2, the Euclidean distance is calculated according to the plurality of charging voltages, including: calculating the average value of the plurality of charging voltages collected at the sampling time; calculating the difference value of the charging voltage of the battery cell to be measured collected at the sampling time minus the average value; and calculating the Euclidean distance according to the plurality of difference values corresponding to the plurality of sampling times. Wherein, the Euclidean distance is calculated according to the plurality of difference values corresponding to the plurality of sampling times, including: d1= ; d1 is the Euclidean distance, is the serial number of the sampling time, n is the number of sampling times (for example, 20), is the charging voltage of the battery cell to be measured collected at the th sampling time, is the average value of all charging voltages of the battery cells collected at the th sampling time.

[0056] The step S3 comprises: respectively downsampling the plurality of sampling time points according to a plurality of different scale factors to obtain a plurality of sampling time points under each scale factor; calculating a sample entropy corresponding to the scale factor according to the plurality of sampling time points under the scale factor; fitting the plurality of scale factors and the plurality of sample entropies corresponding thereto to obtain a sample entropy-scale factor curve; and determining a slope of the sample entropy-scale factor curve.

[0057] The plurality of sampling time points under each scale factor are obtained by respectively downsampling the plurality of sampling time points according to a plurality of different scale factors, and the process of calculating the sample entropy corresponding to the scale factor comprises: = , 1≤b≤n ; b is the serial number of the sampling time, is the bth sampling time, is the scale factor, is the serial number of the sampling time point, is the bth sampling time point, and n is the number of the sampling time points. For example, n=20, =1, 1≤b≤20, = , , = , , , the plurality of sampling time points are the original sampling time points; =2, 1≤b≤10, =( + ) / 2, =( + ) / 2, =( + ) / 2; =3, 1≤b≤20 / 3, =( + + ) / 3, =( + + ) / 3, =( + ) / 3.

[0058] The process of calculating the sample entropy corresponding to the scale factor comprises:

[0059] Supposing that the number of the sampling time points is e, 1≤b≤e, and e sampling time points Reconstruct into e-m+1 m-dimensional vectors X(1), X(2), ..., X(e-m+1), X(b) = { , , ..., For example, if m=2, then e sampling times... Reconstruct into e-1 two-dimensional vectors X(1), X(2), ..., X(e-1). When =2, e=10, then X(1)={ , }、X(2)={ , }、...、X(9)={ , };

[0060] Calculate d2= [ ]; i,j=1,2,...e-m+1,i≠j;d2 is the maximum absolute value of the difference between corresponding elements of two vectors;

[0061] calculate = r is a given threshold, for example, r = 0.2std, where std is the standard deviation of the sampling time;

[0062] Calculate B= ;

[0063] Let f = ,calculate = A= ;

[0064] Sample entropy S= , When taking finite values, S= .

[0065] For each scale factor Calculate the sample entropy S once; this will yield multiple scaling factors. The corresponding entropy S of multiple samples.

[0066] In step S4, the normal battery cluster center includes a normal battery MN cluster center, a normal battery Euclidean distance cluster center and a normal battery slope cluster center; the fault battery cluster center includes a fault battery MN cluster center, a fault battery Euclidean distance cluster center and a fault battery slope cluster center. Step S4 specifically includes: obtaining the sum of MN absolute values, Euclidean distance and sample entropy-scale factor curve slope of each training sample battery; normalizing the sum of MN absolute values, normalizing the Euclidean distance, and normalizing the sample entropy-scale factor curve slope; taking the absolute value of the normalized sample entropy-scale factor curve slope, and weighting the absolute values of the normalized sum of MN absolute values, Euclidean distance and sample entropy-scale factor curve slope; k-means clustering the weighted sum of MN absolute values to obtain the normal battery MN cluster center and the fault battery MN cluster center; k-means clustering the weighted Euclidean distance to obtain the normal battery Euclidean distance cluster center and the fault battery Euclidean distance cluster center; k-means clustering the absolute values of the weighted sample entropy-scale factor curve slope to obtain the normal battery slope cluster center and the fault battery slope cluster center.

[0067] Wherein, steps S1-S3 are performed with each training sample battery instead of the battery to be tested to obtain the sum of MN absolute values, Euclidean distance and sample entropy-scale factor curve slope of each training sample battery; normalization can be performed by Z-score normalization method; taking the absolute value of the normalized sample entropy-scale factor curve slope is to facilitate horizontal comparison; the importance of the sample entropy-scale factor curve slope is high, and the normalized sum of MN absolute values, Euclidean distance and sample entropy-scale factor curve slope can be weighted 1:1:2 respectively.

[0068] Taking the sum of MN absolute values as an example, the process of k-means clustering is as follows: first, randomly select two MN absolute values as initial centroids; second, assign each MN absolute value to the nearest initial centroid to form two clusters; third, calculate the average of all MN absolute values in the cluster where the first initial centroid is located and take it as the new centroid of the cluster, and calculate the average of all MN absolute values in the cluster where the second initial centroid is located and take it as the new centroid of the cluster; fourth, repeat the second and third steps until the centroids of the two clusters no longer change, and the centroids of the two clusters are the normal battery MN cluster center and the fault battery MN cluster center. The k-means clustering processes of Euclidean distance and sample entropy-scale factor curve slope are the same as that of the sum of MN absolute values. Here, the objective function of k-means clustering is SSE.

[0069] Step S5 includes: N1=1- ; N1 is the first Hamming closeness, in turn, the MN absolute value sum of the to-be-tested battery cell, the Euclidean distance, and the sample entropy-scale factor curve slope, in turn, the MN clustering center of the normal battery cell, the Euclidean distance clustering center of the normal battery cell, and the slope clustering center of the normal battery cell.

[0070] Step S6 includes: N2=1- ; N2 is the second Hamming closeness, in turn, the MN absolute value sum of the to-be-tested battery cell, the Euclidean distance, and the sample entropy-scale factor curve slope, in turn, the MN clustering center of the fault battery cell, the Euclidean distance clustering center of the fault battery cell, and the slope clustering center of the fault battery cell.

[0071] In the embodiment of the application, if the second Hamming closeness of the to-be-tested battery cell is not greater than the first Hamming closeness, it is determined that the to-be-tested battery cell is normal.

[0072] From the above, the battery cell fault diagnosis method according to the embodiment of the application calculates the MN absolute value sum and the Euclidean distance of the to-be-tested battery cell according to the charging voltage of the battery cell, calculates the sample entropy-scale factor curve slope of the to-be-tested battery cell according to the sampling time, calculates the first Hamming closeness of the MN absolute value sum, the Euclidean distance, and the sample entropy-scale factor curve slope of the to-be-tested battery cell to the normal battery cell clustering center, calculates the second Hamming closeness of the MN absolute value sum, the Euclidean distance, and the sample entropy-scale factor curve slope of the to-be-tested battery cell to the fault battery cell clustering center, and if the second Hamming closeness is greater than the first Hamming closeness, it is determined that the to-be-tested battery cell is faulty, and the battery cell fault is comprehensively diagnosed from three dimensions.

[0073] As shown in FIG. 8, the embodiment of the application further provides a battery cell fault diagnosis device, which comprises:

[0074] An acquisition module is configured to acquire the charging voltages of a plurality of battery cells collected at a plurality of sampling times.

[0075] A calculation module is configured to calculate the MN absolute value sum and the Euclidean distance of a to-be-tested battery cell according to the plurality of charging voltages.

[0076] The calculation module is further configured to calculate the sample entropy-scale factor curve slope of the to-be-tested battery cell according to the plurality of sampling times.

[0077] ​​​​​​​​The acquisition module is further configured to acquire a preset normal battery cluster center and a preset fault battery cluster center.

[0078] The calculation module is further configured to calculate a first Hamming closeness degree according to the sum of the absolute values of the MNs, the Euclidean distance, the slope of the sample entropy-scale factor curve, and the normal battery cluster center.

[0079] The calculation module is further configured to calculate a second Hamming closeness degree according to the sum of the absolute values of the MNs, the Euclidean distance, the slope of the sample entropy-scale factor curve, and the fault battery cluster center.

[0080] The judgment module is configured to judge that the battery under test is faulty if the second Hamming closeness degree is greater than the first Hamming closeness degree.

[0081] Further, the calculation module is further configured to: determine a maximum value of the plurality of charging voltages collected at the sampling moment; calculate an average value of the plurality of charging voltages collected at the sampling moment; calculate an MN value corresponding to the sampling moment according to the charging voltage of the battery under test collected at the sampling moment, the maximum value, and the average value; and calculate a sum of absolute values of a plurality of MN values corresponding to a plurality of sampling moments, to obtain the sum of the absolute values of the MNs.

[0082] Further, the calculation module is further configured to: calculate an average value of the plurality of charging voltages collected at the sampling moment; calculate a difference value of the charging voltage of the battery under test collected at the sampling moment minus the average value; and calculate the Euclidean distance according to a plurality of difference values corresponding to a plurality of sampling moments.

[0083] Further, the calculation module is further configured to: down-sample a plurality of sampling moments according to a plurality of different scale factors, to obtain a plurality of sampling times under each scale factor; calculate a sample entropy corresponding to the scale factor according to the plurality of sampling times under the scale factor; fit a plurality of scale factors and a plurality of sample entropies corresponding thereto, to obtain a sample entropy-scale factor curve; and determine a slope of the sample entropy-scale factor curve.

[0084] Further, the normal battery cluster center includes a normal battery MN cluster center, a normal battery Euclidean distance cluster center, and a normal battery slope cluster center; and the fault battery cluster center includes a fault battery MN cluster center, a fault battery Euclidean distance cluster center, and a fault battery slope cluster center.

[0085] The acquisition module can further be configured to: acquire the sum of MN absolute values, the Euclidean distance, and the sample entropy-scale factor curve slope of each training sample battery cell in the plurality of training sample battery cells; normalize the plurality of sums of MN absolute values, normalize the plurality of Euclidean distances, and normalize the plurality of sample entropy-scale factor curve slopes; take absolute values of the normalized sample entropy-scale factor curve slopes, and weight the absolute values of the normalized sum of MN absolute values, Euclidean distance, and sample entropy-scale factor curve slope; perform k-means clustering on the weighted plurality of sums of MN absolute values to obtain normal battery cell MN clustering centers and fault battery cell MN clustering centers; perform k-means clustering on the weighted plurality of Euclidean distances to obtain normal battery cell Euclidean distance clustering centers and fault battery cell Euclidean distance clustering centers; and perform k-means clustering on the absolute values of the weighted plurality of sample entropy-scale factor curve slopes to obtain normal battery cell slope clustering centers and fault battery cell slope clustering centers.

[0086] Further, the calculation module can calculate a first Hamming closeness degree according to the sum of MN absolute values, the Euclidean distance, the sample entropy-scale factor curve slope of the battery cell to be tested, and the normal battery cell clustering centers, and can include:

[0087] N1=1- ;

[0088] N1 is the first Hamming closeness degree, 、 、 in turn, the sum of MN absolute values, the Euclidean distance, and the sample entropy-scale factor curve slope of the battery cell to be tested, 、 、 in turn, the normal battery cell MN clustering center, the normal battery cell Euclidean distance clustering center, and the normal battery cell slope clustering center.

[0089] Further, the calculation module can calculate a second Hamming closeness degree according to the sum of MN absolute values, the Euclidean distance, the sample entropy-scale factor curve slope of the battery cell to be tested, and the fault battery cell clustering centers, and can include:

[0090] N2=1- ;

[0091] N2 is the second Hamming closeness degree, 、 、 in turn, the sum of MN absolute values, the Euclidean distance, and the sample entropy-scale factor curve slope of the battery cell to be tested, 、 、 in turn, the fault battery cell MN clustering center, the fault battery cell Euclidean distance clustering center, and the fault battery cell slope clustering center.

[0092] Based on the same inventive concept as the battery cell fault diagnosis method above, the embodiments of the present application also provide a computer device, comprising a memory, a processor and a computer program stored in the memory, the processor executes the computer program to implement the steps of any one of the battery cell fault diagnosis methods above.

[0093] The bus architecture (represented by a bus) can include any number of interconnecting buses and bridges, the bus links various circuits such as one or more processors represented by a processor and a memory represented by a memory together. The bus can also link various other circuits such as peripheral devices, voltage stabilizers and power management circuits, which are well known in the art, and therefore, will not be further described herein. The bus interface provides an interface between the bus and the receiver and the transmitter. The receiver and the transmitter can be the same element, i.e. a transceiver, which provides a unit for communicating with various other devices on a transmission medium. The processor is responsible for managing the bus and general processing, while the memory can be used to store data used by the processor in performing operations.

[0094] Since the computer device introduced in the embodiments of the present application is the computer device used to implement the battery cell fault diagnosis method in the embodiments of the present application, based on the battery cell fault diagnosis method introduced in the embodiments of the present application, those skilled in the art can understand the specific implementation of the computer device of the embodiments of the present application and its various forms, so the implementation of the method in the embodiments of the present application by the computer device will not be described in detail here. As long as the computer device used to implement the battery cell fault diagnosis method in the embodiments of the present application is implemented by those skilled in the art, it belongs to the scope of the present application.

[0095] Based on the same inventive concept as the battery cell fault diagnosis method above, the embodiments of the present application also provide a computer device, comprising a memory, a processor and a computer program stored in the memory, the processor executes the computer program to implement the steps of any one of the battery cell fault diagnosis methods above.

[0096] Those skilled in the art will appreciate that embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) containing computer-usable program code.

[0097] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks.

[0098] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart block or blocks.

[0099] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks.

[0100] Although preferred embodiments of the application have been described herein, it will be apparent to those skilled in the art that various modifications can be made within the scope of the application. Accordingly, it is intended that all content of the appended claims be interpreted to include all such modifications and alterations.

[0101] Obviously, numerous modifications and variations of the present application are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A battery cell fault diagnosis method applied to a diagnosis device of a battery, comprising: obtaining charging voltages of a plurality of battery cells collected at a plurality of sampling time points; calculating a sum of absolute values of MNs and an Euclidean distance of a to-be-tested battery cell according to the charging voltages; calculating a sample entropy-scale factor curve slope of the to-be-tested battery cell according to the sampling time points; obtaining a preset normal battery cell clustering center and a preset fault battery cell clustering center; calculating a first Hamming closeness degree according to the sum of absolute values of MNs, the Euclidean distance, the sample entropy-scale factor curve slope of the to-be-tested battery cell, and the normal battery cell clustering center; calculating a second Hamming closeness degree according to the sum of absolute values of MNs, the Euclidean distance, the sample entropy-scale factor curve slope of the to-be-tested battery cell, and the fault battery cell clustering center; if the second Hamming closeness degree is greater than the first Hamming closeness degree, judging that the to-be-tested battery cell is faulty.

2. The battery cell failure diagnosis method according to claim 1, wherein calculating the sum of absolute values of MNs according to the charging voltages, comprising: determining a maximum value of the charging voltages collected at the sampling time points; calculating an average value of the charging voltages collected at the sampling time points; calculating an MN value corresponding to the sampling time according to the charging voltage of the to-be-tested battery cell, the maximum value and the average value collected at the sampling time; calculating a sum of absolute values of MNs corresponding to the sampling time points, to obtain the sum of absolute values of MNs.

3. The battery cell failure diagnosis method of claim 1, wherein, calculating the Euclidean distance according to the charging voltages, comprising: calculating an average value of the charging voltages collected at the sampling time points; calculating a difference value of the charging voltage of the to-be-tested battery cell collected at the sampling time minus the average value; calculating the Euclidean distance according to a plurality of difference values corresponding to a plurality of sampling time points.

4. The battery cell failure diagnosis method of claim 1, wherein, calculating the sample entropy-scale factor curve slope of the to-be-tested battery cell according to the sampling time points, comprising: down-sampling a plurality of the sampling time points according to a plurality of different scale factors, to obtain a plurality of sampling times under each scale factor; calculating a sample entropy corresponding to the scale factor according to a plurality of the sampling times under the scale factor; fitting a plurality of the scale factors and a plurality of the sample entropies corresponding thereto, to obtain a sample entropy-scale factor curve; determining a slope of the sample entropy-scale factor curve.

5. The battery cell failure diagnosis method of claim 1, wherein, the normal battery cell clustering center comprises a normal battery cell MN clustering center, a normal battery cell Euclidean distance clustering center and a normal battery cell slope clustering center; and the fault battery cell clustering center comprises a fault battery cell MN clustering center, a fault battery cell Euclidean distance clustering center and a fault battery cell slope clustering center; obtaining the preset normal battery cell clustering center and the preset fault battery cell clustering center, comprising: obtaining the sum of absolute values of MNs, the Euclidean distance and the sample entropy-scale factor curve slope of each training sample battery cell in a plurality of training sample battery cells; normalizing a plurality of the sum of absolute values of MNs, a plurality of the Euclidean distances and a plurality of the sample entropy-scale factor curve slopes; and taking absolute values of slopes of the normalized sample entropy-scale factor curves, weighting the sum of the MN absolute values, the Euclidean distances and the absolute values of the slopes of the sample entropy-scale factor curves after normalization; performing k-means clustering on the weighted sum of the MN absolute values to obtain the normal battery cell MN clustering center and the fault battery cell MN clustering center; performing k-means clustering on the weighted Euclidean distances to obtain the normal battery cell Euclidean distance clustering center and the fault battery cell Euclidean distance clustering center; performing k-means clustering on the weighted absolute values of the slopes of the sample entropy-scale factor curves to obtain the normal battery cell slope clustering center and the fault battery cell slope clustering center.

6. The battery cell failure diagnosis method according to claim 5, wherein calculating a first Hamming closeness according to the sum of the MN absolute values, the Euclidean distance, the slope of the sample entropy-scale factor curve of the battery cell to be tested and the normal battery cell clustering center, including: N1=1- ； N1 is the first Hamming proximity, 、 、 The MN absolute value sum, Euclidean distance, sample entropy-scale factor curve slope of the to-be-tested battery cell in turn, 、 、 in turn, the normal battery cell MN clustering center, the normal battery cell Euclidean distance clustering center and the normal battery cell slope clustering center.

7. The battery cell failure diagnosis method according to claim 5, wherein calculating a second Hamming closeness according to the sum of the MN absolute values, the Euclidean distance, the slope of the sample entropy-scale factor curve of the battery cell to be tested and the fault battery cell clustering center, including: N2=1- ； N2 is the second Hamming proximity, 、 、 The MN absolute value sum, Euclidean distance, sample entropy-scale factor curve slope of the to-be-tested battery cell in turn, 、 、 in turn, the fault battery cell MN clustering center, the fault battery cell Euclidean distance clustering center and the fault battery cell slope clustering center. 8.A battery cell fault diagnosis apparatus, comprising: an acquisition module configured to acquire charging voltages of a plurality of battery cells collected at a plurality of sampling time points; a calculation module configured to calculate a sum of MN absolute values and a Euclidean distance of a battery cell to be tested according to the charging voltages; calculate a slope of a sample entropy-scale factor curve of the battery cell to be tested according to the plurality of sampling time points; the acquisition module is further configured to acquire preset normal battery cell clustering centers and fault battery cell clustering centers; the calculation module is further configured to calculate a first Hamming closeness according to the sum of the MN absolute values, the Euclidean distance, the slope of the sample entropy-scale factor curve of the battery cell to be tested and the normal battery cell clustering centers; calculate a second Hamming closeness according to the sum of the MN absolute values, the Euclidean distance, the slope of the sample entropy-scale factor curve of the battery cell to be tested and the fault battery cell clustering centers; a judgment module configured to judge that the battery cell to be tested is faulty if the second Hamming closeness is greater than the first Hamming closeness. 9.A computer device, comprising a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement steps of the method according to any one of claims 1-7. 10.A computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement steps of the method according to any one of claims 1-7.

Images