CN112858919A - Battery system online fault diagnosis method and system based on cluster analysis - Google Patents
Battery system online fault diagnosis method and system based on cluster analysis Download PDFInfo
- Publication number
- CN112858919A CN112858919A CN202110059389.7A CN202110059389A CN112858919A CN 112858919 A CN112858919 A CN 112858919A CN 202110059389 A CN202110059389 A CN 202110059389A CN 112858919 A CN112858919 A CN 112858919A
- Authority
- CN
- China
- Prior art keywords
- battery
- cluster
- voltage
- determining
- monomer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/385—Arrangements for measuring battery or accumulator variables
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/60—Other road transportation technologies with climate change mitigation effect
- Y02T10/70—Energy storage systems for electromobility, e.g. batteries
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Secondary Cells (AREA)
- Tests Of Electric Status Of Batteries (AREA)
Abstract
The invention relates to a method and a system for diagnosing online faults of a battery system based on cluster analysis. According to the method and the system for the online fault diagnosis of the battery system based on the cluster analysis, provided by the invention, based on the acquired running data of the electric vehicle, the battery monomers in the battery system of the electric vehicle are subjected to cluster classification by adopting a K-means clustering algorithm, then abnormal battery monomers are quickly and accurately determined according to the Euclidean distance between two battery monomer clusters obtained by classification, and the serial numbers of the battery monomers are output, so that the difficulty in fault monitoring of the battery monomers in a real vehicle is reduced.
Description
Technical Field
The invention relates to the field of battery monomer fault detection, in particular to a method and a system for diagnosing online faults of a battery system based on cluster analysis.
Background
Lithium ion batteries are mainstream energy storage devices for electric vehicles due to their characteristics of long cycle life, high voltage, large output power and low pollution. In order to obtain sufficient output, a lithium ion power battery system is generally composed of a plurality of battery cells connected in series and parallel, and during vehicle running, each battery cell connected in series has the same current excitation, so that theoretically, the battery cells should have the same voltage variation tendency. However, since the battery cells constituting the battery pack initially have a certain difference in performance and the distribution positions and temperatures of the battery cells in an actual vehicle are different, when the battery is operated for a long time or when the battery is affected by external forces such as impact and extrusion, the difference between the battery cells is more significant, and thus, the voltage variation of each battery cell is significantly inconsistent.
The new energy automobile battery management system can obtain data such as battery temperature and voltage. For the battery model-based diagnosis method, residual errors of battery characteristics such as voltage, temperature and the like are generated by using the battery model and real vehicle data, and whether the battery has a fault is judged according to the residual errors. Although the data-driven fault diagnosis method does not need an accurate battery model, a large amount of sample data is needed for training, and the calculation cost is high.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a method and a system for diagnosing the online faults of a battery system based on cluster analysis.
In order to achieve the purpose, the invention provides the following scheme:
a battery system online fault diagnosis method based on cluster analysis comprises the following steps:
acquiring operation data of the electric automobile; the operational data includes: voltage, current and temperature of each cell;
forming a voltage matrix according to the operation data; the row of the voltage matrix represents the serial number of the battery monomer, and the column of the voltage matrix represents a time sequence;
dividing battery monomers in the electric automobile into an abnormal battery monomer cluster and a normal battery monomer cluster according to the voltage matrix by adopting a K-means clustering algorithm;
determining the number ratio of the battery monomers in the abnormal battery monomer cluster to the normal battery monomer cluster, and respectively determining relevant parameters of a cluster center in the abnormal battery monomer cluster and a cluster center in the normal battery monomer cluster; the relevant parameters include: correlation coefficient and fluctuation variance;
determining the Euclidean distance between the cluster center of the abnormal single battery cluster and the cluster center of the normal single battery cluster according to the relevant parameters;
acquiring a preset threshold value; the preset threshold includes: a quantity ratio threshold and a Euclidean distance threshold;
and determining abnormal single batteries according to the relationship between the number ratio and the number ratio threshold value and the relationship between the Euclidean distance and the Euclidean distance threshold value, and outputting the serial numbers of the abnormal single batteries.
Preferably, the dividing of the battery cells in the electric vehicle into an abnormal battery cell cluster and a normal battery cell cluster according to the voltage matrix by using a K-means clustering algorithm specifically includes:
constructing a sample set according to the voltage matrix; the sample set includes: a plurality of elements consisting of the correlation coefficient of each cell and the fluctuation variance of each cell;
and dividing the battery monomers in the electric automobile into an abnormal battery monomer cluster and a normal battery monomer cluster by adopting the K-means clustering algorithm based on the sample set.
Preferably, the constructing a sample set according to the voltage matrix specifically includes:
determining a Pearson correlation coefficient between two adjacent battery cells in the voltage matrix;
determining the correlation coefficient of the battery monomer according to the determined Pearson correlation coefficient between two adjacent battery monomers;
acquiring a voltage value of each battery monomer and a voltage average value of all the battery monomers;
determining the fluctuation variance of the battery monomers according to the voltage value of each battery monomer and the voltage mean value of all the battery monomers;
and constructing the sample set according to the correlation coefficient of the battery cell and the fluctuation variance of the battery cell.
Preferably, the determining the fluctuation variance of the battery cell according to the voltage value and the voltage mean value specifically includes:
performing trend processing on the battery monomers according to the voltage value of each battery monomer and the voltage average value of all the battery monomers to obtain trend vectors;
and determining the fluctuation variance of the battery cell according to the trend vector.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
according to the battery system online fault diagnosis method based on cluster analysis, the battery monomers in the battery system of the electric vehicle are subjected to cluster classification by adopting a K-means clustering algorithm based on the acquired running data of the electric vehicle, then abnormal battery monomers are quickly and accurately determined according to the Euclidean distance between two battery monomer clusters obtained by classification, and the serial numbers of the battery monomers are output, so that the difficulty in monitoring the faults of the battery monomers in a real vehicle is reduced.
In addition, the invention also provides a battery system online fault diagnosis system based on cluster analysis, which corresponds to the battery system online fault diagnosis method based on cluster analysis. The battery system online fault diagnosis system based on cluster analysis comprises:
the operation data acquisition module is used for acquiring operation data of the electric automobile; the operational data includes: voltage, current and temperature of each cell;
the voltage matrix forming module is used for forming a voltage matrix according to the operation data; the row of the voltage matrix represents the serial number of the battery monomer, and the column of the voltage matrix represents a time sequence;
the cluster classification module is used for classifying the battery monomers in the electric automobile into an abnormal battery monomer cluster and a normal battery monomer cluster according to the voltage matrix by adopting a K-means clustering algorithm;
the parameter determining module is used for determining the number ratio of the battery monomers in the abnormal battery monomer cluster to the normal battery monomer cluster and respectively determining relevant parameters of a cluster center in the abnormal battery monomer cluster and a cluster center in the normal battery monomer cluster; the relevant parameters include: correlation coefficient and fluctuation variance;
the Euclidean distance determining module is used for determining the Euclidean distance between the cluster center of the abnormal single battery cluster and the cluster center of the normal single battery cluster according to the relevant parameters;
the threshold value obtaining module is used for obtaining a preset threshold value; the preset threshold includes: a quantity ratio threshold and a Euclidean distance threshold;
and the abnormal single battery determining module is used for determining the abnormal single battery according to the relationship between the number ratio and the number ratio threshold value and the relationship between the Euclidean distance and the Euclidean distance threshold value, and outputting the serial number of the abnormal single battery.
Preferably, the cluster classification module specifically includes:
the sample set constructing submodule is used for constructing a sample set according to the voltage matrix; the sample set includes: a plurality of elements consisting of the correlation coefficient of each cell and the fluctuation variance of each cell;
and the cluster classification submodule is used for classifying the battery monomers in the electric automobile into an abnormal battery monomer cluster and a normal battery monomer cluster based on the sample set by adopting the K-means clustering algorithm.
Preferably, the sample set constructing submodule specifically includes:
the Pearson correlation coefficient determining unit is used for determining the Pearson correlation coefficient between two adjacent battery cells in the voltage matrix;
the correlation coefficient determining unit is used for determining the correlation coefficient of the battery monomer according to the determined Pearson correlation coefficient between two adjacent battery monomers;
the voltage value acquisition unit is used for acquiring the voltage value of each battery monomer and the voltage average value of all the battery monomers;
the fluctuation variance determining unit is used for determining the fluctuation variance of the battery monomers according to the voltage value of each battery monomer and the voltage mean value of all the battery monomers;
and the sample set construction unit is used for constructing the sample set according to the correlation coefficient of the battery monomer and the fluctuation variance of the battery monomer.
Preferably, the fluctuation variance determining unit specifically includes:
the trend vector determining subunit is used for performing trend processing on the battery monomers according to the voltage value of each battery monomer and the voltage average value of all the battery monomers to obtain a trend vector;
and the fluctuation variance determining subunit is used for determining the fluctuation variance of the battery cell according to the trend vector.
The technical effect of the online battery system fault diagnosis system based on cluster analysis provided by the invention is the same as that of the online battery system fault diagnosis method based on cluster analysis provided by the invention, and the implementation is not repeated herein.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a flow chart of a method for online fault diagnosis of a battery system based on cluster analysis according to the present invention;
FIG. 2 is a schematic diagram of data acquisition provided by an embodiment of the present invention;
fig. 3 is a schematic diagram of calculating a correlation coefficient of a battery cell according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of the cluster analysis-based online battery system fault diagnosis system provided by the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
According to the method and the system for the online fault diagnosis of the battery system based on the cluster analysis, provided by the invention, the fault battery monomer is clustered and identified by calculating the correlation coefficient and fluctuation variance of the battery monomer, and the method and the system have the advantages of high calculation speed, high accuracy, easiness in implementation, and the like, and are used for the online fault diagnosis of the battery on a real vehicle.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
Fig. 1 is a flowchart of a method for diagnosing online faults of a battery system based on cluster analysis according to the present invention, and as shown in fig. 1, the method for diagnosing online faults of a battery system based on cluster analysis includes:
step 100: and acquiring the operation data of the electric automobile. The operational data includes: voltage, current, and temperature of each cell. Specifically, as shown in fig. 2, the process of acquiring the operation data of the present invention is as follows: operational data (e.g., voltage, current, and temperature) of the electric vehicle obtained from the plurality of sensors is collected into a transmission terminal BOX (T-BOX). All data are transmitted to a storage server of a new energy automobile big data platform through a wireless network (4G or 5G), and the network conforms to a transmission protocol named by an electric automobile remote service and management system (GB/T32960).
Step 101: a voltage matrix is formed from the operational data. The rows of the voltage matrix represent the cell numbers and the columns of the voltage matrix represent the time series.
Specifically, the collected operational data needs to be preprocessed before the voltage matrix is formed. The pretreatment process comprises the following steps: and (4) removing the duplication of the multi-frame data collected at the same time point, and only keeping one frame of data. And deleting the data frame when the voltage of the battery cell of a certain frame is less than two.
The process of constructing the voltage matrix based on the data obtained after the preprocessing specifically comprises the following steps:
firstly, n battery cells at the t time are expressed as vectors
From t0Time tsThe cell voltage vector at a time may be expressed as
Thus, the voltage matrix of all the cells can be expressed as
Step 102: and dividing the battery monomers in the electric automobile into an abnormal battery monomer cluster and a normal battery monomer cluster according to the voltage matrix by adopting a K-means clustering algorithm. Specifically, the method comprises the following steps:
step 1021: a sample set is constructed from the voltage matrix. The sample set includes: and a plurality of elements consisting of the correlation coefficient of each battery cell and the fluctuation variance of each battery cell. The construction process of the sample set specifically comprises the following steps:
step 10211: and determining the Pearson correlation coefficient between two adjacent battery cells in the voltage matrix. The specific principle of determining the pearson correlation coefficient is shown in fig. 3, and the process specifically includes:
introduction of concept:
x and y are two vectors, x ═ x1，x2，x3，….，xN]，y＝[y1，y2，y3，….，yN]The original formula for the correlation coefficient is:
wherein sigmaxAnd σyIs the standard deviation of vector x and vector y, respectively, and cov (x, y) is the covariance of the x and y vectors.
According to an original formula, calculating a Pearson correlation coefficient between a battery cell i and a battery cell j:
in the formula (I), the compound is shown in the specification,
and
respectively represent at the sampling time t0To the sampling time tsCell voltage of cell i and cell j in between. Wherein j is defined according to the following formula:
through calculation, the Pearson correlation coefficient vector of two adjacent battery monomers is obtained as follows:
wherein, V in FIG. 31，V2，…VDEach represents a voltage value of the battery cell.
Step 10212: and determining the correlation coefficient of the battery monomer according to the determined Pearson correlation coefficient between two adjacent battery monomers. The specific calculation formula of the correlation coefficient is as follows:
calculating to obtain a correlation coefficient vector of the single battery:
step 10213: and acquiring the voltage value of each battery cell and the voltage average value of all the battery cells.
Step 10214: and determining the fluctuation variance of the battery cells according to the voltage value of each battery cell and the voltage mean value of all the battery cells.
Wherein, the determination process of the fluctuation variance is as follows:
and performing trend processing on the single batteries according to the voltage value of each single battery and the voltage average value of all the single batteries to obtain trend vectors. Specifically, for a certain frame, the voltage average of all the cells in the frame is first determined, and then the voltage average of all the cells in the frame is subtracted from the voltage value of each cell in the frame, which is denoted as DV. For cell i, at t0-tsThe trend vectors formed after detrending over the time period are represented as:
and determining the fluctuation variance of the battery cell according to the trend vector. Specifically, the battery cell i is at t0-tsThe fluctuation variance in the time period is calculated according to the following formula:
wherein the content of the first and second substances,
is a vector
The average of all values in (1), and l represents the number of sampling points. Finally, forming a battery cell fluctuation variance matrix:
step 10215: and constructing a sample set according to the correlation coefficient of the battery monomer and the fluctuation variance of the battery monomer.
Step 1022: and dividing the battery monomers in the electric automobile into an abnormal battery monomer cluster and a normal battery monomer cluster based on the sample set by adopting a K-means clustering algorithm. The clustering process specifically comprises the following steps:
the input is a sample set D ═ q1，q2，...qnWherein q isi＝[Si，ri]Two parameters (fluctuation variance and correlation coefficient) of the battery unit i are shown, the clusters of the default cluster are divided into two types, namely h is 2, and the center of each cluster is respectively used as c1And c2And (4) showing.
Randomly selecting two battery monomers q from all battery monomersiAnd q isjAs a cluster center.
Calculating the Euclidean distance d from each single battery sample to the center of the clusteriEach sample is mapped to a cluster center c1、c2The principle of minimum distance determines the cluster in which it is located. The determination principle is expressed by equation (6). Calculating battery monomeriEuclidean distance of two parameters to the center of two clusters, as shown in formula (7), when c is distance1Is less than the distance c2C, the battery cell i is assigned to c1Cluster being cluster-centered, otherwise due to c2A cluster that is a cluster center.
c(h)：＝arg minh||q(i)-uh||2 h＝1，2 (6)
The cluster center is recalculated using the following formula:
wherein, in formula (7), the denominator represents the number of samples in each cluster.
Changing the cluster center until the cluster center is not changed any more, and obtaining the center of each cluster as c1And c2Wherein
Step 103: determining the number ratio of the single batteries in the abnormal single battery cluster to the normal single battery cluster, and respectively determining the relevant parameters of the cluster center in the abnormal single battery cluster and the cluster center in the normal single battery cluster. The relevant parameters include: correlation coefficient and fluctuation variance.
The number ratio of the battery monomers in the normal battery monomer cluster is determined by adopting the following formula:
in the formula, kaIs a number ratio of nabnormalAnd nnormalRespectively representing the number of battery cells in the abnormal cell cluster and the normal cell cluster.
Step 104: and determining the Euclidean distance between the cluster center of the abnormal single battery cluster and the cluster center of the normal single battery cluster according to the related parameters. The formula for calculating the euclidean distance is as follows:
step 105: and acquiring a preset threshold value. The preset threshold includes: number ratio threshold ksSum Euclidean distance threshold ds。
Step 106: and determining the abnormal single battery according to the relationship between the number ratio and the number ratio threshold value and the relationship between the Euclidean distance and the Euclidean distance threshold value, and outputting the serial number of the abnormal single battery. In particular, if d>dsAnd k isa＜ksThen h is set to 2 and the cell number in the abnormal cluster is alarmed. Otherwise, setting h as 1, and giving no alarm to the abnormal monomer. Preferred k of the inventions＝5％，dsBut is not limited thereto.
In addition, the invention also provides a battery system online fault diagnosis system based on cluster analysis, which corresponds to the battery system online fault diagnosis method based on cluster analysis. As shown in fig. 4, the cluster analysis-based online fault diagnosis system for a battery system includes: the device comprises an operation data acquisition module 1, a voltage matrix forming module 2, a cluster classification module 3, a parameter determination module 4, a Euclidean distance determination module 5, a threshold acquisition module 6 and an abnormal single battery determination module 7.
The operation data acquisition module 1 is used for acquiring operation data of the electric vehicle. The operational data includes: voltage, current, and temperature of each cell.
The voltage matrix forming module 2 is used for forming a voltage matrix according to the operation data. The rows of the voltage matrix represent the cell numbers and the columns of the voltage matrix represent the time series.
The cluster classification module 3 is used for classifying the battery monomers in the electric automobile into an abnormal battery monomer cluster and a normal battery monomer cluster according to the voltage matrix by adopting a K-means clustering algorithm.
The parameter determining module 4 is configured to determine a ratio of the number of the battery cells in the abnormal battery cell cluster to the number of the battery cells in the normal battery cell cluster, and determine related parameters of a cluster center in the abnormal battery cell cluster and a cluster center in the normal battery cell cluster, respectively. The relevant parameters include: correlation coefficient and fluctuation variance.
The Euclidean distance determining module 5 is used for determining the Euclidean distance between the cluster center of the abnormal single battery cluster and the cluster center of the normal single battery cluster according to the relevant parameters.
The threshold obtaining module 6 is configured to obtain a preset threshold. The preset threshold includes: a quantity ratio threshold and a euclidean distance threshold.
The abnormal single battery determining module 7 is configured to determine an abnormal single battery according to a relationship between the number ratio and the number ratio threshold and a relationship between the euclidean distance and the euclidean distance threshold, and output a serial number of the abnormal single battery.
Preferably, the cluster classification module 3 specifically includes: a sample set construction sub-module and a cluster classification sub-module.
And the sample set constructing submodule is used for constructing a sample set according to the voltage matrix. The sample set includes: and a plurality of elements consisting of the correlation coefficient of each battery cell and the fluctuation variance of each battery cell.
The cluster classification submodule is used for classifying the battery monomers in the electric automobile into an abnormal battery monomer cluster and a normal battery monomer cluster based on the sample set by adopting a K-means clustering algorithm.
Preferably, in order to further improve the accuracy of determining the battery cell, the sample set constructing submodule may include: the device comprises a Pearson correlation coefficient determining unit, a voltage value acquiring unit, a fluctuation variance determining unit and a sample set constructing unit.
The Pearson correlation coefficient determining unit is used for determining the Pearson correlation coefficient between two adjacent battery cells in the voltage matrix.
The correlation coefficient determining unit is used for determining the correlation coefficient of the battery cell according to the determined Pearson correlation coefficient between two adjacent battery cells.
The voltage value acquisition unit is used for acquiring the voltage value of each battery cell and the voltage average value of all the battery cells.
The fluctuation variance determining unit is used for determining the fluctuation variance of the battery cells according to the voltage value of each battery cell and the voltage mean value of all the battery cells.
The sample set constructing unit is used for constructing a sample set according to the correlation coefficient of the battery monomer and the fluctuation variance of the battery monomer.
Preferably, the fluctuation variance determining unit specifically includes: a trend vector determination subunit and a fluctuation variance determination subunit.
The trend vector determination subunit is used for performing trend processing on the battery monomers according to the voltage value of each battery monomer and the voltage average value of all the battery monomers to obtain a trend vector.
The fluctuation variance determining subunit is used for determining the fluctuation variance of the battery cell according to the trend vector.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (8)
1. A battery system online fault diagnosis method based on cluster analysis is characterized by comprising the following steps:
acquiring operation data of the electric automobile; the operational data includes: voltage, current and temperature of each cell;
forming a voltage matrix according to the operation data; the row of the voltage matrix represents the serial number of the battery monomer, and the column of the voltage matrix represents a time sequence;
dividing battery monomers in the electric automobile into an abnormal battery monomer cluster and a normal battery monomer cluster according to the voltage matrix by adopting a K-means clustering algorithm;
determining the number ratio of the battery monomers in the abnormal battery monomer cluster to the normal battery monomer cluster, and respectively determining relevant parameters of a cluster center in the abnormal battery monomer cluster and a cluster center in the normal battery monomer cluster; the relevant parameters include: correlation coefficient and fluctuation variance;
determining the Euclidean distance between the cluster center of the abnormal single battery cluster and the cluster center of the normal single battery cluster according to the relevant parameters;
acquiring a preset threshold value; the preset threshold includes: a quantity ratio threshold and a Euclidean distance threshold;
and determining abnormal single batteries according to the relationship between the number ratio and the number ratio threshold value and the relationship between the Euclidean distance and the Euclidean distance threshold value, and outputting the serial numbers of the abnormal single batteries.
2. The cluster analysis-based online fault diagnosis method for the battery system according to claim 1, wherein the battery cells in the electric vehicle are classified into abnormal battery cell clusters and normal battery cell clusters according to the voltage matrix by using a K-means clustering algorithm, and the method specifically comprises the following steps:
constructing a sample set according to the voltage matrix; the sample set includes: a plurality of elements consisting of the correlation coefficient of each cell and the fluctuation variance of each cell;
and dividing the battery monomers in the electric automobile into an abnormal battery monomer cluster and a normal battery monomer cluster by adopting the K-means clustering algorithm based on the sample set.
3. The cluster analysis-based online battery system fault diagnosis method according to claim 2, wherein the constructing a sample set according to the voltage matrix specifically comprises:
determining a Pearson correlation coefficient between two adjacent battery cells in the voltage matrix;
determining the correlation coefficient of the battery monomer according to the determined Pearson correlation coefficient between two adjacent battery monomers;
acquiring a voltage value of each battery monomer and a voltage average value of all the battery monomers;
determining the fluctuation variance of the battery monomers according to the voltage value of each battery monomer and the voltage mean value of all the battery monomers;
and constructing the sample set according to the correlation coefficient of the battery cell and the fluctuation variance of the battery cell.
4. The cluster analysis-based online battery system fault diagnosis method according to claim 3, wherein the determining of the fluctuation variance of the battery cells according to the voltage value and the voltage mean value specifically comprises:
performing trend processing on the battery monomers according to the voltage value of each battery monomer and the voltage average value of all the battery monomers to obtain trend vectors;
and determining the fluctuation variance of the battery cell according to the trend vector.
5. A battery system online fault diagnosis system based on cluster analysis is characterized by comprising:
the operation data acquisition module is used for acquiring operation data of the electric automobile; the operational data includes: voltage, current and temperature of each cell;
the voltage matrix forming module is used for forming a voltage matrix according to the operation data; the row of the voltage matrix represents the serial number of the battery monomer, and the column of the voltage matrix represents a time sequence;
the cluster classification module is used for classifying the battery monomers in the electric automobile into an abnormal battery monomer cluster and a normal battery monomer cluster according to the voltage matrix by adopting a K-means clustering algorithm;
the parameter determining module is used for determining the number ratio of the battery monomers in the abnormal battery monomer cluster to the normal battery monomer cluster and respectively determining relevant parameters of a cluster center in the abnormal battery monomer cluster and a cluster center in the normal battery monomer cluster; the relevant parameters include: correlation coefficient and fluctuation variance;
the Euclidean distance determining module is used for determining the Euclidean distance between the cluster center of the abnormal single battery cluster and the cluster center of the normal single battery cluster according to the relevant parameters;
the threshold value obtaining module is used for obtaining a preset threshold value; the preset threshold includes: a quantity ratio threshold and a Euclidean distance threshold;
and the abnormal single battery determining module is used for determining the abnormal single battery according to the relationship between the number ratio and the number ratio threshold value and the relationship between the Euclidean distance and the Euclidean distance threshold value, and outputting the serial number of the abnormal single battery.
6. The cluster analysis-based online battery system fault diagnosis system of claim 5, wherein the cluster classification module specifically comprises:
the sample set constructing submodule is used for constructing a sample set according to the voltage matrix; the sample set includes: a plurality of elements consisting of the correlation coefficient of each cell and the fluctuation variance of each cell;
and the cluster classification submodule is used for classifying the battery monomers in the electric automobile into an abnormal battery monomer cluster and a normal battery monomer cluster based on the sample set by adopting the K-means clustering algorithm.
7. The cluster analysis-based online battery system fault diagnosis system of claim 6, wherein the sample set construction submodule specifically comprises:
the Pearson correlation coefficient determining unit is used for determining the Pearson correlation coefficient between two adjacent battery cells in the voltage matrix;
the correlation coefficient determining unit is used for determining the correlation coefficient of the battery monomer according to the determined Pearson correlation coefficient between two adjacent battery monomers;
the voltage value acquisition unit is used for acquiring the voltage value of each battery monomer and the voltage average value of all the battery monomers;
the fluctuation variance determining unit is used for determining the fluctuation variance of the battery monomers according to the voltage value of each battery monomer and the voltage mean value of all the battery monomers;
and the sample set construction unit is used for constructing the sample set according to the correlation coefficient of the battery monomer and the fluctuation variance of the battery monomer.
8. The cluster analysis-based online battery system fault diagnosis system of claim 7, wherein the fluctuation variance determination unit specifically comprises:
the trend vector determining subunit is used for performing trend processing on the battery monomers according to the voltage value of each battery monomer and the voltage average value of all the battery monomers to obtain a trend vector;
and the fluctuation variance determining subunit is used for determining the fluctuation variance of the battery cell according to the trend vector.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110059389.7A CN112858919B (en) | 2021-01-18 | 2021-01-18 | Battery system online fault diagnosis method and system based on cluster analysis |
| PCT/CN2021/129524 WO2022151819A1 (en) | 2021-01-18 | 2021-11-09 | Clustering analysis-based battery system online fault diagnosis method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110059389.7A CN112858919B (en) | 2021-01-18 | 2021-01-18 | Battery system online fault diagnosis method and system based on cluster analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112858919A true CN112858919A (en) | 2021-05-28 |
| CN112858919B CN112858919B (en) | 2022-04-01 |
Family
ID=76007154
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110059389.7A Active CN112858919B (en) | 2021-01-18 | 2021-01-18 | Battery system online fault diagnosis method and system based on cluster analysis |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112858919B (en) |
| WO (1) | WO2022151819A1 (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113820333A (en) * | 2021-09-16 | 2021-12-21 | 无锡先导智能装备股份有限公司 | Battery pole piece abnormity detection method and device, upper computer and detection system |
| CN114114039A (en) * | 2021-12-06 | 2022-03-01 | 湖北亿纬动力有限公司 | Method and device for evaluating consistency of single battery cells of battery system |
| CN114264957A (en) * | 2021-12-02 | 2022-04-01 | 东软集团股份有限公司 | Abnormal monomer detection method and related equipment thereof |
| WO2022151819A1 (en) * | 2021-01-18 | 2022-07-21 | 北京理工大学 | Clustering analysis-based battery system online fault diagnosis method and system |
| CN114839541A (en) * | 2022-05-18 | 2022-08-02 | 山东大学 | Power battery pack inconsistency diagnosis method and system based on K-means clustering |
| CN115877222A (en) * | 2023-02-14 | 2023-03-31 | 国网浙江省电力有限公司宁波供电公司 | Energy storage power station fault detection method and device, medium and energy storage power station |
| CN116404186A (en) * | 2023-06-08 | 2023-07-07 | 西安黄河电子技术有限公司 | Power lithium-manganese battery production system |
| CN116482560A (en) * | 2023-06-21 | 2023-07-25 | 中国华能集团清洁能源技术研究院有限公司 | Battery fault detection method and device, electronic equipment and storage medium |
| CN116749812A (en) * | 2023-08-14 | 2023-09-15 | 南通百仕灵新能源科技有限公司 | Self-adaptive charging method of new energy charging equipment |
| CN113820333B (en) * | 2021-09-16 | 2024-06-07 | 无锡先导智能装备股份有限公司 | Battery pole piece abnormality detection method, device, upper computer and detection system |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115061049B (en) * | 2022-08-08 | 2022-11-01 | 山东卓朗检测股份有限公司 | Method and system for rapidly detecting UPS battery fault of data center |
| CN115366683B (en) * | 2022-08-09 | 2024-05-24 | 北京理工大学 | Fault diagnosis strategy for multi-dimensional model fusion of power battery of new energy automobile |
| CN115511013B (en) * | 2022-11-22 | 2023-04-07 | 河海大学 | Large-scale energy storage power station abnormal battery identification method, device and storage medium |
| CN115840157B (en) * | 2022-12-08 | 2023-08-22 | 斯润天朗(合肥)科技有限公司 | Lithium battery electrical performance index coordination analysis system based on EOF analysis |
| CN116106758B (en) * | 2023-03-23 | 2024-01-30 | 华能新能源股份有限公司山西分公司 | Battery fault diagnosis method and system based on data driving |
| CN116502112B (en) * | 2023-06-29 | 2023-10-24 | 深圳市联明电源有限公司 | New energy power supply test data management method and system |
| CN116756595B (en) * | 2023-08-23 | 2023-12-01 | 深圳市森瑞普电子有限公司 | Conductive slip ring fault data acquisition and monitoring method |
| CN117648589B (en) * | 2024-01-30 | 2024-05-14 | 云储新能源科技有限公司 | Energy storage battery thermal runaway early warning method, system, electronic equipment and medium |
| CN117665629A (en) * | 2024-01-31 | 2024-03-08 | 成都深瑞同华科技有限公司 | Method and device for evaluating voltage consistency of energy storage battery bin and storage medium |
| CN117892095B (en) * | 2024-03-14 | 2024-05-28 | 山东泰开电力电子有限公司 | Intelligent detection method for faults of heat dissipation system for energy storage system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007167058A (en) * | 2005-06-22 | 2007-07-05 | Tohoku Univ | Method for forecasting prognosis of cancer |
| CN110398382A (en) * | 2019-06-28 | 2019-11-01 | 南京康尼机电股份有限公司 | A kind of underground automobile door system performance decline detection method based on DPC |
| CN110794305A (en) * | 2019-10-14 | 2020-02-14 | 北京理工大学 | Power battery fault diagnosis method and system |
| CN111090050A (en) * | 2020-01-21 | 2020-05-01 | 合肥工业大学 | Lithium battery fault diagnosis method based on support vector machine and K mean value |
| CN111707951A (en) * | 2020-06-22 | 2020-09-25 | 北京理工大学 | Battery pack consistency evaluation method and system |
| CN111929591A (en) * | 2020-08-21 | 2020-11-13 | 彩虹无线（北京）新技术有限公司 | Fault battery detection method, device, equipment and computer storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102468895B1 (en) * | 2015-07-21 | 2022-11-21 | 삼성전자주식회사 | Method and apparatus for estimating state of battery |
| CN108254689B (en) * | 2016-12-29 | 2020-04-28 | 中国电信股份有限公司 | Method and system for detecting battery pack reverse-pole single battery |
| CN112858919B (en) * | 2021-01-18 | 2022-04-01 | 北京理工大学 | Battery system online fault diagnosis method and system based on cluster analysis |
-
2021
- 2021-01-18 CN CN202110059389.7A patent/CN112858919B/en active Active
- 2021-11-09 WO PCT/CN2021/129524 patent/WO2022151819A1/en active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007167058A (en) * | 2005-06-22 | 2007-07-05 | Tohoku Univ | Method for forecasting prognosis of cancer |
| CN110398382A (en) * | 2019-06-28 | 2019-11-01 | 南京康尼机电股份有限公司 | A kind of underground automobile door system performance decline detection method based on DPC |
| CN110794305A (en) * | 2019-10-14 | 2020-02-14 | 北京理工大学 | Power battery fault diagnosis method and system |
| CN111090050A (en) * | 2020-01-21 | 2020-05-01 | 合肥工业大学 | Lithium battery fault diagnosis method based on support vector machine and K mean value |
| CN111707951A (en) * | 2020-06-22 | 2020-09-25 | 北京理工大学 | Battery pack consistency evaluation method and system |
| CN111929591A (en) * | 2020-08-21 | 2020-11-13 | 彩虹无线（北京）新技术有限公司 | Fault battery detection method, device, equipment and computer storage medium |
Non-Patent Citations (3)
| Title |
|---|
| YANG ZHAO: "Fault and defect diagnosis of battery for electric vehicles based on big data analysis methods", 《APPLIED ENERGY》 * |
| 刘克平: "基于特征聚类与特征选择算法的SOFC 系统故障定位", 《化工自动化及仪表》 * |
| 李晓宇: "基于聚类分析算法电动汽车在线安全性研究", 《昆明理工大学学报》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022151819A1 (en) * | 2021-01-18 | 2022-07-21 | 北京理工大学 | Clustering analysis-based battery system online fault diagnosis method and system |
| CN113820333A (en) * | 2021-09-16 | 2021-12-21 | 无锡先导智能装备股份有限公司 | Battery pole piece abnormity detection method and device, upper computer and detection system |
| CN113820333B (en) * | 2021-09-16 | 2024-06-07 | 无锡先导智能装备股份有限公司 | Battery pole piece abnormality detection method, device, upper computer and detection system |
| CN114264957A (en) * | 2021-12-02 | 2022-04-01 | 东软集团股份有限公司 | Abnormal monomer detection method and related equipment thereof |
| CN114264957B (en) * | 2021-12-02 | 2024-05-07 | 东软集团股份有限公司 | Abnormal monomer detection method and related equipment thereof |
| CN114114039B (en) * | 2021-12-06 | 2023-10-03 | 湖北亿纬动力有限公司 | Method and device for evaluating consistency of single battery cells of battery system |
| CN114114039A (en) * | 2021-12-06 | 2022-03-01 | 湖北亿纬动力有限公司 | Method and device for evaluating consistency of single battery cells of battery system |
| CN114839541A (en) * | 2022-05-18 | 2022-08-02 | 山东大学 | Power battery pack inconsistency diagnosis method and system based on K-means clustering |
| CN115877222A (en) * | 2023-02-14 | 2023-03-31 | 国网浙江省电力有限公司宁波供电公司 | Energy storage power station fault detection method and device, medium and energy storage power station |
| CN116404186A (en) * | 2023-06-08 | 2023-07-07 | 西安黄河电子技术有限公司 | Power lithium-manganese battery production system |
| CN116404186B (en) * | 2023-06-08 | 2023-09-19 | 西安黄河电子技术有限公司 | Power lithium-manganese battery production system |
| CN116482560B (en) * | 2023-06-21 | 2023-09-12 | 中国华能集团清洁能源技术研究院有限公司 | Battery fault detection method and device, electronic equipment and storage medium |
| CN116482560A (en) * | 2023-06-21 | 2023-07-25 | 中国华能集团清洁能源技术研究院有限公司 | Battery fault detection method and device, electronic equipment and storage medium |
| CN116749812A (en) * | 2023-08-14 | 2023-09-15 | 南通百仕灵新能源科技有限公司 | Self-adaptive charging method of new energy charging equipment |
| CN116749812B (en) * | 2023-08-14 | 2023-10-20 | 南通百仕灵新能源科技有限公司 | Self-adaptive charging method of new energy charging equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2022151819A1 (en) | 2022-07-21 |
| CN112858919B (en) | 2022-04-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112858919B (en) | Battery system online fault diagnosis method and system based on cluster analysis | |
| CN111707951B (en) | Battery pack consistency evaluation method and system | |
| CN110068774B (en) | Lithium battery health state estimation method and device and storage medium | |
| CN107422266B (en) | Fault diagnosis method and device for high-capacity battery energy storage system | |
| CN112904219B (en) | Big data-based power battery health state prediction method | |
| CN111257753B (en) | Battery system fault diagnosis method | |
| CN112816881B (en) | Battery pressure difference abnormality detection method, device and computer storage medium | |
| CN112327189B (en) | Comprehensive judging method for health state of energy storage battery based on KNN algorithm | |
| CN114818831B (en) | Bidirectional lithium ion battery fault detection method and system based on multi-source perception | |
| CN113791350B (en) | Battery fault prediction method | |
| CN112345956A (en) | Battery pack charge state detection method and device | |
| CN113189495A (en) | Method and device for predicting health state of battery and electronic equipment | |
| CN115327403A (en) | Power battery safety evaluation method and device based on new energy automobile big data | |
| CN115219903A (en) | Battery self-discharge rate abnormity judgment method and device based on Internet of vehicles data analysis | |
| CN115032556A (en) | Energy storage battery system state evaluation method and device, storage medium and electronic equipment | |
| CN112986839B (en) | Confidence interval-based fault diagnosis method and system for lithium ion power battery pack | |
| CN114200323A (en) | Battery short-circuit fault early warning information generation method and device, equipment and medium | |
| CN113391214A (en) | Battery micro-fault diagnosis method based on battery charging voltage ranking change | |
| CN113379005A (en) | Intelligent energy management system and method for power grid power equipment | |
| CN117148194A (en) | Electric automobile battery pack fault detection method and system | |
| CN116679213A (en) | SOH estimation method for electric vehicle power battery based on integrated deep learning | |
| CN115270983A (en) | Switch cabinet fault prediction method based on AdaBoost-RBF algorithm | |
| CN115840157B (en) | Lithium battery electrical performance index coordination analysis system based on EOF analysis | |
| CN115389947B (en) | Lithium battery health state prediction method and device, electronic equipment and storage medium | |
| CN118132959A (en) | Power battery inconsistency quantification method based on battery cell voltage value |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |