CN111126449A - Battery fault classification diagnosis method based on cluster analysis - Google Patents
Battery fault classification diagnosis method based on cluster analysis Download PDFInfo
- Publication number
- CN111126449A CN111126449A CN201911206611.0A CN201911206611A CN111126449A CN 111126449 A CN111126449 A CN 111126449A CN 201911206611 A CN201911206611 A CN 201911206611A CN 111126449 A CN111126449 A CN 111126449A
- Authority
- CN
- China
- Prior art keywords
- battery
- data
- clustering
- cluster analysis
- cluster
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000007621 cluster analysis Methods 0.000 title claims abstract description 20
- 238000003745 diagnosis Methods 0.000 title claims abstract description 14
- 239000000178 monomer Substances 0.000 claims description 8
- 239000010755 BS 2869 Class G Substances 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000005484 gravity Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000007418 data mining Methods 0.000 abstract description 4
- 230000006698 induction Effects 0.000 abstract description 2
- 238000005065 mining Methods 0.000 abstract description 2
- 238000010606 normalization Methods 0.000 abstract 1
- 238000004458 analytical method Methods 0.000 description 9
- 238000003064 k means clustering Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Secondary Cells (AREA)
- Tests Of Electric Status Of Batteries (AREA)
Abstract
The invention relates to a battery fault classification diagnosis method based on cluster analysis, which takes data such as pressure difference, temperature difference, average current, average vehicle speed and the like recorded by a battery in a period of time as 4 important indexes of cluster induction. After normalization processing, the data mining of cluster analysis can be carried out on various problems recorded by the battery by using a K-means method. The method can be used for mining the association between the battery data from a deeper level and better classifying various batteries. And then the efficiency of the manager is higher during information retrieval, and the obtained result is more complete.
Description
Technical Field
The invention relates to the technical field of batteries, in particular to a battery fault classification diagnosis method based on cluster analysis.
Background
Before the data mining technology is expanded to battery performance induction analysis mining, battery module data is used as the most effective carrier for battery quantitative analysis, and a large amount of technical information is hidden. The traditional patent data mining has the problems of low efficiency, single dimension, small data sample, insufficient deep level and the like, so that the current requirements on the battery performance data mining cannot be met.
Disclosure of Invention
The battery fault classification diagnosis method based on cluster analysis can deeply mine the association among data, better classify the battery performance data and enable the clustering result to be more integral.
In order to achieve the purpose, the invention adopts the following technical scheme:
a battery fault classification diagnosis method based on cluster analysis comprises the following steps:
s100, collecting and sorting battery data, and taking the power battery pressure difference, the temperature difference, the average current and the average speed as clustering variables;
and S200, clustering by using a K-means method.
Further, in the step S100, the differential pressure, the temperature difference, the average current and the average speed of the power battery are used as clustering variables; the method specifically comprises the following steps:
s101, recording the difference between the maximum value and the minimum value of the monomer voltage in the battery module as a pressure difference, and recording the difference between the maximum value and the minimum value of the monomer temperature as a temperature difference; taking the date as a period of the data recorded according to a certain sampling frequency, and calculating an average value as a clustering variable;
and S102, calculating the average value and the average speed value of the current of the battery in each day of use as a clustering variable.
Further, the S200 performs clustering by using a K-means method, which specifically includes:
s201, selecting K initial central points as clustering centers;
s202, in the Nth iteration, calculating the distance from any sample to K centers, and classifying the sample to the class where the center closest to the sample is located;
s203, recalculating the average value of all the points in each cluster, and updating the average value to a new cluster center;
and S204, repeating the processes of the second step and the third step until the cluster center does not change or is smaller than a given threshold value.
Further, in S201, K initial central points are selected as clustering centers; the method specifically comprises the following steps:
and determining the value of K by adopting an SSE method, wherein a specific algorithm is as follows:
wherein, ciIs the ith cluster, p is ciSample point of (1), miIs ciThe centroid of (1), SSE, is the clustering error of all samples, and represents how good the clustering effect is.
Further, in the nth iteration, the distance from any sample to the K centers is calculated, and the sample is classified into the class where the center closest to the sample is located; the method specifically comprises the following steps:
let DklRepresents GkAnd GlThe sum of squared deviations formula is as follows:
Dki＝Wm-Wk-Wi
in the formula:
Further, the step S202 further includes:
converting the original data into a standard Z score by adopting a standard normal conversion mode, wherein the calculation formula is as follows:
the method adopts the squared Euclidean distance, the original data contains p variables, and each sample is a point in a p-dimensional space;
when two samples are represented by x ═ (x1, x2, …, xp) and y ═ y1, y2, …, yp, the squared euclidean distance between the p variables of the two samples is calculated as follows:
further, the step S100 of collecting and arranging battery data includes data of differences of reaction monomers in the battery module, data of use conditions of the electric vehicle, and data of performance of the entire use of the battery.
According to the technical scheme, the battery fault classification diagnosis method based on the cluster analysis is a battery overall performance cluster analysis method based on the K-means method; according to the method, 4 important evaluation indexes of the power battery pressure difference, the temperature difference, the average current and the average speed are simultaneously selected as clustering variables for clustering analysis. The method can deeply mine the correlation among the battery performance data, better classify the patent data, enable the clustering result to be more integral, and make up the defects of the traditional battery fault analysis method judged by experience.
The method can deeply mine the association among the battery data, better classify various batteries, and enable managers to have higher efficiency and more complete results in information retrieval.
Drawings
FIG. 1 is a flow chart of a method of the present invention;
FIG. 2 is a flow diagram of the present invention;
FIG. 3 is a graph of the cluster analysis of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, 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 some, but not all, embodiments of the present invention.
As shown in fig. 1, the method for classifying and diagnosing battery faults based on cluster analysis in this embodiment includes:
the method comprises the following specific steps:
step1, taking the differential pressure, the temperature difference, the average current and the average vehicle speed of the power battery as clustering variables;
step2, K-means clustering;
the clustering variable calculating method in the Step1 comprises the following steps:
the difference between the maximum value and the minimum value of the monomer voltage recorded in the Step1.1 battery module is pressure difference; the difference between the maximum and minimum recorded monomer temperatures is the temperature difference. And taking the date as a period of the data recorded according to a certain sampling frequency, and averaging to obtain the average value as a clustering variable.
The average value and the average speed value of the current of the Step1.2 battery are calculated and used as clustering variables in each day of use.
The K-means clustering in the Step2 comprises the following specific steps:
step2.1, selecting K initial central points as clustering centers;
determining the value of K by adopting an SSE (sum of the squared errors) method, wherein a specific algorithm is as follows:
wherein, ciIs the ith cluster, p is ciSample point of (1), miIs ciThe centroid of (1), SSE, is the clustering error of all samples, and represents how good the clustering effect is.
Step2.2, in the Nth iteration, calculating the distance from any sample to K centers, and classifying the sample into the class where the center closest to the sample is located;
Dklrepresents GkAnd GlThe sum of squared deviations formula is as follows:
Dki＝Wm-Wk-Wi
in the formula:
Because data have different dimensions and different orders of magnitude, in order to make the data comparable, and make the data more equal to perform clustering analysis, it is necessary to perform standardized transformation on the data. Therefore, the raw data is converted into standard Z scores (Z scores) by adopting a standard normal transformation mode, and the calculation formula is as follows:
the squared euclidean distance is used, and the original data contains p variables, and each sample is a point in the p-dimensional space. By x ═ x1，x2，…，xp) And y ═ y1，y2，…，yp) Representing two samples, the squared euclidean distance between p variables of the two samples is calculated as follows:
step2.3, recalculating the average value of all the points in each cluster, and updating the average value to a new cluster center;
and Step2.4, repeating the processes of the second step and the third step until the cluster center does not change or is smaller than a given threshold value.
Application example 1: monthly running condition analysis of overall performance data of certain batch of batteries of electric vehicles
Patent pretreatment: the data preprocessing results of the change situation of the monitoring data of 100 battery modules of the power batteries of the electric automobile in a certain batch are shown in the following table:
the results of performing the K-means clustering analysis are shown in the following table, which shows only a part of the data because of the large amount of data.
| Battery numbering | Class cluster | Distance |
| 1 | ① | 20.569 |
| 5 | ① | 6.73 |
| 6 | ① | 5.54 |
| 12 | ① | 4.21 |
| 2 | ② | 7.61 |
| 4 | ② | 5.22 |
| 7 | ② | 2.69 |
| 11 | ② | 4.87 |
| … | … | … |
And (4) analyzing results: cluster 1 is the first type and comprises about 23 batteries. The 23 batteries which are gathered into one type are shown to have similar numerical values on 4 indexes of pressure difference, temperature difference, average current and average vehicle speed. Further observation shows that the pressure difference-temperature difference of the 23 batteries are mostly concentrated near the center of (0.15, 1) in two-dimensional coordinates. That is, the data attributed to the first type of battery set is characterized by a mean value of the differential pressure around 0.15, and a mean value of the differential pressure around 1. The control effect of the battery is better. Cluster 2 is a second category comprising about 49 batteries. In this divided second battery set, the center point of the current and voltage difference values is located at (2.8, 0.22). That is, the voltage difference and current values attributed to these batteries are characterized by a mean current value of about 2.8A, while the mean voltage difference is significantly greater than that of the first battery by about 0.22 v. It can be seen that the battery control effect of the second cluster is numerically worse than that of the first cluster. Cluster 3 is the third battery Cluster, which includes about 28 batteries. This type of battery data is characterized by a faster average speed, which means a higher average current. The center position of the average current-average speed was about (3.2, 13.1). The vehicle mounted with the battery is shown to run faster in the period of investigation.
In conclusion, the battery data classification and analysis system based on the k-means clustering method can conveniently and quickly find out the electrical performance data characteristics of the battery in operation, and further can help operation and maintenance personnel to quickly find out the battery problem characteristics, so that the reason can be searched purposefully.
The embodiment of the invention selects more important analysis indexes in the overall performance analysis of the battery: the differential pressure, the temperature difference, the average current and the average speed of the power battery are used as clustering variables, the association among data can be deeply excavated, and the classification of the battery data is better performed, so that the clustering result is more integrated, and the defects of the traditional method for analyzing the battery data by depending on experience are overcome.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (7)
1. A battery fault classification diagnosis method based on cluster analysis is characterized in that: the method comprises the following steps:
s100, collecting and sorting battery data, and taking the power battery pressure difference, the temperature difference, the average current and the average speed as clustering variables;
and S200, carrying out cluster analysis on the battery data by using a K-means method.
2. The cluster analysis-based battery fault classification diagnosis method according to claim 1, characterized in that: the S100 takes the power battery pressure difference, the temperature difference, the average current and the average vehicle speed as clustering variables; the method specifically comprises the following steps:
s101, recording the difference between the maximum value and the minimum value of the monomer voltage in the battery module as a pressure difference, and recording the difference between the maximum value and the minimum value of the monomer temperature as a temperature difference; taking the date as a period of the data recorded according to a certain sampling frequency, and calculating an average value as a clustering variable;
and S102, calculating the average value and the average speed value of the current of the battery in each day of use as a clustering variable.
3. The cluster analysis-based battery fault classification diagnosis method according to claim 1, characterized in that: the S200 utilizes a K-means method to perform cluster analysis on the battery data, and specifically comprises the following steps:
s201, selecting K initial central points as clustering centers;
s202, in the Nth iteration, calculating the distance from any sample to K centers, and classifying the sample to the class where the center closest to the sample is located;
s203, recalculating the average value of all the points in each cluster, and updating the average value to a new cluster center;
and S204, repeating the processes of the second step and the third step until the cluster center does not change or is smaller than a given threshold value.
4. The cluster analysis-based battery fault classification diagnosis method according to claim 3, characterized in that: the S201 selects K initial central points as clustering centers; the method specifically comprises the following steps:
and determining the value of K by adopting an SSE method, wherein a specific algorithm is as follows:
wherein, ciIs the ith cluster, p is ciSample point of (1), miIs ciThe centroid of (1), SSE, is the clustering error of all samples, and represents how good the clustering effect is.
5. The cluster analysis-based battery fault classification diagnosis method according to claim 3, characterized in that:
s202, in the Nth iteration, calculating the distance from any sample to K centers, and classifying the sample to the class where the center closest to the sample is located; the method specifically comprises the following steps:
let DklRepresents GkAnd GlThe sum of squared deviations formula is as follows:
Dki＝Wm-Wk-Wi
in the formula:
6. The cluster analysis-based battery fault classification diagnosis method according to claim 5, characterized in that: the step S202 further includes:
converting the original data into a standard Z score by adopting a standard normal conversion mode, wherein the calculation formula is as follows:
the method adopts the squared Euclidean distance, the original data contains p variables, and each sample is a point in a p-dimensional space;
when two samples are represented by x ═ (x1, x2, …, xp) and y ═ y1, y2, …, yp, the squared euclidean distance between the p variables of the two samples is calculated as follows:
7. the cluster analysis-based battery fault classification diagnosis method according to claim 1, characterized in that: and S100, collecting and arranging battery data, wherein the battery data comprises data of reaction monomer difference in the battery module, service condition data of the electric vehicle and performance data of the whole battery.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911206611.0A CN111126449A (en) | 2019-11-29 | 2019-11-29 | Battery fault classification diagnosis method based on cluster analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911206611.0A CN111126449A (en) | 2019-11-29 | 2019-11-29 | Battery fault classification diagnosis method based on cluster analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111126449A true CN111126449A (en) | 2020-05-08 |
Family
ID=70497165
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911206611.0A Pending CN111126449A (en) | 2019-11-29 | 2019-11-29 | Battery fault classification diagnosis method based on cluster analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111126449A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN114942386A (en) * | 2022-07-20 | 2022-08-26 | 湖北工业大学 | Power battery fault online detection method and system |
| CN115782584A (en) * | 2022-11-22 | 2023-03-14 | 重庆长安新能源汽车科技有限公司 | New energy vehicle safety state determination method, system, equipment and medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102245437A (en) * | 2008-12-10 | 2011-11-16 | 本田技研工业株式会社 | Vehicle failure diagnostic device |
| CN105301508A (en) * | 2015-11-09 | 2016-02-03 | 华晨汽车集团控股有限公司 | Prediction method for electric automobile endurance mileage through redial basis function neural network |
| CN107340475A (en) * | 2016-04-29 | 2017-11-10 | 株式会社日立制作所 | Battery fault detection method and battery fault detection device |
| CN109446319A (en) * | 2018-09-29 | 2019-03-08 | 昆明理工大学 | A kind of biological medicine patent clustering method based on K-means |
-
2019
- 2019-11-29 CN CN201911206611.0A patent/CN111126449A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102245437A (en) * | 2008-12-10 | 2011-11-16 | 本田技研工业株式会社 | Vehicle failure diagnostic device |
| CN105301508A (en) * | 2015-11-09 | 2016-02-03 | 华晨汽车集团控股有限公司 | Prediction method for electric automobile endurance mileage through redial basis function neural network |
| CN107340475A (en) * | 2016-04-29 | 2017-11-10 | 株式会社日立制作所 | Battery fault detection method and battery fault detection device |
| CN109446319A (en) * | 2018-09-29 | 2019-03-08 | 昆明理工大学 | A kind of biological medicine patent clustering method based on K-means |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 |
| CN114942386A (en) * | 2022-07-20 | 2022-08-26 | 湖北工业大学 | Power battery fault online detection method and system |
| CN114942386B (en) * | 2022-07-20 | 2022-11-04 | 湖北工业大学 | Power battery fault online detection method and system |
| CN115782584A (en) * | 2022-11-22 | 2023-03-14 | 重庆长安新能源汽车科技有限公司 | New energy vehicle safety state determination method, system, equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112699913B (en) | Method and device for diagnosing abnormal relationship of household transformer in transformer area | |
| CN111126449A (en) | Battery fault classification diagnosis method based on cluster analysis | |
| CN110738232A (en) | grid voltage out-of-limit cause diagnosis method based on data mining technology | |
| CN111667135B (en) | Load structure analysis method based on typical feature extraction | |
| CN112287980B (en) | Power battery screening method based on typical feature vector | |
| CN113289931A (en) | Lithium ion battery echelon utilization and sorting method | |
| CN110610121B (en) | Small-scale source load power abnormal data identification and restoration method based on curve clustering | |
| CN112860183B (en) | Multisource distillation-migration mechanical fault intelligent diagnosis method based on high-order moment matching | |
| CN116780781B (en) | Power management method for smart grid access | |
| CN110659682A (en) | Data classification method based on MCWD-KSMOTE-AdaBoost-DenseNet algorithm | |
| CN112001441A (en) | Power distribution network line loss anomaly detection method based on Kmeans-AHC hybrid clustering algorithm | |
| CN112101765A (en) | Abnormal data processing method and system for operation index data of power distribution network | |
| CN115291006A (en) | Method for identifying environmental protection working condition abnormity of polluted enterprise based on electricity utilization data | |
| CN116796271A (en) | Resident energy abnormality identification method | |
| CN116338454A (en) | New energy automobile electrical property detection method and device | |
| CN112305441A (en) | Power battery health state assessment method under integrated clustering | |
| Li et al. | A Novel Method for Lithium‐Ion Battery Fault Diagnosis of Electric Vehicle Based on Real‐Time Voltage | |
| CN114646888A (en) | Assessment method and system for capacity attenuation of power battery | |
| CN114611604A (en) | User screening method based on electric drive assembly load characteristic fusion and clustering | |
| CN110929761A (en) | Balance method for collecting samples in situation awareness framework of intelligent system security system | |
| CN108090635B (en) | Road performance prediction method based on cluster classification | |
| CN113469252A (en) | Extra-high voltage converter valve operation state evaluation method considering unbalanced samples | |
| CN117517966A (en) | Fault diagnosis method for battery pack | |
| CN116679221A (en) | Battery abnormality detection and battery maintenance management method and device | |
| CN110163437B (en) | Day-ahead photovoltaic power generation power prediction method based on DPK-means |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200508 |
|
| RJ01 | Rejection of invention patent application after publication |