CN110826645A - Adaboost algorithm-based lithium battery retirement detection method and system - Google Patents
Adaboost algorithm-based lithium battery retirement detection method and system Download PDFInfo
- Publication number
- CN110826645A CN110826645A CN201911155104.9A CN201911155104A CN110826645A CN 110826645 A CN110826645 A CN 110826645A CN 201911155104 A CN201911155104 A CN 201911155104A CN 110826645 A CN110826645 A CN 110826645A
- Authority
- CN
- China
- Prior art keywords
- retirement
- parameter information
- samples
- battery
- battery parameter
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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]
- G01R31/392—Determining battery ageing or deterioration, e.g. state of health
-
- 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]
- G01R31/396—Acquisition or processing of data for testing or for monitoring individual cells or groups of cells within a battery
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Abstract
The invention relates to the technical field of battery management systems, aims to solve the problem that the mode of judging whether a lithium battery reaches a decommissioning standard in the prior art is complex, and provides a lithium battery decommissioning detection method based on an Adaboost algorithm, which comprises the following steps: selecting a plurality of samples from known battery parameter information, wherein the samples at least comprise samples which do not reach a retirement standard and samples which reach the retirement standard, and training by adopting an Adaboost algorithm to obtain a strong classifier for lithium battery retirement detection based on the battery parameter information in the samples and the retirement standard information corresponding to the battery parameter information; the method comprises the steps of collecting battery parameter information of a lithium battery to be tested, inputting the battery parameter information into a strong classifier, and distinguishing the input battery parameter information by the strong classifier so as to judge whether the lithium battery to be tested reaches a retirement standard. The invention can judge whether the lithium battery in the using process reaches the retirement standard without calculating the SOH value, and is more convenient, quicker and more accurate.
Description
Technical Field
The invention relates to the technical field of battery management systems, in particular to a lithium battery retirement detection method and system.
Background
With the increasing prominence of energy crisis and environmental pollution problems, power machines mainly using clean energy and electric power are more favored. Lithium ion batteries are widely used as new energy mechanical power sources and energy storage devices due to their characteristics of large specific energy, high working voltage, long cycle life and the like. However, abuse, misuse and severe working conditions of the lithium battery can lead to gradual battery failure, and continuous use of the failed battery can bring serious consequences to a machine using the lithium battery and can cause greater damages such as explosion, spontaneous combustion and the like. The State Of Health (SOH) Of the lithium battery is accurately estimated, so that the use safety Of the new energy power source can be improved, and the safe and reliable operation Of the new energy power source is guaranteed.
The SOH is the percentage of the full charge capacity of the battery relative to the rated capacity, the new battery is 100 percent, the complete scrappage is 0 percent, and according to the IEEE standard, when the SOH value of the battery is reduced to 80 percent, the battery needs to be replaced, namely the battery is judged to be out of service. Since the battery does not have laboratory conditions in the use process, the SOH value of the battery cannot be directly measured, and whether the battery reaches the retirement standard cannot be judged, methods for estimating the SOH value in real time based on a kalman filter method, an artificial neural network method and the like are provided in the prior art, and then the retirement state of the battery is detected, but the methods are complex.
Disclosure of Invention
The invention aims to solve the problem that the mode of judging whether a lithium battery reaches a decommissioning standard in the prior art is complex, and provides a lithium battery decommissioning detection method and system based on an Adaboost algorithm.
The technical scheme adopted by the invention for solving the technical problems is as follows: the lithium battery retirement detection method based on the Adaboost algorithm comprises the following steps:
step 1, selecting a plurality of samples from known battery parameter information, wherein the samples at least comprise samples which do not reach a retirement standard and samples which reach the retirement standard, and training by adopting an Adaboost algorithm based on the battery parameter information in the samples and the retirement standard information corresponding to the battery parameter information to obtain a strong classifier for lithium battery retirement detection;
and 2, acquiring battery parameter information of the lithium battery to be tested, inputting the battery parameter information into the strong classifier, and distinguishing the input battery parameter information by the strong classifier so as to judge whether the lithium battery to be tested reaches the retirement standard.
As a further optimization, in step 1, the selecting a plurality of samples from the known battery parameter information includes:
respectively selecting a certain number of samples which do not reach the retirement standard and samples which reach the retirement standard from known battery parameter information, forming a multidimensional data vector by using the battery parameter information of the samples which do not reach the retirement standard as a positive sample of Adaboost algorithm training, and forming a multidimensional data vector by using the battery parameter information of the samples which reach the retirement standard as a negative sample of the Adaboost algorithm training.
As a further optimization, in step 1, the training by using the Adaboost algorithm based on the samples to obtain the strong classifier includes:
step 11, initializing the sample weights of the positive sample and the negative sample;
step 12, the number of times of loop iteration is specified, and each iteration is completed, wherein the step of each iteration comprises the following steps:
step 121, carrying out normalization processing on the sample weight to make each iteration weight obey probability distribution;
step 122, training a weak classifier corresponding to the multi-dimensional data feature vector;
step 123, selecting the classifier with the minimum classification error relative to each sample as an optimal weak classifier;
step 124, judging whether the specified iteration times are reached, if so, entering step 13, otherwise, updating the sample weight, and returning to step 121;
and 13, combining the optimal weak classifiers obtained by each iteration to obtain a strong classifier.
As a further optimization, the battery parameter information includes: battery current, load voltage and cycle number, the multi-dimensional data vector is a three-dimensional data vector.
For further optimization, the number of the samples which do not reach the retirement standard and the number of the samples which reach the retirement standard are respectively 100.
As a further optimization, the step 2 further comprises:
and (3) acquiring the accuracy of the strong classifier for distinguishing the battery parameter information, if the accuracy is greater than a preset value, entering the step (2), and otherwise, increasing the iteration times and continuing training the strong classifier.
As a further optimization, the obtaining the accuracy of the strong classifier for distinguishing the battery parameter information includes:
and inputting the battery parameter information of the test samples into the strong classifier, verifying the output result of the strong classifier, and counting the accuracy of distinguishing the battery parameter information by the strong classifier according to a certain number of test samples.
On the other hand, the invention also provides a lithium battery retirement detection system based on the Adaboost algorithm, which comprises the following steps:
the training unit is used for selecting a plurality of samples from known battery parameter information, wherein the samples at least comprise samples which do not reach the retirement standard and samples which reach the retirement standard, and training by adopting an Adaboost algorithm based on the battery parameter information in the samples and the retirement standard information corresponding to the battery parameter information to obtain a strong classifier for the retirement detection of the lithium battery;
and the acquisition unit is used for acquiring the battery parameter information of the lithium battery to be tested and inputting the battery parameter information into the strong classifier, and the strong classifier distinguishes the input battery parameter information so as to judge whether the lithium battery to be tested reaches the retirement standard.
As a further optimization, the training unit is further configured to:
respectively selecting a certain number of samples which do not reach the retirement standard and samples which reach the retirement standard from known battery parameter information, forming a multidimensional data vector by using the battery parameter information of the samples which do not reach the retirement standard as a positive sample of Adaboost algorithm training, and forming a multidimensional data vector by using the battery parameter information of the samples which reach the retirement standard as a negative sample of the Adaboost algorithm training.
As a further optimization, the battery parameter information includes: battery current, load voltage and cycle number, the multi-dimensional data vector is a three-dimensional data vector.
The invention has the beneficial effects that: according to the lithium battery decommissioning detection method and system based on the Adaboost algorithm, the Adaboost algorithm is adopted for training to obtain the strong classifier for lithium battery decommissioning detection, the strong classifier is used for distinguishing the input battery parameter information of the lithium battery, and the detection whether the lithium battery reaches the decommissioning standard is completed.
Drawings
Fig. 1 is a schematic flow chart of a lithium battery retirement detection method based on the Adaboost algorithm according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a lithium battery retirement detection system based on the Adaboost algorithm according to an embodiment of the present invention.
Detailed Description
Embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
The lithium battery retirement detection method based on the Adaboost algorithm, disclosed by the embodiment of the invention, as shown in FIG. 1, comprises the following steps:
s1, selecting a plurality of samples from known battery parameter information, wherein the samples at least comprise samples which do not reach a retirement standard and samples which reach the retirement standard, and training by adopting an Adaboost algorithm to obtain a strong classifier for lithium battery retirement detection based on the battery parameter information in the samples and the retirement standard information corresponding to the battery parameter information;
and S2, collecting battery parameter information of the lithium battery to be tested, inputting the battery parameter information into the strong classifier, and distinguishing the input battery parameter information by the strong classifier so as to judge whether the lithium battery to be tested reaches the retirement standard.
Firstly, selecting a plurality of samples from known battery parameter information, namely selecting a group of lithium batteries which do not reach the retirement standard, such as 100 lithium batteries, respectively obtaining the battery parameter information of the group of lithium batteries, then selecting a group of lithium batteries which reach the retirement standard, such as 100 lithium batteries, respectively obtaining the battery parameter information of the group of lithium batteries, then training based on the battery parameter information corresponding to the lithium batteries which do not reach the retirement standard and the battery parameter information corresponding to the lithium batteries which reach the retirement standard to obtain a strong classifier, and finally distinguishing the input parameter information of the lithium batteries to be tested according to the strong classifier, and further judging whether the lithium batteries to be tested reach the retirement standard.
Wherein the battery parameter information may include: the battery current, the load voltage and the cycle number, and the battery parameter information can be collected when the system runs and the collected data is stored. The retirement standard information comprises that the retirement standard is not met and the retirement standard is met, a three-dimensional data vector is formed by battery current, load voltage and cycle frequency corresponding to the lithium battery which does not meet the retirement standard and serves as a positive sample of Adaboost algorithm training, a three-dimensional data vector is formed by battery current, load voltage and cycle frequency corresponding to the lithium battery which meets the retirement standard and serves as a negative sample of the Adaboost algorithm training, and the positive sample and the negative sample form a training sample group (x)1,y1),(x2,y2),...(xn,yn) Wherein x is a three-dimensional data vector, y { -1, +1}, and yi+1 represents a positive sample, yi-1 represents negative samples and n represents the total number of samples.
The method for obtaining the strong classifier by adopting Adaboost algorithm training based on the sample comprises the following specific steps:
step 11, initializing sample weights of the positive sample and the negative sample, wherein the weight W is 1/n;
step 12, the number T of times of loop iteration is specified, and each iteration is completed, wherein the step of each iteration comprises the following steps:
(1) the sample weight is normalized and processed,making each iteration weight obey a probability distribution;
(2) training a weak classifier corresponding to the feature vector of each sample, wherein the error of the classifier is as follows:
(3) selecting the classifier with the least classification error with respect to each sample as the optimal weak classifier ht；
(4) Judging whether the specified iteration times T are reached, if so, entering the step 13, otherwise, according to a formula wt+1＝wi,texp(-αtyiht(xi) Update sample weights and return to step (1) where αt＝ln((1-εt)/εt)。
And 13, combining the optimal weak classifiers obtained by each iteration to generate a strong classifier:
in addition, the iteration times T of the Adaboost algorithm can be set according to the precision requirement of the product, and the detection precision is higher as the iteration times are more.
Optionally, in order to improve the accuracy of the decommissioning judgment of the lithium battery, the method further comprises a step of verifying the strong classifier:
and obtaining the accuracy of the strong classifier for distinguishing the battery parameter information, if the accuracy is greater than a preset value, indicating that the strong classifier passes the verification, and entering the step S2, otherwise, increasing the iteration number T and continuing to train the strong classifier. Specifically, the battery parameter information of the test sample can be input into the strong classifier, the output result of the strong classifier is verified, and the accuracy of distinguishing the battery parameter information by the strong classifier is counted according to a certain number of test samples.
And finally, the strong classifier codes obtained by training can be embedded into a Battery Management System (BMS), namely, the lithium battery system in operation can be judged, and when the strong classifier judges that the lithium battery reaches the retirement standard, a user is prompted to replace the battery.
Based on the above technical solution, the present invention further provides a lithium battery retirement detection system based on the Adaboost algorithm, as shown in fig. 2, including:
the training unit is used for selecting a plurality of samples from known battery parameter information, wherein the samples at least comprise samples which do not reach the retirement standard and samples which reach the retirement standard, and training by adopting an Adaboost algorithm based on the battery parameter information in the samples and the retirement standard information corresponding to the battery parameter information to obtain a strong classifier for the retirement detection of the lithium battery;
and the acquisition unit is used for acquiring the battery parameter information of the lithium battery to be tested and inputting the battery parameter information into the strong classifier, and the strong classifier distinguishes the input battery parameter information so as to judge whether the lithium battery to be tested reaches the retirement standard.
Optionally, the training unit is further configured to: respectively selecting a certain number of samples which do not reach the retirement standard and samples which reach the retirement standard from known battery parameter information, forming a multidimensional data vector by using the battery parameter information of the samples which do not reach the retirement standard as a positive sample of Adaboost algorithm training, and forming a multidimensional data vector by using the battery parameter information of the samples which reach the retirement standard as a negative sample of the Adaboost algorithm training.
Optionally, the battery parameter information includes: battery current, load voltage and cycle number, the multi-dimensional data vector is a three-dimensional data vector.
Because the lithium battery decommissioning detection system based on the Adaboost algorithm is a system for realizing the lithium battery decommissioning detection method based on the Adaboost algorithm, for the disclosed system, the description is simpler as the system corresponds to the disclosed method, and the relevant points can be referred to the partial description of the method.
Claims (10)
1. The lithium battery retirement detection method based on the Adaboost algorithm is characterized by comprising the following steps of:
step 1, selecting a plurality of samples from known battery parameter information, wherein the samples at least comprise samples which do not reach a retirement standard and samples which reach the retirement standard, and training by adopting an Adaboost algorithm based on the battery parameter information in the samples and the retirement standard information corresponding to the battery parameter information to obtain a strong classifier for lithium battery retirement detection;
and 2, acquiring battery parameter information of the lithium battery to be tested, inputting the battery parameter information into the strong classifier, and distinguishing the input battery parameter information by the strong classifier so as to judge whether the lithium battery to be tested reaches the retirement standard.
2. The lithium battery retirement detection method based on Adaboost algorithm as claimed in claim 1, wherein in step 1, the selecting a plurality of samples from known battery parameter information comprises:
respectively selecting a certain number of samples which do not reach the retirement standard and samples which reach the retirement standard from known battery parameter information, forming a multidimensional data vector by using the battery parameter information of the samples which do not reach the retirement standard as a positive sample of Adaboost algorithm training, and forming a multidimensional data vector by using the battery parameter information of the samples which reach the retirement standard as a negative sample of the Adaboost algorithm training.
3. The lithium battery retirement detection method based on Adaboost algorithm as claimed in claim 2, wherein in step 1, the obtaining of the strong classifier by training with Adaboost algorithm based on the sample comprises:
step 11, initializing the sample weights of the positive sample and the negative sample;
step 12, the number of times of loop iteration is specified, and each iteration is completed, wherein the step of each iteration comprises the following steps:
step 121, carrying out normalization processing on the sample weight to make each iteration weight obey probability distribution;
step 122, training a weak classifier corresponding to the multi-dimensional data feature vector;
step 123, selecting the classifier with the minimum classification error relative to each sample as an optimal weak classifier;
step 124, judging whether the specified iteration times are reached, if so, entering step 13, otherwise, updating the sample weight, and returning to step 121;
and 13, combining the optimal weak classifiers obtained by each iteration to obtain a strong classifier.
4. The Adaboost algorithm-based lithium battery retirement detection method of claim 2, wherein the battery parameter information comprises: battery current, load voltage and cycle number, the multi-dimensional data vector is a three-dimensional data vector.
5. The Adaboost algorithm-based lithium battery retirement detection method according to claim 2, wherein the number of samples not meeting the retirement standard and the number of samples meeting the retirement standard are 100 respectively.
6. The lithium battery retirement detection method based on Adaboost algorithm as claimed in claim 1, wherein step 2 further comprises, before:
and (3) acquiring the accuracy of the strong classifier for distinguishing the battery parameter information, if the accuracy is greater than a preset value, entering the step (2), and otherwise, increasing the iteration times and continuing training the strong classifier.
7. The lithium battery retirement detection method based on the Adaboost algorithm as claimed in claim 6, wherein the obtaining of the accuracy rate of the strong classifier for distinguishing the battery parameter information comprises:
and inputting the battery parameter information of the test samples into the strong classifier, verifying the output result of the strong classifier, and counting the accuracy of distinguishing the battery parameter information by the strong classifier according to a certain number of test samples.
8. Lithium battery retirement detection system based on Adaboost algorithm is characterized by comprising:
the training unit is used for selecting a plurality of samples from known battery parameter information, wherein the samples at least comprise samples which do not reach the retirement standard and samples which reach the retirement standard, and training by adopting an Adaboost algorithm based on the battery parameter information in the samples and the retirement standard information corresponding to the battery parameter information to obtain a strong classifier for the retirement detection of the lithium battery;
and the acquisition unit is used for acquiring the battery parameter information of the lithium battery to be tested and inputting the battery parameter information into the strong classifier, and the strong classifier distinguishes the input battery parameter information so as to judge whether the lithium battery to be tested reaches the retirement standard.
9. The Adaboost algorithm-based lithium battery decommissioning detection system of claim 8, wherein the training unit is further configured to:
respectively selecting a certain number of samples which do not reach the retirement standard and samples which reach the retirement standard from known battery parameter information, forming a multidimensional data vector by using the battery parameter information of the samples which do not reach the retirement standard as a positive sample of Adaboost algorithm training, and forming a multidimensional data vector by using the battery parameter information of the samples which reach the retirement standard as a negative sample of the Adaboost algorithm training.
10. The Adaboost algorithm-based lithium battery decommissioning detection system of claim 9, wherein the battery parameter information comprises: battery current, load voltage and cycle number, the multi-dimensional data vector is a three-dimensional data vector.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911155104.9A CN110826645A (en) | 2019-11-22 | 2019-11-22 | Adaboost algorithm-based lithium battery retirement detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911155104.9A CN110826645A (en) | 2019-11-22 | 2019-11-22 | Adaboost algorithm-based lithium battery retirement detection method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110826645A true CN110826645A (en) | 2020-02-21 |
Family
ID=69558413
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911155104.9A Pending CN110826645A (en) | 2019-11-22 | 2019-11-22 | Adaboost algorithm-based lithium battery retirement detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110826645A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115330275A (en) * | 2022-10-13 | 2022-11-11 | 深圳市杰成镍钴新能源科技有限公司 | Echelon utilization method and device for retired battery |
| CN116429339A (en) * | 2023-06-07 | 2023-07-14 | 海斯坦普汽车组件(北京)有限公司 | Leakage detection method and system for new energy battery box |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190212391A1 (en) * | 2016-06-30 | 2019-07-11 | Robert Bosch Gmbh | Method for monitoring a battery |
| CN110058178A (en) * | 2019-05-14 | 2019-07-26 | 郑州大学 | A kind of lithium battery method for detecting health status and system |
| CN110146817A (en) * | 2019-05-13 | 2019-08-20 | 上海博强微电子有限公司 | The diagnostic method of lithium battery failure |
| CN110441706A (en) * | 2019-08-23 | 2019-11-12 | 优必爱信息技术（北京）有限公司 | A kind of battery SOH estimation method and equipment |
-
2019
- 2019-11-22 CN CN201911155104.9A patent/CN110826645A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190212391A1 (en) * | 2016-06-30 | 2019-07-11 | Robert Bosch Gmbh | Method for monitoring a battery |
| CN110146817A (en) * | 2019-05-13 | 2019-08-20 | 上海博强微电子有限公司 | The diagnostic method of lithium battery failure |
| CN110058178A (en) * | 2019-05-14 | 2019-07-26 | 郑州大学 | A kind of lithium battery method for detecting health status and system |
| CN110441706A (en) * | 2019-08-23 | 2019-11-12 | 优必爱信息技术（北京）有限公司 | A kind of battery SOH estimation method and equipment |
Non-Patent Citations (3)
| Title |
|---|
| 天泽28: "AdaBoost算法（一）——基础知识篇", 《HTTPS://BLOG.CSDN.NET/U012328159/ARTICLE/DETAILS/89847193》 * |
| 杨丽: "基于模型驱动的锂离子电池剩余寿命预测方法研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
| 王玉斐: "基于模型的锂离子电池剩余寿命预测方法", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115330275A (en) * | 2022-10-13 | 2022-11-11 | 深圳市杰成镍钴新能源科技有限公司 | Echelon utilization method and device for retired battery |
| CN115330275B (en) * | 2022-10-13 | 2023-04-07 | 深圳市杰成镍钴新能源科技有限公司 | Echelon utilization method and device for retired battery |
| CN116429339A (en) * | 2023-06-07 | 2023-07-14 | 海斯坦普汽车组件(北京)有限公司 | Leakage detection method and system for new energy battery box |
| CN116429339B (en) * | 2023-06-07 | 2023-09-01 | 海斯坦普汽车组件(北京)有限公司 | Leakage detection method and system for new energy battery box |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110068774B (en) | Lithium battery health state estimation method and device and storage medium | |
| CN110752410B (en) | Method for rapidly sorting and recombining retired lithium battery | |
| CN109165687B (en) | Vehicle lithium battery fault diagnosis method based on multi-classification support vector machine algorithm | |
| US10209314B2 (en) | Systems and methods for estimation and prediction of battery health and performance | |
| CN107422266B (en) | Fault diagnosis method and device for high-capacity battery energy storage system | |
| CN111584952A (en) | Method and system for online evaluation of electrochemical cells of energy storage power station | |
| CN108931725B (en) | Method and device for detecting battery fault | |
| CN116502112A (en) | New energy power supply test data management method and system | |
| CN110826645A (en) | Adaboost algorithm-based lithium battery retirement detection method and system | |
| CN113655385A (en) | Lithium battery SOC estimation method and device and computer readable storage medium | |
| CN111064253A (en) | Battery health degree rapid evaluation method based on average discrete Frechet distance | |
| CN112305441B (en) | Power battery health state assessment method under integrated clustering | |
| Zhou et al. | A study of online state-of-health estimation method for in-use electric vehicles based on charge data | |
| Chen et al. | An approach for state of charge estimation of Li-ion battery based on Thevenin equivalent circuit model | |
| Qaisar et al. | Machine Learning Assistive State of Charge Estimation of Li-Ion Battery | |
| Shuai et al. | Intelligent diagnosis of abnormal charging for electric bicycles based on improved dynamic time warping | |
| CN116224127A (en) | Battery state of health estimation method based on big data analysis | |
| Han et al. | State of Charge estimation of Li-ion battery in EVs based on second-order sliding mode observer | |
| CN114280490B (en) | Lithium ion battery state of charge estimation method and system | |
| CN115600143A (en) | Energy storage lithium battery fault diagnosis method and device | |
| Cui et al. | An Online State of Health Estimation Method for Lithium-Ion Battery Based on ICA and TPA-LSTM | |
| CN115980607A (en) | Battery state of health (SOH) estimation method based on incremental capacity curve characteristic points | |
| CN112611972A (en) | Method for estimating SOC (state of charge) of lithium battery under condition of low-frequency sampling data | |
| CN114200310B (en) | Lithium precipitation detection method for electrochemical device and electronic equipment | |
| Chen et al. | Sorting Strategy for Retired Batteries Based on Voltage Segments |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200221 |