CN114200327B - Method for estimating single battery SOC in battery pack by considering multi-factor influence - Google Patents
Method for estimating single battery SOC in battery pack by considering multi-factor influence Download PDFInfo
- Publication number
- CN114200327B CN114200327B CN202210055336.2A CN202210055336A CN114200327B CN 114200327 B CN114200327 B CN 114200327B CN 202210055336 A CN202210055336 A CN 202210055336A CN 114200327 B CN114200327 B CN 114200327B
- Authority
- CN
- China
- Prior art keywords
- domain data
- matrix
- source domain
- data
- soc
- 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.)
- Active
Links
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]
- G01R31/385—Arrangements for measuring battery or accumulator variables
- G01R31/387—Determining ampere-hour charge capacity or SoC
-
- 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)
Abstract
A single battery SOC estimation method in a battery pack considering multi-factor influence relates to a battery SOC estimation method. Carrying out experiments on a single battery of the same type in the battery pack to obtain source domain data; and utilizing a migration learning framework to migrate and transform the source domain data and the target domain data, wherein the migration learning framework comprises: feature enhancement, feature compression and MPD adaptation; modeling the transformed source domain data by using a data driving algorithm, and predicting the SOC of the transformed target domain data to obtain an SOC estimated value. Based on a data driving algorithm and adopting a transfer learning method, adverse effects of temperature, battery aging state and other factors on SOC estimation are solved, and a transfer learning framework which is simultaneously adaptive to MPD and CPD is provided for solving the problems of the existing transfer learning method, so that the method is simple and less experimental data are required.
Description
Technical Field
The invention relates to a battery SOC estimation method, in particular to a single battery SOC estimation method in a battery pack considering multi-factor influence, and belongs to the technical field of battery research and development.
Background
The rapid development of the electric automobile effectively relieves the energy crisis and environmental deterioration, and the power source of the electric automobile is a battery pack consisting of a large number of single batteries, wherein the State of Charge (SOC) of each single battery in the battery pack is accurately estimated, the safety of the electric automobile is vital, and accurate information can be provided for designing an equalization strategy and guaranteeing the endurance mileage of the electric automobile.
Due to the different production processes and the differences in the environments to which the battery packs are subjected, there is inevitably an inconsistency between the unit cells. Current methods of estimating the SOC of a unit cell include an equivalent circuit model (Equivalent Circuit Model, ECM) based method, an electrochemical model based method, a data-driven based method, etc., which require a large amount of computing power and memory space if applied to all unit cells in one battery pack. At present, the SOC estimation of each single battery in a battery pack is most commonly used in a method based on an average difference model, but due to the inconsistency of the single batteries, the "average battery" model cannot well represent the whole battery pack, and most importantly, the method is proposed and verified under the conditions of constant temperature, fixed aging state and the like, however, different influencing factors such as temperature, battery aging state and the like exist in practical application, so that the existing estimation method generates larger deviation and even is not applicable any more, for example, different influencing factors can cause the parameter of ECM to change, and in addition, although the data driving method does not need to consider the complex reaction principle inside the battery, the different influencing factors need to mix all data for modeling, which greatly increases the complexity of the model and the storage space of the model. In practice, different influencing factors change the edge probability distribution (Marginal Probability Distribution, MPD) and the conditional probability distribution (Conditional Probability Distribution, CPD) of the data, so that a single data-driven model is no longer applicable to other influencing factors.
The core of the migration learning is to find the similarity between the source domain data and the target domain data, so that the knowledge migration is smoothly realized. However, at present, the example-based migration learning method is effective when the distribution difference between two data domains is small, the feature-based migration learning method assumes that the MPD of the data domains is the same, only adapts to the CPD of the data domains, the model-based migration learning method needs to adjust the super-parameters in the model, and still assumes that the MPD between the data domains is the same, so that it is difficult to obtain a good migration effect in practical application for the above problems. Therefore, there is a need to improve the existing migration learning method and provide an SOC estimation method based on the existing migration learning method, which considers the influence of multiple factors, so as to solve the defects of the existing SOC estimation method.
Disclosure of Invention
In order to solve the defects in the prior art, the invention provides a single battery SOC estimation method in a battery pack considering multi-factor influence, which is based on a data driving algorithm and adopts a transfer learning method to solve adverse effects of temperature, battery aging state and other multi-factor on SOC estimation, and provides a transfer learning framework simultaneously adapting to MPD and CPD aiming at the problems of the existing transfer learning method.
In order to achieve the above purpose, the invention adopts the following technical scheme: a single battery SOC estimation method in a battery pack considering multi-factor influence comprises the following steps:
step one: carrying out experiments on a single battery of the same type in the battery pack to obtain source domain data;
step two: migrating and transforming the obtained source domain data and the target domain data of the single battery by utilizing a migration learning framework;
the migration learning framework specifically comprises:
1) Feature enhancement
In order to obtain an implicit mapping relation between input variables and target variables in source domain data and target domain data, combining the input variables containing MPD information with the target variables containing CPD information to obtain a characteristic enhancement matrix containing CPD and MPD information at the same time, wherein the characteristic enhancement matrix comprises the following steps:
in the method, in the process of the invention,representing an input variable having m n-dimensional features, < >>Representing an output variable having m 1-dimensional features,representing an enhancement matrix having m n+1-dimensional features;
2) Feature compression
Based on principal componentsPrinciple of the partial analysis feature enhancement matrix by transforming matrix WMapping to a new space to explicitly obtain implicit mapping relationship between input variable and output variable in source domain data and target domain data, maximizingIs defined as an optimization problem, namely:
solving the optimization problem by using a Lagrangian function to obtain a transformation matrix W, and converting the source domain data and the target domain data through the transformation matrix W to obtain a characteristic compression matrix as follows:
in which W is s Transform matrix representing source domain data, W t A transformation matrix representing the data of the target domain,enhancement matrix representing source domain data, +.>Enhancement matrix representing target domain data, +.>Compression matrix representing source domain data, +.>A compression matrix representing target domain data;
3) MPD adaptation
Performing MPD adaptation on the transformed source domain data and target domain data by using an existing transfer learning method;
step three: modeling the source domain data transformed in the second step by using a data driving algorithm, and predicting the SOC of the target domain data transformed in the second step, so as to obtain the SOC estimated value of the single battery in the battery pack.
Compared with the prior art, the invention has the beneficial effects that: the invention provides a migration learning framework which is simultaneously adapted to the problems of the existing migration learning method, the migration learning framework is used for effectively transforming the source domain data and the target domain data, the migration learning framework is suitable for SOC estimation under different influencing factors and different working conditions, migration and prediction effects are better, the migrated source domain data are modeled based on a data driving algorithm, the migrated target domain data are predicted to obtain the SOC of the single battery, the complex reaction mechanism in the battery is not required to be considered, adverse influences of influencing factors such as different temperatures and different aging states are not required to be considered, the adaptability is strong, the source domain data are only required to be obtained through a small number of experiments under any environment of single batteries of the same type in a battery pack, the method is not required to be proposed and verified under the conditions of constant temperature and fixed aging states and the like, compared with the existing method, the method is simple and efficient, and has good application prospect, and resources and the complexity and occupied resource space of the model are greatly reduced.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a graph showing comparison of SOC estimation results of various methods at different temperatures in the examples;
FIG. 3 is a graph showing comparison of SOC estimation results of various methods under different aging conditions in the examples;
FIG. 4 is a graph showing comparison of SOC estimation results of various methods at different temperatures and aging conditions in the examples.
Detailed Description
The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the invention, but not all embodiments, and all other embodiments obtained by those skilled in the art without making creative efforts based on the embodiments of the present invention are all within the protection scope of the present invention.
Referring to the flowchart shown in fig. 1, a method for estimating SOC of a unit cell in a battery pack in consideration of multi-factor influence includes the steps of:
step one: carrying out experiments on a single battery of the same type in a battery pack to obtain source domain data, wherein the source domain data comprises accurate SOC target variable values;
step two: migrating and transforming the obtained source domain data and the target domain data of the single battery by utilizing a migration learning framework;
the migration learning framework specifically comprises:
1) Feature enhancement
In order to obtain an implicit mapping relation between input variables and target variables in source domain data and target domain data, combining the input variables containing MPD information with the target variables containing CPD information to obtain a characteristic enhancement matrix containing CPD and MPD information at the same time, wherein the characteristic enhancement matrix comprises the following steps:
in the method, in the process of the invention,representing an input variable having m n-dimensional features, < >>Representing an output variable having m 1-dimensional features,representing an enhancement matrix having m n+1-dimensional features;
2) Feature compression
In the process of obtaining the characteristic enhancement matrixThen, based on principle of principal component analysis, the feature enhancement matrix is +.>Mapping to a new space to explicitly obtain implicit mapping relation between input variable and output variable in source domain data and target domain data, and maximizing (or maximizing) most of information of source domain data and target domain data in the new space>And thus can be defined as an optimization problem, namely:
solving the optimization problem by using Lagrangian function, namely:
by deriving W, it is possible to obtain:
wherein W represents a compound represented byMatrix of eigenvectors after eigenvalue decomposition, - λ represents a matrix of +.>A matrix of eigenvalues after feature decomposition,
after the transformation matrix W is obtained, the source domain data and the target domain data are converted through the transformation matrix W to obtain a characteristic compression matrix as follows:
in which W is s Transform matrix representing source domain data, W t A transformation matrix representing the data of the target domain,enhancement matrix representing source domain data, +.>Enhancement matrix representing target domain data, +.>Compression matrix representing source domain data, +.>A compression matrix representing target domain data;
3) MPD adaptation
After obtaining a feature compression matrix with an implicit mapping relation between input variables and output variables in source domain data and target domain data, performing MPD (multi-domain data storage) adaptation on the transformed source domain data and target domain data by using an existing migration learning method, such as migration component analysis (Transfer Component Analysis, TCA), geodesic flow core (Geodesic Flow Kernel, GFK), maximum independent domain adaptation (Maximum Independent Domain Adaptation, MIDA) and the like;
step three: through the migration learning framework of the second step, firstly, characteristic enhancement is carried out on input variables (temperature, voltage and current) of source domain data and target variables represented by SOC calculated by an ampere-hour integration method, wherein the target variables of the source domain data are accurate SOC values, then, after characteristic compression and MPD adaptation are carried out on a characteristic enhancement matrix, a data driving algorithm is used for modeling the source domain data transformed in the second step, and then, SOC prediction is carried out on the target domain data transformed in the second step, so that an SOC estimated value of a single battery in a battery pack is obtained.
Examples
The embodiment performs verification based on practical battery pack experimental data, wherein the number of single batteries in the battery pack is 100, the rated capacity of the battery pack is 96Ah, the charge and discharge cut-off voltages are 420v and 250v respectively, the rated capacity of the single batteries is 3.2Ah, the battery type is loose NCR18650BD, and the anode and cathode materials are Li (NiCoMn) O respectively 2 And graphite, experiments including capacity test experiments, pulse test experiments, and NEDC condition test experiments at different temperatures (-10 ℃,15 ℃, and 45 ℃) and different aging states (100% soh (State of Health) and 95% soh).
The data-driven algorithm based on long-short-term memory (Long Short Term Memory, LSTM) neural network, the geodesic streaming kernel (Geodesic Flow Kernel, GFK) and maximum independent domain adaptation (Maximum Independent Domain Adaptation, MIDA) based MPD adaptation methods, experimental verification at different temperatures and/or different aging conditions, and the mean absolute error percentage (Mean Absolute Percentage Error, MAPE) and standard root mean square error (Normalized Root Mean Square Error, NRMSE) were used to evaluate the performance between the ampere-hour integration method (Ah), LSTM method, GFK-based LSTM (gfk+lstm) method, GFK-based LSTM (igfk+lstm) method, MIDA-based LSTM (mida+lstm) method, and MIDA (imida+lstm) method.
Referring to fig. 2, SOC estimation results of each method using-10 ℃ cell 1 (100% soh) experimental data as source domain data and 15 ℃ cell 1 (100% soh) experimental data as target domain data are shown in table 1 below:
TABLE 1 SOC estimation results for each method at different temperatures
Referring to fig. 3, the SOC estimation results of the respective methods for the 15 ℃ cell 10 (100% soh) experimental data as the source domain data and the 15 ℃ cell 10 (95% soh) experimental data as the target domain data are shown in table 2 below:
TABLE 2 SOC estimation results for each method under different aging conditions
Referring to fig. 4, SOC estimation results of each method using-10 ℃ cell 10 (100% soh) experimental data as source domain data and 45 ℃ cell 80 (95% soh) experimental data as target domain data are shown in table 3 below:
TABLE 3 SOC estimation results for each method at different temperatures and different aging conditions
The SOC estimation method provided by the invention firstly obtains source domain data through experiments of single batteries of the same type in a battery pack under any environment, and then uses a proposed migration learning framework to migrate and convert the source domain data and target domain data of the single batteries in the battery pack, wherein the migration learning framework comprises the following components: and finally modeling the migrated source domain data by using a data driving algorithm, and predicting the migrated target domain data to obtain the SOC estimation result of the single battery. As can be seen from fig. 2-4 and tables 1-3, the present invention can obtain accurate SOC estimation results under the condition of considering the influence of multiple factors, and the migration effect of the proposed migration learning framework is far better than that of the existing migration learning method.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Furthermore, it should be understood that although the present disclosure describes embodiments, not every embodiment is provided with a separate embodiment, and that this description is provided for clarity only, and that the disclosure is not limited to the embodiments described in detail below, and that the embodiments described in the examples may be combined as appropriate to form other embodiments that will be apparent to those skilled in the art.
Claims (3)
1. A single battery SOC estimation method in a battery pack considering multi-factor influence is characterized in that: the method comprises the following steps:
step one: carrying out experiments on a single battery of the same type in the battery pack to obtain source domain data;
step two: migrating and transforming the obtained source domain data and the target domain data of the single battery by utilizing a migration learning framework;
the migration learning framework specifically comprises:
1) Feature enhancement
In order to obtain an implicit mapping relation between input variables and target variables in source domain data and target domain data, combining the input variables containing MPD information with the target variables containing CPD information to obtain a characteristic enhancement matrix containing CPD and MPD information at the same time, wherein the characteristic enhancement matrix comprises the following steps:
in the method, in the process of the invention,representing an input variable having m n-dimensional features, < >>Representing an output variable having m 1-dimensional features, < >>Representing an enhancement matrix having m n+1-dimensional features;
2) Feature compression
Feature enhancement matrix by transforming matrix W based on principle of principal component analysisMapping to a new space to explicitly obtain implicit mapping relationship between input variable and output variable in source domain data and target domain data, maximizingIs defined as an optimization problem, namely:
solving the optimization problem by using a Lagrangian function to obtain a transformation matrix W, and converting the source domain data and the target domain data through the transformation matrix W to obtain a characteristic compression matrix as follows:
in which W is s Transform matrix representing source domain data, W t A transformation matrix representing the data of the target domain,enhancement matrix representing source domain data, +.>Enhancement matrix representing target domain data, +.>Compression matrix representing source domain data, +.>A compression matrix representing target domain data;
3) MPD adaptation
Performing MPD adaptation on the transformed source domain data and target domain data by using an existing transfer learning method;
step three: modeling the source domain data transformed in the second step by using a data driving algorithm, and predicting the SOC of the target domain data transformed in the second step, so as to obtain the SOC estimated value of the single battery in the battery pack.
2. The method for estimating SOC of a battery cell in a battery pack taking into account multi-factor effects as set forth in claim 1, wherein: in the feature compression of the second step, solving the optimization problem by using the lagrangian function to obtain a transformation matrix W specifically includes:
by deriving W, it is possible to obtain:
wherein W represents a compound represented byMatrix of eigenvectors after eigenvalue decomposition, - λ represents a matrix of +.>And a matrix composed of the characteristic values after characteristic decomposition.
3. The method for estimating SOC of a battery cell in a battery pack taking into account multi-factor effects as set forth in claim 1, wherein: in the MPD adaptation in the second step, the existing migration learning method adopts migration component analysis TCA, geodesic flow core GFK or maximum independent domain adaptation MIDA.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210055336.2A CN114200327B (en) | 2022-01-18 | 2022-01-18 | Method for estimating single battery SOC in battery pack by considering multi-factor influence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210055336.2A CN114200327B (en) | 2022-01-18 | 2022-01-18 | Method for estimating single battery SOC in battery pack by considering multi-factor influence |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114200327A CN114200327A (en) | 2022-03-18 |
CN114200327B true CN114200327B (en) | 2023-07-28 |
Family
ID=80658619
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210055336.2A Active CN114200327B (en) | 2022-01-18 | 2022-01-18 | Method for estimating single battery SOC in battery pack by considering multi-factor influence |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114200327B (en) |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102608542A (en) * | 2012-04-10 | 2012-07-25 | 吉林大学 | Method for estimating charge state of power cell |
CN105425154A (en) * | 2015-11-02 | 2016-03-23 | 北京理工大学 | Method for estimating charge state of power cell set of electric vehicle |
US10180460B1 (en) * | 2012-04-20 | 2019-01-15 | Motiv Power Systems, Inc. | Performing active interrogation of battery packs in situ to obtain precise SOC and SOH estimates |
CN109917223A (en) * | 2019-03-08 | 2019-06-21 | 广西电网有限责任公司电力科学研究院 | A kind of transmission line malfunction current traveling wave feature extracting method |
CN110391401A (en) * | 2019-06-25 | 2019-10-29 | 肇庆理士电源技术有限公司 | OMC based combined electrode and lead-acid battery |
CN112540298A (en) * | 2020-12-26 | 2021-03-23 | 福州大学 | Lithium ion battery SOC estimation method based on deep-migration learning |
CN113128672A (en) * | 2021-04-16 | 2021-07-16 | 昆明理工大学 | Lithium ion battery pack SOH estimation method based on transfer learning algorithm |
CN113420444A (en) * | 2021-06-23 | 2021-09-21 | 南通大学 | Lithium ion battery SOC estimation method based on parameter online identification |
CN113740736A (en) * | 2021-08-31 | 2021-12-03 | 哈尔滨工业大学 | Electric vehicle lithium battery SOH estimation method based on deep network self-adaptation |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8922217B2 (en) * | 2012-05-08 | 2014-12-30 | GM Global Technology Operations LLC | Battery state-of-charge observer |
US9077182B2 (en) * | 2013-01-29 | 2015-07-07 | Mitsubishi Electric Research Laboratories, Inc. | Method for estimating state of charge for lithium-ion batteries |
US9834112B2 (en) * | 2016-03-09 | 2017-12-05 | Ford Global Technologies, Llc | Battery state of charge estimation based on reduced order electrochemical models |
-
2022
- 2022-01-18 CN CN202210055336.2A patent/CN114200327B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102608542A (en) * | 2012-04-10 | 2012-07-25 | 吉林大学 | Method for estimating charge state of power cell |
US10180460B1 (en) * | 2012-04-20 | 2019-01-15 | Motiv Power Systems, Inc. | Performing active interrogation of battery packs in situ to obtain precise SOC and SOH estimates |
CN105425154A (en) * | 2015-11-02 | 2016-03-23 | 北京理工大学 | Method for estimating charge state of power cell set of electric vehicle |
CN109917223A (en) * | 2019-03-08 | 2019-06-21 | 广西电网有限责任公司电力科学研究院 | A kind of transmission line malfunction current traveling wave feature extracting method |
CN110391401A (en) * | 2019-06-25 | 2019-10-29 | 肇庆理士电源技术有限公司 | OMC based combined electrode and lead-acid battery |
CN112540298A (en) * | 2020-12-26 | 2021-03-23 | 福州大学 | Lithium ion battery SOC estimation method based on deep-migration learning |
CN113128672A (en) * | 2021-04-16 | 2021-07-16 | 昆明理工大学 | Lithium ion battery pack SOH estimation method based on transfer learning algorithm |
CN113420444A (en) * | 2021-06-23 | 2021-09-21 | 南通大学 | Lithium ion battery SOC estimation method based on parameter online identification |
CN113740736A (en) * | 2021-08-31 | 2021-12-03 | 哈尔滨工业大学 | Electric vehicle lithium battery SOH estimation method based on deep network self-adaptation |
Non-Patent Citations (3)
Title |
---|
"State-of-Charge Estimation of Lithium Battery Based on Transfer Learning";Zhenzhen Zhang;《2021 International Conference on Artificial Intelligence, Big Data and Algorithms (CAIBDA)》;全文 * |
"基于梯度提升决策树的动力锂离子电池SOH预测";秦鹏亮;《中国优秀硕士学位论文全文数据库》;全文 * |
"基于贝叶斯优化和迁移学习的电池荷电状态估计方法研究";张珍珍;《中国优秀硕士学位论文全文数据库》;全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN114200327A (en) | 2022-03-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP7095110B2 (en) | Battery status estimation method | |
Sheng et al. | Small sample state of health estimation based on weighted Gaussian process regression | |
Che et al. | Semi-supervised self-learning-based lifetime prediction for batteries | |
Farag et al. | Continuous piecewise-linear, reduced-order electrochemical model for lithium-ion batteries in real-time applications | |
CN113128672B (en) | Lithium ion battery pack SOH estimation method based on transfer learning algorithm | |
WO2023030024A1 (en) | Electrochemical model-based method and system for estimating state of solid-state lithium battery | |
CN111985156A (en) | Method for predicting health state of battery | |
CN113640674B (en) | Lithium ion battery available capacity estimation method based on optimized Gaussian process regression | |
Wang et al. | Battery full life cycle management and health prognosis based on cloud service and broad learning | |
CN111914485B (en) | Adaptive power battery SOC estimation method and system based on fractional order technology | |
Liu et al. | Review on degradation mechanism and health state estimation methods of lithium-ion batteries | |
Fortenbacher et al. | Battery degradation maps for power system optimization and as a benchmark reference | |
CN114839537A (en) | Battery state of charge prediction method and device | |
CN114200327B (en) | Method for estimating single battery SOC in battery pack by considering multi-factor influence | |
CN110850315A (en) | Method and device for estimating state of charge of battery | |
Wang et al. | A novel capacity estimation method for li-ion battery cell by applying ensemble learning to extremely sparse significant points | |
CN112327165A (en) | Battery SOH prediction method based on unsupervised transfer learning | |
CN116819372A (en) | Method for calculating health state of energy storage battery by neural network | |
CN113702836B (en) | Lithium ion battery state of charge estimation method based on EMD-GRU | |
CN116774045A (en) | Lithium battery health state prediction method based on HHO-SVR | |
Upashrutti et al. | Estimation of State of Charge of EV Batteries-A Machine Learning Approach | |
CN112285565B (en) | Method for predicting SOH (State of health) of battery by transfer learning based on RKHS (remote keyless entry) domain matching | |
Pannala et al. | Consistently Tuned Battery Lifetime Predictive Model of Capacity Loss, Resistance Increase, and Irreversible Thickness Growth | |
Channegowda et al. | An attention‐based synthetic battery data augmentation technique to overcome limited dataset challenges | |
Forman et al. | Reduction of an electrochemistry-based Li-ion battery health degradation model via constraint linearization and Padé approximation |
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 |