CN115436814A - Probability prediction method for residual life of lithium ion battery - Google Patents
Probability prediction method for residual life of lithium ion battery Download PDFInfo
- Publication number
- CN115436814A CN115436814A CN202211268274.XA CN202211268274A CN115436814A CN 115436814 A CN115436814 A CN 115436814A CN 202211268274 A CN202211268274 A CN 202211268274A CN 115436814 A CN115436814 A CN 115436814A
- Authority
- CN
- China
- Prior art keywords
- lithium ion
- probability
- ion battery
- prediction
- battery
- 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 56
- HBBGRARXTFLTSG-UHFFFAOYSA-N Lithium ion Chemical compound [Li+] HBBGRARXTFLTSG-UHFFFAOYSA-N 0.000 title claims abstract description 35
- 229910001416 lithium ion Inorganic materials 0.000 title claims abstract description 35
- 238000012360 testing method Methods 0.000 claims abstract description 37
- 238000007637 random forest analysis Methods 0.000 claims abstract description 17
- 238000012549 training Methods 0.000 claims abstract description 16
- 230000032683 aging Effects 0.000 claims abstract description 10
- 238000007781 pre-processing Methods 0.000 claims abstract description 5
- 238000005070 sampling Methods 0.000 claims description 11
- 238000007619 statistical method Methods 0.000 claims description 5
- 238000010187 selection method Methods 0.000 claims description 3
- 108090000672 Annexin A5 Proteins 0.000 claims 1
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000012423 maintenance Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000005562 fading Methods 0.000 description 2
- 239000008358 core component Substances 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000000178 monomer Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/392—Determining battery ageing or deterioration, e.g. state of health
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Tests Of Electric Status Of Batteries (AREA)
Abstract
The invention discloses a probability prediction method for the residual life of a lithium ion battery, which relates to the technical field of lithium ion batteries and comprises the following steps: acquiring aging test data of the lithium ion battery, and preprocessing the test data; extracting the characteristics of the preprocessed data, and selecting the characteristics by utilizing a pinballloss loss function; selecting a model hyper-parameter by using a method of minimizing a pinballloss loss function or minimizing a coverage probability deviation index, establishing a quantile regression random forest model QRRF based on the model hyper-parameter, and training the model; and inputting the test data into the battery residual life probability prediction model, outputting a target quantile prediction value of the battery residual life, and obtaining a probability prediction interval of the battery residual life.
Description
Technical Field
The invention relates to the technical field of lithium ion batteries, in particular to a probability prediction method for the residual life of a lithium ion battery.
Background
The safety of a power battery serving as one of the core components of a pure electric vehicle is subject to the users' problems. Lithium ion batteries are mostly used as power batteries of electric vehicles, and in the research on safety of the batteries, remaining Useful Life (RUL) is one of the important points of research. RUL refers to the number of cycles that the battery undergoes during the charge and discharge cycles before the battery capacity declines to 80% of the rated capacity.
Most of research on the RUL adopts a deterministic method to predict the expected value, namely, a deterministic value is used to predict the remaining useful life of the lithium ion battery through an algorithm, but because of the measurement error of historical data and the influence of factors which are not sensed/predicted in environment and working condition, the actual remaining life of the battery is necessarily deviated from the predicted value, the action characteristics of uncertainty on an individual cannot be fully embodied only through evaluation indexes of deterministic estimation, such as root mean square error, average absolute percentage error and the like, so that users and vehicle production enterprises cannot effectively utilize the deterministic method, definite battery maintenance or replacement suggestions are assigned, and the safety of the battery in the using process is improved. Some researches consider introducing a probability prediction model into the residual effective life prediction, but in order to simplify the algorithm, the probability distribution of the residual effective life is directly assumed to be normal distribution, which is not consistent with the actual situation, and a large number of experiments show that the probability distribution of the residual effective life of the battery is asymmetric distribution.
The service life prediction method mainly comprises a model method and a data driving method. The model method relates to the physical and chemical reaction in the battery, the work load of parameter identification is large, and the consideration to the actual use environment is difficult, so the practical application is less. The data-driven model predicts the service life of the lithium ion battery mostly based on the expected value of the service life of the battery at the current focus point, namely, the deterministic prediction is influenced by uncertainty factors, and the actual service life of the battery is obviously different from the prediction, so that the application of the residual effective service life prediction result is limited to a great extent. For example, the patent "a method for estimating lifetime of lithium ion battery based on relaxation time distribution" provides a method for estimating lifetime of lithium ion battery that correlates a relaxation time distribution function of ac impedance with lifetime, but this method can only obtain a deterministic estimation result, and cannot realize probabilistic predictive prediction. There are also several patents that propose methods for predicting the remaining useful life of a battery, such as: the patent 'a method for predicting the remaining life probability of a lithium ion battery based on a gray model' provides a lithium ion battery life probability estimation method combining the gray model with a Markov chain and a related vector machine, but historical data of the method only uses a battery capacity fading sequence as a feature, and the accuracy of probability prediction is influenced due to the shortage of input features of the battery capacity fading sequence in consideration of the high complexity of uncertainty influence factors of a battery attenuation process. In fact, probabilistic predictions have become a trend in recent technological advances, such as: the patent 'wind power probability prediction method of fractal point regression forest and variable bandwidth evaluation' provides a wind power probability prediction method of fractal point regression forest and variable bandwidth evaluation, but the method is only applied to wind power prediction, the wind power correlation of different seasons, months and dates is weak, and in the practical application of the remaining effective life of a battery, the attenuation process historical data of the battery has strong correlation to a prediction target, how to consider the correlation among the data of the whole life cycle of the battery, and how to select reasonable data characteristics is an unsolved key problem in the technology of predicting the remaining effective life of a lithium ion battery.
Disclosure of Invention
Aiming at the problems, the invention provides a probability prediction method based on Quantile Regression Random Forest (QRRF for short), based on any life cycle historical data (more than 100 cycles) of the battery, the interval distribution of the RUL under different confidence probabilities when the battery runs to the appointed cycle number is predicted, more accurate and comprehensive information is provided for the maintenance or replacement of the battery, and the method has important significance for improving the use safety of the battery of the electric vehicle and improving the use experience of users.
The data driving method utilized by the invention can directly apply actual data, thereby better considering the reaction mechanism of the model which is not modeled and the imperceptible influence in the actual application, weakening the theoretical interpretability, but being more beneficial to the efficient engineering practice application.
The invention provides a probability prediction method for the residual life of a lithium ion battery, which specifically comprises the following steps:
acquiring aging test data of the lithium ion battery;
determining an alternative feature set consisting of the following features and statistical analysis thereof: the method comprises the following steps of (1) obtaining a discharge capacity difference value delta Q under corresponding voltage of two cycles, an internal resistance difference value of the two cycles, a polynomial fitting parameter of a curve of discharge voltage changing along with the number of cycles, and a discharge capacity of a specific cycle;
selecting a characteristic subset from the candidate characteristic set by using a pinball loss function, inputting the characteristic subset into a quantile regression random forest model, outputting pinball loss of quantile prediction results, and determining the characteristic subset which enables pinball loss to be minimum as a finally selected characteristic set;
selecting a hyperparameter of a quantile regression random forest model by utilizing a minimized pinball loss function or a minimized coverage probability deviation index method, inputting a finally selected characteristic set into the quantile regression random forest model, changing the hyperparameter to minimize pinball loss or CPDI of a quantile prediction result output by the model, thereby determining the optimal model hyperparameter, and constructing a battery residual life probability prediction model based on the quantile regression random forest;
inputting the test data into a battery residual life probability prediction model, outputting a target quantile prediction value of the battery residual life, obtaining a probability prediction interval of the battery residual life, and expressing a probability prediction result by using a confidence interval which is symmetrical about a median.
Further, the acquiring aging test data of the lithium ion battery specifically includes:
and acquiring voltage, capacity, internal resistance and temperature data of each test cycle of the lithium ion battery, wherein the voltage, the discharge capacity and the internal resistance are necessary data, the charging capacity and the temperature are optional data, and the sampling frequency is f.
Further, after the aging test data of the lithium ion battery is acquired, the test data needs to be preprocessed, and the method specifically includes:
for all aging test data, the data with the division ratio p is a training set, and the rest data are test sets, wherein p is more than 0 and less than 1.
Further, the preprocessing the test data further includes:
sampling is carried out for c cycles at sampling intervals, and oversampling of test data is achieved.
Further, the statistical analysis quantity comprises a mean value, a maximum value, a variance, a skewness and a kurtosis;
the total number of features in the candidate feature set is denoted as k.
Further, the selecting a feature subset from the candidate feature set specifically includes:
selecting a feature subset comprising m to n features from the candidate feature set, wherein 1 < m < n < k, to obtain
A subset of features.
Further, the pinball loss function is defined as
Wherein q is the quantile,
is the prediction of the qth quantile of y.
Further, the constructing of the prediction model of the battery remaining life probability based on the quantile regression random forest specifically includes:
and establishing a battery residual life probability prediction model based on quantile regression random forest QRRF by setting proper hyper-parameters such as the number of trees in the forest and the maximum depth of the trees.
Furthermore, the selection method of the hyper-parameters comprises a minimized pinball loss function method or a minimized coverage probability deviation index method;
wherein, the coverage probability deviation index is defined as: CPDI = | PICP-Confidence Level | non-ventilatedIn the formula, PICP is the proportion of the observed value of the target variable within the prediction interval:
wherein N is the number of samples, U i ,L i Is the upper and lower limits of the prediction interval, y, for sample i i Is the target variable observation for sample i.
Compared with the prior art, the method for predicting the probability of the residual life of the lithium ion battery has the beneficial effects that:
the method uses pinball loss function to select the characteristics, uses quantile regression random forest to obtain quantile predicted value of the remaining effective life of the battery, and estimates the remaining effective life interval of the battery under different confidence probabilities based on the quantile predicted value, thereby not only effectively utilizing data, but also obtaining the characteristic subset more suitable for actual data; and because the model has a simple structure, all target quantiles can be obtained by only one training (compared with a common quantile regression algorithm), and finally, accurate battery residual life intervals under different confidence probabilities can be obtained.
Drawings
Fig. 1 is a flowchart of a method for predicting the remaining life of a lithium ion battery according to the present invention;
FIG. 2 is a schematic diagram showing the result of probability prediction of the battery 1 in example 2 of the present invention;
FIG. 3 is a diagram illustrating a probability prediction result of the battery 2 according to embodiment 2 of the present invention;
fig. 4 is a schematic diagram of a probability prediction result of the battery 3 in embodiment 2 of the present invention.
Detailed Description
The following describes the present invention with reference to fig. 1 to 4. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.
Example 1: as shown in fig. 1, the present invention provides a method for predicting the remaining life of a lithium ion battery with a probability, which includes the following steps:
1. performing aging test on the lithium ion battery, and acquiring data such as voltage, capacity, internal resistance, temperature and the like of each test cycle, wherein the voltage, the discharge capacity and the internal resistance are necessary data, the charge capacity and the temperature are optional data, and the sampling frequency is f;
2. preprocessing test data: (1) dividing data with the proportion of p into a training set, and data with the proportion of (1-p) into a test set, wherein p is more than 0 and less than 1, and (2) sampling by taking a sampling interval as c cycles to realize oversampling of the test data;
3. and (3) performing feature extraction and selection in the preprocessed data:
(1) extracting features, and determining an alternative feature set consisting of the following features and statistical analysis quantities (such as mean, maximum, variance, skewness, kurtosis and the like) thereof: the discharge capacity difference value delta Q (V) under the voltage corresponding to the two cycles, the internal resistance difference value of the two cycles, polynomial fitting parameters of a curve of the discharge voltage changing along with the number of cycles, the discharge capacity of the specific cycle, and the total characteristic number in the characteristic set are marked as k;
(2) selecting features by using pinball loss function, selecting a feature subset containing m to n features in the alternative feature set in the step, wherein 1 < m < n < k, obtaining the total
The characteristic subsets sequentially input data contained in the characteristic subsets into a QRRF model, and respectively calculate pinball loss of quantile prediction results output by the model, so that the characteristic subset with the minimum pinball loss is determined as a finally selected characteristic set to serve as the input of the model;
pinball loss function is defined as
Wherein q is the quantile,
is the prediction of the qth quantile of y;
4. establishing a battery remaining life probability prediction model based on Quantile Regression Random Forest (QRRF), and completing modeling by setting appropriate model hyperparameters such as the number of trees in the forest, the maximum depth of the trees and the like;
the selection method of the hyper-parameter is two, (1) minimizing pinball loss function, or (2) minimizing Coverage Probability Deviation Index (CPDI): presetting a value set of model hyper-parameters, inputting the selected feature set in the step 3(2) into the model, and changing the hyper-parameters to minimize pinball loss or CPDI of quantile prediction results output by the model, thereby determining the optimal model hyper-parameters and completing modeling;
the definition of CPDI is
CPDI＝|PICP-Confidence Level|#(2)
Wherein, PICP (prediction interval coverage probability) is the proportion of the observed value of the target variable in the prediction interval:
wherein N is the number of samples, U i ,L i Is the upper and lower limits of the prediction interval for sample i, y i Is the target variable observation for sample i.
5. Training a model and predicting RUL, inputting a training set into a QRRF model for training, inputting a test set into the model after training is finished, and outputting a target quantile predicted value of the RUL, so that a probability prediction interval of the RUL is obtained, a probability prediction result is represented by a confidence interval which is symmetrical about a median, and the prediction result is compared with a label value of the RUL in the test set to evaluate the accuracy of the model.
Example 2: as shown in fig. 2-4, the feasibility and effectiveness of the method of the present invention were verified using the battery test data set disclosed in the ministry of technology MIT of Majordomo (MIT) comprising 124 cells in total for a total of about 96700 test cycles, according to the procedure of example 1, and the detailed information of the data set is shown in fig. 1https:// www.nature.com/articles/s41560-019-0356-8。
Data preprocessing: in step 2, take p =0.75, and put the battery in the data set as 3:1 into a training set and a test set, and sampling each monomer by taking c =100 cycles as sampling intervals, so as to realize oversampling of the data set; thus, the training set contains 707 cycles of data for 83 cells in total, and the test set contains 242 cycles of data for 21 cells in total.
Feature extraction and selection: based on the teaching in step 3, a total of 12 features having strong correlation with the battery life are extracted, i.e., k =12; taking m =5,n =7, traversing the obtained feature subset, and finally determining 6 features which minimize pinball loss function as the input of the model.
Establishing a model: and (3) determining the hyper-parameters of the model by minimizing a pinball loss function based on the method (1) in the step 4, and completing modeling.
Model training and prediction: in this example, q =1,2, … … and 98,99 are taken, training is performed by inputting training set data into a model, after training is completed, prediction is performed by inputting a test set into the model, and the prediction value is compared with a label value in the test set to evaluate the accuracy of the model. In this example, the probabilistic predictors are represented by 90% confidence intervals, which are intervals determined by the 5 th quantile and the 95 th quantile.
Schematic diagrams of the probability prediction results for three cells in the test set are shown in fig. 2-4. It can be seen that the distribution of RUL is asymmetric and does not fit the normal distribution assumption. The significance of the probability interval predicted by the invention is that the deterministic prediction result of a certain battery RUL is 500 cycles, the error is 100 cycles, but because the RUL distribution is asymmetric, the user is difficult to arrange the battery maintenance according to the information, and the prediction interval with 90% confidence coefficient provided by the invention is 480-580 cycles, and the user can know that the maintenance is more stable when 450 cycles are carried out according to the information.
In summary, the invention uses pinball loss function to select features, uses quantile regression random forest to obtain quantile predicted value of the remaining effective service life of the battery, and estimates the remaining effective service life interval of the battery under different confidence probabilities based on the quantile predicted value, thereby not only effectively utilizing data, but also obtaining a feature subset more suitable for actual data; and because the model has a simple structure, all target quantiles can be obtained by only one training (compared with a common quantile regression algorithm), and finally, accurate battery residual life intervals under different confidence probabilities can be obtained.
The above-mentioned embodiments are only preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, and any simple modifications or equivalent substitutions of the technical solutions that can be obviously obtained by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention.
Claims (9)
1. A probability prediction method for the residual life of a lithium ion battery is characterized by comprising the following steps:
acquiring aging test data of the lithium ion battery;
determining an alternative feature set consisting of the following features and statistical analysis thereof: the discharge capacity difference value delta Q under the corresponding voltage of the two cycles, the internal resistance difference value of the two cycles, polynomial fitting parameters of a curve of the discharge voltage changing along with the number of cycles, and the discharge capacity of the specific cycle;
selecting a characteristic subset from the candidate characteristic set by using a pinball loss function, inputting the characteristic subset into a quantile regression random forest model, outputting pinball loss of quantile prediction results, and determining the characteristic subset which enables pinball loss to be minimum as a finally selected characteristic set;
selecting a hyperparameter of a quantile regression random forest model by utilizing a minimized pinball loss function or a minimized coverage probability deviation index method, inputting a finally selected characteristic set into the quantile regression random forest model, changing the hyperparameter to minimize pinball loss or CPDI of a quantile prediction result output by the model, thereby determining the optimal model hyperparameter, and constructing a battery residual life probability prediction model based on the quantile regression random forest;
inputting the test data into a battery residual life probability prediction model, outputting a target quantile prediction value of the battery residual life, obtaining a probability prediction interval of the battery residual life, and expressing a probability prediction result by using a confidence interval which is symmetrical about a median.
2. The method for predicting the probability of the remaining life of the lithium ion battery according to claim 1, wherein the obtaining aging test data of the lithium ion battery specifically comprises:
and acquiring voltage, capacity, internal resistance and temperature data of each test cycle of the lithium ion battery, wherein the voltage, the discharge capacity and the internal resistance are necessary data, the charging capacity and the temperature are optional data, and the sampling frequency is f.
3. The method for predicting the probability of the remaining life of the lithium ion battery according to claim 1, wherein after the aging test data of the lithium ion battery is obtained, the test data needs to be preprocessed, and specifically the method comprises:
for all aging test data, the data with the division ratio p is a training set, and the rest data are test sets, wherein p is more than 0 and less than 1.
4. The method according to claim 3, wherein the preprocessing the test data further comprises:
sampling is carried out for c cycles at sampling intervals, and oversampling of test data is achieved.
5. The method according to claim 1, wherein the method for predicting the probability of the remaining life of the lithium ion battery comprises:
the statistical analysis quantity comprises a mean value, a maximum value, a variance, a skewness and a kurtosis;
the total number of features in the candidate feature set is denoted as k.
6. The method according to claim 5, wherein the selecting the feature subset from the candidate feature set specifically includes:
selecting a feature subset comprising m to n features from the candidate feature set, wherein 1 < m < n < k, to obtain
A subset of features.
7. The method according to claim 1, wherein the method for predicting the lithium ion battery residual life probability comprises the following steps:
the definition of the pinball loss function is
Wherein q is the quantile,
is the prediction of the qth quantile of y.
8. The method for predicting the probability of the remaining life of the lithium ion battery according to claim 1, wherein the step of constructing a probability prediction model of the remaining life of the battery based on a quantile regression random forest specifically comprises the following steps:
and establishing a battery residual life probability prediction model based on quantile regression random forest QRRF by setting proper hyper-parameters such as the number of trees in the forest and the maximum depth of the trees.
9. The method according to claim 8, wherein the method for predicting the remaining life of the lithium ion battery comprises:
the selection method of the over-parameters comprises a minimized pinball loss function method or a minimized coverage probability deviation index method;
wherein, the coverage probability deviation index is defined as: CPDI = | PICP-Confidence Level |, where PICP is the proportion of the observed value of the target variable falling within the prediction interval:
wherein N is the number of samples, U i ,L i Is the upper and lower limits of the prediction interval for sample i, y i Is the target variable observation for sample i.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211268274.XA CN115436814A (en) | 2022-10-17 | 2022-10-17 | Probability prediction method for residual life of lithium ion battery |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211268274.XA CN115436814A (en) | 2022-10-17 | 2022-10-17 | Probability prediction method for residual life of lithium ion battery |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115436814A true CN115436814A (en) | 2022-12-06 |
Family
ID=84251300
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211268274.XA Pending CN115436814A (en) | 2022-10-17 | 2022-10-17 | Probability prediction method for residual life of lithium ion battery |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115436814A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115639480A (en) * | 2022-12-21 | 2023-01-24 | 中创新航科技股份有限公司 | Method and device for detecting health state of battery |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106291372A (en) * | 2016-07-22 | 2017-01-04 | 南京工业大学 | Novel method for predicting residual life of lithium ion power battery |
| CN108846517A (en) * | 2018-06-12 | 2018-11-20 | 清华大学 | A kind of probability short-term electric load prediction integrated approach of quantile |
| CN109558975A (en) * | 2018-11-21 | 2019-04-02 | 清华大学 | A kind of integrated approach of a variety of prediction results of electric load probability density |
| CN110619360A (en) * | 2019-09-09 | 2019-12-27 | 国家电网有限公司 | Ultra-short-term wind power prediction method considering historical sample similarity |
| CN112215393A (en) * | 2020-08-29 | 2021-01-12 | 复旦大学 | Rainfall numerical prediction post-processing correction method based on adaptive space-time scale selection |
| CN114545274A (en) * | 2022-01-26 | 2022-05-27 | 湖州学院 | Lithium battery residual life prediction method |
-
2022
- 2022-10-17 CN CN202211268274.XA patent/CN115436814A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106291372A (en) * | 2016-07-22 | 2017-01-04 | 南京工业大学 | Novel method for predicting residual life of lithium ion power battery |
| CN108846517A (en) * | 2018-06-12 | 2018-11-20 | 清华大学 | A kind of probability short-term electric load prediction integrated approach of quantile |
| CN109558975A (en) * | 2018-11-21 | 2019-04-02 | 清华大学 | A kind of integrated approach of a variety of prediction results of electric load probability density |
| CN110619360A (en) * | 2019-09-09 | 2019-12-27 | 国家电网有限公司 | Ultra-short-term wind power prediction method considering historical sample similarity |
| CN112215393A (en) * | 2020-08-29 | 2021-01-12 | 复旦大学 | Rainfall numerical prediction post-processing correction method based on adaptive space-time scale selection |
| CN114545274A (en) * | 2022-01-26 | 2022-05-27 | 湖州学院 | Lithium battery residual life prediction method |
Non-Patent Citations (1)
| Title |
|---|
| 黄星知，等: "基于小波分解技术和随机森林算法的负荷概率预测", 《电力与能源》, vol. 42, no. 3, 30 June 2021 (2021-06-30), pages 280 - 286 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115639480A (en) * | 2022-12-21 | 2023-01-24 | 中创新航科技股份有限公司 | Method and device for detecting health state of battery |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112731159B (en) | Method for pre-judging and positioning battery faults of battery compartment of energy storage power station | |
| CN109061504B (en) | Method and system for predicting remaining service life of lithium ion batteries of same type and different types | |
| CN114114049B (en) | Lithium ion battery life prediction method based on sample migration | |
| CN111220921A (en) | Lithium battery capacity estimation method based on improved convolution-long-and-short-term memory neural network | |
| CN113917334B (en) | Battery health state estimation method based on evolution LSTM self-encoder | |
| CN110082682B (en) | Lithium battery state of charge estimation method | |
| CN102749589A (en) | Recession-mode predicting method of power battery of electric automobile | |
| CN111999648A (en) | Lithium battery residual life prediction method based on long-term and short-term memory network | |
| CN108009585B (en) | Lead-acid battery health state prediction method based on local information fusion | |
| US20230305073A1 (en) | Method and apparatus for providing a predicted aging state of a device battery based on a predicted usage pattern | |
| CN104680024A (en) | Method for predicting remaining useful life of lithium ion battery based on GA (Genetic Algorithms) and ARMA (Auto Regressive and Moving Average) models | |
| CN115308628A (en) | Battery health state monitoring method | |
| CN114462259B (en) | SOC-based battery remaining life prediction method and system, automobile and medium | |
| CN114035098A (en) | Lithium battery health state prediction method integrating future working condition information and historical state information | |
| CN115436814A (en) | Probability prediction method for residual life of lithium ion battery | |
| CN118169582B (en) | Lithium ion battery health state and residual life prediction method | |
| CN113176518A (en) | Lithium battery health state estimation method | |
| CN116381545A (en) | Method and device for determining consistency grade of energy storage battery pack | |
| CN115994441A (en) | Big data cloud platform online battery life prediction method based on mechanism information | |
| CN115792680A (en) | Retired battery health state assessment method based on Bayes deep learning | |
| CN115201692A (en) | Method for analyzing and predicting residual life of lithium battery | |
| US11733313B2 (en) | Method and apparatus for operating a system for providing states of health of electrical energy stores for a multiplicity of devices with the aid of machine learning methods | |
| CN112327169B (en) | Lithium battery residual life prediction method | |
| CN116653608B (en) | Electric automobile charging protection and control method, device and storage medium | |
| CN116908727A (en) | Multi-feature combination-based energy storage power station lithium battery capacity estimation method |
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 |