CN115684971A - Lithium ion battery health state estimation method based on fragment multi-charging feature fusion - Google Patents
Lithium ion battery health state estimation method based on fragment multi-charging feature fusion Download PDFInfo
- Publication number
- CN115684971A CN115684971A CN202211276570.4A CN202211276570A CN115684971A CN 115684971 A CN115684971 A CN 115684971A CN 202211276570 A CN202211276570 A CN 202211276570A CN 115684971 A CN115684971 A CN 115684971A
- Authority
- CN
- China
- Prior art keywords
- data
- voltage
- health state
- lithium ion
- ion 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
Images
Landscapes
- Secondary Cells (AREA)
Abstract
The invention relates to a lithium ion battery health state estimation method based on fragment multi-charging feature fusion. The method comprises the following steps: acquiring charging voltage data, corresponding time data and maximum discharge capacity data in the charging and discharging cycle of the lithium ion battery; respectively extracting equal voltage difference time data, equal time difference voltage data and battery health state data aiming at the voltage and time data of each cycle; processing the extracted isoelectric voltage difference time data and the extracted isoelectric voltage difference voltage data by adopting a typical correlation analysis method, extracting fusion characteristic data, forming a lithium ion battery health state data set with battery health state data, and dividing the lithium ion battery health state data set into a training set and a test set; establishing a long-short term memory cyclic neural network model; and training the model by using a training set, adjusting the parameters of the model, and testing the model by using a test set to estimate the precision. The method improves the estimation precision of the health state of the lithium ion battery, reduces the influence of data acquisition errors and the complexity of a model, and accelerates the estimation speed.
Description
Technical Field
The invention relates to the technical field of lithium ion batteries, in particular to a lithium ion battery health state estimation method based on fragment multi-charging feature fusion.
Background
Lithium ion batteries are widely used in various fields, such as electric vehicles, electrochemical energy storage, aerospace, and the like. However, in the practical application of lithium ion batteries, the performance of the batteries is degraded continuously, and the practically usable capacity is reduced continuously. Therefore, there is a need for a continuous estimation of the state of health of a battery during battery operation.
The current estimation method for the state of health of the lithium ion battery mainly comprises a model-based method and a data driving method. The model-based method needs to construct a battery equivalent circuit model or an electrochemical model, realize model parameter identification through a least square method and the like, and estimate the battery health state by using a Kalman filtering method, a particle filtering method and the like. The estimation accuracy of such methods depends on the accuracy of the model and its parameter identification. The data driving method does not require the establishment of a battery equivalent circuit model. The existing data-driven lithium ion battery health state estimation method based on segment charging data generally extracts an equal voltage difference time characteristic or an equal time difference voltage characteristic as an input of a deep learning model, and then estimates the lithium ion battery health state. However, the time characteristic or the voltage characteristic is extracted independently, the obtained characteristics are single, and the characteristics have large similarity, so that model overfitting is caused, and the estimation accuracy of the health state is influenced.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a lithium ion battery health state estimation method based on fragment multi-charging feature fusion.
In order to achieve the purpose, the technical scheme of the invention is as follows: a lithium ion battery health state estimation method based on fragment multi-charging feature fusion comprises the following steps:
step 3, processing the isoelectrical pressure difference time data and the isoelectrical time difference voltage data extracted in the step 2 by adopting a typical correlation analysis method, extracting fusion characteristic data, combining the fusion characteristic data with corresponding circulating battery health state data to form a lithium ion battery health state data set, and dividing the data set into a training set and a testing set;
In an embodiment of the present invention, the step 1 includes the following sub-steps:
step 101, performing charge and discharge circulation on a newly-shipped lithium ion battery by using a battery charge and discharge tester, wherein the charge mode is constant-current constant-voltage charge, the discharge mode is constant-current discharge, and the charge and discharge circulation test is ended until the maximum discharge capacity of the lithium ion battery is reduced to 70% of the rated capacity, and the total cycle number is defined as D;
and 102, recording charging voltage data and corresponding time data in the constant-current charging process of each cycle, and recording the maximum discharging capacity data of the current cycle.
In an embodiment of the present invention, the step 2 includes the following sub-steps:
step 201, when the isoelectric voltage difference time data is extracted, determining the initial voltage V of the segment charging voltage for each cycle 0 End voltage V n Extracting the initial voltage V 0 End voltage V n Corresponding time T 0 And
and determining a sampling voltage interval Deltav of
Wherein n is the characteristic quantity of the isoelectric pressure difference time data;
from a starting voltage V 0 At the beginning, for each additional sampling voltage interval Δ v, the voltage value and the corresponding time are recorded, i.e. for
Obtaining a sequence of voltage data
And corresponding time series
Calculating the time difference corresponding to the equal voltage interval in the charging process according to the time sequence, namely
Thereby extracting the isoelectric pressure difference time data characteristics of single charging circulation
Step 202, when the voltage data with equal time difference is extracted, determining the initial voltage V of the segment charging voltage for each cycle 0 And its corresponding time T 0 And determining the sampling time interval Deltat, rootDetermining the termination time according to the number n of features in step 201
From a starting voltage V 0 At the beginning, for each additional sampling time interval Δ t, the time and the corresponding voltage value are recorded, i.e. for
Obtaining corresponding time series
And voltage data sequence
Calculating the voltage difference corresponding to the equal time interval in the charging process according to the voltage data sequence, namely
Thereby extracting the equal time difference voltage data characteristics of a single charging cycle
Step 203, extracting the health state data of the battery, aiming at each circulation, and according to the maximum discharge capacity C of each circulation max And rated capacity C of battery 0 And calculating the health state of the lithium ion battery in each cycle as follows:
in an embodiment of the present invention, the step 3 includes the following sub-steps:
step 301, 2n features extracted in step 2, namely isoelectric pressure difference time data features
Sum equal time difference voltage data characterization
Performing dimensionality reduction on the feature data by adopting a typical correlation analysis method, and constructing a multivariate random variable Z = (X, Y), wherein X is a feature matrix, the dimensionality is (2n, D), 2n is the feature quantity, D is the cycle number, Y is the health state data of the lithium ion battery, and the dimensionality is (1, D);
step 302, construct linear transformation P = a T X and Q = b T Y, carrying out standardization processing on the original data to obtain a standardized matrix Z of the multivariate random variable Z = (X, Y) * And calculating a normalized matrix Z * The covariance matrix of (a) yields:
wherein S is XX 、S XY 、S YX 、S YY Calculating a result for the covariance;
to maximize the correlation between P and Q, the vectors a and b need to be chosen to maximize the objective function, i.e.:
the Rayleigh entropy matrix is defined as:
definition u j And λ j The jth eigenvector and square root of eigenvalues for R, solving the maximization problem, representing vectors a and b as:
from vectors a and b, P = a is calculated T X and Q = b T Y, to obtain the final meltA sum-feature vector FF with dimension D, i.e.:
FF＝XP+YQ
step 303, constructing a lithium ion battery health state data set according to the fusion characteristics and the lithium ion battery health state data:
n Data samples in the Data set Data are used as a training set Data1, and the rest Data samples are used as a testing set Data2.
In an embodiment of the present invention, in the step 4, the number of input nodes of the long-short term memory recurrent neural network model is 1, the number of output nodes of the long-short term memory recurrent neural network model is 1, and the optimization algorithm is Adam.
In an embodiment of the present invention, the step 5 includes the following sub-steps:
step 501, performing normalization processing on the fusion characteristic Data and the health state Data in the training set Data1, taking the normalized fusion characteristic Data as the input of the long-short term memory cyclic neural network model, taking the normalized health state Data as the output of the long-short term memory cyclic neural network model, and training the model;
step 502, after normalization processing is carried out on the fusion characteristic Data in the test set Data2, the fusion characteristic Data are input into a trained long-short term memory recurrent neural network model, anti-normalization processing is carried out on a health state estimation value output by the model, the health state estimation value is compared with the health state Data in the test set Data2, and a Root Mean Square Error (RMSE) and an average absolute error (MAE) are calculated so as to evaluate the precision of the long-short term memory recurrent neural network model;
and 503, when the RMSE and the MAE calculated in the step 502 do not meet the expected target, returning to the step 4 to readjust the model parameters, and training until the test error of the model meets the expected target.
Compared with the prior art, the invention has the following beneficial effects:
according to the invention, the equal-time-difference time characteristic data and the equal-time-difference voltage characteristic data are extracted from the segment charging data of the lithium ion battery, compared with a single time characteristic or voltage characteristic, more battery aging information can be extracted, and the influence of the acquisition error of the charging data of the battery is not easy to occur; the typical correlation analysis is adopted to perform dimension reduction processing on the feature data to obtain fusion features, so that the redundancy of the feature data can be reduced, and the estimation precision of the health state of the lithium ion battery can be improved; because the fusion characteristics are one-dimensional data, the input node of the long-term and short-term memory recurrent neural network model only needs to be set to 1, the complexity of the model can be effectively reduced, and the estimation speed of the model is accelerated.
Drawings
Fig. 1 is a flowchart of a lithium ion battery state of health estimation method based on segment multi-charging feature fusion according to the present invention.
Detailed Description
For a more detailed description of the advantages and features of the present invention, reference is now made to the following description, taken in conjunction with the accompanying drawings. The embodiment is an embodiment of the present invention, and the present invention may also be implemented in other ways, which also belong to the protection scope of the present invention.
In a specific embodiment, as shown in fig. 1, a method for estimating a state of health of a lithium ion battery based on segment multi-charging feature fusion includes the following steps:
1. performing charge and discharge circulation on a newly-delivered lithium ion battery by using a battery charge and discharge tester, wherein the charge mode is constant-current constant-voltage charge, the discharge mode is constant-current discharge, and the charge and discharge circulation test is ended until the maximum discharge capacity of the lithium ion battery is reduced to 70% of the rated capacity, and the total cycle number is defined as D; and recording charging voltage data and corresponding time data in the constant-current charging process of each cycle, and recording the maximum discharge capacity data of the cycle.
2. And respectively extracting equal voltage difference time data, equal time difference voltage data and battery health state data aiming at the voltage and time data of each cycle.
Determining the starting voltage V of the segment charging voltage for each cycle when extracting isoelectric focusing time data 0 End voltage V n Extracting the initial voltage V 0 End voltage V n Corresponding time T 0 And
and determining a sampling voltage interval Deltav of
Wherein n is the characteristic quantity of the isoelectric pressure difference time data;
from a starting voltage V 0 At the beginning, for each additional sampling voltage interval Δ v, the voltage value and the corresponding time are recorded, i.e. for
Voltage data sequences can be obtained
And corresponding time series
Calculating the time difference corresponding to the equal voltage interval in the charging process according to the time sequence, namely
Thereby extracting the isoelectric pressure difference time data characteristics of single charging circulation
Determining the starting voltage V of the segment charging voltage for each cycle when extracting the voltage data with equal time difference 0 And its corresponding time T 0 And determining a sampling time interval delta t, and determining a termination time according to the characteristic number n in the step 201
From a starting voltage V 0 At the beginning, every time one is addedAt sampling time intervals Δ t, recording time and corresponding voltage value, i.e. for
Corresponding time series can be obtained
And voltage data sequence
Calculating the voltage difference corresponding to the equal time interval in the charging process according to the voltage data sequence, namely
Thereby extracting the equal time difference voltage data characteristics of a single charging cycle
Extracting the state of health data of the battery according to the maximum discharge capacity C of each cycle for each cycle max And rated capacity C of battery 0 And calculating the health state of the lithium ion battery in each cycle as follows:
3. for 2n features extracted in 2, namely isoelectric pressure difference time data features
Sum equal time difference voltage data characterization
And (3) performing dimension reduction on the feature data by adopting a typical correlation analysis method, and constructing a multivariate random variable Z = (X, Y), wherein X is a feature matrix, the dimension of the multivariate random variable is (2n, D), 2n is the feature quantity, D is the cycle number, and Y is the health state data of the lithium ion battery.
Construction lineSexual transformation P = a T X and Q = b T Y, carrying out standardization processing on the original data to obtain a standardized matrix Z of the multivariate random variable Z = (X, Y) * And calculating a normalized matrix Z * The covariance matrix of (a) yields:
wherein S is XX 、S XY 、S YX 、S YY The result is calculated for covariance.
In order to maximize the correlation between P and Q, it is necessary to select the appropriate vectors a and b to maximize the objective function, i.e.:
the Rayleigh entropy matrix is defined as:
definition u j And λ j The jth eigenvector and square root of eigenvalues for R, solving the maximization problem, representing vectors a and b as:
from vectors a and b, P = a can be calculated T X and Q = b T Y, resulting in a final fused feature vector FF with dimensions D, i.e.:
FF＝XP+YQ
according to the fusion characteristics and the lithium ion battery health state data, a lithium ion battery health state data set is constructed:
n Data samples in the Data set Data are used as a training set Data1, and the rest Data samples are used as a testing set Data2.
4. Establishing a long-short term memory cyclic neural network model, setting the input of the model as fusion characteristic data, setting the input node as 1, setting the output as the health state of a lithium ion battery, setting the output node as 1, setting initial parameters of the long-short term memory cyclic neural network model, setting a single hidden layer, setting the initial nodes as 250, setting the optimization algorithm as Adam, setting the initial iteration times as 120 times, setting the initial learning rate as 0.005, and after iterating for 90 times, changing the learning rate into 0.2.
5. And training the long-term and short-term memory cyclic neural network model by using the training set, and adjusting the model parameters according to the estimation error so as to reduce the estimation error of the model. And carrying out normalization processing on the fusion characteristic Data and the health state Data in the training set Data1, taking the normalized fusion characteristic Data as the input of the long-short term memory cyclic neural network model, taking the normalized health state Data as the output of the long-short term memory cyclic neural network model, and training the model.
The data normalization method comprises the following steps:
wherein x is k Is the kth value of the x sequence, namely characteristic data and health state data in the text, k is more than 0 and less than or equal to D, x min And x max Respectively, the minimum value and the maximum value in the x sequence.
And evaluating the estimation effect of the model by using the estimation precision of the test set test model. And after normalization processing is carried out on the fusion characteristic Data in the test set Data2, the fusion characteristic Data is input into the trained long-short term memory cyclic neural network model, the health state estimation value output by the model is subjected to inverse normalization processing, the health state estimation value is compared with the health state Data in the test set Data2, and the Root Mean Square Error (RMSE) and the Mean Absolute Error (MAE) are calculated according to the following formula, so that the precision of the long-short term memory cyclic neural network model is evaluated.
Wherein
Is an estimated value of the state of health, and S is an actual value of the state of health.
And when the calculated RMSE and MAE do not meet the expected target, returning to step 4 to readjust the model parameters and training until the test error of the model meets the expected target.
The above are preferred embodiments of the present invention, and all changes made according to the technical solutions of the present invention that produce functional effects do not exceed the scope of the technical solutions of the present invention belong to the protection scope of the present invention.
Claims (6)
1. A lithium ion battery health state estimation method based on segment multi-charging feature fusion is characterized by comprising the following steps:
step 1, performing a plurality of charging and discharging cycles on a lithium ion battery, and collecting charging voltage data and corresponding time data in each charging and discharging cycle and maximum discharging capacity data of the current cycle;
step 2, for the charging voltage data and the time data of each charging and discharging cycle, respectively extracting equal voltage difference time data, equal time difference voltage data and battery health state data, wherein: in order to extract the isoelectronic voltage difference time data, firstly determining the initial voltage, the final voltage and the sampling voltage interval of extraction, and extracting time data once every other sampling voltage from the initial voltage until the final voltage, thereby obtaining the isoelectronic voltage difference time data; in order to extract the voltage data with equal time difference, firstly determining the initial voltage, the final voltage and the sampling time interval to be extracted, and extracting the voltage data once every other sampling time from the initial voltage until the final voltage, thereby obtaining the voltage data with equal time difference; in order to extract the battery health state data, the maximum discharge capacity of each cycle is divided by the rated capacity to obtain the battery health state data of the current cycle;
step 3, processing the isoelectrical pressure difference time data and the isoelectrical time difference voltage data extracted in the step 2 by adopting a typical correlation analysis method, extracting fusion characteristic data, combining the fusion characteristic data with corresponding circulating battery health state data to form a lithium ion battery health state data set, and dividing the data set into a training set and a testing set;
step 4, establishing a long-short term memory cyclic neural network model, setting the input of the model as fusion characteristic data, setting the output of the model as the health state of the lithium ion battery, and setting initial parameters of the long-short term memory cyclic neural network model;
step 5, training the long-term and short-term memory cyclic neural network model by using the training set, and adjusting model parameters according to estimation errors to reduce the estimation errors of the model; and evaluating the estimation effect of the model by using the estimation precision of the test set test model.
2. The method for estimating the state of health of a lithium ion battery based on segment multi-charging feature fusion according to claim 1, wherein the step 1 comprises the following substeps:
step 101, using a battery charge-discharge tester to perform charge-discharge circulation on a newly-delivered lithium ion battery, wherein the charge mode is constant-current constant-voltage charge, the discharge mode is constant-current discharge, and the charge-discharge circulation test is ended until the maximum discharge capacity of the lithium ion battery is reduced to 70% of the rated capacity, and the total circulation frequency is defined as D;
and 102, recording charging voltage data and corresponding time data in the constant-current charging process of each cycle, and recording the maximum discharging capacity data of the current cycle.
3. The method for estimating the state of health of a lithium ion battery based on segment multi-charging feature fusion according to claim 1, wherein the step 2 comprises the following substeps:
step 201, when the isoelectric voltage difference time data is extracted, the initial voltage V of the segment charging voltage is determined for each cycle 0 End voltage V n Extracting the initial voltage V 0 End voltage V n Corresponding time T 0 And T n (1) And determining a sampling voltage interval Deltav of
Wherein n is the characteristic quantity of the isoelectric pressure difference time data;
from a starting voltage V 0 At the beginning, for each increment of the sampling voltage interval Δ v, the voltage value and the corresponding time are recorded, i.e. for
Obtaining a sequence of voltage data
And corresponding time series
Calculating the time difference corresponding to the equal voltage interval in the charging process according to the time sequence, namely
Thereby extracting the isoelectric pressure difference time data characteristics of single charging circulation
Step 202, when the voltage data with equal time difference is extracted, the initial voltage V of the charging voltage of the segment is determined for each cycle 0 And its corresponding time T 0 And determining a sampling time interval Δ t according to the characteristics in step 201Number n, determining the end time
From a starting voltage V 0 At the beginning, for each additional sampling time interval Δ t, the time and the corresponding voltage value are recorded, i.e. for
Obtaining corresponding time series
And voltage data sequence
Calculating the voltage difference corresponding to the equal time interval in the charging process according to the voltage data sequence, namely
Thereby extracting the equal time difference voltage data characteristics of a single charging cycle
Step 203, extracting the health state data of the battery, aiming at each circulation, and according to the maximum discharge capacity C of each circulation max And rated capacity C of battery 0 And calculating the health state of the lithium ion battery in each cycle as follows:
4. the method according to claim 3, wherein the step 3 comprises the following sub-steps:
step 301, 2n features extracted for step 2, i.e.Isoelectric pressure time data characterization
Sum equal time difference voltage data characterization
Performing dimensionality reduction on the feature data by adopting a typical correlation analysis method, and constructing a multivariate random variable Z = (X, Y), wherein X is a feature matrix, the dimensionality of the X is (2n, D), 2n is the feature quantity, D is the cycle number, Y is the lithium ion battery health state data, and the dimensionality of the Y is (1, D);
step 302, construct linear transformation P = a T X and Q = b T Y standardizing the original data to obtain a standardized matrix Z of multivariate random variables Z = (X, Y) * And calculating a normalized matrix Z * The covariance matrix of (a) yields:
wherein S is XX 、S XY 、S YX 、S YY Calculating a result for the covariance;
to maximize the correlation between P and Q, vectors a and b need to be selected to maximize the objective function, i.e.:
the Rayleigh entropy matrix is defined as:
definition u j And λ j The jth eigenvector and square root of eigenvalues for R, solving the maximization problem, representing vectors a and b as:
from vectors a and b, P = a is calculated T X and Q = b T Y, obtaining a final fusion feature vector FF, wherein the dimensionality of the final fusion feature vector FF is D, namely:
FF＝XP+YQ
step 303, constructing a lithium ion battery health state data set according to the fusion characteristics and the lithium ion battery health state data:
n Data samples in the Data set Data are used as a training set Data1, and the rest Data samples are used as a testing set Data2.
5. The method for estimating the state of health of the lithium ion battery based on segment multi-charging feature fusion according to claim 1, wherein the number of input nodes of the long-term and medium-term and short-term memory recurrent neural network model in the step 4 is 1, the number of output nodes of the long-term and medium-term memory recurrent neural network model is 1, and an optimization algorithm is Adam.
6. The method according to claim 4, wherein the step 5 comprises the following sub-steps:
step 501, performing normalization processing on the fusion characteristic Data and the health state Data in the training set Data1, taking the normalized fusion characteristic Data as the input of the long-short term memory cyclic neural network model, taking the normalized health state Data as the output of the long-short term memory cyclic neural network model, and training the model;
step 502, after normalization processing is carried out on the fusion characteristic Data in the test set Data2, the fusion characteristic Data are input into a trained long-short term memory recurrent neural network model, anti-normalization processing is carried out on a health state estimation value output by the model, the health state estimation value is compared with the health state Data in the test set Data2, and a Root Mean Square Error (RMSE) and an average absolute error (MAE) are calculated so as to evaluate the precision of the long-short term memory recurrent neural network model;
step 503, when the RMSE and the MAE calculated in step 502 do not meet the expected target, returning to step 4 to readjust the model parameters, and performing training until the test error of the model meets the expected target.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211276570.4A CN115684971A (en) | 2022-10-18 | 2022-10-18 | Lithium ion battery health state estimation method based on fragment multi-charging feature fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211276570.4A CN115684971A (en) | 2022-10-18 | 2022-10-18 | Lithium ion battery health state estimation method based on fragment multi-charging feature fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115684971A true CN115684971A (en) | 2023-02-03 |
Family
ID=85066904
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211276570.4A Pending CN115684971A (en) | 2022-10-18 | 2022-10-18 | Lithium ion battery health state estimation method based on fragment multi-charging feature fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115684971A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116774075A (en) * | 2023-08-28 | 2023-09-19 | 清华四川能源互联网研究院 | Lithium ion battery health state evaluation method and system |
| CN117706406A (en) * | 2024-02-05 | 2024-03-15 | 安徽布拉特智能科技有限公司 | Lithium battery health state monitoring model, method, system and storage medium |
| CN117930028A (en) * | 2024-03-21 | 2024-04-26 | 成都赛力斯科技有限公司 | Method, system, equipment and medium for predicting thermal failure of new energy vehicle battery |
-
2022
- 2022-10-18 CN CN202211276570.4A patent/CN115684971A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116774075A (en) * | 2023-08-28 | 2023-09-19 | 清华四川能源互联网研究院 | Lithium ion battery health state evaluation method and system |
| CN117706406A (en) * | 2024-02-05 | 2024-03-15 | 安徽布拉特智能科技有限公司 | Lithium battery health state monitoring model, method, system and storage medium |
| CN117706406B (en) * | 2024-02-05 | 2024-04-16 | 安徽布拉特智能科技有限公司 | Lithium battery health state monitoring model, method, system and storage medium |
| CN117930028A (en) * | 2024-03-21 | 2024-04-26 | 成都赛力斯科技有限公司 | Method, system, equipment and medium for predicting thermal failure of new energy vehicle battery |
| CN117930028B (en) * | 2024-03-21 | 2024-05-17 | 成都赛力斯科技有限公司 | Method, system, equipment and medium for predicting thermal failure of new energy vehicle battery |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108896914B (en) | Gradient lifting tree modeling and prediction method for health condition of lithium battery | |
| CN115684971A (en) | Lithium ion battery health state estimation method based on fragment multi-charging feature fusion | |
| CN110398697B (en) | Lithium ion health state estimation method based on charging process | |
| CN113917337A (en) | Battery health state estimation method based on charging data and LSTM neural network | |
| CN108490365B (en) | Method for estimating residual life of power battery of electric automobile | |
| CN112067998A (en) | Lithium ion battery state of charge estimation method based on deep neural network | |
| CN109543317B (en) | Method and device for predicting remaining service life of PEMFC | |
| CN103954913A (en) | Predication method of electric vehicle power battery service life | |
| CN111999649A (en) | XGboost algorithm-based lithium battery residual life prediction method | |
| CN114861527A (en) | Lithium battery life prediction method based on time series characteristics | |
| CN114545274A (en) | Lithium battery residual life prediction method | |
| CN113901707A (en) | SOH time sequence prediction method for vehicle lithium ion power battery | |
| CN114970332A (en) | Lithium battery model parameter identification method based on chaotic quantum sparrow search algorithm | |
| CN110646708A (en) | 10kV single-core cable early state identification method based on double-layer long-and-short-term memory network | |
| CN113721159A (en) | Data-driven battery health estimation method oriented to random local charging | |
| CN115586452A (en) | Lithium ion battery health state estimation method based on novel health characteristics | |
| CN115389946A (en) | Lithium battery health state estimation method based on isobaric rise energy and improved GRU | |
| CN115146723A (en) | Electrochemical model parameter identification method based on deep learning and heuristic algorithm | |
| Vilsen et al. | Log-linear model for predicting the lithium-ion battery age based on resistance extraction from dynamic aging profiles | |
| CN117554846B (en) | Lithium battery life prediction method and system considering constraint conditions | |
| Wei et al. | State of health and remaining useful life prediction of lithium-ion batteries with conditional graph convolutional network | |
| CN112580211A (en) | Lead-acid storage battery SOH estimation method based on SA and ANN algorithm | |
| CN116736171A (en) | Lithium ion battery health state estimation method based on data driving | |
| CN116298935A (en) | Lithium ion battery health state estimation method based on countermeasure encoder network | |
| CN116381505A (en) | Lithium battery health state estimation method based on dynamic working condition of variation modal decomposition |
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 |