CN115980588A - Lithium ion battery health state estimation method based on self-encoder extraction features - Google Patents
Lithium ion battery health state estimation method based on self-encoder extraction features Download PDFInfo
- Publication number
- CN115980588A CN115980588A CN202211674538.1A CN202211674538A CN115980588A CN 115980588 A CN115980588 A CN 115980588A CN 202211674538 A CN202211674538 A CN 202211674538A CN 115980588 A CN115980588 A CN 115980588A
- Authority
- CN
- China
- Prior art keywords
- constant
- lithium ion
- ion battery
- voltage
- current charging
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- 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
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/10—Energy storage using batteries
Landscapes
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
The invention discloses a lithium ion battery health state estimation method based on self-encoder extraction features, which comprises the following steps: carrying out cycle aging and capacity calibration tests on the lithium ion battery in sequence to obtain a constant voltage and constant current charging curve of the lithium ion battery; preprocessing a constant voltage and constant current charging curve of the lithium ion battery to obtain a preprocessed constant voltage and constant current charging curve; performing automatic feature extraction processing on the preprocessed constant-voltage constant-current charging curve through a convolution automatic encoder, and constructing a battery SOH estimation model based on a self-attention mechanism; and evaluating the lithium ion battery based on a battery SOH estimation model of a self-attention mechanism to obtain SOH data of the lithium ion battery. By using the method and the device, the SOH of the lithium ion battery can be accurately evaluated according to the charging characteristic data of the lithium ion battery. The lithium ion battery health state estimation method based on the self-encoder extracted features can be widely applied to the technical field of battery health state estimation.
Description
Technical Field
The invention relates to the technical field of battery health state evaluation, in particular to a lithium ion battery health state estimation method based on self-encoder extracted features.
Background
Lithium ion batteries, which are well known as energy storage devices, have many advantages, such as high energy density and high power density; the method plays an important role in a plurality of applications such as transportation, aerospace industry and portable electronic products, and during the charge and discharge cycle period of the lithium ion battery, the energy storage capacity is reduced along with the increase of the number of cycles experienced by the lithium ion battery, and the discharge and energy storage capacity is also reduced, when the battery health condition (SOH) is reduced to a certain degree, accidents such as battery leakage, insulation damage and partial short circuit problems may generate a plurality of potential safety hazards, so that efficient SOH estimation of lithium ions is necessary to remind users of the time for replacement and maintenance.
Disclosure of Invention
In order to solve the above technical problems, an object of the present invention is to provide a lithium ion battery state of health estimation method based on features extracted from an encoder, which can accurately estimate the SOH of a lithium ion battery according to charging feature data of the lithium ion battery.
The first technical scheme adopted by the invention is as follows: the lithium ion battery health state estimation method based on the feature extraction of the self-encoder comprises the following steps:
sequentially carrying out cyclic aging and capacity calibration tests on the lithium ion battery to obtain a constant-voltage constant-current charging curve of the lithium ion battery;
preprocessing a constant-voltage constant-current charging curve of the lithium ion battery to obtain a preprocessed constant-voltage constant-current charging curve;
performing automatic feature extraction processing on the preprocessed constant-voltage constant-current charging curve through a convolution automatic encoder and constructing a battery SOH estimation model based on a self-attention mechanism;
and evaluating the lithium ion battery based on a battery SOH estimation model of a self-attention mechanism to obtain SOH data of the lithium ion battery.
Further, the step of sequentially performing cycle aging and capacity calibration tests on the lithium ion battery to obtain a constant voltage and constant current charging curve of the lithium ion battery specifically includes:
carrying out cyclic aging treatment on the lithium ion battery to obtain a decommissioned aged lithium ion battery;
and carrying out capacity calibration treatment on the retired aged lithium ion battery to obtain a constant voltage and constant current charging curve of the lithium ion battery.
Further, the step of performing cyclic aging treatment on the lithium ion battery to obtain a decommissioned aged lithium ion battery specifically includes:
performing constant current charging processing on the lithium ion battery until the cut-off voltage of the lithium ion battery reaches a first preset voltage value, and stopping the constant current charging processing to obtain a first-stage lithium ion battery;
performing constant-voltage charging treatment on the lithium ion battery in the first stage until the current of the lithium ion battery is reduced to a preset current value, and stopping the constant-voltage charging treatment to obtain the lithium ion battery in the second stage;
performing constant current discharge treatment on the lithium ion battery in the second stage until the cut-off voltage of the lithium ion battery is reduced to a second preset voltage value, and stopping the constant current discharge treatment to obtain the lithium ion battery in the third stage;
and (4) circulating the constant-current charging, constant-voltage charging and constant-current discharging processes for the lithium ion battery in the third stage until the maximum available capacity of the lithium ion battery is lower than a preset threshold value, and obtaining the retired aged lithium ion battery.
Further, the step of performing capacity calibration processing on the retired aged lithium ion battery to obtain a constant voltage and constant current charging curve of the lithium ion battery specifically includes:
carrying out constant-current constant-voltage charging treatment on the retired aged lithium ion battery until the voltage and current of the retired aged lithium ion battery reach a preset threshold value, and stopping the constant-current constant-voltage charging treatment to obtain a first retired aged lithium ion battery;
and carrying out constant-current discharge treatment on the first retired aged lithium ion battery until the voltage of the first retired aged lithium ion battery reaches a preset threshold value, stopping the constant-current discharge treatment, and outputting a constant-voltage constant-current charging curve of the lithium ion battery.
Further, the step of preprocessing the constant voltage and constant current charging curve of the lithium ion battery to obtain a preprocessed constant voltage and constant current charging curve specifically includes:
intercepting the constant voltage and constant current charging curve of the lithium ion battery according to a preset range to obtain the intercepted constant voltage and constant current charging curve of the lithium ion battery;
inverting the coordinate axis of the intercepted constant voltage and constant current charging curve of the lithium ion battery to obtain an inverted constant voltage and constant current charging curve of the lithium ion battery;
aligning the inverted constant-voltage constant-current charging curve of the lithium ion battery according to a preset sampling data point to obtain an aligned constant-voltage constant-current charging curve of the lithium ion battery;
and splicing the aligned constant voltage and constant current charging curves of the lithium ion battery to obtain a preprocessed constant voltage and constant current charging curve.
Further, the preprocessed constant-voltage constant-current charging curve is three-dimensional data with two channels, wherein the first dimension represents data of different periods, the second dimension represents two channels formed by the time-current data and the time-voltage data after being intercepted and aligned, and the third dimension represents a sampling data point in each segment of data.
Further, the step of performing automatic feature extraction processing on the preprocessed constant-voltage constant-current charging curve through a convolution automatic encoder and constructing a battery SOH estimation model based on an attention-free mechanism specifically includes:
inputting the preprocessed constant-voltage constant-current charging curve into a convolution automatic encoder to perform unsupervised learning training, wherein the convolution automatic encoder comprises an encoder module and a decoder module;
the preprocessed constant-voltage constant-current charging curve is coded based on the coder module, and the coded constant-voltage constant-current charging curve is obtained;
decoding the encoded constant-voltage constant-current charging curve based on a decoder module to obtain a constant-voltage constant-current charging curve with a network middle layer;
splitting the network middle layer with the constant-voltage constant-current charging curve of the network middle layer, and extracting characteristic data of the constant-voltage constant-current charging curve;
performing self-attention mechanism calculation on the characteristic data of the constant-voltage constant-current charging curve to obtain a calculation result;
and inputting the calculation result into a convolutional neural network model, and constructing a battery SOH estimation model based on a self-attention mechanism through a back propagation algorithm.
Further, the calculation formula of the unsupervised learning training of the convolutional automatic encoder is as follows:
in the above equation, d represents the input data, f (-) represents the encoder function, h represents the hidden layer output, g (-) represents the decoder function,
representing the input data output from the encoder, i.e. its reconstruction.
The method has the beneficial effects that: according to the method, the charging data of the retired aged lithium ion battery is utilized, and the convolution self-encoder is adopted to automatically extract the characteristics so as to obtain the characteristics highly related to the SOH of the lithium ion battery, so that the requirements of a large amount of expert experience and the expenditure of time and labor cost of the traditional SOH estimation method are optimized and overcome; a battery SOH estimation model is constructed by adopting a self-attention mechanism with strong learning capacity, so that the problem of inaccurate estimation of the SOH of the lithium ion battery is solved; the method is beneficial to accurately and reliably evaluating the SOH of the lithium ion battery, provides reliable reference for the management, maintenance and replacement of the lithium ion battery and avoids unexpected safety accidents.
Drawings
Fig. 1 is a flow chart illustrating the steps of the lithium ion battery state of health estimation method based on the feature extracted from the encoder according to the present invention.
Detailed Description
The invention is described in further detail below with reference to the figures and the specific embodiments. The step numbers in the following embodiments are provided only for convenience of illustration, the order between the steps is not limited at all, and the execution order of each step in the embodiments can be adapted according to the understanding of those skilled in the art.
Referring to fig. 1, the present invention provides a lithium ion battery state of health estimation method based on self-encoder extracted features, the method includes the following steps:
s1, carrying out cyclic aging and capacity calibration tests on a lithium ion battery;
specifically, in the cycle aging test in this embodiment, a 18650 lithium ion (NCM 523) battery is used, a soft-package or square lithium battery may also be used, and conventional power batteries (ternary lithium and lithium iron phosphate batteries) in the material aspect may be applied to the present invention, and the cycle aging is performed on the battery and in the reverse direction until the maximum available capacity of the battery is lower than 80% of the initial capacity, which means that the lithium ion battery reaches the decommissioning condition and the life before decommissioning is terminated;
the cycle aging process is as follows: the battery is charged by constant current at 1C to reach the cut-off voltage of 4.2V at each period, then a CV charging stage is carried out, and the charging is stopped and the battery is placed when the current is reduced to 0.02C; then, constant current discharge treatment is carried out, 3C constant current discharge is carried out until the voltage reaches 2.75V, and cut-off discharge and standing are carried out;
the capacity calibration process comprises the following steps: firstly, constant-current and constant-voltage charging treatment is carried out, the battery is fully charged with constant current and constant voltage to 4.2V, the cut-off current is 0.02C, and the battery is placed; and then performing constant current discharge treatment, discharging to a cut-off voltage of 2.75V at a constant current of 1C, standing for 10 minutes, discharging to a cut-off voltage of 2.75V at a constant current of 0.5C, wherein the sum of the discharge capacities of the two stages of constant current discharge is the current maximum available capacity of the battery, and the SOH of the lithium ion battery is defined as follows:
in the above formula, Q rated Indicating the rated capacity, Q, of the battery aged Representing the maximum available capacity of the current battery, voltage, current and time data are recorded during the capacity calibration process.
S2, carrying out relevant preprocessing on voltage and current data which change along with time in the charging process of the lithium ion battery;
in particular, the pre-processing of the current and voltage data collected by the sensors over time during the charging phase of the battery cycle may be performed by finding the range of data most representative of the operating conditions and intercepting the data within that range, with the precondition that the current or voltage data for different periods contain the same range of voltages where the voltage is [ V ] during the constant current charging phase l ,V h ]In the constant voltage charging stage is [ I ] l ,I h ]；
Because the ranges of the captured charging current and charging voltage data in different periods are the same, and the corresponding charging time is different, for the captured data in different periods, the mapping relation between the original independent variable and the original dependent variable should be inverted to be aligned, and the original current-time and voltage-time data are converted into time-current and time-voltage data, that is, the current and the voltage in the same range in different periods are used as the independent variable, and the charging time changing along with the periods is used as the dependent variable to be remapped for data alignment;
the aligned data has two groups, namely current-time data and voltage-time data, and the two groups of data need to be uniformly utilized, so that the two groups of data are combined into two-channel three-dimensional data, and the specific form of the data is as follows:
d＝(cycle,2,samples)
in the above formula, cycle represents the first dimension, which represents different periods, 2 represents the second dimension, which represents two channels of current and voltage, and samples represents the third dimension, which represents sampled data points.
S3, inputting the preprocessed data into a convolution self-encoder to perform automatic feature extraction;
specifically, the convolutional auto-encoder model is trained using the data d obtained after preprocessing, and in general, the auto-encoder can be divided into two parts, an encoder and a decoder, and the formula is described as follows:
in the above equation, d represents the input data, f (-) represents the encoder function, h represents the hidden layer output, g (-) represents the decoder function,
representing the input data output from the encoder, i.e. its reconstruction;
the training target of the self-encoder is at the output
The input data d is reconstructed, so that other labels are not needed in training, after the training is finished, the decoder of the self-encoder is removed, the data can be input into the encoder, the automatically extracted features can be obtained in the hidden layer h, and the convolution self-encoder is the self-encoder constructed by using the convolution network. After the self-encoder model is built, data of charging current and voltage which are subjected to preprocessing and change at any time are used as input, the input data are used as targets to train the convolutional self-encoder model, and after training is finished, required health state characteristics are obtained in a hidden layer.
S4, constructing and training an SOH estimation model based on an attention-free mechanism through a back propagation algorithm;
specifically, the construction process updates the neural network model by using a back propagation algorithm by calculating the gradient obtained by the SOH label and the neural network, and the back propagation algorithm comprises the following steps:
(1) Inputting a training set D = { (x (n), y (n)) }, a verification set V, a learning rate alpha, a network layer number L and a neuron number L;
(2) Randomly initializing a network weight w and a weight b;
(3) Repeating the following steps;
-randomly reordering the samples in the training set D;
-for n＝1…N do；
-selecting samples (x (n), y (n)) from the training set D;
-feedforward calculating the net input z (l) and activation value a (l) for each layer until the last layer;
-calculating the error δ (l) for each layer by back propagation;
-calculating a derivative of each layer parameter;
-updating the parameters;
(4) Finishing;
(5) Until the error rate on the set V is verified to not decrease again by the neural network model, outputting w and b;
the characteristic automatically extracted by the convolution self-encoder is an abstract expression of a section of lithium ion battery SOH, a model is needed to map and process the SOH to convert the SOH into a concrete SOH value, and a one-dimensional convolution model is used as the data shape structure is suitable for a convolution network, and the convolution network model is suitable for processing data with a specific grid topological structure;
further, a self-attention mechanism is added into the convolutional network to further improve the learning performance of the model so as to ensure that the features automatically extracted by the self-encoder can be fully learned, the self-attention mechanism adopts query-key-value to process data, and firstly, the input of the self-attention mechanism module is mapped into three parts: series, keys and values, queries and keyss has a common dimension d k And values have a dimension d v These three parts of a batch of data acquisition can be formed as three matrices, respectively: q, K, and V, the process of obtaining output through these three matrices from the attention mechanism can be expressed as:
in the above equation, Q, K and V represent three matrices mapped from the attention mechanism model input, d k Represents the dimension of the matrix K, softmax (·) represents a function, attention (Q, K, V) represents the output from the Attention mechanism module;
wherein, the first and the second end of the pipe are connected with each other,
mapping the automatically extracted features from the encoder using 3 different one-dimensional convolution layers of kernel size 1 when mapping to queries, keys and values;
after the model is established, the SOH label of the battery is used as a target to train the SOH estimation model based on the self-attention mechanism by using the characteristics automatically extracted by the self-encoder as input so as to achieve the purpose of accurately estimating the SOH of the lithium ion battery.
And S5, evaluating the lithium ion battery based on a battery SOH estimation model of the self-attention mechanism to obtain SOH data of the lithium ion battery.
Specifically, capacity calibration process data is utilized to train a lithium ion battery SOH estimation model, the trained model and charging and discharging data measurement system hardware are integrated into a battery system, the system hardware part further comprises a memory and a processor, the memory is used for storing the model, and the processor executes the lithium ion battery health state evaluation method based on self-encoder feature extraction.
While the preferred embodiments of the present invention have been illustrated and described, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (8)
1. The lithium ion battery health state estimation method based on the feature extraction of the self-encoder is characterized by comprising the following steps of:
sequentially carrying out cyclic aging and capacity calibration tests on the lithium ion battery to obtain a constant-voltage constant-current charging curve of the lithium ion battery;
preprocessing a constant voltage and constant current charging curve of the lithium ion battery to obtain a preprocessed constant voltage and constant current charging curve;
performing automatic feature extraction processing on the preprocessed constant-voltage constant-current charging curve through a convolution automatic encoder and constructing a battery SOH estimation model based on a self-attention mechanism;
and evaluating the lithium ion battery based on a battery SOH estimation model of a self-attention mechanism to obtain SOH data of the lithium ion battery.
2. The method for estimating the health status of the lithium ion battery based on the extracted features of the self-encoder according to claim 1, wherein the step of sequentially performing cyclic aging and capacity calibration tests on the lithium ion battery to obtain a constant voltage and constant current charging curve of the lithium ion battery specifically comprises:
carrying out cyclic aging treatment on the lithium ion battery to obtain a decommissioned aged lithium ion battery;
and carrying out capacity calibration treatment on the retired aged lithium ion battery to obtain a constant voltage and constant current charging curve of the lithium ion battery.
3. The method for estimating the state of health of a lithium ion battery based on the extracted features of the self-encoder according to claim 2, wherein the step of performing cyclic aging processing on the lithium ion battery to obtain an aged-out lithium ion battery specifically comprises:
performing constant current charging processing on the lithium ion battery until the cut-off voltage of the lithium ion battery reaches a first preset voltage value, and stopping the constant current charging processing to obtain a first-stage lithium ion battery;
performing constant voltage charging treatment on the lithium ion battery in the first stage until the current of the lithium ion battery is reduced to a preset current value, and stopping the constant voltage charging treatment to obtain the lithium ion battery in the second stage;
performing constant current discharge treatment on the lithium ion battery in the second stage until the cut-off voltage of the lithium ion battery is reduced to a second preset voltage value, and stopping the constant current discharge treatment to obtain the lithium ion battery in the third stage;
and (4) circulating the constant-current charging, constant-voltage charging and constant-current discharging processes for the lithium ion battery in the third stage until the maximum available capacity of the lithium ion battery is lower than a preset threshold value, and obtaining the retired aged lithium ion battery.
4. The lithium ion battery state of health estimation method based on self-encoder extracted features as claimed in claim 3, wherein the step of performing capacity calibration processing on the retired aged lithium ion battery to obtain a constant voltage and constant current charging curve of the lithium ion battery specifically comprises:
carrying out constant-current constant-voltage charging treatment on the retired aged lithium ion battery until the voltage and current of the retired aged lithium ion battery reach a preset threshold value, and stopping the constant-current constant-voltage charging treatment to obtain a first retired aged lithium ion battery;
and carrying out constant-current discharge treatment on the first retired aged lithium ion battery until the voltage of the first retired aged lithium ion battery reaches a preset threshold value, stopping the constant-current discharge treatment, and outputting a constant-voltage constant-current charging curve of the lithium ion battery.
5. The lithium ion battery state of health estimation method based on self-encoder extracted features as claimed in claim 4, wherein the step of preprocessing the constant voltage and constant current charging curve of the lithium ion battery to obtain a preprocessed constant voltage and constant current charging curve specifically comprises:
intercepting the constant-voltage constant-current charging curve of the lithium ion battery according to a preset range to obtain the intercepted constant-voltage constant-current charging curve of the lithium ion battery;
inverting the coordinate axis of the intercepted constant voltage and constant current charging curve of the lithium ion battery to obtain an inverted constant voltage and constant current charging curve of the lithium ion battery;
aligning the inverted constant-voltage constant-current charging curve of the lithium ion battery according to a preset sampling data point to obtain an aligned constant-voltage constant-current charging curve of the lithium ion battery;
and splicing the aligned constant voltage and constant current charging curves of the lithium ion battery to obtain a preprocessed constant voltage and constant current charging curve.
6. The lithium ion battery state of health estimation method based on self-encoder extracted features of claim 5, wherein the preprocessed constant voltage constant current charging curve is three-dimensional data with two channels, wherein a first dimension represents data of different periods used, a second dimension represents two channels formed by the intercepted and aligned time-current data and time-voltage data, and a third dimension represents a sampling data point in each segment of data.
7. The lithium ion battery state of health estimation method based on self-encoder extracted features of claim 6, wherein the step of performing automatic feature extraction processing on the preprocessed constant voltage and constant current charging curve through a convolution automatic encoder and constructing a battery SOH estimation model based on a self-attention mechanism specifically comprises:
inputting the preprocessed constant-voltage constant-current charging curve into a convolution automatic encoder to perform unsupervised learning training, wherein the convolution automatic encoder comprises an encoder module and a decoder module;
the preprocessed constant-voltage constant-current charging curve is coded based on a coder module, and the coded constant-voltage constant-current charging curve is obtained;
decoding the encoded constant-voltage constant-current charging curve based on a decoder module to obtain a constant-voltage constant-current charging curve with a network middle layer;
splitting the network middle layer with the constant-voltage constant-current charging curve of the network middle layer, and extracting characteristic data of the constant-voltage constant-current charging curve;
performing self-attention mechanism calculation on the characteristic data of the constant-voltage constant-current charging curve to obtain a calculation result;
and inputting the calculation result into a convolutional neural network model, and constructing a battery SOH estimation model based on a self-attention mechanism through a back propagation algorithm.
8. The lithium ion battery state of health estimation method based on self-encoder extracted features as claimed in claim 7, wherein the calculation formula of the unsupervised learning training of the convolutional auto-encoder is as follows:
in the above equation, d represents the input data, f (-) represents the encoder function, h represents the hidden layer output, g (-) represents the decoder function,
representing the input data output from the encoder, i.e. its reconstruction. />
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211674538.1A CN115980588A (en) | 2022-12-26 | 2022-12-26 | Lithium ion battery health state estimation method based on self-encoder extraction features |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211674538.1A CN115980588A (en) | 2022-12-26 | 2022-12-26 | Lithium ion battery health state estimation method based on self-encoder extraction features |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115980588A true CN115980588A (en) | 2023-04-18 |
Family
ID=85971794
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211674538.1A Pending CN115980588A (en) | 2022-12-26 | 2022-12-26 | Lithium ion battery health state estimation method based on self-encoder extraction features |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115980588A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116298936A (en) * | 2023-05-19 | 2023-06-23 | 河南科技学院 | Intelligent lithium ion battery health state prediction method in incomplete voltage range |
| CN117872204A (en) * | 2024-01-31 | 2024-04-12 | 武汉全日行特种新能源有限公司 | Lithium ion battery health state evaluation method and device and electronic equipment |
-
2022
- 2022-12-26 CN CN202211674538.1A patent/CN115980588A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116298936A (en) * | 2023-05-19 | 2023-06-23 | 河南科技学院 | Intelligent lithium ion battery health state prediction method in incomplete voltage range |
| CN117872204A (en) * | 2024-01-31 | 2024-04-12 | 武汉全日行特种新能源有限公司 | Lithium ion battery health state evaluation method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110568361B (en) | Method for predicting health state of power battery | |
| CN115980588A (en) | Lithium ion battery health state estimation method based on self-encoder extraction features | |
| CN110146822A (en) | A kind of Vehicular dynamic battery capacity On-line Estimation method based on constant-current charge process | |
| CN101946187A (en) | Battery hysteresis based on model is estimated | |
| CN113917334B (en) | Battery health state estimation method based on evolution LSTM self-encoder | |
| CN111257754B (en) | Battery SOC robust evaluation method based on PLSTM sequence mapping | |
| CN112834927A (en) | Lithium battery residual life prediction method, system, device and medium | |
| CN104569844A (en) | Valve control seal type lead-acid storage battery health condition monitoring method | |
| CN113743661A (en) | Method, system, equipment and storage medium for predicting online capacity of lithium ion battery | |
| CN113406503A (en) | Lithium battery SOH online estimation method based on deep neural network | |
| CN114726045A (en) | Lithium battery SOH estimation method based on IPEA-LSTM model | |
| CN106443496A (en) | Battery charge state estimation method with improved noise estimator | |
| CN115389946A (en) | Lithium battery health state estimation method based on isobaric rise energy and improved GRU | |
| CN116298936A (en) | Intelligent lithium ion battery health state prediction method in incomplete voltage range | |
| CN115389954A (en) | Battery capacity estimation method, electronic equipment and readable storage medium | |
| CN113791351B (en) | Lithium battery life prediction method based on transfer learning and difference probability distribution | |
| CN114280490B (en) | Lithium ion battery state of charge estimation method and system | |
| CN117171588A (en) | Method for detecting gradient utilization faults of power battery | |
| CN116577686B (en) | Multi-working condition SOH estimation method and system based on local stage charging data | |
| CN117471320A (en) | Battery state of health estimation method and system based on charging fragments | |
| CN116679231A (en) | Lithium battery SoH estimation method based on gram angle field and VGG16 model | |
| CN116908727A (en) | Multi-feature combination-based energy storage power station lithium battery capacity estimation method | |
| CN116224085A (en) | Lithium battery health state assessment method based on data driving | |
| CN117074955A (en) | Cloud-end correction OCV-based lithium battery state joint estimation method | |
| CN116953554A (en) | Multi-fragment data-based method and device for estimating SOH of lithium battery of energy storage power station |
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 |