CN116338501A - Lithium ion battery health detection method based on neural network prediction relaxation voltage - Google Patents
Lithium ion battery health detection method based on neural network prediction relaxation voltage Download PDFInfo
- Publication number
- CN116338501A CN116338501A CN202211635410.4A CN202211635410A CN116338501A CN 116338501 A CN116338501 A CN 116338501A CN 202211635410 A CN202211635410 A CN 202211635410A CN 116338501 A CN116338501 A CN 116338501A
- Authority
- CN
- China
- Prior art keywords
- neural network
- battery
- error
- voltage
- lithium ion
- 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.)
- Granted
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 45
- HBBGRARXTFLTSG-UHFFFAOYSA-N Lithium ion Chemical compound [Li+] HBBGRARXTFLTSG-UHFFFAOYSA-N 0.000 title claims abstract description 33
- 229910001416 lithium ion Inorganic materials 0.000 title claims abstract description 33
- 230000036541 health Effects 0.000 title claims abstract description 26
- 238000001514 detection method Methods 0.000 title claims abstract description 15
- 238000000034 method Methods 0.000 claims abstract description 23
- 238000012360 testing method Methods 0.000 claims abstract description 19
- 238000012549 training Methods 0.000 claims description 47
- 238000012795 verification Methods 0.000 claims description 32
- 238000003062 neural network model Methods 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 10
- 210000002569 neuron Anatomy 0.000 claims description 7
- 238000004806 packaging method and process Methods 0.000 claims description 4
- 238000012546 transfer Methods 0.000 claims description 4
- 238000011478 gradient descent method Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 210000005036 nerve Anatomy 0.000 claims 3
- 230000008859 change Effects 0.000 abstract description 2
- 239000010410 layer Substances 0.000 description 17
- 230000008569 process Effects 0.000 description 5
- 238000011161 development Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 229910052799 carbon Inorganic materials 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000011897 real-time detection Methods 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 210000002364 input neuron Anatomy 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000011229 interlayer Substances 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
- 238000012544 monitoring process Methods 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000006386 neutralization reaction Methods 0.000 description 1
- 210000004205 output neuron Anatomy 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 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]
- G01R31/392—Determining battery ageing or deterioration, e.g. state of health
-
- 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
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Secondary Cells (AREA)
Abstract
The invention discloses a lithium ion battery health detection method based on neural network predicted relaxation voltage, which uses the neural network predicted relaxation voltage as a core method, predicts the relaxation voltage through a neural network by using information such as voltage change, temperature, multiplying power and the like of a battery in a short time after charging, and evaluates the battery health state by combining the correlation relation of the relaxation voltage and the battery capacity. The method combines two methods of relaxation voltage prediction and neural network prediction, realizes the relaxation voltage acquisition in a short time, and further accurately predicts the health state of the battery, and has the characteristics of wide application range (suitable for various commercial lithium ion batteries at present), short test time and good detection precision.
Description
Technical Field
The invention relates to a method for detecting the health state of a lithium ion battery, in particular to a method for detecting and evaluating the health state of the lithium ion battery by utilizing a neural network.
Background
With the advent of the "carbon peak", "carbon neutralization" policies and the increasing exhaustion of traditional fossil energy sources, the search for an alternative energy source is an important issue in the development of the world today. The lithium ion battery is a choice with wide application and good development prospect.
Lithium ion batteries have been widely paid attention since the advent of the prior art, and with the development of recent years, lithium ion batteries are widely used in various fields such as electric automobiles, 3C numbers, aerospace national defense, and the like. However, the problems of performance attenuation, capacity water jump and the like of the lithium ion battery can occur in the using process, and the current detection means for the health state of the lithium ion battery are relatively few. In research in recent years, a machine learning method applied to the AI industry is introduced into the field of lithium ion battery detection, and a large amount of battery data can be learned by a computer, so that the prediction of a lithium ion battery with higher precision is realized, meanwhile, relaxation voltage is considered to have strong correlation with capacity, and the combination of the relaxation voltage and the capacity is one of the current research hotspots. However, the relaxation voltage can wait for a long time in the measurement process, which cannot meet the requirement of real-time detection of battery health.
The machine learning method is usually data driven, and can specifically comprise a neural network, a decision tree, clustering, a support vector machine and the like, wherein the neural network is most widely applied, but the selection of input values can influence the prediction accuracy of the neural network, the current low-cost battery value extraction is single, the traditional data prediction accuracy of voltage, current and the like is poor, and industrialization is difficult to realize. The relaxation voltage value is simpler, and meanwhile, research shows that a good linear relation exists between the relaxation voltage and the battery capacity, but the relaxation voltage is usually selected to be a value waiting for at least 30min after charging, and the purpose of real-time detection of the lithium ion battery cannot be met.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention provides a lithium ion battery health detection method based on neural network prediction relaxation voltage. The method combines two methods of relaxation voltage prediction and neural network prediction, realizes that relaxation voltage is obtained in a short time, and further accurately predicts the health state of the battery, and has the characteristics of wide application range (applicable to various commercial lithium ion batteries at present), short test time and good detection precision.
The invention aims at realizing the following technical scheme:
a lithium ion battery health detection method based on neural network prediction relaxation voltage comprises the following steps:
step 1: carrying out charge and discharge tests on the commercial lithium ion battery under different conditions, and collecting battery data after each charge;
step 2: classifying the battery data acquired in the step 1 under different conditions, randomly selecting part of the data as a training set, and the rest as a verification set;
step 3: training the neural network by using the training set sample obtained in the step 2, comparing the predicted value output by training with the true value in the training set, and updating the neuron weight in the network by back propagation to obtain the weight distribution with the minimum training error;
step 4: substituting the verification set into the neural network obtained in the step 3, calculating the error of the verification set, comparing the error of the training set with the error of the verification set, and if the error of the training set is obviously smaller than the error of the verification set, finishing training in time to prevent the occurrence of overfitting;
step 5: comparing the training error of the verification set with the error of the verification set of the previous cycle, if the training error is smaller than the error of the previous cycle, storing the neural network, and storing the error of the verification set of the current cycle as the current minimum error until the cycle is finished, thereby obtaining an optimal neural network model under the cycle number;
step 6: storing after the neural network model training is finished to obtain a relaxation voltage prediction model of the model lithium ion battery, and determining a linear mathematical expression containing temperature and multiplying power parameters according to the corresponding mathematical relation between the relaxation voltage and capacity of the model lithium ion battery;
step 7: and (3) packaging the neural network model trained in the step (5) and the linear mathematical expression determined in the step (6), and outputting the packaged neural network model into a general desktop applet so as to realize the rapid detection of the health state of the battery.
Compared with the prior art, the invention has the following advantages:
1. the invention predicts the battery health status indirectly by predicting the battery relaxation voltage for the first time. The adopted neural network is BP neural network, so that the calculation speed is high.
2. The temperature, multiplying power and short-time pressure drop information utilized by the method have a strong correlation with relaxation voltage, so that the data required by prediction is less and is relatively easier to obtain.
3. The relaxation voltage can be obtained in a short time by using the neural network, and long-time test is not needed; the required input information is easy and convenient to test, and the labor and time cost can be effectively reduced; the relaxation voltage and the capacity have strong linear relation, and the test precision is higher.
4. Compared with the current capacity prediction method based on relaxation voltage, the method effectively solves the problems that the test time is long and real-time monitoring cannot be achieved, simultaneously retains the advantage of strong linear relation between relaxation voltage and capacity, and achieves the goal of simply and rapidly detecting the health state of the battery.
Drawings
FIG. 1 is a flow chart for indirectly detecting battery state of health based on a neural network predicted relaxation voltage.
Fig. 2 is a graph of battery relaxation voltage versus capacity at different temperatures and different rates.
FIG. 3 is a graph comparing predicted results of a test dataset with actual results.
FIG. 4 is a diagram of a program interface for code packaging integration.
Detailed Description
The following description of the present invention is provided with reference to the accompanying drawings, but is not limited to the following description, and any modifications or equivalent substitutions of the present invention should be included in the scope of the present invention without departing from the spirit and scope of the present invention.
The invention provides a lithium ion battery health detection method based on a neural network for predicting relaxation voltage, which uses the neural network for predicting relaxation voltage as a core method, predicts relaxation voltage through the neural network by using information such as voltage change, temperature, multiplying power and the like of a battery in a short time after charging, and evaluates the health state of the battery by combining the correlation relation between the relaxation voltage and the battery capacity, and specifically comprises the following steps:
step 1: and (3) carrying out charge and discharge tests on the commercial lithium ion battery under different conditions, and collecting battery data after each charge.
In this step, the charge and discharge test for a commercial lithium ion battery of a certain model includes: charge and discharge tests at different temperatures and different charge rates (same discharge rate); the collected battery data includes: the battery voltage after 120s of each charge, the temperature of the charge, the charging multiplying power, the charge cut-off voltage, the battery voltage after 30min of relaxation and the corresponding battery capacity after each charge.
Step 2: classifying the battery data collected in the step 1 under different conditions, randomly selecting part of the data as a training set, and the rest as a verification set.
Step 3: establishing a BP neural network, comprising the number of network layers, the number of input neurons and hidden layers, the number of neurons contained in each hidden layer and the number of output neurons, and setting a transfer function and a training method of the BP neural network.
In the step, the BP neural network adopts a five-layer structure, namely 1 input layer, 3 hidden layers and 1 output layer, wherein the input layer comprises 120s of battery voltage, charging temperature, charging multiplying power and charging cut-off voltage, the total number of the hidden layers comprises 16-64 neural units, the output layer comprises 1 neuron for predicting relaxation voltage, the transfer function of the hidden layers selects a relu function, and the normalization method selects a Standard scaler method; the algorithm adopts a gradient descent method, the learning rate is set to 0.001, the number of cycles is selected 100 times, and the error function selects MSE.
Step 4: training the BP neural network established in the step 3 by using the training set sample obtained in the step 2, comparing the predicted value output by training with the true value in the training set, and updating the neuron weight in the network by back propagation to obtain the weight distribution with the minimum training error.
Step 5: substituting the verification set into the neural network obtained in the step 4, calculating the error of the verification set, comparing the error of the training set with the error of the verification set, and if the error of the training set is obviously smaller than the error of the verification set, finishing training in time, thereby preventing the occurrence of overfitting.
Step 6: and comparing the training error of the verification set with the error of the verification set of the previous cycle, if the training error is smaller than the error of the previous cycle, storing the neural network, and storing the verification set error of the current cycle as the current minimum error until the cycle is finished, thereby obtaining the optimal neural network model under the cycle times.
Step 7: and after the neural network model is trained, storing to obtain a relaxation voltage prediction model of the model lithium ion battery, and determining a linear mathematical expression containing temperature and multiplying power parameters according to the corresponding mathematical relation between the relaxation voltage and capacity of the model lithium ion battery.
In the step, a linear expression of relaxation voltage and capacity of a battery 30min after each charge is fitted according to test data, a functional relation of temperature and multiplying power on the slope and intercept influence of the linear expression is obtained, and a mathematical relation of the relaxation voltage and the capacity which finally contain the temperature and multiplying power as variables is written.
Step 8: and (3) packaging the neural network model trained in the step (6) and the linear mathematical expression determined in the step (7), and outputting the packaged neural network model into a general desktop applet so as to realize the rapid detection of the health state of the battery.
Examples:
first, this example was tested using NCA battery data of model 18650 from Jiangong Zhu et al in the zenodo database. After the battery data are obtained, the battery data are extracted, and charging multiplying power, the temperature of the battery, charging cut-off voltage, battery voltage after charging for 120s, battery voltage after charging for 30min and capacity information after battery charging are collected in each charging and discharging process of the battery. Each cycle under each charge and discharge condition serves as a set of data. The charging rate, the temperature of the battery, the charging cut-off voltage and the battery voltage after the charging is completed for 120s are used as input information, and the battery voltage after the charging is completed for 30min is used as a mark. After a complete set of data sets is obtained, 20 sets of data are randomly extracted as a test set for subsequent model verification. The rest data sets are 80% used as training sets, and 20% used as verification sets are substituted into the neural network for training.
The BP neural network is designed, and the neural network has five layers in total, namely an input layer, 3 hidden layers and an output layer. The input layer comprises 4 neurons in total, namely a battery voltage (X1) charged for 120s, a charging multiplying power (X2), a temperature (X3) of the battery and a charging cut-off voltage (X4), and the number of neurons in the hidden layer is between 16 and 64. The implicit inter-layer transfer function selects the relu function. The data normalization method selects the Standard scaler method, the algorithm adopts the gradient descent method, and the learning rate is set to be 0.001. The error function selects the MSE function.
Substituting the data used as the network training into the data to perform training, wherein the training process is shown in fig. 1, calculating the error of the verification set each time, comparing the error of the training set with the error of the verification set, and if the error of the training set is obviously smaller than the error of the verification set, finishing the training in time and preventing the occurrence of over fitting. And comparing the training error of the verification set with the error of the verification set in the previous cycle, if the training error is smaller than the error of the previous cycle, storing the neural network, and storing the verification set error in the current cycle as the minimum error until the cycle is finished, thereby obtaining the optimal neural network model under the cycle times.
After the optimal neural network is obtained, annotating the network training process, and only reserving the trained neural network as a subsequent prediction tool.
And then fitting a linear expression of relaxation voltage and capacity of the battery 30min after each charge according to the test data to obtain a functional relation of temperature and multiplying power on the slope and intercept influence of the expression. The relationship between capacity and relaxation voltage is shown in FIG. 2, and the relationship between relaxation voltage and capacity of the battery under the test conditions of 0.25-25 ℃, 0.5-35 ℃ and 0.5-45 ℃ is shown in (a) - (d), and the relationship between relaxation voltage and capacity of the battery is found to have a good linear relationship except the influence caused by activation when the battery is started for several times. In this example, it was found by analysis that the influence of the battery magnification on the slope and intercept satisfies the linear relationship, whereas the influence of the temperature on the slope and intercept satisfies the e-exponential relationship at the same magnification. And combining the two influencing factors, and writing a mathematical relation of relaxation voltage and capacity which finally contains temperature and multiplying power as variables. In this embodiment, the mathematical expression is as follows:
wherein T represents the battery ambient temperature, R represents the battery charging multiplying power, V R Representing predicted battery relaxation voltage, C 1 、C 2 Representing a constant term.
After obtaining the optimal neural network and mathematical expression, substituting 4 pieces of input information of 20 groups of data randomly extracted in the first step to obtain a predicted result of relaxation voltage, substituting the result of relaxation voltage into the mathematical expression to obtain a final capacity predicted value, dividing the final capacity predicted value by a standard capacity of a battery to obtain battery health state information, and comparing the battery health state information with a true value, wherein the obtained result is shown in fig. 3, fig. 3 (a) shows that the predicted value of the battery health state is compared with the true value, a triangle represents the true value, and a circle represents the predicted value; fig. 3 (b) shows the relative error of the predicted value, and it can be seen from the figure that the predicted deviation is maintained around ±4% except for the extreme individual samples, and that the prediction accuracy is good.
After the work is finished, the completed neural network model is packaged by using a tkilter tool kit, and the packaged neural network model is output to become a visual operable desktop program. The program interface is shown in fig. 4, and the interface content includes text lines of "please input battery relaxation 120s voltage, charging rate, charging temperature, charging cut-off voltage, 4 user input boxes and 1 prediction interaction button respectively. And finally obtaining the generalized software of the method.
Claims (4)
1. A lithium ion battery health detection method based on neural network prediction relaxation voltage is characterized by comprising the following steps:
step 1: carrying out charge and discharge tests on the commercial lithium ion battery under different conditions, and collecting battery data after each charge;
step 2: classifying the battery data acquired in the step 1 under different conditions, randomly selecting part of the data as a training set, and the rest as a verification set;
step 3: training the neural network by using the training set sample obtained in the step 2, comparing the predicted value output by training with the true value in the training set, and updating the neuron weight in the network by back propagation to obtain the weight distribution with the minimum training error;
step 4: substituting the verification set into the neural network obtained in the step 3, calculating the error of the verification set, comparing the error of the training set with the error of the verification set, and if the error of the training set is obviously smaller than the error of the verification set, finishing training in time to prevent the occurrence of overfitting;
step 5: comparing the training error of the verification set with the error of the verification set of the previous cycle, if the training error is smaller than the error of the previous cycle, storing the neural network, and storing the error of the verification set of the current cycle as the current minimum error until the cycle is finished, thereby obtaining an optimal neural network model under the cycle number;
step 6: storing after the neural network model training is finished to obtain a relaxation voltage prediction model of the model lithium ion battery, and determining a linear mathematical expression containing temperature and multiplying power parameters according to the corresponding mathematical relation between the relaxation voltage and capacity of the model lithium ion battery;
step 7: and (3) packaging the neural network model trained in the step (5) and the linear mathematical expression determined in the step (6), and outputting the packaged neural network model into a general desktop applet so as to realize the rapid detection of the health state of the battery.
2. The method for detecting the health of the lithium ion battery based on the predicted relaxation voltage of the neural network according to claim 1, wherein in the step 1, the charge and discharge test for a commercial lithium ion battery of a certain model comprises: charge and discharge tests at different temperatures and different charge rates; the collected battery data includes: the battery voltage after 120s of each charge, the temperature of the charge, the charging multiplying power, the charge cut-off voltage, the battery voltage after 30min of relaxation and the corresponding battery capacity after each charge.
3. The lithium ion battery health detection method based on the neural network predicted relaxation voltage according to claim 1, wherein in the step 3, the neural network is a BP neural network, a five-layer structure is adopted, namely 1 input layer, 3 hidden layers and 1 output layer, the input layer comprises 120s battery voltage, charging temperature, charging multiplying power and charging cut-off voltage, 4 nerve units are added in total, the hidden layers comprise 16-64 nerve units, the output layer comprises 1 nerve unit of the predicted relaxation voltage, the hidden layer transfer function selects a relu function, and the normalization method selects a Standard scaler method; the algorithm adopts a gradient descent method, the learning rate is set to 0.001, the number of cycles is selected 100 times, and the error function selects MSE.
4. The method for detecting the health of the lithium ion battery based on the neural network predicted relaxation voltage according to claim 1, wherein in the step 6, a linear expression of the relaxation voltage and the capacity of the battery 30min after each charging is fitted according to test data, a functional relation of the influence of temperature and multiplying power on the slope and intercept of the linear expression is obtained, and a mathematical relation of the relaxation voltage and the capacity which finally contain the temperature and the multiplying power as variables is written.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211635410.4A CN116338501B (en) | 2022-12-19 | 2022-12-19 | Lithium ion battery health detection method based on neural network prediction relaxation voltage |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211635410.4A CN116338501B (en) | 2022-12-19 | 2022-12-19 | Lithium ion battery health detection method based on neural network prediction relaxation voltage |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116338501A true CN116338501A (en) | 2023-06-27 |
CN116338501B CN116338501B (en) | 2023-09-12 |
Family
ID=86882834
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211635410.4A Active CN116338501B (en) | 2022-12-19 | 2022-12-19 | Lithium ion battery health detection method based on neural network prediction relaxation voltage |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116338501B (en) |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103645372A (en) * | 2013-12-27 | 2014-03-19 | 哈尔滨工业大学 | Method for quickly estimating open circuit voltage of secondary battery |
US9063018B1 (en) * | 2012-10-22 | 2015-06-23 | Qnovo Inc. | Method and circuitry to determine temperature and/or state of health of a battery/cell |
CN110709717A (en) * | 2017-04-17 | 2020-01-17 | 密歇根大学董事会 | Method for estimating battery health of mobile device based on relaxation voltage |
CN112051504A (en) * | 2020-08-13 | 2020-12-08 | 联合汽车电子有限公司 | Method and device for predicting battery capacity, terminal and computer-readable storage medium |
CN112180274A (en) * | 2020-09-28 | 2021-01-05 | 上海理工大学 | Rapid detection and evaluation method for power battery pack |
CN112924878A (en) * | 2021-01-26 | 2021-06-08 | 同济大学 | Battery safety diagnosis method based on relaxation voltage curve |
CN113743682A (en) * | 2021-11-03 | 2021-12-03 | 中国科学院精密测量科学与技术创新研究院 | NMR (nuclear magnetic resonance) relaxation time inversion method based on supervised deep neural network |
CN114487846A (en) * | 2022-01-14 | 2022-05-13 | 中国人民解放军国防科技大学 | Method and device for estimating electrochemical impedance spectrum of battery on line |
CN115453399A (en) * | 2022-08-26 | 2022-12-09 | 广东工业大学 | Battery pack SOH estimation method considering inconsistency |
CN115932611A (en) * | 2022-10-10 | 2023-04-07 | 中国科学技术大学 | Lithium ion battery internal short circuit fault diagnosis method based on relaxation process |
CN116027218A (en) * | 2021-10-27 | 2023-04-28 | 比亚迪股份有限公司 | Battery state evaluation method and electronic equipment |
-
2022
- 2022-12-19 CN CN202211635410.4A patent/CN116338501B/en active Active
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9063018B1 (en) * | 2012-10-22 | 2015-06-23 | Qnovo Inc. | Method and circuitry to determine temperature and/or state of health of a battery/cell |
CN103645372A (en) * | 2013-12-27 | 2014-03-19 | 哈尔滨工业大学 | Method for quickly estimating open circuit voltage of secondary battery |
CN110709717A (en) * | 2017-04-17 | 2020-01-17 | 密歇根大学董事会 | Method for estimating battery health of mobile device based on relaxation voltage |
US20200191876A1 (en) * | 2017-04-17 | 2020-06-18 | The Regents Of The University Of Michigan | Method to estimate battery health for mobile devices based on relaxing voltages |
CN112051504A (en) * | 2020-08-13 | 2020-12-08 | 联合汽车电子有限公司 | Method and device for predicting battery capacity, terminal and computer-readable storage medium |
CN112180274A (en) * | 2020-09-28 | 2021-01-05 | 上海理工大学 | Rapid detection and evaluation method for power battery pack |
CN112924878A (en) * | 2021-01-26 | 2021-06-08 | 同济大学 | Battery safety diagnosis method based on relaxation voltage curve |
CN116027218A (en) * | 2021-10-27 | 2023-04-28 | 比亚迪股份有限公司 | Battery state evaluation method and electronic equipment |
CN113743682A (en) * | 2021-11-03 | 2021-12-03 | 中国科学院精密测量科学与技术创新研究院 | NMR (nuclear magnetic resonance) relaxation time inversion method based on supervised deep neural network |
CN114487846A (en) * | 2022-01-14 | 2022-05-13 | 中国人民解放军国防科技大学 | Method and device for estimating electrochemical impedance spectrum of battery on line |
CN115453399A (en) * | 2022-08-26 | 2022-12-09 | 广东工业大学 | Battery pack SOH estimation method considering inconsistency |
CN115932611A (en) * | 2022-10-10 | 2023-04-07 | 中国科学技术大学 | Lithium ion battery internal short circuit fault diagnosis method based on relaxation process |
Non-Patent Citations (4)
Title |
---|
JIANGONG ZHU 等: "Data-driven capacity estimation of commercial lithium-ion batteries from voltage relaxation", 《 NATURE COMMUNICATIONS 》, pages 1 - 10 * |
JIANGONG ZHU 等: "Data-driven lithium-ion battery capacity estimation from voltage relaxation", 《NATURE PORTFOLIO》, pages 1 - 18 * |
汤依伟;艾亮;程昀;王安安;李书国;贾明;: "锂离子动力电池高倍率充放电过程中弛豫行为的仿真", 物理学报, no. 05, pages 340 - 349 * |
魏学哲 等: "基于电压弛豫模型的锂离子电池老化状态估计", 《同济大学学报(自然学科版)》, vol. 46, no. 1, pages 171 - 177 * |
Also Published As
Publication number | Publication date |
---|---|
CN116338501B (en) | 2023-09-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Ma et al. | State of health estimation and remaining useful life prediction for lithium-ion batteries by improved particle swarm optimization-back propagation neural network | |
CN111443293A (en) | Lithium battery state of health (SOH) estimation method based on data driving | |
CN112630662B (en) | Power battery SOH estimation method based on data driving and multi-parameter fusion | |
CN113219343A (en) | Lithium battery health state prediction method, system, equipment and medium based on elastic network | |
CN110806541B (en) | AD-BAS-based lithium battery model parameter identification method | |
CN113917334B (en) | Battery health state estimation method based on evolution LSTM self-encoder | |
CN114818831B (en) | Bidirectional lithium ion battery fault detection method and system based on multi-source perception | |
CN112684363A (en) | Lithium ion battery health state estimation method based on discharge process | |
CN113109715A (en) | Battery health condition prediction method based on feature selection and support vector regression | |
CN115718263B (en) | Attention-based lithium ion battery calendar aging prediction model and method | |
CN116298934B (en) | Modeling method of prediction network for lithium battery health state estimation | |
CN116609668B (en) | Lithium ion battery health state and residual life prediction method | |
CN111983474A (en) | Lithium ion battery life prediction method and system based on capacity decline model | |
CN114966436A (en) | Lithium battery state of charge prediction method, device, equipment and readable storage medium | |
CN112580211A (en) | Lead-acid storage battery SOH estimation method based on SA and ANN algorithm | |
CN115994441A (en) | Big data cloud platform online battery life prediction method based on mechanism information | |
CN109346787A (en) | A kind of electric automobile power battery adaptive optimization charging method | |
CN114460481A (en) | Energy storage battery thermal runaway early warning method based on Bi-LSTM and attention mechanism | |
Ganesh et al. | Prediction of residual energy in batteries using CNN-BiGRU and attention mechanism model | |
Kuang et al. | State-of-charge estimation hybrid method for lithium-ion batteries using BiGRU and AM co-modified Seq2Seq network and H-infinity filter | |
CN113721149A (en) | Lithium battery capacity prediction method based on semi-supervised transfer learning | |
CN117169743A (en) | Battery health state estimation method and device based on partial data and model fusion | |
CN116338501B (en) | Lithium ion battery health detection method based on neural network prediction relaxation voltage | |
CN116679208A (en) | Lithium battery residual life estimation method | |
CN116029183A (en) | Power battery temperature prediction method based on iPSO-LSTM model |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |