CN114720882B - Reconstruction method of maximum capacity fading curve of lithium ion battery - Google Patents

Reconstruction method of maximum capacity fading curve of lithium ion battery Download PDF

Info

Publication number
CN114720882B
CN114720882B CN202210549795.6A CN202210549795A CN114720882B CN 114720882 B CN114720882 B CN 114720882B CN 202210549795 A CN202210549795 A CN 202210549795A CN 114720882 B CN114720882 B CN 114720882B
Authority
CN
China
Prior art keywords
curve
capacity
lithium ion
ion battery
output
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202210549795.6A
Other languages
Chinese (zh)
Other versions
CN114720882A (en
Inventor
孙立
杜建成
苏志刚
钱俊良
童雨晨
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Liyang Research Institute of Southeast University
Original Assignee
Liyang Research Institute of Southeast University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Liyang Research Institute of Southeast University filed Critical Liyang Research Institute of Southeast University
Priority to CN202210549795.6A priority Critical patent/CN114720882B/en
Publication of CN114720882A publication Critical patent/CN114720882A/en
Application granted granted Critical
Publication of CN114720882B publication Critical patent/CN114720882B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/385—Arrangements for measuring battery or accumulator variables
    • G01R31/387—Determining ampere-hour charge capacity or SoC
    • G01R31/388—Determining ampere-hour charge capacity or SoC involving voltage measurements
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • 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)
  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Charge And Discharge Circuits For Batteries Or The Like (AREA)
  • Secondary Cells (AREA)

Abstract

The invention discloses a reconstruction method of a maximum capacity fading curve of a lithium ion battery, which comprises the following steps: s1: based on the existing accelerated aging data set, performing data processing to obtain a relation curve between the incremental capacity of the lithium ion battery and the voltage change in a single charging process; s2: determining input and output variables of a neural network according to a curve of the relationship between incremental capacity and voltage change in a single charging process, substituting accelerated aging test data into the neural network, and constructing a basic model for reconstructing a decay curve of the maximum capacity of the lithium ion battery; s3: selecting a reference capacity point according to the basic model and the charging and discharging data under the normal industrial use condition, and establishing a migration model; s4: and the model after the migration is used for reconstructing a maximum capacity fading curve of the lithium ion battery during normal aging, and carrying out error analysis. The method has the advantages of less data requirement, high precision and small error.

Description

Reconstruction method of maximum capacity fading curve of lithium ion battery
Technical Field
The invention relates to the field of lithium ion battery power supplies, in particular to a reconstruction method of a lithium ion battery storage capacity decline curve based on a neural network and a migration model.
Background
Lithium batteries have been widely used in vehicles, power supplies, secondary charging and energy storage devices by virtue of their advantages of high energy density, long service life, low self-discharge rate and cleanliness and reliability. The existing research results show. The percentage of lithium batteries in electrochemical energy storage is the largest, reaching 86%. In combination with the current requirements for energy conservation and emission reduction and the continuous development of comprehensive energy systems, lithium batteries play an increasingly important role in the systems, and the performance and service life of the lithium batteries also become more and more of the aspects of concern.
At present, the evaluation of the state of charge and the state of health of the lithium battery mainly depends on SOC and SOH, the research on SOC has been developed for a long time, and SOH has more profound and realistic significance as an evaluation index of the state of health of the lithium battery. For example, in the process of charge and discharge cycle of a lithium battery, the battery is worn, and the aging process of the battery is aggravated continuously. When the battery is aged to some extent, there is a great safety risk if it continues to operate, which may cause an accident. Therefore, the research on the SOH is very practical, and the accurate SOH estimation can provide a basis for the estimation of the health state of the lithium battery and the replacement of the lithium battery. Therefore, how to accurately estimate the SOH of the lithium battery is a very important research.
The conventional lithium battery SOH prediction method mainly comprises an electrochemical model and a data driving model, various aging conditions in a lithium battery do not need to be considered in the data driving method, and the method is a relatively popular lithium battery health state prediction method, but the conventional data driving method needs a large amount of accurate laboratory data as support to obtain a relatively sufficient training set and a relatively sufficient test set. Current experimental dataset acquisition faces a number of problems such as: the time of charge-discharge cycle is long, and the experiment time required by the complete aging of one battery is long; the test environment of the laboratory test is single, and the multi-working condition problem in the real use environment of the lithium ion battery cannot be well simulated.
At present, with the development of the electric automobile industry, lithium ion batteries are widely applied to electric automobiles, and the electric automobiles need to be charged and discharged every day, so that a large amount of relevant industrial data sets can be generated. However, the electric vehicle may have the problem of incomplete charging and discharging in the charging and discharging process, the generated data set is not complete enough, and if the industrial data set is used to replace laboratory data, the problems need to be solved urgently.
Disclosure of Invention
In order to solve the problems, the invention provides a reconstruction method of a lithium ion battery maximum capacity fading curve based on a neural network and a migration model, which has low data requirement and high prediction precision.
In order to achieve the purpose, the invention is realized by the following technical scheme:
the invention relates to a reconstruction method of a lithium ion battery maximum capacity decline curve based on a neural network and a migration model, which comprises the following steps:
s1: based on the existing accelerated aging data set, performing data processing to obtain a relation curve between the incremental capacity of the lithium ion battery and the voltage change in a single charging process;
s2: determining input and output variables of a neural network according to a curve of the relationship between incremental capacity and voltage change in a single charging process, substituting accelerated aging data into the neural network, and constructing a basic model for reconstructing a decay curve of the maximum capacity of the lithium ion battery;
s3: selecting a small number of reference capacity points according to the basic model and the charging and discharging data under the normal industrial use condition, and establishing a migration model;
s4: and the model after the migration is used for reconstructing a maximum capacity fading curve of the lithium ion battery during normal aging, and carrying out error analysis.
The invention is further improved in that: in S1, calculating a relation curve between incremental capacity of a lithium battery and terminal voltage change in a single charging process by using the capacity obtained by the lithium battery in accelerated aging constant current charging, the number of charging and discharging cycles, the charging voltage in the constant current charging process, the charging current in the constant current charging process and the charging capacity in the constant current charging process, wherein the incremental capacity is expressed as:
Figure GDA0003997362590000021
where Q refers to the charged capacity of the battery during a certain period of time, V refers to the terminal voltage of the battery during charging, Δ k Is a finite difference interval, and k is a representative k-th charge-discharge cycle.
The invention is further improved in that: s2, training a neural network to obtain a basic model, wherein the neural network has a three-layer structure of an input layer, a hidden layer and an output layer, wherein the input layer comprises four inputs which are sequentially peak values IC of an IC curve pk Terminal voltage value V corresponding to IC curve peak value pk 、(V pk Area C enclosed by IC curve in +/-10 mV) interval A1 IC curve and 85% IC pk Area C enclosed between values A2 The activation function of each neuron in the hidden layer is a poslin function, the activation function in the output layer is a purlin function, a bias node is arranged in the hidden layer, the value of the node is 1, and the output function of the finally constructed basic model is as follows:
Figure GDA0003997362590000031
where h is the output layer excitation function, f is the hidden layer excitation function, W 1 Is the weight of random initialization input to the hidden layer, k is the number of neurons in the hidden layer, W 2 The weights are the weights of random initialization from the hidden layer to the output layer, i is the number of neurons in the output layer, and the two weights are continuously corrected by a rainlm method.
The invention is further improved in that: (V) pk Area C enclosed by IC curve in +/-10 mV) interval A1 The expression of (a) is:
Figure GDA0003997362590000032
IC＝y＝g(v) (4)
wherein, deltav is a voltage interval and is 10mv.
The invention is further improved in that: IC curve and 85% IC pk Area enclosed between values C A2 The expression of (a) is:
Figure GDA0003997362590000033
wherein v is l Is the point to the left of the intersection of 85% of the straight line with the IC curve, v r Refers to the point to the right of the intersection of 85% of the straight line with the IC curve.
The invention is further improved in that: selecting a small number of normal aging data reference points, substituting normal aging input data into a basic neural network model to obtain biased output and measured accurate output, and performing linear interpolation to obtain a migration model based on an interpolation method, wherein the output of the migration model is as follows:
Figure GDA0003997362590000034
wherein:
Figure GDA0003997362590000035
wherein,
Figure GDA0003997362590000036
refers to the capacity of the Nth reference capacity point, l N It is referred to as the N-th cycle,
Figure GDA0003997362590000037
refers to the biased output of bringing the collected data into the underlying neural network model.
The invention is further improved in that: and S4, comparing the sample capacity output obtained by calculating the migration model with the real capacity output, and calculating RMSE or MSE to obtain the prediction precision.
The invention has the beneficial effects that: the method can establish a basic neural network model by a small amount of experimental data and training samples, and reconstruct the maximum capacity fading curve of a certain batch of lithium ion batteries by using a migration model and a related industrial data set in the electric automobile industry. Data can be continuously collected during the use of the batch of lithium ion batteries to update the maximum capacity fading curve. The method has the advantages of less data requirement, high prediction precision, strong real-time performance and good convergence.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention.
FIG. 2 is a graph showing the IC-V curve of the present invention.
FIG. 3 is a schematic diagram of the basic neural network constructed by the present invention.
FIG. 4 is a migration diagram of the migration model of the present invention.
FIG. 5 is a graph of the predicted results of the neural network incorporating the migration model of the present invention.
Detailed Description
The method of the present invention will be described in further detail with reference to the accompanying drawings and examples;
as shown in fig. 1, the method for reconstructing the maximum capacity fading curve of the lithium ion battery based on the neural network and the migration model comprises the following specific steps:
s1: based on the existing accelerated aging data set in the database, carrying out data processing to obtain a relation curve between the incremental capacity of the lithium ion battery and the voltage change in a single charging process;
s2: determining input and output variables of a neural network according to a curve of the relationship between incremental capacity and voltage change in a single charging process, substituting accelerated aging data into the neural network, and constructing a basic model for reconstructing a decay curve of the maximum capacity of the lithium ion battery;
s3: selecting a small number of reference capacity points according to the basic model and the charging and discharging data under the normal industrial use condition, and establishing a migration model;
s4: and the model after the migration is used for reconstructing a maximum capacity fading curve of the lithium ion battery during normal aging, and carrying out error analysis.
The data in the database comprises the capacity obtained by constant current charging of the lithium battery under the conditions of accelerated aging and normal aging, the times of charge-discharge cycles, the charging voltage in the constant current charging process, the charging current in the constant current charging process and the charging capacity in the constant current charging process. In the step S1, 50 groups of accelerated aging experimental data are used for calculation, and a relation curve between the incremental capacity of the lithium battery and the terminal voltage change in a single charging process is obtained.
The computational expression for incremental capacity is:
Figure GDA0003997362590000041
where Q refers to the charged capacity of the battery for a certain period of time, V refers to the terminal voltage of the battery, Δ k Is the finite difference interval and k is the represented kth cycle.
The specific operation of the step S2 is as follows:
s2.1: determining input and output variables of the neural network according to the constructed Incremental Capacity (IC) and voltage change relation curve;
s2.2: determining the structure of a neural network, and training the neural network according to the experimental data of accelerated aging to obtain a basic model; the structure of the neural network is shown in fig. 3, and the neural network has three layers of structures, namely an input layer, a hidden layer and an output layer. Wherein the input layer comprises four inputs: IC (integrated circuit) pk 、V pk 、C A1 、C A2 . Wherein the first one is input to IC pk Is the peak value of the curve of Incremental Capacity (IC) versus voltage variation, the second input V pk Is incremental capacity: (IC) and the terminal voltage value corresponding to the peak value of the voltage change relation curve, and the third input is C A1 Is a number V pk The area enclosed by the Incremental Capacity (IC) and voltage variation curve in the range of +/-10 mV, and the fourth input is C A2 The Incremental Capacity (IC) versus voltage curve is 85% IC pk The area enclosed between the values.
The specific determination steps of the four inputs are:
1): first, the first input IC is determined pk Is the IC value corresponding to the peak of the IC curve;
2): then determining a second input V pk Is the corresponding terminal voltage value at the peak value of the IC curve;
3): determining a third input C A1 Is a number V pk And (3) approximately calculating the area enclosed by the IC curve in the +/-10 mV interval by using a infinitesimal method, wherein the calculation formula is as follows:
IC＝y＝g(v) (2)
Figure GDA0003997362590000051
delta v is a voltage interval of 10mv; considering that the expression of g (v) is a form difficult to express by a continuous fixed function, a numerical integration method is used, and a complex product formula is used for approximate solution; the solving formula is as follows:
Figure GDA0003997362590000052
and m is equally divided on the integration interval, a trapezoidal integration formula is used for solving the integration in each cell, i corresponds to the ith equally dividing point, and h is the length of each cell. Finally, adding the obtained integrals in each cell to obtain the final integral to be obtained;
4): determining a fourth input C A2 Is the IC curve with 85% IC pk The area enclosed between the values.
Figure GDA0003997362590000053
v l Is the point to the left of the intersection of 85% of the straight line with the IC curve, v r Refers to the point to the right of the intersection of 85% of the straight line with the IC curve.
After determining the input and the output of the neural network, the hidden layer and the output layer need an activation function to predict the whole output, the activation function h (x) of the hidden layer is determined as a linear transfer function, and the transfer function f (x) of the output layer is a purlin function, that is:
h(x)＝x (6)
f(x)＝max{0,x} (7)
and setting a bias node in the hidden layer, wherein the value of the node is 1. The final output function is:
Figure GDA0003997362590000061
where h is the excitation function of the output layer, f is the excitation function of the hidden layer, W 1 Is the weight of random initialization input to the hidden layer, k is the number of neurons in the hidden layer, W 2 The weights are the weights of random initialization from the hidden layer to the output layer, i is the number of neurons in the output layer, and the two weights are continuously corrected by a rainlm method.
The step S3 specifically comprises the following steps:
and substituting the normal aging data obtained by industry into the deviation capacity output and the real capacity output obtained by the basic model to perform linear interpolation fitting to obtain the migration model. The obtained migration model can migrate the offset output obtained by the accelerated aging base model to the normal output obtained under the normal aging condition. The output of the migration model is:
Figure GDA0003997362590000062
wherein:
Figure GDA0003997362590000063
wherein,
Figure GDA0003997362590000064
refers to the capacity of the Nth reference capacity point, l N It is referred to as the N-th cycle,
Figure GDA0003997362590000065
refers to the biased output of bringing the collected data into the underlying neural network model. And obtaining a transfer function L (-) by using a linear interpolation method, and obtaining more accurate lithium ion battery capacity output through transfer.
In the step S4, the sample capacity output and the real capacity output obtained by using the migration model are compared by calculation, and an error is calculated. If RMSE or MSE is calculated, the correlation error value is obtained and analyzed.
The beneficial effects of the invention are: the method can utilize the curve of the relationship between the incremental capacity and the terminal voltage of the graph 2 to extract a large amount of effective information from a small amount of experimental data and training samples, establish a basic neural network model by utilizing the basic logic of the graph 3, and then utilize a migration model from the aging acceleration offset data to accurate measurement data represented by the graph 4 and a related industrial data set in the electric automobile industry to reconstruct the maximum capacity decline curve of the lithium ion batteries of a certain batch of models to obtain the reconstruction curve of the maximum capacity decline shown in the graph 5. Data can be continuously collected during the use of the batch of lithium ion batteries to update the maximum capacity fading curve. The method has the advantages of less data requirement, high prediction precision, strong real-time performance and good convergence.
The technical means disclosed in the invention scheme are not limited to the technical means disclosed in the above embodiments, but also include the technical scheme formed by any combination of the above technical features. It should be noted that numerous modifications and adaptations can be made by those skilled in the art without departing from the principles of the present invention. Such modifications and refinements are also considered to be within the scope of the present invention.

Claims (5)

1. The method for reconstructing the maximum capacity fading curve of the lithium ion battery is characterized by comprising the following steps of:
s1: based on the existing accelerated aging data set in the database, carrying out data processing to obtain a relation curve between the incremental capacity of the lithium ion battery and the voltage change in a single charging process;
s2: determining input and output variables of a neural network according to a curve of the relationship between incremental capacity and voltage change in a single charging process, substituting accelerated aging data into the neural network, and constructing a basic model for reconstructing a decay curve of the maximum capacity of the lithium ion battery;
s3: selecting a reference capacity point according to the basic model and the charging and discharging data under the normal industrial use condition, and establishing a migration model;
s4: the model after the migration is used for reconstructing a maximum capacity fading curve of the lithium ion battery during normal aging, and error analysis is carried out;
s2, training a neural network to obtain a basic model, wherein the neural network has a three-layer structure of an input layer, a hidden layer and an output layer, wherein the input layer comprises four inputs which are sequentially peak values IC of an IC curve pk Terminal voltage value V corresponding to IC curve peak value pk 、V pk Area C enclosed by IC curve in +/-10 mV interval A1 IC curve and 85% IC pk Area C enclosed between values A2 The activation function of each neuron in the hidden layer is a poslin function, the activation function in the output layer is a purlin function, a bias node is arranged in the hidden layer, the value of the node is 1, and the output function of the finally constructed basic neural network model is as follows:
Figure FDA0003969672280000011
where h is the output layer excitation function, f is the hidden layer excitation function, W 1 Is the weight of random initialization input to the hidden layer, k is the number of neurons in the hidden layer, W 2 From a hidden layer to an output layerI is the number of neurons in the output layer, and the two weights are continuously corrected by a rainlm method;
selecting a plurality of normal aging data reference points, substituting normal aging input data into a basic neural network model to obtain biased output and measured accurate output, and performing linear interpolation to obtain a migration model based on an interpolation method, wherein the output of the migration model is as follows:
Figure FDA0003969672280000012
wherein:
Figure FDA0003969672280000013
wherein,
Figure FDA0003969672280000014
refers to the capacity of the Nth reference capacity point, l N It is referred to as the N-th cycle,
Figure FDA0003969672280000015
refers to the biased output of bringing the collected data into the underlying neural network model.
2. The method for reconstructing the maximum capacity fading curve of the lithium ion battery according to claim 1, wherein: in S1, calculating a relation curve between incremental capacity of a lithium battery and terminal voltage change in a single charging process by using the capacity obtained by the lithium battery in accelerated aging constant current charging, the number of charging and discharging cycles, the charging voltage in the constant current charging process, the charging current in the constant current charging process and the charging capacity in the constant current charging process, wherein the incremental capacity is expressed as:
Figure FDA0003969672280000021
where Q refers to the charged capacity of the battery during a certain period of time, V refers to the terminal voltage of the battery during charging, Δ k Is a finite difference interval, and k is a representative k-th charge-discharge cycle.
3. The method for reconstructing the maximum capacity fading curve of the lithium ion battery according to claim 1, wherein: v pk Area C enclosed by IC curve in +/-10 mV interval A1 The expression of (a) is:
Figure FDA0003969672280000022
IC＝y＝g(v) (4)
wherein, deltav is a voltage interval and is 10mv.
4. The method for reconstructing the maximum capacity fading curve of the lithium ion battery according to claim 1, wherein: IC curve and 85% IC pk Area C enclosed between values A2 The expression of (c) is:
Figure FDA0003969672280000023
wherein v is l Is the point to the left of the intersection of 85% of the straight line with the IC curve, v r Refers to the point to the right of the intersection of 85% of the straight line with the IC curve.
5. The method for reconstructing the maximum capacity fading curve of the lithium ion battery according to claim 1, wherein: and S4, comparing the sample capacity output obtained by calculating the migration model with the real capacity output, and calculating RMSE or MSE to obtain the prediction precision.
CN202210549795.6A 2022-05-20 2022-05-20 Reconstruction method of maximum capacity fading curve of lithium ion battery Active CN114720882B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210549795.6A CN114720882B (en) 2022-05-20 2022-05-20 Reconstruction method of maximum capacity fading curve of lithium ion battery

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210549795.6A CN114720882B (en) 2022-05-20 2022-05-20 Reconstruction method of maximum capacity fading curve of lithium ion battery

Publications (2)

Publication Number Publication Date
CN114720882A CN114720882A (en) 2022-07-08
CN114720882B true CN114720882B (en) 2023-02-17

Family

ID=82231138

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210549795.6A Active CN114720882B (en) 2022-05-20 2022-05-20 Reconstruction method of maximum capacity fading curve of lithium ion battery

Country Status (1)

Country Link
CN (1) CN114720882B (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111638465A (en) * 2020-05-29 2020-09-08 浙大宁波理工学院 Lithium battery health state estimation method based on convolutional neural network and transfer learning
CN114114049A (en) * 2021-09-30 2022-03-01 北京交通大学 Lithium ion battery life prediction method based on sample migration
WO2022100230A1 (en) * 2020-11-16 2022-05-19 北京理工大学 Lithium-ion battery charging curve reconstruction and state estimation method based on artificial intelligence

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10393813B2 (en) * 2013-08-27 2019-08-27 The Regents Of The University Of Michigan On-board state of health monitoring of batteries using incremental capacity analysis

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111638465A (en) * 2020-05-29 2020-09-08 浙大宁波理工学院 Lithium battery health state estimation method based on convolutional neural network and transfer learning
WO2022100230A1 (en) * 2020-11-16 2022-05-19 北京理工大学 Lithium-ion battery charging curve reconstruction and state estimation method based on artificial intelligence
CN114114049A (en) * 2021-09-30 2022-03-01 北京交通大学 Lithium ion battery life prediction method based on sample migration

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
Analysis of Optimal Machine Learning Approach for Battery Life Estimation of Li-Ion Cell;PRAKASH VENUGOPAL 等;《IEEE Access》;20211125;第9卷;第159616-159626页 *
Health State Prediction of Lithium Ion Battery Based On Deep Learning Method;Sheng Lu 等;《IOP Conference Series: Materials Science and Engineering》;20201231;第1-8页 *
Lithium-ion battery capacity estimation—A pruned convolutional neural network approach assisted with transfer learning;Yihuan Li 等;《Applied Energy》;20210104;第1-13页 *
Time-Frequency Image Analysis and Transfer Learning for Capacity Prediction of Lithium-Ion Batteries;Ma’d El-Dalahmeh 等;《Energies》;20201016;第1-19页 *
Transfer learning LSTM model for battery useful capacity fade prediction;Aniruddha Gupta 等;《2021 24th International Conference on Mechatronics Technology (ICMT)》;20220101;全文 *

Also Published As

Publication number Publication date
CN114720882A (en) 2022-07-08

Similar Documents

Publication Publication Date Title
Jiang et al. A review on the state of health estimation methods of lead-acid batteries
CN112034356B (en) GP-UKF-based online SOH estimation method for electric vehicle power battery
CN111323719A (en) Method and system for online determination of health state of power battery pack of electric automobile
CN105974327A (en) Lithium battery pack SOC prediction method based on neural network and UKF
CN111426957B (en) SOC estimation optimization method for power battery under simulated vehicle working condition
CN114861527A (en) Lithium battery life prediction method based on time series characteristics
CN112051511A (en) Power battery state of health estimation method and system based on multichannel technology
CN112557907A (en) SOC estimation method of electric vehicle lithium ion battery based on GRU-RNN
CN115308606B (en) Lithium ion battery health state estimation method based on adjacent characteristics
CN115186579A (en) Lithium battery health state estimation method based on capacity increment variable quantity curve
CN112098874A (en) Lithium ion battery electric quantity prediction method considering aging condition
CN114839536A (en) Lithium ion battery health state estimation method based on multiple health factors
CN113687242A (en) Lithium ion battery SOH estimation method for optimizing and improving GRU neural network based on GA algorithm
CN110673037A (en) Battery SOC estimation method and system based on improved simulated annealing algorithm
CN115389946A (en) Lithium battery health state estimation method based on isobaric rise energy and improved GRU
Bak et al. Accurate estimation of battery SOH and RUL based on a progressive lstm with a time compensated entropy index
CN115963407A (en) ICGWO (intensive care unit) optimization ELM (element-based robust model) based lithium battery SOC estimation method
CN116298936A (en) Intelligent lithium ion battery health state prediction method in incomplete voltage range
CN114861545A (en) Lithium battery SOP online estimation method based on RNN neural network and multi-parameter constraint
Xu et al. State-of-charge estimation and health prognosis for lithium-ion batteries based on temperature-compensated Bi-LSTM network and integrated attention mechanism
CN114355218A (en) Lithium ion battery charge state prediction method based on multi-feature quantity screening
CN110232432B (en) Lithium battery pack SOC prediction method based on artificial life model
CN114720882B (en) Reconstruction method of maximum capacity fading curve of lithium ion battery
CN116679213A (en) SOH estimation method for electric vehicle power battery based on integrated deep learning
Li et al. Evaluation and analysis of circuit model for lithium batteries

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