CN113504482B - Lithium ion battery health state estimation and life prediction method considering mechanical strain - Google Patents

Lithium ion battery health state estimation and life prediction method considering mechanical strain Download PDF

Info

Publication number
CN113504482B
CN113504482B CN202110681576.9A CN202110681576A CN113504482B CN 113504482 B CN113504482 B CN 113504482B CN 202110681576 A CN202110681576 A CN 202110681576A CN 113504482 B CN113504482 B CN 113504482B
Authority
CN
China
Prior art keywords
battery
lithium ion
ion battery
wavelet
threshold
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
CN202110681576.9A
Other languages
Chinese (zh)
Other versions
CN113504482A (en
Inventor
曲杰
韩孝耀
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.)
South China University of Technology SCUT
Original Assignee
South China University of Technology SCUT
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 South China University of Technology SCUT filed Critical South China University of Technology SCUT
Priority to CN202110681576.9A priority Critical patent/CN113504482B/en
Publication of CN113504482A publication Critical patent/CN113504482A/en
Application granted granted Critical
Publication of CN113504482B publication Critical patent/CN113504482B/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]
    • G01R31/392—Determining battery ageing or deterioration, e.g. state of health
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
    • G01D21/00—Measuring or testing not otherwise provided for
    • G01D21/02—Measuring two or more variables by means not covered by a single other subclass
    • 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]
    • G01R31/3644—Constructional arrangements
    • G01R31/3648—Constructional arrangements comprising digital calculation means, e.g. for performing an algorithm
    • 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]
    • G01R31/378—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC] specially adapted for the type of battery or accumulator
    • 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
    • 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
    • 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]
    • G01R31/389—Measuring internal impedance, internal conductance or related variables
    • 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]
    • G01R31/396—Acquisition or processing of data for testing or for monitoring individual cells or groups of cells within a battery
    • 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

Abstract

The invention discloses a lithium ion battery health state estimation and life prediction method considering mechanical strain, which comprises the following steps: s1, carrying out an aging degradation experiment of the lithium ion battery to obtain external characteristic parameters and degradation data of the battery; s2, denoising external characteristic parameters by using a multi-resolution wavelet denoising model, and extracting typical characteristic parameters of denoised experimental data by using a characteristic parameter extraction model based on wavelet analysis; s3, establishing a typical characteristic parameter state process model; s4, establishing a battery typical characteristic parameter-aging degradation mapping relation model; s5, estimating the health state of the lithium ion battery on line by using an on-line typical characteristic parameter and a battery typical characteristic parameter-aging degradation mapping relation model; and S6, online predicting the remaining service life of the battery by using the online typical characteristic parameters, historical online data, a typical characteristic parameter state process model and a battery typical characteristic parameter-aging degradation mapping relation model.

Description

Lithium ion battery health state estimation and life prediction method considering mechanical strain
Technical Field
The invention relates to the technical field of lithium ion batteries, in particular to a lithium ion battery health state estimation and life prediction method considering mechanical strain.
Background
The secondary lithium ion battery has the advantages of high energy and capacity density, long cycle life, no memory effect and the like, and is widely applied to products such as electric automobiles, communication base stations, mobile phones, notebook computers, aviation emergency power supplies and the like. However, during the use process, the lithium ion battery inevitably has aging degradation phenomenon, which is mainly manifested as degradation of battery capacity, increase of internal resistance, degradation of power performance and the like. The aging and fading mechanism of the battery is complex, and the aging process is influenced by a plurality of factors such as temperature, charge and discharge multiplying power, service time and the like. Accurate estimation of the state of health of a lithium ion battery is one of the important functions of a battery management system. Any system that uses a lithium-ion battery as a power source must know the energy that the battery can store and the power that can be provided at any point in time.
At present, when the lithium ion battery is subjected to online state of health estimation and life prediction, parameters which can be monitored by a battery management system and can reflect the state of health of the battery are limited, wherein external characteristics which can be directly measured are terminal voltage, current and temperature.
The patent literature (Chinese patent: CN103308864B, 2015-06-24) adopts temperature, voltage, current, resistance, humidity, salt fog and vibration as monitoring variables for estimating the health state and predicting the service life of the secondary battery, although the method considers various factors to improve the prediction accuracy, the modeling process becomes complicated by using various variables, the uncertainty of the model becomes large, in addition, the humidity, the salt fog and the vibration are covariates, no coupling exists with the electrochemical reaction in the battery, and the space for improving the prediction accuracy is limited; journal literature (John canarella, craig b. Arnold. State of health and charge measures in lithium-ion batteries using mechanical stress [ J ]. Journal of Power Sources,2014,269.) proposes a method of estimating the state of health of a battery using the stress monitored during the charge and discharge of a lithium battery in an optimal preload state, but this method requires a lot of experiments in the early stage of determining the optimal preload of the battery, and the optimal preload may change depending on the size and type of the battery, and in addition, in practical application, the batteries are mostly used in groups, and it is not easy to monitor the mechanical stress of the battery.
Therefore, there is a need to add other parameters that are easy to measure online, and should be coupled with the electrochemical degradation mechanism of the lithium ion battery, and the accuracy of the estimation of the state of health and the lifetime prediction of the lithium ion battery can be further improved by combining with other parameters.
Disclosure of Invention
In order to overcome the technical problem that the battery health state estimation and service life prediction accuracy of the conventional lithium ion battery management system is low, the invention provides a lithium ion battery health state estimation and service life prediction method considering mechanical strain.
The invention is realized by at least one of the following technical schemes.
A lithium ion battery state of health estimation and life prediction method that accounts for mechanical strain, the method comprising the steps of:
s1, performing an aging degradation experiment on the lithium ion battery under different charging and discharging working conditions to obtain mechanical strain, voltage, current and temperature data of the lithium ion battery and corresponding aging degradation data of the battery;
s2, denoising the experimental data obtained in the step S1 by using a multi-resolution wavelet denoising model, and extracting typical characteristic parameters of the denoised experimental data by using a characteristic parameter extraction model based on wavelet analysis;
s3, establishing a typical characteristic parameter state process model by using the typical characteristic parameters extracted in the step S2 and identifying model parameters;
s4, establishing a typical characteristic parameter-aging degradation mapping relation model of the lithium ion battery according to the battery aging degradation data obtained in the step S1 and the typical characteristic parameters obtained in the step S2;
s5, inputting typical characteristic parameters obtained by using the external characteristic parameters of the battery monitored in real time on line and through a multi-resolution wavelet denoising model and a characteristic parameter extraction model based on wavelet analysis into a typical characteristic parameter-aging degradation mapping relation model of the lithium ion battery, and estimating the health state of the current lithium ion battery on line;
s6, inputting typical characteristic parameters obtained by using external characteristic parameters of the battery monitored in real time on line through a multi-resolution wavelet denoising model and a characteristic parameter extraction model based on wavelet analysis into a typical characteristic parameter state process model by combining historical on-line data; and inputting the result into a typical characteristic parameter-aging degradation mapping relation model of the lithium ion battery, and predicting the residual service life of the lithium ion battery on line.
Preferably, the lithium ion battery is a secondary lithium ion battery pack formed by combining rechargeable and recyclable secondary lithium ion battery cells, rechargeable and recyclable secondary lithium ion battery cells in a series-parallel connection mode, and a secondary lithium ion battery system constructed by the secondary lithium ion battery pack in a module-assembling mode.
Preferably, the step S1 specifically includes the following steps:
s11, presetting at least two battery charging and discharging test working conditions, and presetting one or more aging decline index failure threshold values of the batteries;
s12, testing at least two lithium ion batteries according to a preset charging and discharging test working condition, and simultaneously recording mechanical strain, voltage, current and temperature data of the batteries and corresponding battery aging decline index data;
s13, judging whether one or more aging decline indexes of the current lithium ion battery reach a failure threshold value; if yes, stopping the test; if not, the step S12 is repeated.
Preferably, the aging degradation index in step S11 includes more than one of a ratio of the maximum capacity of the battery currently in the experiment to the maximum capacity of the original battery, a ratio of the maximum energy of the battery currently in the experiment to the maximum energy of the original battery, a ratio of the maximum output power of the battery currently in the experiment to the maximum output power of the original battery, and a ratio of the internal resistance of the battery currently in the experiment to the internal resistance of the original battery.
Preferably, the mechanical strain of the battery in step S12 refers to the strains in two mutually perpendicular directions, i.e., the length and the width of the lithium ion battery cell, and the resultant strains in the two directions, which are generated by the electrochemical reaction during the charge and discharge of the prismatic winding type lithium ion battery cell or the prismatic stacking type lithium ion battery cell, on the plane formed by the maximum area of the battery.
Preferably, the mechanical strain of the battery is measured by using a biaxial strain gauge, and the position of the biaxial strain gauge is arranged at the central position on a plane formed by the largest area of the lithium ion battery.
Preferably, the parameters in the multiresolution wavelet denoising model include a wavelet function, a decomposition layer number, a threshold and a threshold, the wavelet function, the decomposition layer number, the threshold and the threshold are obtained through optimization, the optimization problem converts the determination problem of the wavelet type, the decomposition layer number, the threshold and the threshold processing method into a mixed integer optimization problem, the objective function is determined through a cross-check method, and the optimization problem is solved through a genetic algorithm.
Preferably, the optimization problem is established as follows:
let x i I =1,2, …, N, wherein i represents a signal sequence number and N represents a signal length;
firstly, generating an even point signal fe and an odd point signal fo from an experimental data signal according to the parity of i, and then smoothing the odd point signal fo to obtain a uniform estimation value fe * Expressed as:
Figure BDA0003122830690000031
wherein N is an even number, and the even point signal fe is subjected to two-point smoothing to obtain the even estimation value fo of the odd point signal * (ii) a Respectively reducing the even point signal fe and the odd point signal foIf the noise is found to be the even dot signal approximation fe and the odd dot signal approximation fo, the estimated value of the Relative Squared Error (RSE) is expressed as follows:
Figure BDA0003122830690000032
in the wavelet denoising parameters, both wavelet types and threshold processing strategies are non-numerical variables, a wavelet type candidate set and a threshold processing strategy candidate set are established, and an integer variable I is respectively used ψ And I ρ Representing wavelet type candidate set and threshold processing strategy candidate set elements, selecting a hard threshold processing strategy and a soft threshold processing strategy by the threshold processing strategy, I ρ The value and corresponding threshold processing strategy is:
Figure BDA0003122830690000041
assuming that the number of signals in the sequence of approximated signals at the lowest resolution is not lower than the minimum number of signals N in the sequence of approximated signals at the lowest resolution 0 Based on the multi-resolution wavelet denoising theory, the number of decomposition layers J in wavelet denoising is an integer and needs to satisfy the constraint shown as the following formula:
Figure BDA0003122830690000042
determining the upper and lower limits of the threshold based on a universal threshold method: firstly based on a general threshold method and an original signal x i Calculating the threshold corresponding to each wavelet in the wavelet candidate set and constructing a vector
Figure BDA0003122830690000043
Wherein T is Iψ Denotes the reference number I ψ The second to find the upper limit T min = minT and lower limit T max = maxT, aT last respectively aT min And beta T max As upper and lower limits of the parameter threshold T, whichWherein the coefficients a and beta are 0 < a < 1 and beta > 1, the problem of determining the wavelet type, the decomposition layer number, the threshold processing strategy and the threshold of the denoising parameter in the wavelet denoising model is converted into a mixed integer optimization problem bearing simple constraints as shown below:
Figure BDA0003122830690000044
where s.t. means constraint, Z denotes integer set, UI ψ Representing the number of wavelets, UI, in a wavelet candidate set ρ Representing the number of thresholding strategies, N 0 Representing the minimum number of signals in the sequence of approximation signals at the lowest resolution, let N 0 ＝8，aT min And beta T max Representing the upper and lower limits of the threshold T.
Preferably, the process of extracting the typical characteristic parameters of the denoised experimental data by the characteristic parameter extraction model based on wavelet analysis is as follows:
calculating an envelope feature function (DF) by multi-scale envelope superposition of Continuous Wavelet Transform (CWT) coefficients, the DF being:
Figure BDA0003122830690000045
in the formula, α represents a continuous wavelet analysis scale, n a Is the number of scales, τ is the continuous wavelet analysis time transfer window size, n represents the total number of samples in the discrete signal, and the envelope function of the Continuous Wavelet Transform (CWT) coefficients is
Figure BDA0003122830690000051
Wherein W s (α, τ) represents the time-frequency domain resulting from subjecting the original signal sequence to successive wavelet transforms,
Figure BDA0003122830690000052
is the result of the hilbert transform of the continuous wavelet transform coefficients;
calculating an operating energy proportion using the obtained envelope feature function DFER 1 Said operating energy ratio ER 1 Comprises the following steps:
Figure BDA0003122830690000053
wherein L is the length of the energy collection window before and after the time transfer window size is tau;
using the resulting operating energy ratio ER 1 Obtaining a characteristic parameter ER characterizing aging-related chemical reactions inside a lithium ion battery 2 The ER 2 Comprises the following steps:
ER 2 (τ)＝ER 1 (τ)|DF(α)| (8)。
preferably, the remaining service life of the lithium ion battery in step S6 includes a remaining calendar life of the battery in terms of calendar time and a remaining cycle life in terms of the number of charging and discharging times.
Compared with the prior art, the invention has the beneficial effects that: according to the invention, the mechanical strain of the battery which is easy to monitor online in real time is brought into the estimation of the health state and the life prediction of the lithium ion battery, so that more effective information can be provided for a battery management system, the accuracy and the reliability of the estimation of the health state and the life prediction of the lithium ion battery are improved, the use and maintenance efficiency of the battery is improved, and the service life of the battery is prolonged.
Drawings
FIG. 1 is a flow chart of a method and system for estimating the state of health and predicting the life of a lithium ion battery according to an embodiment of the present invention;
fig. 2 is a schematic flowchart of step S1 of a lithium ion battery state of health estimation and life prediction method according to an embodiment of the present invention;
fig. 3 is a typical strain distribution diagram of the lithium ion battery according to the embodiment of the present invention in the front view direction in the full charge state;
FIG. 4 is a diagram of a typical strain profile of a lithium-ion battery of an embodiment of the present invention in a side view direction at full charge;
fig. 5 is a typical strain distribution diagram of a lithium ion battery according to an embodiment of the present invention in a top view direction under a full charge state;
fig. 6 is a schematic diagram of measuring mechanical strain of the lithium ion battery according to the embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and embodiments.
It is to be noted that, unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.
Fig. 1 shows a flow chart of a method and a system for estimating a state of health and predicting a lifetime of a lithium ion battery considering mechanical strain, which includes the following steps:
s1, performing an aging degradation experiment on the lithium ion battery under different charging and discharging working conditions to obtain mechanical strain, voltage, current and temperature data of the lithium ion battery and corresponding aging degradation data of the battery;
the lithium ion battery is a secondary lithium ion battery pack formed by combining secondary lithium ion battery monomers capable of being used in a charging and discharging cycle manner and secondary lithium ion battery monomers capable of being used in a charging and discharging cycle manner in a series-parallel connection manner, and a secondary lithium ion battery system formed by constructing the secondary lithium ion battery pack in a module-assembling manner. That is, any battery system constructed by connecting lithium ion battery cells in any connection manner is within the scope of the present invention.
As shown in fig. 2, step S1 specifically includes the following steps:
s11, presetting at least two battery charging and discharging test working conditions, and presetting one or more aging decline index failure threshold values of the batteries;
the charging and discharging working conditions of the battery can reflect the running characteristics of the lithium ion battery under the actual working conditions, and the types of the charging and discharging working conditions of the battery are not lower than two. Furthermore, the charging and discharging working condition of the battery is a constant-current constant-voltage charging working condition and a constant-current discharging working condition; constant-current constant-voltage charging, namely a constant-current discharging working condition with pulse discharging; the charging and discharging working conditions are equivalently converted from the driving working conditions of the automobile.
The battery charging and discharging working condition can also be a working condition developed and designed by technicians according to the actual operation working condition of the battery.
It should be noted that the aging degradation index failure threshold refers to an index value to be referred to when a battery charge/discharge cycle is performed to determine whether an experiment needs to be terminated.
The battery aging degradation index in step S11 includes one or more of a ratio of a maximum capacity of the battery currently in the experiment to a maximum capacity of the original battery, a ratio of a maximum energy of the battery currently in the experiment to a maximum energy of the original battery, a ratio of a maximum output power of the battery currently in the experiment to a maximum output power of the original battery, and a ratio of an internal resistance of the battery currently in the experiment to an internal resistance of the original battery.
S12, testing at least two lithium ion batteries according to a preset charging and discharging test working condition, and simultaneously recording mechanical strain, voltage, current and temperature data of the batteries and corresponding battery aging decline index data;
the mechanical strain of the battery in step S12 refers to the strains along two mutually perpendicular directions, i.e., the length and the width of the lithium ion battery cell, and the resultant strains in the two directions, which are generated by the electrochemical reaction of the square wound lithium ion battery cell or the square stacked lithium ion battery cell during the charging and discharging processes, on the plane formed by the maximum area of the battery.
Fig. 3, 4 and 5 show typical strain distribution diagrams of the lithium ion battery at different viewing angles when the lithium ion battery is fully charged, and it is obvious that the strain of the lithium ion battery at the top view angle is more changed than that at other viewing angles, particularly reaches the maximum at the central area, and strains are generated in the x-axis direction and the y-axis direction. From the results analyzed in fig. 3 to 5, the lithium ion battery mechanical strain measurement method shown in fig. 6 was set. It should be noted that the views indicated in fig. 3 to fig. 5 all correspond to the coordinate system in fig. 6 one-to-one, and are one or more viewing angles of the lithium ion battery coordinate system in fig. 6.
As shown in fig. 6, the mechanical strain of the battery was measured using a biaxial strain gauge, the position of which was arranged at the center on the plane constituted by the largest area of the lithium ion battery.
S13, judging whether one or more aging decline indexes of the current lithium ion battery reach a failure threshold value; if yes, stopping the test; if not, step S12 is repeated.
If the adopted aging degradation index is the ratio of the current maximum capacity of the battery to the original maximum capacity when the battery is not used, and the failure threshold value is preset to be 75%, when the ratio of the current maximum capacity of the tested lithium ion battery to the original maximum capacity is less than 75%, stopping the test; and when the ratio of the maximum capacity of the currently tested lithium ion battery to the original maximum capacity is more than or equal to 75%, repeating the step S12. In other embodiments of the invention, the failure threshold may of course also be set to 80% or 85%.
S2, denoising the experimental data obtained in the step S1 by using a multi-resolution wavelet denoising model, and extracting typical characteristic parameters of the denoised experimental data by using a characteristic parameter extraction model based on wavelet analysis;
the parameters in the multi-resolution wavelet denoising model in the step S2 include a wavelet function, a decomposition layer number, a threshold value and a threshold value, the processing methods of the wavelet function, the decomposition layer number, the threshold value and the threshold value are obtained through an optimization method, the optimization problem converts the determination problem of the wavelet type, the decomposition layer number, the threshold value and the threshold value processing method into a mixed integer optimization problem, the objective function is determined through a cross check method, and the optimization problem is solved through a genetic algorithm.
Specifically, the process for establishing the multi-resolution wavelet denoising model parameter optimization problem is as follows:
the objective function is determined by a cross-check method.
Let x i I =1,2, …, N for the original signal, where i denotes the signal sequence number and N denotes the signal length.
Firstly, the first step is toGenerating an even point signal fe and an odd point signal fo according to the parity of the experimental data signal i, and then smoothing the odd point signal fo to obtain a uniform estimation value fe * Expressed as:
Figure BDA0003122830690000071
wherein N is an even number, and the even point signal fe is subjected to two-point smoothing to obtain the even estimation value fo of the odd point signal * . If the even dot signal fe and the odd dot signal fo are respectively denoised to obtain an even dot signal approximation fe and an odd dot signal approximation fo, the estimated value of the Relative Squared Error (RSE) is as follows:
Figure BDA0003122830690000081
in the wavelet denoising parameters, the wavelet type and the threshold processing strategy are both non-numerical variables, so a wavelet type candidate set and a threshold processing strategy candidate set are established and respectively used as integer variables I ψ And I ρ Characterizing wavelet type candidate set and thresholding strategy candidate set elements, wherein the wavelet type candidate set I ψ The correspondence between values and wavelet types is shown in table 1.
TABLE 1 wavelet types in wavelet candidate set and their corresponding integers
Figure BDA0003122830690000082
The threshold processing strategy selects a hard threshold processing strategy and a soft threshold processing strategy. I.C. A ρ The value and corresponding threshold processing strategy is:
Figure BDA0003122830690000083
since the signal length N is limited, even at the lowest resolutionA certain resolution of the signal is required. Assuming that the number of signals in the sequence of approximated signals at the lowest resolution is not lower than the minimum number of signals N in the sequence of approximated signals at the lowest resolution 0 . Based on the multi-resolution wavelet denoising theory, the number of decomposition layers J in the wavelet denoising is an integer and needs to satisfy the constraint shown in the following formula:
Figure BDA0003122830690000084
since the optimal threshold in multi-resolution wavelet denoising is typically lower than the common threshold, the upper and lower limits of the threshold are determined based on the common thresholding method. Firstly based on a general threshold method and an original signal x i Calculating the threshold corresponding to each wavelet in the wavelet candidate set and constructing a vector
Figure BDA0003122830690000085
Wherein
Figure BDA0003122830690000086
Denotes the reference number I ψ The second to find the upper limit T min = minT and lower limit T max = maxT, aT last respectively aT min And beta T max As the upper and lower limits of the parameter threshold T, 0 < a < 1 and β > 1. The determination of the denoising parameter wavelet type, the number of decomposition layers, the thresholding strategy and the threshold in the wavelet denoising model can be converted into a mixed integer optimization problem subject to simple constraints as follows:
Figure BDA0003122830690000091
where s.t. means constraint, Z denotes integer set, UI ψ Representing the number of wavelets, UI, in a wavelet candidate set ρ Representing the number of thresholding strategies, N 0 Representing the minimum number of signals in the sequence of approximation signals at the lowest resolution, let N 0 ＝8，aT min And beta T max Representing the upper and lower limits of the threshold TBased on experience, coefficients a =0.2 and β =5 are set.
And solving the obtained optimization problem by adopting a genetic algorithm. In other embodiments of the present invention, other effective swarm intelligence optimization algorithms, such as a particle swarm algorithm and an ant colony algorithm, may also be employed; deterministic constraint optimization algorithms such as sequential quadratic programming algorithms and penalty function methods may also be used.
Specifically, the process of extracting the typical characteristic parameters of the denoised experimental data by the characteristic parameter extraction model based on the wavelet analysis is as follows:
after the experimental data are processed by a multi-resolution wavelet denoising model, calculating an envelope characteristic function DF by multi-scale envelope superposition of Continuous Wavelet Transform (CWT) coefficients, wherein the DF is as follows:
Figure BDA0003122830690000092
in the formula, α represents a continuous wavelet analysis scale, n a Is the number of scales, τ is the continuous wavelet analysis time transfer window size, n represents the total number of samples in the discrete signal, and the envelope function of the Continuous Wavelet Transform (CWT) coefficients is
Figure BDA0003122830690000093
Wherein W s And (alpha, tau) represents a time-frequency domain obtained by subjecting an original signal sequence to continuous wavelet transform,
Figure BDA0003122830690000094
is the result of the hilbert transform of the continuous wavelet transform coefficients;
calculating an operating energy ratio ER using the DF obtained above 1 Said operating energy ratio ER 1 Comprises the following steps:
Figure BDA0003122830690000095
wherein L is the length of the energy collection window before and after the time transfer window size is tau;
using ER obtained as described above 1 Calculating a characteristic parameter ER characterizing the aging-related chemical reactions inside a lithium ion battery 2 The ER 2 Comprises the following steps:
ER 2 (τ)＝ER 1 (τ)|DF(α)| (8)
s3, establishing a typical characteristic parameter state process model by using the typical characteristic parameters extracted in the step S2 and identifying model parameters;
in this embodiment, the typical characteristic parameter state process model is established by polynomial fitting, the model parameters to be identified are polynomial orders and polynomial coefficients, and the polynomial orders and the polynomial coefficients are determined by minimizing the model error. In other embodiments of the present invention, the model of the characteristic parameter state process may be established by fitting other functions, such as an exponential function, a trigonometric function, or a complex function of a combination of the two.
S4, establishing a typical characteristic parameter-aging degradation mapping relation model of the lithium ion battery by a mathematical method according to the aging degradation data of the lithium ion battery obtained in the step S1 and the typical characteristic parameters obtained in the step S2;
in this embodiment, the model of the lithium ion battery typical characteristic parameter-aging degradation mapping relationship is established by using a support vector regression algorithm, and in other embodiments of the present invention, the model of the lithium ion battery mechanical strain may also be established by using other data-driven algorithms, such as an artificial neural network algorithm, a sparse bayes learning method, and a fuzzy logic method.
S5, obtaining typical characteristic parameters ER by using external characteristic parameters of the battery monitored in real time on line and through a multi-resolution wavelet denoising model and a characteristic parameter extraction model 2 Inputting a typical characteristic parameter-aging decay mapping relation model of the lithium ion battery, and estimating the health state of the current lithium ion battery on line;
the health state of the battery refers to one or more of the ratio of the current maximum capacity of the battery to the maximum capacity of the original battery, the ratio of the maximum energy to the maximum energy of the original battery, the ratio of the maximum output power to the maximum output power of the original battery, and the ratio of the internal resistance to the internal resistance of the original battery.
S6, utilizing the external characteristic parameters of the battery monitored in real time on line and extracting the typical characteristic parameters ER from the model through a multi-resolution wavelet denoising model and a characteristic parameter extraction model based on wavelet analysis 2 Inputting a typical characteristic parameter state process model by combining historical online data; and inputting the result into a typical characteristic parameter-aging degradation mapping relation model of the lithium ion battery, and predicting the residual service life of the lithium ion battery on line. The remaining service life of the lithium ion battery comprises the remaining calendar life of the battery counted by calendar time and the remaining cycle life counted by charging and discharging times.
In this embodiment, the health state of the battery is the ratio of the current maximum capacity of the battery to the maximum capacity of the original battery; the remaining service life of the battery is the remaining cycle life counted by the number of charging and discharging. The specific processes of the lithium ion battery for on-line health state estimation and life prediction are as follows:
assume that the current number of charge and discharge times of the battery is M. Firstly, inputting the typical characteristic parameters extracted during the Mth charging and discharging into a lithium ion battery typical characteristic parameter-aging degradation mapping relation model, thereby obtaining the ratio of the maximum capacity of the current online battery to the maximum capacity of the original battery, namely completing the online health state estimation of the battery;
secondly, inputting a typical characteristic parameter in a typical characteristic parameter state process model according to the currently extracted typical characteristic parameter and combining historical online data to estimate to obtain a typical characteristic parameter during the M +1 th charging and discharging, inputting the typical characteristic parameter during the M +1 th charging and discharging into a lithium ion battery typical characteristic parameter-aging degradation mapping relation model, estimating a ratio of the maximum capacity of the battery during the M +1 th charging and discharging to the maximum capacity of the original battery, repeating the estimation process until the estimated ratio of the maximum capacity of the battery during the M + k +1 th charging and discharging to the maximum capacity of the original battery is firstly lower than a set failure threshold, so that the remaining service life of the current battery is k charging and discharging cycles, and the life prediction of the battery is completed.
The mechanical strain of the battery which is easy to monitor online in real time is brought into the estimation of the health state and the prediction of the service life of the lithium ion battery, because the mechanical strain of the battery is coupled with the electrochemical reaction of the battery in the charging and discharging process, more effective information can be provided for a battery management system, so that the accuracy and the reliability of the estimation of the health state and the prediction of the service life of the lithium ion battery are improved, the use and maintenance efficiency of the battery are improved, and the service life of the battery is prolonged.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; although the present invention has been described with a certain degree of particularity with reference to the foregoing embodiments, those skilled in the art should understand that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.

Claims (10)

1. A method for state of health estimation and life prediction of a lithium ion battery taking into account mechanical strain, the method comprising the steps of:
s1, carrying out an aging degradation experiment of the lithium ion battery under different charging and discharging working conditions to obtain mechanical strain, voltage, current and temperature data of the lithium ion battery and corresponding battery aging degradation data;
s2, denoising the experimental data obtained in the step S1 by using a multi-resolution wavelet denoising model, and extracting typical characteristic parameters of the denoised experimental data by using a characteristic parameter extraction model based on wavelet analysis;
s3, establishing a typical characteristic parameter state process model by using the typical characteristic parameters extracted in the step S2 and identifying model parameters;
s4, establishing a typical characteristic parameter-aging degradation mapping relation model of the lithium ion battery according to the battery aging degradation data obtained in the step S1 and the typical characteristic parameters obtained in the step S2;
s5, inputting typical characteristic parameters obtained by using the external characteristic parameters of the battery monitored in real time on line and through a multi-resolution wavelet denoising model and a characteristic parameter extraction model based on wavelet analysis into a typical characteristic parameter-aging degradation mapping relation model of the lithium ion battery, and estimating the health state of the current lithium ion battery on line;
s6, inputting typical characteristic parameters obtained by using external characteristic parameters of the battery monitored in real time on line through a multi-resolution wavelet denoising model and a characteristic parameter extraction model based on wavelet analysis into a typical characteristic parameter state process model by combining historical on-line data; and inputting the result into a typical characteristic parameter-aging degradation mapping relation model of the lithium ion battery, and predicting the residual service life of the lithium ion battery on line.
2. The method according to claim 1, wherein the lithium ion battery is a secondary lithium ion battery cell capable of being used in a charging and discharging cycle, a secondary lithium ion battery pack formed by combining the secondary lithium ion battery cells capable of being used in the charging and discharging cycle in a series-parallel manner, or a secondary lithium ion battery system formed by building the secondary lithium ion battery pack in a modular manner.
3. The method for estimating state of health and predicting life of a lithium ion battery considering mechanical strain according to claim 2, wherein the step S1 specifically comprises the steps of:
s11, presetting at least two battery charging and discharging test working conditions, and presetting one or more aging decline index failure threshold values of the batteries;
s12, testing at least two lithium ion batteries according to a preset charging and discharging test working condition, and simultaneously recording mechanical strain, voltage, current and temperature data of the batteries and corresponding battery aging decline index data;
s13, judging whether one or more aging decline indexes of the current lithium ion battery reach a failure threshold value; if yes, stopping the test; if not, step S12 is repeated.
4. The method of claim 3, wherein the aging degradation indicator in step S11 comprises at least one of a ratio of a maximum capacity of the battery currently under test to a maximum capacity of the original battery, a ratio of a maximum energy of the battery currently under test to a maximum energy of the original battery, a ratio of a maximum output power of the battery currently under test to a maximum output power of the original battery, and a ratio of an internal resistance of the battery currently under test to an internal resistance of the original battery.
5. The method of claim 4, wherein the mechanical strain of the battery in step S12 refers to the strains along two mutually perpendicular directions, namely the length and the width of the lithium ion battery cell, and the resultant strains along the two directions, on the plane formed by the maximum area of the battery, generated by electrochemical reactions during charging and discharging of the square wound lithium ion battery cell or the square stacked lithium ion battery cell.
6. The method of claim 5, wherein the mechanical strain of the battery is measured by using a biaxial strain gauge, and the position of the biaxial strain gauge is arranged at a central position on a plane formed by the largest area of the lithium ion battery.
7. The method for estimating the state of health and predicting the life of the lithium ion battery considering the mechanical strain according to claim 6, wherein the parameters in the multi-resolution wavelet denoising model comprise a wavelet function, a decomposition layer number and a threshold value, the wavelet function, the decomposition layer number and the threshold value are obtained through optimization, the optimization problem converts the determination problem of the wavelet type, the decomposition layer number, the threshold value and the threshold value processing method into a mixed integer optimization problem, the objective function is determined through a cross-checking method, and the optimization problem is solved through a genetic algorithm.
8. The method of claim 7, wherein the optimization problem is established as follows:
let x i I =1,2, …, N, wherein i represents a signal sequence number and N represents a signal length;
firstly, generating an even point signal fe and an odd point signal fo from an experimental data signal according to the parity of i, and then smoothing the odd point signal fo to obtain a uniform estimation value fe * Expressed as:
Figure FDA0003680720580000021
wherein N is an even number, and the even point signal fe is subjected to two-point smoothing to obtain the even estimation value fo of the odd point signal * (ii) a If the even dot signal fe and the odd dot signal fo are respectively denoised to obtain an even dot signal approximation fe and an odd dot signal approximation fo, the estimated value of the Relative Squared Error (RSE) is as follows:
Figure FDA0003680720580000022
in the wavelet denoising parameters, both wavelet types and threshold processing strategies are non-numerical variables, a wavelet type candidate set and a threshold processing strategy candidate set are established, and an integer variable I is respectively used ψ And I ρ Representing wavelet type candidate set and threshold processing strategy candidate set elements, selecting a hard threshold processing strategy and a soft threshold processing strategy by the threshold processing strategy, I ρ The value and corresponding threshold processing strategy is:
Figure FDA0003680720580000031
assuming that the number of signals in the sequence of approximated signals at the lowest resolution is not lower than the minimum number of signals N in the sequence of approximated signals at the lowest resolution 0 Based on the multiresolution wavelet denoising theory, the number of decomposition layers J in wavelet denoising is an integer and needs to satisfy the constraint as shown in the following formula:
Figure FDA0003680720580000032
determining the upper and lower limits of the threshold based on a universal threshold method: firstly based on a general threshold method and an original signal x i Calculating the threshold corresponding to each wavelet in the wavelet candidate set and constructing a vector
Figure FDA0003680720580000033
Wherein
Figure FDA0003680720580000034
Denotes the reference number I ψ The second to find the upper limit T min = minT and lower limit T max = maxT, finally separately α T min And beta T max As the upper limit and the lower limit of the parameter threshold T, wherein the coefficients α and β are 0 < α < 1 and β > 1, the problem of determining the wavelet type, the number of decomposition layers, the threshold processing strategy and the threshold of the denoising parameter in the wavelet denoising model is transformed into a mixed integer optimization problem subject to simple constraints as shown below:
Figure FDA0003680720580000035
where s.t. means constraint, Z denotes integer set, UI ψ Representing the number of wavelets, UI, in a wavelet candidate set ρ Representing the number of thresholding strategies, N 0 Representing the minimum number of signals in the sequence of approximation signals at the lowest resolution, let N 0 ＝8，αT min And beta T max Representing the upper and lower limits of the threshold T.
9. The method for estimating the state of health and predicting the life of a lithium ion battery considering mechanical strain according to claim 8, wherein the process of extracting the typical characteristic parameters of the denoised experimental data by the characteristic parameter extraction model based on wavelet analysis comprises the following steps:
calculating an envelope feature function (DF) by multi-scale envelope superposition of Continuous Wavelet Transform (CWT) coefficients, the DF being:
Figure FDA0003680720580000036
in the formula, α represents a continuous wavelet analysis scale, n a Is the number of scales, τ is the continuous wavelet analysis time transfer window size, n represents the total number of samples in the discrete signal, and the envelope function of the Continuous Wavelet Transform (CWT) coefficients is
Figure FDA0003680720580000041
Wherein W s (α, τ) represents the time-frequency domain resulting from subjecting the original signal sequence to successive wavelet transforms,
Figure FDA0003680720580000042
is the result of the hilbert transform of the continuous wavelet transform coefficients;
calculating an operating energy ratio ER using the obtained envelope feature function DF 1 Said operating energy ratio ER 1 Comprises the following steps:
Figure FDA0003680720580000043
wherein L is the length of the energy collection window before and after the time transfer window size is tau;
using the resulting operating energy ratio ER 1 Obtaining characterization lithium ionsCharacteristic parameter ER typical of chemical reactions related to aging inside a battery 2 The ER 2 Comprises the following steps:
ER 2 (τ)＝ER 1 (τ)|DF(α)| (8)。
10. the method of claim 9, wherein the remaining useful life of the lithium ion battery in step S6 comprises a calendar time-based battery remaining calendar life and a number of charge/discharge times-based remaining cycle life.
CN202110681576.9A 2021-06-18 2021-06-18 Lithium ion battery health state estimation and life prediction method considering mechanical strain Active CN113504482B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110681576.9A CN113504482B (en) 2021-06-18 2021-06-18 Lithium ion battery health state estimation and life prediction method considering mechanical strain

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110681576.9A CN113504482B (en) 2021-06-18 2021-06-18 Lithium ion battery health state estimation and life prediction method considering mechanical strain

Publications (2)

Publication Number Publication Date
CN113504482A CN113504482A (en) 2021-10-15
CN113504482B true CN113504482B (en) 2022-10-25

Family

ID=78010368

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110681576.9A Active CN113504482B (en) 2021-06-18 2021-06-18 Lithium ion battery health state estimation and life prediction method considering mechanical strain

Country Status (1)

Country Link
CN (1) CN113504482B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11443569B1 (en) * 2021-10-30 2022-09-13 Beta Air, Llc Systems and methods for battery management for a vehicle
CN114236415A (en) * 2021-12-16 2022-03-25 湖北亿纬动力有限公司 Power battery health state detection method, device, equipment and storage medium
CN114918159A (en) * 2022-05-16 2022-08-19 杭州煦达新能源科技有限公司 Screening method of retired lithium ion power battery
CN116593895B (en) * 2023-06-16 2024-02-23 中国科学技术大学 Strain-based lithium ion battery pack current detection method and system

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103308864A (en) * 2013-07-09 2013-09-18 中国人民解放军国防科学技术大学 Method for estimating secondary cell SOH value and testing residual service life
CN106033112A (en) * 2014-11-06 2016-10-19 现代自动车株式会社 System and method for estimating state of health using battery model parameter
CN106093796A (en) * 2016-08-09 2016-11-09 北京航空航天大学 Based on regression criterion reconstruction in frequency-domain have lithium battery capacity and life-span prediction method in the case of capacity regeneration
CN106920997A (en) * 2017-03-17 2017-07-04 成都新柯力化工科技有限公司 A kind of method that utilization piezoelectric ceramics improves cycle life of lithium ion battery
CN107064800A (en) * 2016-11-29 2017-08-18 北京交通大学 The real-time predicting method of lithium ion battery remaining life
CN107768752A (en) * 2016-10-27 2018-03-06 杨春晓 Raising extends lead-acid accumulator or the device of battery pack service life
CN111426955A (en) * 2020-04-23 2020-07-17 华南理工大学 Lithium ion battery fault diagnosis method
CN112098847A (en) * 2020-08-25 2020-12-18 华南理工大学 Lithium ion battery SOC estimation method considering mechanical strain
WO2021056102A1 (en) * 2019-09-23 2021-04-01 Gbatteries Energy Canada Inc. Adaptive battery charging based on battery measurements during discharging pulse
CN112949060A (en) * 2021-03-01 2021-06-11 中国人民解放军火箭军工程大学 Lithium battery health state estimation and residual life prediction method under time-varying temperature

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103308864A (en) * 2013-07-09 2013-09-18 中国人民解放军国防科学技术大学 Method for estimating secondary cell SOH value and testing residual service life
CN106033112A (en) * 2014-11-06 2016-10-19 现代自动车株式会社 System and method for estimating state of health using battery model parameter
CN106093796A (en) * 2016-08-09 2016-11-09 北京航空航天大学 Based on regression criterion reconstruction in frequency-domain have lithium battery capacity and life-span prediction method in the case of capacity regeneration
CN107768752A (en) * 2016-10-27 2018-03-06 杨春晓 Raising extends lead-acid accumulator or the device of battery pack service life
CN107064800A (en) * 2016-11-29 2017-08-18 北京交通大学 The real-time predicting method of lithium ion battery remaining life
CN106920997A (en) * 2017-03-17 2017-07-04 成都新柯力化工科技有限公司 A kind of method that utilization piezoelectric ceramics improves cycle life of lithium ion battery
WO2021056102A1 (en) * 2019-09-23 2021-04-01 Gbatteries Energy Canada Inc. Adaptive battery charging based on battery measurements during discharging pulse
CN111426955A (en) * 2020-04-23 2020-07-17 华南理工大学 Lithium ion battery fault diagnosis method
CN112098847A (en) * 2020-08-25 2020-12-18 华南理工大学 Lithium ion battery SOC estimation method considering mechanical strain
CN112949060A (en) * 2021-03-01 2021-06-11 中国人民解放军火箭军工程大学 Lithium battery health state estimation and residual life prediction method under time-varying temperature

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
State of health and charge measurements in lithium-ion batteries using mechanical stress;John Cannarella 等;《Journal of Power Sources》;20141231;第7-14页 *
锂离子电池健康评估和寿命预测综述;刘大同 等;《仪器仪表学报》;20150131;第36卷(第1期);第1-16页 *

Also Published As

Publication number Publication date
CN113504482A (en) 2021-10-15

Similar Documents

Publication Publication Date Title
CN113504482B (en) Lithium ion battery health state estimation and life prediction method considering mechanical strain
Guo et al. Online estimation of SOH for lithium-ion battery based on SSA-Elman neural network
US11598817B2 (en) Storage cell diagnostic device and storage cell diagnostic method, and storage cell control system
CN112986830A (en) Lithium battery capacity estimation method based on convolution time memory neural network
CN113740736A (en) Electric vehicle lithium battery SOH estimation method based on deep network self-adaptation
CN112433170B (en) Method for identifying parameter difference of single batteries of series battery pack
CN111579993B (en) Lithium battery capacity online estimation method based on convolutional neural network
CN113917334B (en) Battery health state estimation method based on evolution LSTM self-encoder
CN112327193B (en) Lithium battery capacity diving early warning method
CN110712528B (en) Real-time monitoring method and device for power battery pack
CN116609676B (en) Method and system for monitoring state of hybrid energy storage battery based on big data processing
CN111983474A (en) Lithium ion battery life prediction method and system based on capacity decline model
CN115308628A (en) Battery health state monitoring method
CN116125306A (en) Power battery thermal runaway diagnosis method and system based on hybrid prediction framework
CN113945852B (en) Method for evaluating inconsistency of storage battery pack
Wang et al. Dynamic early recognition of abnormal lithium-ion batteries before capacity drops using self-adaptive quantum clustering
CN112327169B (en) Lithium battery residual life prediction method
Singh et al. State of charge estimation techniques of Li-ion battery of electric vehicles
Selvabharathi et al. Estimating the state of health of lead-acid battery using feed-forward neural network
US20230324463A1 (en) Method and Apparatus for Operating a System for Detecting an Anomaly of an Electrical Energy Store for a Device by Means of Machine Learning Methods
CN113687237B (en) Lithium battery residual charging time prediction method for guaranteeing electrical safety
Li et al. A state-of-health estimation method of lithium-ion batteries using ICA and SVM
Ang et al. Linear model for online state of health estimation of lithium-ion batteries using segmented discharge profiles
CN114966436A (en) Lithium battery state of charge prediction method, device, equipment and readable storage medium
Wei et al. Two-phase early prediction method for remaining useful life of lithium-ion batteries based on a neural network and Gaussian process regression

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