CN115494400B - Lithium battery lithium separation state online monitoring method based on ensemble learning - Google Patents
Lithium battery lithium separation state online monitoring method based on ensemble learning Download PDFInfo
- Publication number
- CN115494400B CN115494400B CN202211381818.3A CN202211381818A CN115494400B CN 115494400 B CN115494400 B CN 115494400B CN 202211381818 A CN202211381818 A CN 202211381818A CN 115494400 B CN115494400 B CN 115494400B
- Authority
- CN
- China
- Prior art keywords
- lithium battery
- lithium
- identification parameters
- positive
- soc
- 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
Links
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 title claims abstract description 162
- 229910052744 lithium Inorganic materials 0.000 title claims abstract description 162
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000012544 monitoring process Methods 0.000 title claims abstract description 20
- 238000000926 separation method Methods 0.000 title claims description 10
- 238000004458 analytical method Methods 0.000 claims abstract description 36
- 230000006870 function Effects 0.000 claims abstract description 20
- 230000015556 catabolic process Effects 0.000 claims abstract description 17
- 238000006731 degradation reaction Methods 0.000 claims abstract description 17
- 238000002474 experimental method Methods 0.000 claims abstract description 13
- 238000007600 charging Methods 0.000 claims description 25
- 238000009792 diffusion process Methods 0.000 claims description 25
- 238000007599 discharging Methods 0.000 claims description 25
- 230000010287 polarization Effects 0.000 claims description 24
- 239000007791 liquid phase Substances 0.000 claims description 18
- 238000004422 calculation algorithm Methods 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 16
- 238000009830 intercalation Methods 0.000 claims description 13
- 230000002687 intercalation Effects 0.000 claims description 13
- 239000002245 particle Substances 0.000 claims description 13
- 238000010206 sensitivity analysis Methods 0.000 claims description 13
- 238000006243 chemical reaction Methods 0.000 claims description 11
- 239000007790 solid phase Substances 0.000 claims description 11
- 238000003066 decision tree Methods 0.000 claims description 6
- 238000007637 random forest analysis Methods 0.000 claims description 6
- 238000003487 electrochemical reaction Methods 0.000 claims description 5
- HBBGRARXTFLTSG-UHFFFAOYSA-N Lithium ion Chemical compound [Li+] HBBGRARXTFLTSG-UHFFFAOYSA-N 0.000 claims description 3
- 150000001768 cations Chemical class 0.000 claims description 3
- 239000003792 electrolyte Substances 0.000 claims description 3
- 229910001416 lithium ion Inorganic materials 0.000 claims description 3
- 101150092509 Actn gene Proteins 0.000 claims 2
- 101150015939 Parva gene Proteins 0.000 claims 2
- 238000010281 constant-current constant-voltage charging Methods 0.000 claims 1
- 238000004364 calculation method Methods 0.000 abstract description 6
- 230000008859 change Effects 0.000 description 4
- 238000004088 simulation Methods 0.000 description 3
- 238000010280 constant potential charging Methods 0.000 description 2
- 238000010277 constant-current charging Methods 0.000 description 2
- 230000002427 irreversible effect Effects 0.000 description 2
- 238000001556 precipitation Methods 0.000 description 2
- 206010000369 Accident Diseases 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 125000004432 carbon atom Chemical group C* 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003411 electrode reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000001453 impedance spectrum Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- 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
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Secondary Cells (AREA)
Abstract
The invention provides an integrated learning-based lithium battery lithium analysis state online monitoring method, which comprises the steps of modeling a lithium battery through an SP + model, determining importance scores of identification parameters according to a degradation experiment, finally adjusting the SP + model in real time through a fitting function, calculating the current negative electrode potential of the lithium battery, and judging whether the lithium battery is close to or in a lithium analysis state according to the negative electrode potential. According to the method, the importance of each identification parameter is sequenced through a degradation experiment, so that the accuracy of lithium analysis state prediction of the lithium battery is improved on one hand; on the other hand, the calculation amount is greatly reduced, the timeliness of lithium analysis state prediction of the lithium battery is improved, and online monitoring of the lithium analysis state of the lithium battery is really realized.
Description
Technical Field
The invention relates to the field of lithium batteries, in particular to a lithium battery lithium analysis state online monitoring method based on ensemble learning.
Background
In practical use, an excessively high charging current or an excessively high ambient temperature may cause an irreversible lithium precipitation phenomenon in the lithium battery. The occurrence of the phenomenon of lithium precipitation not only can reduce the service life of the lithium battery, but also can cause irreversible damage to the lithium battery. In addition, the lithium battery is in a lithium analysis state for a long time, so that internal short circuit can be caused, and thermal runaway can occur in serious people, and further, fire accidents are caused. Therefore, if the lithium analysis condition of the lithium battery can be monitored on line, the safety of the lithium battery, vehicles and passengers is guaranteed.
Usually, need to disassemble the lithium cell and can learn through manual observation whether the lithium cell takes place to analyse the lithium phenomenon, however the way is difficult to in time obtain the lithium condition of analysing of lithium cell like this, moreover also inadvisable in practical application. At present, a method for calculating the potential of a negative electrode of a lithium battery through an RC equivalent model so as to judge whether the lithium battery is in a lithium analysis state exists, but the method is large in calculation amount and low in efficiency, and if the accuracy of lithium analysis state prediction is guaranteed, the timeliness of prediction cannot be guaranteed, so that the method cannot be really used in a vehicle-mounted scene.
Disclosure of Invention
In order to overcome the defects in the background art, the invention discloses an integrated learning-based lithium battery lithium separation state online monitoring method, which aims to: the method improves the accuracy of lithium battery lithium analysis state prediction, considers the operation efficiency and really realizes the online monitoring of the lithium battery lithium analysis state.
In order to achieve the purpose, the invention adopts the following technical scheme:
a lithium battery lithium analysis state online monitoring method based on ensemble learning comprises the following steps:
the method comprises the following steps: modeling the lithium battery by using an SP + model, and determining identification parameters required by modeling;
step two: setting pulse working conditions under various different charging and discharging multiplying powers, carrying out degradation experiments on the lithium battery at different temperatures, identifying and obtaining identification parameters under different SOH, different SOC, different temperatures and different charging and discharging multiplying powers according to the pulse working conditions set for the lithium battery at different degradation stages, and carrying out sensitivity analysis on the identification parameters respectively to obtain importance scores of the SOH, the SOC, the temperatures and the charging and discharging multiplying powers on the identification parameters;
step three: according to the variation ranges of the identification parameters in different SOH and SOC stages and under the variation of temperature and charge-discharge multiplying power, respectively carrying out sensitivity analysis on the identification parameters to obtain the importance scores of the identification parameters to terminal voltage;
step four: the method comprises the steps of conducting importance sorting on importance scores of SOH, SOC, temperature and charge-discharge multiplying power for identification parameters and importance scores of the identification parameters for terminal voltage to obtain important identification parameters, and fitting the important identification parameters into fitting functions relevant to the SOH, the SOC, the temperature and the charge-discharge multiplying power;
step five: and taking the real-time temperature and the real-time current of the current lithium battery as input, adjusting the SP + model in real time through a fitting function, calculating the current negative electrode potential of the lithium battery, and judging whether the lithium battery is close to or in a lithium analysis state according to the negative electrode potential.
Further improving the technical scheme, in the step one, the SP + model includes the following formula:
the basic process of the electrochemical reaction is described by the following formula:
wherein,for positive open circuit potentialAboutAs a function of (a) or (b),is negative open circuit potentialAboutAs a function of (a) or (b),、respectively the lithium concentration fractions embedded on the surfaces of the positive electrode active particles and the negative electrode active particles of the lithium battery,、respectively, the average lithium intercalation concentration fractions in the positive and negative electrode active particles,、respectively the maximum variation range of the lithium intercalation concentration fraction of the positive electrode and the negative electrode,for the load current, the discharge is specified to be positive,the total capacity (C) of the lithium battery.
The solid phase diffusion process is described by the following equation:
The liquid phase concentration polarization process is described by the following equation:
wherein,the gas constant is an ideal gas constant,the internal temperature (K) of the lithium battery,is a function of the faraday constant and is,the initial lithium ion concentration of the electrolyte is generally set to 1000,the cation transport number is generally 0.3 to 0.4,is the time constant of the diffusion of the liquid phase,is the liquid phase diffusion proportionality coefficient.
The reactive polarization process is described by the following equation:
in the formula, the reaction polarization overpotential of the negative electrode is as follows:
The ohmic polarization process is described by the following equation: a
wherein, the identification parameters required by modeling are 13, which are respectively the positive and negative active particles of the batterySub-initial lithium intercalation concentration、Maximum variation range of positive and negative electrode lithium intercalation concentration fractions、Positive and negative electrode capacity、Time constant of solid phase diffusion of positive and negative electrodes、Time constant of liquid phase diffusionCoefficient of liquid phase diffusion ratioPositive and negative reaction polarization coefficients、Equivalent ohmic internal resistance。
Further improving the technical scheme, in the second step, after the battery is fully charged, the battery is subjected to cross discharge charging with the magnification from small to large and with the discharge first and the charge later according to the actually acceptable charge-discharge magnification of the battery, the discharge and charge duration time is minutes, and the battery needs to be placed for minutes after each charge or discharge is finished; each charging multiplying factor and each discharging multiplying factor are subjected to 1 cycle from small to large, and the discharging cut-off voltage is reached after a plurality of cycles.
And in the second step, performing the degradation experiment on the battery under the working conditions that constant current and constant voltage charging with fixed multiplying power is carried out to the charging cut-off voltage and constant current discharging with fixed multiplying power is carried out to the discharging cut-off voltage, performing pulse working conditions in each cycle, and recording the current corresponding SOH of the battery.
And in the second step, performing sensitivity analysis on the identification parameters by using an AdaBoost algorithm, a decision tree algorithm and a random forest algorithm to obtain the importance of SOH, SOC, temperature and charge-discharge multiplying power on the identification parameters, and obtaining the importance scores of the identification parameters by weighting.
And in the third step, performing sensitivity analysis on the identification parameters by using an AdaBoost algorithm, a decision tree algorithm and a random forest algorithm to obtain the importance of the identification parameters to the terminal voltage, and obtaining the importance scores of the identification parameters by weighting.
The technical scheme is further improved, and the importance scores of identification parameters which are sensitive to temperature and charge-discharge multiplying power and have important influence on terminal voltage are increased.
According to the further improved technical scheme, in the fifth step, the negative electrode potentialDescribed by the following formula:
The technical scheme is further improved, the UKF is used for updating in real time to obtain the accurate SOC, the battery simulation terminal voltage obtained by the SP + model is used as an estimated value, the actually measured terminal voltage is used as an observed value, the relative error between the estimated value and the observed value is calculated, and the relative error is updated; when the relative error is <1%, the SOC is considered accurate.
Further improve the technical scheme whenJudging that the battery is in a normal state; when the temperature is higher than the set temperatureJudging that the battery is close to a lithium separation state; when the temperature is higher than the set temperatureAnd judging that the battery is in a lithium separation state.
Due to the adoption of the technical scheme, compared with the background technology, the invention has the following beneficial effects:
the invention provides an integrated learning-based lithium battery lithium analysis state online monitoring method, which is characterized in that a lithium battery is modeled through an SP + model, the importance scores of identification parameters are determined according to a degradation experiment, finally, the SP + model is adjusted in real time through a fitting function, the current negative pole potential of the lithium battery is calculated, and whether the lithium battery is close to or in a lithium analysis state is judged according to the negative pole potential.
According to the method, the importance of each identification parameter is sequenced through a degradation experiment, so that the accuracy of lithium analysis state prediction of the lithium battery is improved, the calculated amount is greatly reduced, the timeliness of lithium analysis state prediction of the lithium battery is improved, and the online monitoring of the lithium analysis state of the lithium battery is really realized.
Drawings
Fig. 1 shows a flow chart of the lithium battery lithium analysis state online monitoring method.
FIG. 2 is a schematic diagram illustrating the analysis and ranking of the sensitivity of the algorithms to identifying parameters.
FIG. 3 is a flow chart showing the updating of SOC by UKF according to the present invention.
Fig. 4 is a graph showing a comparison of a simulated voltage and a measured voltage obtained by using the existing RC equivalent model.
Fig. 5 shows a comparison of the simulated voltage and the measured voltage obtained by the method.
Detailed Description
Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only for explaining the technical principle of the present invention, and are not intended to limit the scope of the present invention.
The invention provides an integrated learning-based lithium battery lithium analysis state online monitoring method, which comprises the following steps of:
the method comprises the following steps: and modeling the lithium battery by using the SP + model, and determining identification parameters required by modeling.
The SP + model comprises five processes of a lithium battery in the charging and discharging processes, namely an electrochemical reaction basic process, a particle internal solid phase diffusion process, a liquid phase concentration polarization process, reaction polarization and ohmic polarization. The basic process of the electrochemical reaction is described by the following formula:
wherein,for positive open circuit potentialAboutAs a function of (a) or (b),is negative open circuit potentialAboutIs a function of (a) a function of (b),、respectively the lithium concentration fractions embedded on the surfaces of the positive electrode active particles and the negative electrode active particles of the lithium battery,、respectively, the average lithium intercalation concentration fractions in the positive and negative electrode active particles,、respectively the maximum variation range of the lithium intercalation concentration fraction of the positive electrode and the negative electrode,for the load current, the discharge is specified to be positive,the total capacity (C) of the lithium battery.
The solid phase diffusion process is described by the following formula:
The liquid phase concentration polarization process is described by the following equation:
wherein,the gas constant is an ideal gas constant,the internal temperature (K) of the lithium battery,is a function of the faraday constant and is,the initial lithium ion concentration of the electrolyte is generally set to 1000,the cation transference number is generally 0.3 to 0.4,is the time constant of the diffusion of the liquid phase,is the liquid phase diffusion proportionality coefficient.
The reactive polarization process is described by the following equation:
the negative electrode reaction polarization overpotential in the above formula is:
The ohmic polarization process is described by the following equation:
wherein, the identification parameters required by modeling are 13, and the identification parameters are respectively the initial lithium intercalation concentration of the positive and negative active particles of the lithium battery、Maximum variation range of positive and negative electrode lithium intercalation concentration fraction、Positive and negative electrode capacity、Time constant of solid phase diffusion of positive and negative electrodes、Time constant of liquid phase diffusionCoefficient of liquid phase diffusion ratioPositive and negative reaction polarization coefficients、Equivalent ohmic internal resistance。
Step two: setting pulse working conditions under various different charging and discharging multiplying powers, carrying out degradation experiments on the battery at different temperatures, identifying and obtaining identification parameters under different SOH, different SOC, different temperatures and different charging and discharging multiplying powers according to the pulse working conditions set for the battery at different degradation stages, and carrying out sensitivity analysis on the identification parameters respectively to obtain the importance scores of the SOH, the SOC, the temperatures and the charging and discharging multiplying powers on the identification parameters.
Specifically, the pulse working condition is set according to the actually acceptable charge-discharge multiplying power of the lithium battery. In this embodiment, the maximum acceptable charge rate of the lithium battery is 4C, the maximum acceptable discharge rate is 4C, four charge rates are set to be 1C, 2C, 3C, and 4C, and four discharge rates are set to be 1C, 2C, 3C, and 4C. And then carrying out pulse discharge charging working condition on the lithium battery in a full charging state. The lithium battery is subjected to cross discharging charging with the multiplying power from small to large and charging after discharging, the discharging and charging duration time is 1 minute each time, and the lithium battery needs to be placed for 30 minutes after charging or discharging is completed each time. After the four kinds of charge rates and discharge rates were performed one round from small to large, 1 cycle was considered to be completed, and the cycle was continued until the discharge cutoff voltage was reached.
And performing degradation experiments on the lithium battery at different temperatures, and setting the pulse working conditions for the lithium battery at different degradation stages. The degradation experiment refers to the degradation experiment carried out in a constant temperature box with different temperatures (5 ℃, 10 ℃, 25 ℃, 30 ℃ and 40 ℃) under the working conditions that 1C constant current and constant voltage charging is carried out till the charging cut-off voltage and 1C constant current discharging is carried out till the discharging cut-off voltage, the pulse working condition is carried out every 50 times of circulation, and the SOH corresponding to the lithium battery at present is recorded.
Based on a pulse excitation response method and according to the following sequence, identification parameters under different SOH, different SOC, different temperatures and different multiplying power are obtained through identification:
(1) preparing a three-electrode lithium battery, fitting an inherent characteristic curve of a positive electrode and a negative electrode, and identifying to obtain basic characteristic parameters of the lithium battery, namely the initial lithium intercalation concentration of active particles of the positive electrode and the negative electrode by combining with an electrochemical reaction basic equation of the lithium battery、And maximum variation range of lithium intercalation concentration fractions of positive and negative electrodes、;
(2) Measuring the AC impedance spectrum of the lithium battery by using an electrochemical workstation, and calculating to obtain the equivalent ohmic internal resistance;
(3) Measuring and calculating reaction polarization overpotential at pulse rising edge, and identifying to obtain positive and negative reaction polarization coefficients、;
(4) Calculating concentration polarization overpotential at each steady-state working point, and identifying to obtain the time constants of the solid phase diffusion of the positive electrode and the negative electrode、Time constant of liquid phase diffusionAnd liquid phase diffusion proportionality coefficient;
As shown in the attached figure 2, sensitivity analysis is carried out on the identification parameters by using AdaBoost, a decision tree and a random forest algorithm, and the importance of SOH, SOC, temperature and charge-discharge multiplying power on 13 identification parameters is obtained. And giving equal weight weighting calculation for the three algorithms, wherein the sum of the weights is 1, and comprehensively obtaining importance scores of different identification parameters.
Step three: and respectively carrying out sensitivity analysis on the identification parameters according to the variation ranges of the identification parameters in different SOH and SOC stages under the variation of temperature and charge-discharge multiplying power, and obtaining the importance scores of the identification parameters to the terminal voltage.
Specifically, when sensitivity analysis is performed on a certain identification parameter, the other identification parameters are kept as identification values under the conditions of current SOH, current SOC, current temperature and current multiplying power of 0.5C, the identification parameter to be analyzed is changed from low temperature to high temperature and from low multiplying power to high multiplying power by taking a 1% change interval as a change value in a change interval of the identification parameter, a corresponding simulation terminal voltage curve is calculated through an SP + model and compared with an actual terminal voltage curve under the conditions of current SOH, current SOC, current temperature and current multiplying power of 0.5C, and the maximum error in the curves corresponding to different change values and the average error of the curves are obtained through calculation. Then, sensitivity analysis is carried out on the identification parameters by using AdaBoost, a decision tree and a random forest algorithm to obtain the importance of the identification parameters to terminal voltage, three methods are given for equal weight weighted calculation, wherein the total weight is 1, and importance scores of different parameters are obtained comprehensively.
Step four: and carrying out importance sorting on the importance scores of the SOH, the SOC, the temperature and the charge-discharge multiplying factor for the identification parameters and the importance scores of the identification parameters for the terminal voltage to obtain important identification parameters, and fitting the important identification parameters into a fitting function related to the SOH, the SOC, the temperature and the charge-discharge multiplying factor.
Specifically, the weight of each identification parameter is calculated according to the importance scores of the SOH, the SOC, the temperature and the charge-discharge multiplying power on the identification parameters and the importance scores of the identification parameters on the terminal voltage obtained in the step 2 and the step 3 to obtain the importance ranking. Wherein the weight of the importance in the range of 0.8 to 1.0 is 0.7, the weight of the importance in the range of 0.5 to 0.8 is 0.2, the weight of the importance below 0.5 is 0.1, and then the identification parameters are classified according to the importance scores and the result of importance ranking.
It is worth noting that for identification parameters which are sensitive to temperature and charge-discharge multiplying power and have important influence on terminal voltage, importance scores of the identification parameters are properly increased according to specific conditions in actual working conditions, and therefore model accuracy is improved. And the identification parameters which are sensitive to temperature and charge-discharge multiplying power but have no important influence on terminal voltage, or have important influence on terminal voltage but are insensitive to temperature and charge-discharge multiplying power are set to be identification values at the room temperature of 25 ℃ and the charge-discharge multiplying power of 0.5C, so that the parameter calculation amount in the actual working condition is reduced.
After obtaining the important identification parameters, the important identification parameters are fit to a fitting function related to SOH, SOC, temperature and charge-discharge multiplying power. For example, a high-order polynomial is used to fit the important identification parameters into a function related to SOH, SOC, temperature, and charge/discharge rate.
Step five: and taking the real-time temperature and the real-time current of the current battery as input, adjusting the SP + model in real time through a fitting function, calculating the current negative electrode potential of the battery, and judging whether the battery is close to or in a lithium analysis state according to the negative electrode potential.
FIG. 3 is a flow chart showing the updating of SOC by UKF according to the present invention. And (4) in an actual working condition, taking the real-time temperature and the real-time current of the lithium battery measured by the sensor as input, and adjusting the identification parameters in real time according to the function obtained in the step (4). Because lithium cell SOC is great to electrochemical model precision influence, consequently usable UKF discerns in real time and obtains accurate SOC. Meanwhile, battery simulation terminal voltage obtained by the SP + model is used as an estimation value, actual terminal voltage is used as an observation value, a relative error between the estimation value and the observation value is calculated, and the relative error is updated. When the relative error is <1%, the SOC is considered accurate.
And then, calculating the current negative electrode potential through an SP + model, and judging whether the lithium battery is close to or in a lithium separation state at the current moment according to the negative electrode potential. Wherein the negative pole potentialDescribed by the following formula:
in the above formula, the first and second carbon atoms are,is a negative open circuit potential. If it is notIf so, judging that the battery is in a normal state; if it is usedIf so, judging that the battery is close to a lithium separation state; if it is usedIf so, the battery is judged to be in a lithium separation state. When the lithium battery is in a lithium analysis state or is close to the lithium analysis state, an alarm is triggered, and at the moment, the current needs to be reduced or the temperature of the lithium battery needs to be reduced through thermal management.
Fig. 4 is a graph showing a comparison of a simulated voltage and a measured voltage obtained by using a conventional RC equivalent model. Fig. 5 shows a comparison of the simulated voltage and the measured voltage obtained by the method. Comparing fig. 4 and 5, it can be seen that the simulated voltage obtained by the present invention is closer to the measured voltage. Moreover, the method greatly reduces the calculated amount, improves the timeliness of the lithium analysis state prediction of the lithium battery, and really realizes the online monitoring of the lithium analysis state of the lithium battery.
The details of which are not described in the prior art. Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (7)
1. An integrated learning-based lithium battery lithium analysis state online monitoring method is characterized by comprising the following steps:
the method comprises the following steps: modeling the lithium battery by using an SP + model, and determining identification parameters required by modeling;
the SP + model includes the following formula:
the basic process of the electrochemical reaction is described by the following formula:
E ocv (t)=U p [y surf (t)]-U n [x surf (t)]
y surf (t)=y avg (t)+Δy(t)=y 0 +D y (1-SoC(t))
x surf (t)=x avg (t)-Δx(t)=x 0 -D x (1-SoC(t))
wherein, U p [y surf (t)]For positive pole open circuit potential U p About y surf Function of (t), U n [x surf (t)]Is negative open circuit potential U n With respect to x surf Function of (t), y surf (t)、x surf (t) the concentration fractions of lithium intercalated into the surfaces of positive and negative active particles of a lithium battery, y avg (t)、x avg (t) average embedded lithium concentration fractions in positive and negative electrode active particles, respectively, D y 、D x The maximum variation ranges of the lithium intercalation concentration fractions of the positive and negative electrodes, I is the load current, the discharge is defined as positive, and Q all The total capacity (C) of the lithium battery;
the solid phase diffusion process is described by the following formula:
the liquid phase concentration polarization process is described by the following equation:
wherein R is an ideal gas constant, T is the internal temperature (K) of the lithium battery, F is a Faraday constant, c 0 The initial lithium ion concentration of the electrolyte is generally set to 1000,t + The cation transport number is generally 0.3 to 0.4, τ e Is the liquid phase diffusion time constant, P con Is the liquid phase diffusion proportionality coefficient;
the reactive polarization process is described by the following equation:
in the formula, the reaction polarization overpotential of the negative electrode is as follows:
wherein, P actp 、P actn Positive and negative reaction polarization coefficients, respectively;
the ohmic polarization process is described by the following equation:
η ohm (t)=I(t)R ohm (t)
wherein R is ohm Equivalent ohmic internal resistance;
terminal voltage U app Described by the following formula:
U app (t)=E ocv (t)-η con (t)-η act (t)-η ohm (t)
the identification parameters required by modeling are 13, namely the initial lithium intercalation concentration x of the positive and negative active particles of the lithium battery 0 、y 0 Maximum variation range D of positive and negative electrode lithium intercalation concentration fractions x 、D y Positive and negative electrode capacity Q p 、Q n Time constant of solid phase diffusion of positive and negative electrodesLiquid phase diffusion time constant τ e Liquid phase diffusion proportionality coefficient P con Positive and negative reaction polarization coefficient P actp 、P actn Equivalent ohmic internal resistance R ohm ;
Step two: setting pulse working conditions under various different charging and discharging multiplying powers, carrying out degradation experiments on the lithium battery at different temperatures, identifying and obtaining identification parameters under different SOH, different SOC, different temperatures and different charging and discharging multiplying powers according to the pulse working conditions set for the lithium battery at different degradation stages, and carrying out sensitivity analysis on the identification parameters respectively to obtain the importance scores of the SOH, the SOC, the temperatures and the charging and discharging multiplying powers on the identification parameters;
step three: according to the variation ranges of the identification parameters in different SOH and SOC stages and under the variation of temperature and charge-discharge multiplying power, respectively carrying out sensitivity analysis on the identification parameters to obtain the importance scores of the identification parameters to terminal voltage;
step four: the method comprises the steps of conducting importance sorting on importance scores of SOH, SOC, temperature and charge-discharge multiplying power for identification parameters and importance scores of the identification parameters for terminal voltage to obtain important identification parameters, and fitting the important identification parameters into fitting functions relevant to the SOH, the SOC, the temperature and the charge-discharge multiplying power;
step five: taking the real-time temperature and the real-time current of the current lithium battery as input, adjusting the SP + model in real time through a fitting function, calculating the current negative electrode potential of the lithium battery, and judging whether the lithium battery is close to or in a lithium analysis state according to the negative electrode potential;
the negative pole potential U appn (t) is described by the following formula:
wherein, U n (t) is the negative open circuit potential.
2. The integrated learning-based lithium battery lithium analysis state online monitoring method according to claim 1, wherein in the second step, after the lithium battery is fully charged, the lithium battery is charged by cross discharge with the magnification from small to large and the lithium battery is charged first and then charged according to the actually acceptable charging and discharging magnification of the lithium battery, the discharging and charging duration time of each time is 1 minute, and the lithium battery needs to be left for 30 minutes after each charging or discharging is completed; each charge multiplying factor and discharge multiplying factor are subjected to 1 cycle from small to large, and the discharge cut-off voltage is reached after a plurality of cycles.
3. The lithium battery lithium analysis state online monitoring method based on integrated learning of claim 1, wherein in the second step, the degradation experiment is performed in a thermostat, the degradation experiment is performed on the lithium battery under the working conditions that constant-current constant-voltage charging at a fixed rate is performed to a charging cut-off voltage, and constant-current discharging at the fixed rate is performed to a discharging cut-off voltage, a pulse working condition is performed every cycle, and the SOH corresponding to the lithium battery at present is recorded.
4. The lithium battery lithium analysis state online monitoring method based on ensemble learning of claim 1, wherein in the second step, an AdaBoost algorithm, a decision tree algorithm and a random forest algorithm are used for carrying out sensitivity analysis on identification parameters to obtain the importance of SOH, SOC, temperature and charge-discharge rate on the identification parameters, and importance scores of the identification parameters are obtained through weighting.
5. The lithium battery lithium analysis state online monitoring method based on ensemble learning according to claim 1, wherein in the third step, an AdaBoost algorithm, a decision tree algorithm and a random forest algorithm are used for carrying out sensitivity analysis on the identification parameters to obtain the importance of the identification parameters to terminal voltage, and the importance scores of the identification parameters are obtained through weighting.
6. The lithium battery lithium analysis state online monitoring method based on ensemble learning as claimed in claim 1, wherein UKF is used to update in real time to obtain accurate SOC, simulated terminal voltage of lithium battery obtained by SP + model is used as estimated value, measured terminal voltage is used as observed value, relative error between the estimated value and the observed value is calculated, and the relative error is updated; when the relative error is <1%, the SOC is considered accurate.
7. The integrated learning-based lithium battery lithium analysis state online monitoring method as claimed in claim 1, wherein when U is detected, U is detected appn If the (t) is more than 0.2, judging that the lithium battery is in a normal state; when 0.2 > U appn When the (t) > 0, judging that the lithium battery is close to a lithium separation state; when U is turned appn And (t) < 0, judging that the lithium battery is in a lithium separation state.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211381818.3A CN115494400B (en) | 2022-11-07 | 2022-11-07 | Lithium battery lithium separation state online monitoring method based on ensemble learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211381818.3A CN115494400B (en) | 2022-11-07 | 2022-11-07 | Lithium battery lithium separation state online monitoring method based on ensemble learning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115494400A CN115494400A (en) | 2022-12-20 |
CN115494400B true CN115494400B (en) | 2023-03-28 |
Family
ID=85116167
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211381818.3A Active CN115494400B (en) | 2022-11-07 | 2022-11-07 | Lithium battery lithium separation state online monitoring method based on ensemble learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115494400B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116559757B (en) * | 2023-07-04 | 2023-10-27 | 江苏天合储能有限公司 | Verification method and device for battery lithium-precipitation potential prediction accuracy and electronic equipment |
CN117633498B (en) * | 2024-01-25 | 2024-04-23 | 湖北工业大学 | Lithium battery electrochemical model parameter identification method |
Citations (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104899439A (en) * | 2015-06-02 | 2015-09-09 | 吉林大学 | Mechanism modeling method for lithium ion battery |
CN105223508A (en) * | 2015-07-14 | 2016-01-06 | 上海空间电源研究所 | Inside lithium ion cell performance state lossless detection method |
CN105891724A (en) * | 2016-05-05 | 2016-08-24 | 南京航空航天大学 | On-line estimation method for state of charge of lithium ion battery based on extended single particle model |
CN105932349A (en) * | 2016-06-07 | 2016-09-07 | 哈尔滨工业大学 | Long-life rapid charging method for lithium ion battery |
CN106450536A (en) * | 2016-11-09 | 2017-02-22 | 清华大学 | Quick charging method for lithium ion battery |
CN108761341A (en) * | 2018-06-01 | 2018-11-06 | 哈尔滨工业大学 | A kind of lithium ion battery battery chemical modeling parameter acquisition methods |
CN109004694A (en) * | 2017-06-07 | 2018-12-14 | 宁德新能源科技有限公司 | A kind of charging method and device |
CN111082173A (en) * | 2019-12-06 | 2020-04-28 | 中国第一汽车股份有限公司 | Lithium ion battery rapid charging method based on lithium separation prevention |
WO2020116853A1 (en) * | 2018-12-06 | 2020-06-11 | 주식회사 엘지화학 | Device and method for charging secondary battery |
KR20210014000A (en) * | 2019-07-29 | 2021-02-08 | 부산대학교 산학협력단 | Method of Predicting Battery Performance by Mathematical Modeling and Simulation |
CN112436202A (en) * | 2020-10-22 | 2021-03-02 | 中车长春轨道客车股份有限公司 | Stepped current charging method for preventing lithium precipitation of lithium ion battery cathode |
WO2021035736A1 (en) * | 2019-08-30 | 2021-03-04 | Oppo广东移动通信有限公司 | Charging control method and apparatus, charging test method and system, and electronic device |
CN112673266A (en) * | 2020-04-30 | 2021-04-16 | 华为技术有限公司 | Lithium analysis detection method and device, and polarization ratio acquisition method and device |
CN112703125A (en) * | 2020-08-10 | 2021-04-23 | 华为技术有限公司 | Lithium analysis detection method and device for lithium battery |
CN113659245A (en) * | 2021-08-11 | 2021-11-16 | 东莞新能安科技有限公司 | Electrochemical device heating method, electrochemical device and electric equipment |
CN114158276A (en) * | 2020-02-05 | 2022-03-08 | 株式会社Lg新能源 | Method of detecting lithium plating and method and apparatus for managing battery by using the same |
CN114221049A (en) * | 2021-11-19 | 2022-03-22 | 东莞维科电池有限公司 | Method for judging lithium precipitation of battery cell |
WO2022063236A1 (en) * | 2020-09-27 | 2022-03-31 | 比亚迪股份有限公司 | Battery charging method and system based on lithium plating detection, and automobile and medium |
WO2022063234A1 (en) * | 2020-09-27 | 2022-03-31 | 比亚迪股份有限公司 | Battery lithium precipitation state detection method and system, vehicle, device, and storage medium |
CN114552038A (en) * | 2022-02-24 | 2022-05-27 | 中山大学 | Lithium battery lithium-analysis-free quick charging method and system based on dynamic programming |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9077182B2 (en) * | 2013-01-29 | 2015-07-07 | Mitsubishi Electric Research Laboratories, Inc. | Method for estimating state of charge for lithium-ion batteries |
-
2022
- 2022-11-07 CN CN202211381818.3A patent/CN115494400B/en active Active
Patent Citations (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104899439A (en) * | 2015-06-02 | 2015-09-09 | 吉林大学 | Mechanism modeling method for lithium ion battery |
CN105223508A (en) * | 2015-07-14 | 2016-01-06 | 上海空间电源研究所 | Inside lithium ion cell performance state lossless detection method |
CN105891724A (en) * | 2016-05-05 | 2016-08-24 | 南京航空航天大学 | On-line estimation method for state of charge of lithium ion battery based on extended single particle model |
CN105932349A (en) * | 2016-06-07 | 2016-09-07 | 哈尔滨工业大学 | Long-life rapid charging method for lithium ion battery |
CN106450536A (en) * | 2016-11-09 | 2017-02-22 | 清华大学 | Quick charging method for lithium ion battery |
CN109004694A (en) * | 2017-06-07 | 2018-12-14 | 宁德新能源科技有限公司 | A kind of charging method and device |
CN108761341A (en) * | 2018-06-01 | 2018-11-06 | 哈尔滨工业大学 | A kind of lithium ion battery battery chemical modeling parameter acquisition methods |
WO2020116853A1 (en) * | 2018-12-06 | 2020-06-11 | 주식회사 엘지화학 | Device and method for charging secondary battery |
KR20210014000A (en) * | 2019-07-29 | 2021-02-08 | 부산대학교 산학협력단 | Method of Predicting Battery Performance by Mathematical Modeling and Simulation |
WO2021035736A1 (en) * | 2019-08-30 | 2021-03-04 | Oppo广东移动通信有限公司 | Charging control method and apparatus, charging test method and system, and electronic device |
CN111082173A (en) * | 2019-12-06 | 2020-04-28 | 中国第一汽车股份有限公司 | Lithium ion battery rapid charging method based on lithium separation prevention |
CN114158276A (en) * | 2020-02-05 | 2022-03-08 | 株式会社Lg新能源 | Method of detecting lithium plating and method and apparatus for managing battery by using the same |
CN112673266A (en) * | 2020-04-30 | 2021-04-16 | 华为技术有限公司 | Lithium analysis detection method and device, and polarization ratio acquisition method and device |
WO2021217662A1 (en) * | 2020-04-30 | 2021-11-04 | 华为技术有限公司 | Lithium plating detection method and apparatus, and polarization proportion acquisition method and apparatus |
CN112703125A (en) * | 2020-08-10 | 2021-04-23 | 华为技术有限公司 | Lithium analysis detection method and device for lithium battery |
WO2022032460A1 (en) * | 2020-08-10 | 2022-02-17 | 华为技术有限公司 | Lithium plating detection method and apparatus for lithium battery |
WO2022063236A1 (en) * | 2020-09-27 | 2022-03-31 | 比亚迪股份有限公司 | Battery charging method and system based on lithium plating detection, and automobile and medium |
WO2022063234A1 (en) * | 2020-09-27 | 2022-03-31 | 比亚迪股份有限公司 | Battery lithium precipitation state detection method and system, vehicle, device, and storage medium |
CN112436202A (en) * | 2020-10-22 | 2021-03-02 | 中车长春轨道客车股份有限公司 | Stepped current charging method for preventing lithium precipitation of lithium ion battery cathode |
CN113659245A (en) * | 2021-08-11 | 2021-11-16 | 东莞新能安科技有限公司 | Electrochemical device heating method, electrochemical device and electric equipment |
CN114221049A (en) * | 2021-11-19 | 2022-03-22 | 东莞维科电池有限公司 | Method for judging lithium precipitation of battery cell |
CN114552038A (en) * | 2022-02-24 | 2022-05-27 | 中山大学 | Lithium battery lithium-analysis-free quick charging method and system based on dynamic programming |
Non-Patent Citations (5)
Title |
---|
Xiaojun Tan.Lithium Plating as Limiting Phenomena for Estimating Safety during Lithium-Ion Battery Charging.《International Journal of ELECTROCHEMICAL SCIENCE》.2020,全文. * |
XIAOJUN TAN.Real-Time State-of-Health Estimation of Lithium-Ion Batteries Based on the Equivalent Internal Resistance.《IEEE Access》.2020,全文. * |
吕 超.锂离子电池热耦合 SP+模型及其参数化简.《电源学报》.2015,第第13卷卷(第第13卷期),全文. * |
廖严文鑫.三元锂离子电池脉冲充电技术研究.《电源技术》.2019,全文. * |
董 鹏.基于电化学阻抗谱的锂离子电池析锂检测方法.《汽车安全与节能学报》.2021,第第12 卷卷(第第12 卷期),全文. * |
Also Published As
Publication number | Publication date |
---|---|
CN115494400A (en) | 2022-12-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Qiao et al. | Online quantitative diagnosis of internal short circuit for lithium-ion batteries using incremental capacity method | |
CN110501652B (en) | Rapid assessment method and device for available capacity of retired lithium battery | |
CN107066722B (en) | Electrochemical model-based combined estimation method for state of charge and state of health of power battery system | |
CN106716158B (en) | Battery charge state evaluation method and device | |
CN102468521B (en) | Method and apparatus for assessing battery state of health | |
CN112615075B (en) | Battery quick charging method and computer equipment | |
CN111856282B (en) | Vehicle-mounted lithium battery state estimation method based on improved genetic unscented Kalman filtering | |
CN112464571B (en) | Lithium battery pack parameter identification method based on multi-constraint-condition particle swarm optimization algorithm | |
CN115494400B (en) | Lithium battery lithium separation state online monitoring method based on ensemble learning | |
CN112180274B (en) | Rapid detection and evaluation method for power battery pack | |
CN109358293B (en) | Lithium ion battery SOC estimation method based on IPF | |
CN107192956B (en) | A kind of battery short circuit leakage on-line monitoring method and device | |
CN111208438B (en) | Method for cooperatively estimating residual capacity of lithium-ion battery and sensor deviation based on neural network and unscented Kalman filter | |
CN111366864B (en) | Battery SOH on-line estimation method based on fixed voltage rise interval | |
CN111929602A (en) | Single battery leakage or micro short circuit quantitative diagnosis method based on capacity estimation | |
CN110795851A (en) | Lithium ion battery modeling method considering environmental temperature influence | |
CN108872869A (en) | A kind of lithium ion battery deterioration classification method based on BP neural network | |
CN115639481B (en) | Battery data preprocessing system and method based on big data prediction SOC | |
CN113687251B (en) | Double-model-based lithium ion battery pack voltage abnormality fault diagnosis method | |
CN113484771A (en) | Method for estimating wide-temperature full-life SOC and capacity of lithium ion battery | |
CN111983464B (en) | Lithium battery lithium separation testing method based on pure electric vehicle | |
CN116754971A (en) | Method, device and storage medium for detecting lithium precipitation of battery | |
CN112733427A (en) | Method for establishing negative electrode potential estimation model of lithium ion battery and computer equipment | |
Tang et al. | An aging-and load-insensitive method for quantitatively detecting the battery internal-short-circuit resistance | |
CN115327415A (en) | Lithium battery SOC estimation method based on limited memory recursive least square algorithm |
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 |