CN108037464B - Battery pack SOC estimation method based on IMM-EKF - Google Patents

Battery pack SOC estimation method based on IMM-EKF Download PDF

Info

Publication number
CN108037464B
CN108037464B CN201711395497.1A CN201711395497A CN108037464B CN 108037464 B CN108037464 B CN 108037464B CN 201711395497 A CN201711395497 A CN 201711395497A CN 108037464 B CN108037464 B CN 108037464B
Authority
CN
China
Prior art keywords
soc
model
interaction model
moment
battery pack
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
CN201711395497.1A
Other languages
Chinese (zh)
Other versions
CN108037464A (en
Inventor
朱浩
陈华
邓元望
江银锋
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.)
Hunan Hongxun Yi'an New Energy Technology Co ltd
Original Assignee
Hunan Hongxun Yi'an New Energy Technology Co ltd
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 Hunan Hongxun Yi'an New Energy Technology Co ltd filed Critical Hunan Hongxun Yi'an New Energy Technology Co ltd
Priority to CN201711395497.1A priority Critical patent/CN108037464B/en
Publication of CN108037464A publication Critical patent/CN108037464A/en
Application granted granted Critical
Publication of CN108037464B publication Critical patent/CN108037464B/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]
    • G01R31/382—Arrangements for monitoring battery or accumulator variables, e.g. SoC
    • G01R31/3835—Arrangements for monitoring battery or accumulator variables, e.g. SoC involving only voltage measurements

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Secondary Cells (AREA)

Abstract

The application discloses a battery pack SOC estimation method based on IMM-EKF, which comprises the following steps: establishing a maximum cell voltage interaction model, a minimum cell voltage interaction model and an average voltage interaction model according to the voltage of the battery pack; estimating the SOC of the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model by utilizing an IMM-EKF filtering program; and calculating the information distribution factors of the models, and performing probability fusion on the SOC of the models according to the information distribution factors to obtain the whole SOC of the battery pack. The method divides the battery pack into three models, then carries out probability fusion on the SOC of each model according to the information distribution factors to obtain the whole SOC of the battery pack with higher precision, and solves the problem of larger estimation error of a single model. The application also provides a system for estimating the SOC of the battery pack based on the IMM-EKF, and the system has the beneficial effects.

Description

Battery pack SOC estimation method based on IMM-EKF
Technical Field
The application relates to the technical field of electric automobiles, in particular to a method and a system for estimating SOC of a battery pack based on IMM-EKF.
Background
A Battery Management System (BMS) is an important component of an electric vehicle, and an estimation of a State of Charge (SOC) of a Battery is a core of the BMS, and the SOC represents a ratio of a remaining capacity of the Battery to a capacity of a full Charge State thereof, and precision directly affects a service life, safety performance, equalization control, and customization of a thermal Management strategy of the Battery, so that accurate SOC estimation is very important for the BMS.
Due to the voltage and energy requirements of the electric automobile in the driving process, hundreds of single batteries need to be connected in series or in parallel in the battery management system. However, due to the difference of materials in the battery production process and the variation of battery parameters in the charging and discharging process, the battery cells in the same battery pack have a certain degree of inconsistency, and further the SOC of each battery cell is different, so that the SOC of the whole battery pack is difficult to estimate.
In the prior art, the overall SOC of the battery pack is usually estimated by using the minimum SOC or the average SOC instead of the overall SOC, and if the minimum SOC is used as the overall SOC, the overall overcharge of the battery pack is caused, or the SOC of the battery pack is not 100% when the voltage of the battery cell reaches the end of charging; if the average SOC is taken as the entire SOC, the uniformity of the battery is deteriorated with the use of the assembled battery, and the problem of overcharge or overdischarge of the battery cells also occurs.
Another estimation method in the existing algorithm is to estimate the whole battery group as a large single battery, but the inside of the large battery is more complex and more difficult to model than the single battery, and the consistency between the whole battery group is poorer than that between the single batteries, and the estimation difficulty is higher.
Therefore, how to improve the estimation accuracy of the overall SOC of the battery pack is a technical problem that needs to be solved by those skilled in the art.
Disclosure of Invention
The application aims to provide a method and a system for estimating the SOC of a battery pack based on IMM-EKF, and the method can improve the estimation accuracy of the whole SOC of the battery pack.
In order to solve the technical problem, the application provides a method for estimating the SOC of a battery pack based on IMM-EKF, which comprises the following steps:
establishing a maximum cell voltage interaction model, a minimum cell voltage interaction model and an average voltage interaction model according to the voltage of the battery pack;
estimating the SOC of the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model respectively by utilizing an IMM-EKF filtering program to obtain SOCmax, SOCmin and SOCaverage correspondingly;
calculating respective information distribution factors of the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model;
and performing probability fusion on the SOCmax, the SOCmin and the SOCaverage according to the information distribution factors to obtain the overall SOC of the battery pack.
Optionally, the estimating the SOCs of the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model by using an IMM-EKF filtering program to obtain SOCmax, SOCmin and soceverage correspondingly, including:
discretizing the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model to obtain a target state equation and an observation equation based on the IMM-EKF filtering program:
the state equation of the battery pack model is as follows: xk+1,i＝A×Xk,i+B×(Ik-Ik,drf)；
The observation equation of the battery pack model is as follows: u shapek,i＝Uk,oc-V1k,i-V2k,i-ItrueR0,i；
Wherein i is max, min or average, corresponding to the maximum cell voltage interaction model, the minimum cell voltage interaction model or the average voltage interaction model, Xk+1,iFor the battery state at the moment corresponding to model k +1, Xk,iBattery state for the moment of the corresponding model k, IkFor the value of the collected current at time k, Ik,drfThe current drift amount at the moment k is shown, and A and B are system parameters; u shapek,iV1 for terminal voltage at time corresponding to model kk,iPolarization capacitance C of battery pack for moment corresponding to model k1Polarization voltage of V2k,iPolarization capacitance C of battery pack for moment corresponding to model k2Polarization voltage of ItrueIs the true value of the current, and Itrue＝Ik-Ik,drf，R0,iThe internal resistance of the battery pack corresponding to the model.
Optionally, the battery pack state X at the moment corresponding to the model k +1k+1,iIn particular to
Figure BDA0001518381060000031
The battery state X corresponding to the model k momentk,iIn particular to
Figure BDA0001518381060000032
The estimating the SOC of the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model by using an IMM-EKF filtering program to correspondingly obtain SOCmax, SOCmin and SOCaverage comprises the following steps:
according to the formula SOCk,i＝Xk,i×[1 0 0 0]TCalculating to obtain SOCk,max、SOCk,minAnd SOCk,average；
Therein, SOCk+1,iFor SOC values corresponding to model k +1 time, including SOCk+1,max、SOCk+1,minAnd SOCk+1,average，SOCk,iFor SOC values corresponding to model k times, including SOCk,max、SOCk,minAnd SOCk,average，Ik+1,drfThe amount of current drift at time k +1, V1k+1,iPolarization capacitance C of battery pack corresponding to time k +1 of model1Polarization voltage of V2k+1,iPolarization capacitance C of battery pack corresponding to time k +1 of model2The polarization voltage of (1).
Optionally, X isk，iThe calculation process of (2) includes:
according to the formula
Figure BDA0001518381060000033
Calculating a Kalman gain matrix K corresponding to a model K momentk,i；
According to formula Kk,i＝[KSOC,iKV1,iKV2,iKI]TThe Kalman gain values K corresponding to V1 and V2V1And KV2Multiplying by corresponding suppression coefficients K1 and K2 and adding Kk,iIs updated to
Kk,i＝[KSOC,ik1×KV1,ik2×KV2,iKI]T；
According to the formula
Figure BDA0001518381060000034
Calculating to obtain the battery pack state X at the moment of the corresponding model kk,i；
Wherein, P- k,iError covariance estimation for the corresponding model, h (X)k,Itrue) An observation equation for the battery model, Ck,iAn observation equation h (X) for the battery modelk,Itrue) Partial derivative of, KSOC,iKalman gain values, K, corresponding to model SOC valuesV1,iIs a V1Corresponding Kalman gain value, KV2,iIs a V2Corresponding Kalman gain value, KIIs the current drift value IdrfCorresponding Kalman gain values, k1 and k2 are suppression coefficients,
Figure BDA0001518381060000035
for the state estimate at the moment of the corresponding model K, Kk,iKalman gain matrix, e, for the moment of model kk,iIs the innovation of the corresponding model k moment.
Optionally, the formula
Figure BDA0001518381060000036
Calculating to obtain the battery pack state X at the moment of the corresponding model kk,iThe method comprises the following steps:
according to the formula ek,i＝U(t,k)i-h(Xk,Itrue) Calculating innovation e of the battery pack modelk,i；
Judgment | ek,iWhether | is less than M;
if yes, let Kk,i＝0；
If not, according to formula Ek,i＝ek,i×ek-1,iCalculation of Ek,iAnd E is judgedk,iWhether greater than 0;
if Ek,iIf greater than 0, then K isk,iIs updated to Kk,i×Kstrength；
If Ek,iIf not greater than 0, then K isk,iIs updated to Kk,i×Kcontrol；
Wherein P-k,iError covariance estimation for the corresponding model, h (X)k,Itrue) An observation equation for the battery model，Ck,iAn observation equation h (X) for the battery modelk,Itrue) First derivative of, U(t,k)iFor the voltage value, K, acquired at the moment of the corresponding model KstrengthAs gain enhancement factor, KcontrolIs the suppression factor.
Optionally, the system parameter a is specifically
Figure BDA0001518381060000041
The system parameterBIn particular to
Figure BDA0001518381060000042
Where t is the sampling time, C1,iAnd C2,iFor the polarization capacitance, R, of the corresponding model battery1,iAnd R2,iFor the polarization internal resistance of the corresponding model battery, η is the coulombic efficiency, and Ca is the maximum discharge capacity of the battery.
Optionally, the R is1,i、C1,i、R2,i、C2,i、R0,iThe calculation process comprises the following steps:
establishing a battery pack parameter estimation model and setting a trigger condition;
obtaining the R according to the obtained1,i、C1,i、R2,i、C2,i、R0,iDetermining model parameters in the battery parameter estimation model;
updating model parameters in the battery pack parameter estimation model when the trigger condition is triggered;
calculating the R using the updated battery parameter estimation model1,i、C1,i、R2,i、C2,i、R0,iThe current value of (a).
Optionally, the calculating information distribution factors of the maximum cell voltage interaction model, the minimum cell voltage interaction model, and the average voltage interaction model includes:
judging whether the terminal voltage at the moment of the maximum monomer voltage interaction model k is smaller than the maximum charging cut-off voltage or not and whether the terminal voltage at the moment of the minimum monomer voltage interaction model k is larger than the discharging cut-off voltage or not;
if yes, according to the formula
Figure BDA0001518381060000051
Calculating an information distribution factor omega of the mean voltage interaction model at the k momentk,averageAccording to the formula
Figure BDA0001518381060000052
Calculating an information distribution factor omega of the maximum monomer voltage interaction model at the k momentk,maxAnd the information distribution factor omega of the minimum monomer voltage interaction model at the moment kk,min；
Therein, SOCk,midIs SOCk,maxAnd SOCk,minAverage value of (1), SOCk-1,packThe overall SOC of the battery pack at the time k-1.
Optionally, performing probability fusion on the SOCmax, the SOCmin, and the SOCaverage according to each information distribution factor to obtain the overall SOC of the battery pack, where the probability fusion includes:
according to the formula
Figure BDA0001518381060000053
Obtaining the total state estimation value X of the battery pack at the moment kk,jAnd the total covariance P of the k time of the battery packk,j；
According to the formula SOCk,pack＝Xk,j×[1 0 0 0]TBattery pack overall SOC (State of Charge) for calculating k timek,pack。
The present application further provides a system for battery pack SOC estimation based on IMM-EKF, the system comprising:
the modeling module is used for establishing a maximum cell voltage interaction model, a minimum cell voltage interaction model and an average voltage interaction model according to the voltage of the battery pack;
the estimation module is used for respectively estimating the SOC of the maximum cell voltage interaction model, the SOC of the minimum cell voltage interaction model and the SOC of the average voltage interaction model by utilizing an IMM-EKF filter program to correspondingly obtain SOCmax, SOCmin and SOCaverage;
the calculation module is used for calculating respective information distribution factors of the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model;
and the fusion module is used for performing probability fusion on the SOCmax, the SOCmin and the SOCaverage according to each information distribution factor to obtain the whole SOC of the battery pack.
The method for estimating the SOC of the battery pack based on the IMM-EKF comprises the steps of establishing a maximum cell voltage interaction model, a minimum cell voltage interaction model and an average voltage interaction model according to the voltage of the battery pack; estimating the SOC of the maximum monomer voltage interaction model, the SOC of the minimum monomer voltage interaction model and the SOC of the average voltage interaction model respectively by using an interactive multi-model extended Kalman filter (IMM-EKF) program to correspondingly obtain SOCmax, SOCmin and SOCaperage; calculating respective information distribution factors of the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model; and performing probability fusion on the SOCmax, the SOCmin and the SOCaverage according to the information distribution factors to obtain the whole SOC of the battery pack.
In the prior art, the overall SOC of the battery pack is usually estimated by using the minimum SOC or the average SOC instead of the overall SOC (for example, in patent CN105445665A, the SOC is estimated only based on Vmin, so that the estimated SOC is lower than the actual SOC of the battery pack), and if the minimum SOC is used as the overall SOC, the overall overcharge of the battery pack is caused, or the SOC of the battery pack is not 100% when the voltage of the battery cell reaches the end of charging; if the average SOC is taken as the entire SOC, the uniformity of the battery is deteriorated with the use of the assembled battery, and the problem of overcharge or overdischarge of the battery cells also occurs.
Another estimation method in the existing algorithm is to estimate the whole battery group as a large single battery, but the inside of the large battery is more complex and more difficult to model than the single battery, and the consistency between the whole battery groups is poorer than that between the single batteries, and the estimation difficulty is higher.
The method comprises the steps of dividing a battery pack into three models according to the maximum voltage, the minimum voltage and the average voltage of a battery monomer in the battery pack, utilizing an IMM-EKF filtering program to estimate the SOC of a maximum monomer voltage interaction model, a minimum monomer voltage interaction model and an average voltage interaction model respectively, calculating information distribution factors of the maximum monomer voltage interaction model, the minimum monomer voltage interaction model and the average voltage interaction model respectively, considering the overcharge and overdischarge conditions simultaneously through the calculation of the information distribution factors, and finally performing probability fusion on the SOC of each model according to each information distribution factor to obtain the whole SOC of the battery pack with higher precision, so that the problem of larger estimation error of the single model is solved. The application also provides a system for estimating the SOC of the battery pack based on the IMM-EKF, which has the beneficial effects and is not repeated herein.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flow chart of a method for estimating SOC of a battery pack based on IMM-EKF according to an embodiment of the present disclosure;
FIG. 2 is a schematic structural diagram of a second-order RC battery equivalent circuit model;
FIG. 3 is a flow chart of another method for IMM-EKF based SOC estimation of a battery pack provided by an embodiment of the present application;
FIG. 4 is a flow chart of an actual representation of S203 of the alternative IMM-EKF-based battery SOC estimation method of FIG. 3;
FIG. 5 is a flow chart of yet another method for IMM-EKF based battery SOC estimation provided by an embodiment of the present application;
fig. 6 is a schematic structural diagram of a system for estimating SOC of a battery pack based on IMM-EKF according to an embodiment of the present disclosure.
Detailed Description
The core of the application is to provide a battery pack SOC estimation method and system based on IMM-EKF, and the method can improve the estimation accuracy of the whole SOC of the battery pack.
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Referring to fig. 1 and fig. 2, fig. 1 is a flowchart illustrating a method for estimating SOC of a battery pack based on IMM-EKF according to an embodiment of the present disclosure; fig. 2 is a schematic structural diagram of a second-order RC battery equivalent circuit model.
The method specifically comprises the following steps:
s101: establishing a maximum cell voltage interaction model, a minimum cell voltage interaction model and an average voltage interaction model according to the voltage of the battery pack;
based on the problems that the overall SOC of the battery pack is usually replaced by the minimum SOC or the average SOC in the prior art, so that the overall overcharge or overdischarge of the battery pack and the like are caused, the method for estimating the SOC of the battery pack based on the IMM-EKF is provided, and the overall SOC of the battery pack with higher precision can be obtained;
establishing a maximum cell voltage interaction model, a minimum cell voltage interaction model and an average voltage interaction model according to the voltage of the battery pack, wherein the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model can be specifically as follows:
acquiring the maximum voltage V of each single battery in the battery pack through the BMS voltage monitoring modulemaxAnd a minimum voltage Vmin；
According to the formula
Figure BDA0001518381060000081
Calculating the average voltageVaverage；
According to Vmax、VminAnd VaverageEstablishing a maximum cell voltage interaction model, a minimum cell voltage interaction model and an average voltage interaction model; wherein, VtotalThe total voltage of the battery pack is shown, and n is the number of single batteries in the battery pack;
optionally, a maximum cell voltage interaction model, a minimum cell voltage interaction model, and an average voltage interaction model may be established according to a second-order RC battery equivalent circuit model shown in fig. 2, where the second-order RC battery equivalent circuit model mentioned here is specifically a second-order RC battery equivalent circuit model
Figure BDA0001518381060000082
Wherein, UOCFor the open-circuit voltage of the battery, R has a non-linear relationship with SOC1And R2For internal polarization resistance of the cell, C1And C2For polarizing the capacitance of the battery, R0Is the ohmic internal resistance, V, of the battery1And V2Corresponds to C1And C2Polarization voltage at both ends, U is battery terminal voltage;
step S101 may be executed only once for the same battery pack.
S102: estimating the SOC of the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model respectively by utilizing an IMM-EKF filtering program to correspondingly obtain SOCmax, SOCmin and SOCaverage;
optionally, in consideration of colored noise caused by hall sensor drift, that is, a current drift amount, discretization may be performed on the maximum cell voltage interaction model, the minimum cell voltage interaction model, and the average voltage interaction model established in S101, so as to obtain a target state equation and an observation equation based on the IMM-EKF filtering program:
the state equation of the battery pack model is as follows: xk+1,i＝A×Xk,i+B×(Ik-Ik,drf)；
The observation equation of the battery pack model is as follows: u shapek,i＝Uk,oc-V1k,i-V2k,i-ItrueR0,i；
Wherein i is max, min or average, corresponding to the maximum cell voltage interaction model, the minimum cell voltage interaction model or the average voltage interaction model, Xk+1,iFor the battery state at the moment corresponding to model k +1, Xk,iBattery state for the moment of the corresponding model k, IkFor the value of the collected current at time k, Ik,drfThe current drift amount at the moment k is shown, and A and B are system parameters; u shapek,iV1 for terminal voltage at time corresponding to model kk,iPolarization capacitance C of battery pack for moment corresponding to model k1Polarization voltage of V2k,iPolarization capacitance C of battery pack for moment corresponding to model k2Polarization voltage of ItrueIs the true value of the current, and Itrue＝Ik-Ik,drfThe current true value is calculated by subtracting the current drift value from the acquired current value, so that a short plate of the Kalman filtering for processing the colored noise is effectively compensated; r0,iThe internal resistance of the battery pack corresponding to the model;
alternatively, it can be according to a formula
Figure BDA0001518381060000091
Calculating to obtain the battery pack state X at the moment of corresponding model kk,i；
Wherein the content of the first and second substances,
Figure BDA0001518381060000092
for the state estimate at the moment of the corresponding model K, Kk,iKalman gain matrix, e, for the moment of model kk,iIs the innovation of the corresponding model k moment;
optionally, considering the IMM-EKF to process the colored noise short board, the method of state expansion can be adopted to process IdrfAs state values, i.e. Xk+1,iSpecifically can be
Figure BDA0001518381060000093
Xk,iCorrespond to
Figure BDA0001518381060000094
On the basis, SOC of the maximum monomer voltage interaction model, SOC of the minimum monomer voltage interaction model and SOC of the average voltage interaction model are respectively estimated by utilizing an IMM-EKF filter program to correspondingly obtain
The SOCmax, SOCmin and SOCaverage can be specifically as follows:
according to the formula SOCk,i＝Xk,i×[1 0 0 0]TCalculating to obtain SOCk,max、SOCk,minAnd SOCk,average；
Therein, SOCk+1,iFor SOC values corresponding to model k +1 time, including SOCk+1,max、SOCk+1,minAnd SOCk+1,average，SOCk,iFor SOC values corresponding to model k times, including SOCk,max、SOCk,minAnd SOCk,average，Ik+1,drfThe amount of current drift at time k +1, V1k+1,iPolarization capacitance C of battery pack corresponding to time k +1 of model1Polarization voltage of V2k+1,iPolarization capacitance C of battery pack corresponding to time k +1 of model2The polarization voltage of (1).
S103: calculating respective information distribution factors of the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model;
in the prior art, the weight coefficient of the battery pack SOC estimation using the maximum and minimum SOC weighting fusion is mostly in a discrete breakpoint form, so that the estimated SOC value is unsmooth and jumps; meanwhile, the existing weight method for estimating the SOC of the battery pack divides the unit estimation and the weight fusion into two independent parts, so that the accuracy of the whole SOC (namely SOCpack) of the battery pack directly depends on the selection of the weight, but the used weight is not proved by theory, and based on the weight, the self-adaptive calculation information distribution factor is set, and the values of the information distribution factors of the three models at the time k can be determined according to the whole SOC of the battery pack at the time k-1;
optionally, the calculation process of the information distribution factor may specifically be:
judging whether the terminal voltage at the moment of the maximum monomer voltage interaction model k is smaller than the maximum charging cut-off voltage or not and whether the terminal voltage at the moment of the minimum monomer voltage interaction model k is larger than the discharging cut-off voltage or not;
if yes, according to the formula
Figure BDA0001518381060000101
Calculating an information distribution factor omega at the k moment of the average voltage interaction modelk,averageAccording to the formula
Figure BDA0001518381060000102
Calculating information distribution factor omega of maximum monomer voltage interaction model k momentk,maxAnd information distribution factor omega of minimum monomer voltage interaction model k momentk,min；
Therein, SOCk,midIs SOCk,maxAnd SOCk,minAverage value of (1), SOCk-1,packThe SOC of the whole battery pack at the moment k-1;
before calculating the value of the information distribution factor, firstly, judging whether the current battery pack is in an overcharge or overdischarge state, namely judging whether the terminal voltage at the moment of the maximum monomer voltage interaction model k is smaller than the maximum charge cut-off voltage or not and whether the terminal voltage at the moment of the minimum monomer voltage interaction model k is larger than the discharge cut-off voltage or not;
calculating a median SOCmid according to the SOCmax and the SOCmin, and comparing the SOCmid with the SOCaverage to determine that the SOCmax or the SOCmin deviates more from the whole SOC, and further making corresponding adjustment, such as: if SOCmid>The SOCaverage indicates that the maximum SOC deviates from the whole SOC more, and the information distribution factor of the SOCaverag is made
Figure BDA0001518381060000103
So that the information of the SOCaverag is distributed with a factor omegak,averageThe relative height is increased, so that the functions of regulation and balance are achieved; similarly, if SOCmid<The SOCaverage indicates that the minimum SOC is more deviated from the whole SOC, and the distribution factor of the SOCaverage is
Figure BDA0001518381060000111
Optionally, the SOC may be considered as 100% and 0%, that is:
when the terminal voltage at the moment of the maximum cell voltage model k is greater than or equal to the maximum charge cut-off voltage, let omegak,max＝1，ωk,min＝0，ωk,average＝0；
When the terminal voltage at the moment of the minimum cell voltage model k is less than or equal to the discharge cut-off voltage, let omegak,max＝0，ωk,min＝0，ωk,average＝1；
The overall SOC of the battery pack obtained by utilizing the information distribution factor of the self-adaptive calculation is smoother than the result obtained by the conventional simple weight SOC calculation, and the SOC at two ends is more consistent with the conventional principle.
S104: and performing probability fusion on the SOCmax, the SOCmin and the SOCaverage according to the information distribution factors to obtain the whole SOC of the battery pack.
Alternatively, it can be according to a formula
Figure BDA0001518381060000112
Obtaining the total state estimation value X of the battery pack at the moment kk,jAnd the total covariance P of the k time of the battery packk,j；
According to the formula SOCk,pack＝Xk,j×[1 0 0 0]TBattery pack overall SOC (State of Charge) for calculating k timek,pack。
Based on the technical scheme, the method for estimating the SOC of the battery pack based on the IMM-EKF comprises the steps of dividing the battery pack into three models according to the maximum voltage, the minimum voltage and the average voltage of a battery monomer in the battery pack, estimating the SOC of the maximum monomer voltage interaction model, the minimum monomer voltage interaction model and the average voltage interaction model respectively by utilizing an IMM-EKF filtering program, calculating information distribution factors of the maximum monomer voltage interaction model, the minimum monomer voltage interaction model and the average voltage interaction model, considering the situations of overcharge and overdischarge simultaneously through the calculation of the information distribution factors, and finally performing probability fusion on the SOC of the models according to the information distribution factors to obtain the whole SOC of the battery pack with higher precision, so that the problem of larger estimation error of a single model is solved.
The existing battery pack SOC estimation does not consider the difference of the state value magnitude in the state vector (such as the magnitude difference between the SOC value and the polarization voltage Up value, the change interval of the SOC is between 0 and 1, the values of V1 and V2 are below 0.1V, and I isdrfThe value of (a) may reach more than 1A with the use of the sensor), but the overshoot of all states is easy to occur by using the same error to check, which may cause the overshoot of a state quantity with a small magnitude order, resulting in divergence of an algorithm, based on this, on the basis of the above embodiments, the present application provides another method for estimating the SOC of the battery pack based on the IMM-EKF, please refer to fig. 3, where fig. 3 is a flowchart of another method for estimating the SOC of the battery pack based on the IMM-EKF provided by the embodiments of the present application;
the method specifically comprises the following steps:
s201: according to the formula
Figure BDA0001518381060000121
Calculating a Kalman gain matrix K corresponding to a model K momentk,i；
Wherein, P- k,iFor error covariance estimation of the corresponding model, it can be based on the formula
Figure BDA0001518381060000122
iCalculating Q to obtaink,iFor the system parameters mentioned above, Q is the state noise variance, R is the measurement noise variance, Ck,iObservation equation h (X) for a battery modelk,Itrue) Partial derivatives of (a).
S202: according to formula Kk,i＝[KSOC,iKV1,iKV2,iKI]TThe Kalman gain values K corresponding to V1 and V2V1And KV2Multiplying by corresponding suppression coefficients K1 and K2 and adding Kk,iIs updated to
Kk,i＝[KSOC,ik1×KV1,ik2×KV2,iKI]T；
Wherein, KSOC,iKalman gain values, K, corresponding to model SOC valuesV1,iIs a V1Corresponding Kalman gain value, KV2,iIs a V2Corresponding Kalman gain value, KIIs the current drift value IdrfCorresponding kalman gain values, k1 and k2 are inhibition coefficients;
alternatively, k1, k2 ∈ (0, 1);
optionally, in the use process of the real vehicle, if the iteration step length is small, the whole kalman gain can be multiplied by the suppression coefficient 0.1, so that on one hand, the correction fluctuation can be reduced, and on the other hand, the correction degree can be reduced, so that the error detection is prevented from algorithm divergence in a controllable range.
S203: according to the formula
Figure BDA0001518381060000123
Calculating to obtain the battery pack state X at the moment of corresponding model kk,i。
Wherein the content of the first and second substances,
Figure BDA0001518381060000124
for the state estimate at the moment of the corresponding model K, Kk,iKalman gain matrix, e, for the moment of model kk,iIs the innovation of the corresponding model k moment.
Based on the technical scheme, the calculation method of the battery pack state corresponding to the model k moment in the battery pack SOC estimation method based on IMM-EKF provided by the application can reasonably set different suppression coefficients for each parameter according to the amplitude required to be adjusted of each parameter when the Kalman gain is correspondingly adjusted according to the change interval of each state value in the state equation, so as to achieve the effect of avoiding algorithm divergence.
The existing SOC estimation of the battery pack does not consider the problem of algorithm divergence, and effective measure control cannot be adopted for a divergence process, such as: based on the fact that Q and R in the conventional kalman filter algorithm have a great influence on the iterative process, and unsuitable Q and R directly influence the accuracy of the algorithm and even cause divergence, please refer to fig. 4, where fig. 4 is a flowchart of an actual representation manner of S203 in another method for estimating the SOC of the battery pack based on the IMM-EKF, which is provided in fig. 3.
The method specifically comprises the following steps:
s301: according to the formula ek,i＝U(t,k)i-h(Xk,Itrue) Calculating innovation of battery set model ek,i；
Innovation e of battery pack model mentioned herek,iThe difference between the collected voltage value and the observed voltage value can be understood as an error;
wherein, h (X)k,Itrue) For the observed equation of the battery model, U(t,k)iThe voltage values collected at the moment of the corresponding model k.
S302: judgment | ek,iWhether | is less than M;
if yes, go to step S303; if not, go to step S304;
wherein M is the maximum deviation value of the corresponding model precision.
S303: let Kk,i＝0；
If | ek,iIf | is less than M, it indicates that the corrected equilibrium state has been reached, let the Kalman gain Kk,iAnd (5) 0, namely, the iteration step does not carry out parameter updating, so that model errors are prevented from being introduced.
S304: according to formula Ek,i＝ek,i×ek-1,iCalculation of Ek,iA value of (d);
if | ek,iIf | is not less than M, then by analysis Ek,iAnd ek,iTo adjust the Kalman gain Kk,iTo prevent divergence.
S305: judgment Ek,iWhether greater than 0;
if yes, go to step S306; if not, the process proceeds to step S307.
S306: if Ek,iIf greater than 0, then K isk,iIs updated to Kk,i×Kstrength；
If Ek,iIf the error value is more than 0, the last correction intensity is not enough, the trend of the error changing to the direction of absolute value reduction is slow, and at the moment, a stronger control action is needed to change Kk,iIs updated to Kk,i×Kstrength；
Optionally, Kstrength∈(1,10)；
S307: if Ek,iIf not greater than 0, then K isk,iIs updated to Kk,i×Kcontrol。
If Ek,iIf not greater than 0, this indicates that the intensity of the last correction was excessive, and in this case a weaker control action is required, Kk,iIs updated to Kk,i×Kcontrol；
Optionally, Kcontrol∈(0,1)；
Wherein, KstrengthAs gain enhancement factor, KcontrolIs the suppression factor.
Based on the technical scheme, the problem of divergence of the filter program can be effectively solved by carrying out innovation detection on the battery model, meanwhile, due to the fact that error detection is introduced to carry out corresponding enhancement and suppression effects on Kalman gain, the condition that divergence is brought to an algorithm by two parameters of process noise Q and measurement noise R is weakened, iterative calculation can be carried out on the values of the two parameters in the Kalman filter program expansion as long as the values of the two parameters are in the same order of magnitude, and the complicated process of repeatedly adjusting the parameters is avoided.
Based on the above embodiment, the system parameter a may specifically be
Figure BDA0001518381060000141
The system parameter B may specifically be
Figure BDA0001518381060000142
Where t is the sampling time, C1,iAnd C2,iFor the polarization capacitance, R, of the corresponding model battery1,iAnd R2,iFor the polarization internal resistance of the corresponding model battery, η is the coulombic efficiency, and Ca is the maximum discharge capacity of the battery, please refer to fig. 5, and fig. 5 is a flowchart of another method for estimating the SOC of the battery based on the IMM-EKF according to the embodiment of the present disclosure.
The method specifically comprises the following steps:
s401: establishing a battery pack parameter estimation model and setting a trigger condition;
will be two ordersRC battery equivalent circuit model
Figure BDA0001518381060000143
Discretizing to obtain:
Uk＝UOC,k-b1UOC,k-1-b2UOC,k-2+b1Uk-1+b2Uk-2+b3Ik+b4Ik-1+b5Ik-2；
the data matrix defining the battery parameter estimation model is:
Φ＝[Uk-1-UOC,k-1Uk-2-UOC,k-2I Ik-1Ik-2]T；
the parameter matrix defining the battery pack parameter estimation model is as follows: theta ═ b1b2b3b4b5]T；
And (3) carrying out a recursive least square method aiming at the parameter matrix:
Figure BDA0001518381060000151
where t is the sampling time, b1、b2、b3、b4、b5Is an element in the system parameter matrix;
the setting triggering condition mentioned here may be specifically determined according to at least one of the charge-discharge history number h, the temperature difference, and the SOC value;
optionally, the trigger condition may specifically be set as:
1) starting parameter estimation every 10 times of charge-discharge circulation;
2) starting parameter estimation every 5 ℃;
3) parameter estimation was initiated every 10% SOC interval.
S402: according to the obtained acquisition R1,i、C1,i、R2,i、C2,i、R0,iDetermining model parameters in a battery pack parameter estimation model;
alternatively, the initial R can be calculated by HPPC experiments1,i、C1,i、R2,i、C2,i、R0,iFor initial system startup.
S403: updating the model parameters in the battery pack parameter estimation model when the trigger condition is triggered;
optionally, the updating of the model parameters in the battery pack parameter estimation model specifically includes:
updating the gain K: kk＝Pk-1Φk T(ΦkPk-1Φk T+1)；
Updating the parameter theta: thetak＝θk-1+Kk[Ut-Φkθk-1]；
Covariance P update: pk＝[1-KkΦk]Pk-1；
Wherein: p0＝1000×eye(5)，θ＝[0.1 0.1 0.1 0.1 0.1]T；
S404: calculating R by using the updated battery parameter estimation model1,i、C1,i、R2,i、C2,i、R0,iThe current value of (a).
Based on the technical scheme, the battery pack SOC estimation method based on IMM-EKF adopts multi-scale data driving and model fusion to update the battery parameter list, so that parameter changes caused by aging and operating environment are prevented from influencing the accuracy of the battery model, and errors are introduced into the SOC; meanwhile, the updating of multiple time scales can effectively reduce the calculation time in the driving process and improve the real-time performance of SOC estimation.
Based on the above embodiments, please refer to fig. 6, fig. 6 is a block diagram of a system for estimating SOC of a battery pack based on IMM-EKF according to an embodiment of the present disclosure.
The system may include:
the modeling module 100 is used for establishing a maximum cell voltage interaction model, a minimum cell voltage interaction model and an average voltage interaction model according to the voltage of the battery pack;
the estimation module 200 is used for respectively estimating the SOC of the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model by utilizing an IMM-EKF filtering program to correspondingly obtain SOCmax, SOCmin and SOCaverage;
the calculation module 300 is configured to calculate respective information distribution factors of the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model;
and the fusion module 400 is configured to perform probability fusion on the SOCmax, SOCmin, and socalerge according to each information distribution factor to obtain the overall SOC of the battery pack.
The components of the above system can be applied to one practical process as follows:
the modeling module establishes a maximum monomer voltage interaction model, a minimum monomer voltage interaction model and an average voltage interaction model according to the voltage of the battery pack; the estimation module is used for respectively estimating the SOC of the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model by utilizing an IMM-EKF filtering program to correspondingly obtain SOCmax, SOCmin and SOCaverage; the calculation module calculates respective information distribution factors of the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model; and the fusion module performs probability fusion on the SOCmax, the SOCmin and the SOCaverage according to the information distribution factors to obtain the whole SOC of the battery pack.
The embodiments are described in a progressive manner in the specification, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The method and system for estimating the SOC of the battery pack based on the IMM-EKF provided by the present application are described in detail above. The principles and embodiments of the present application are explained herein using specific examples, which are provided only to help understand the method and the core idea of the present application. It should be noted that, for those skilled in the art, it is possible to make several improvements and modifications to the present application without departing from the principle of the present application, and such improvements and modifications also fall within the scope of the claims of the present application.
It is further noted that, in the present specification, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

Claims (8)

1. A method for battery SOC estimation based on IMM-EKF, comprising:
establishing a maximum cell voltage interaction model, a minimum cell voltage interaction model and an average voltage interaction model according to the voltage of the battery pack;
estimating the SOC of the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model respectively by utilizing an IMM-EKF filtering program to correspondingly obtain the SOCmax、SOCminAnd SOCaverage；
Judging whether the terminal voltage at the moment of the maximum monomer voltage interaction model k is smaller than the maximum charging cut-off voltage or not and whether the terminal voltage at the moment of the minimum monomer voltage interaction model k is larger than the discharging cut-off voltage or not;
if yes, according to the formula
Figure FDA0002463275550000011
Calculating an information distribution factor omega of the mean voltage interaction model at the k momentk,averageAccording to the formula
Figure FDA0002463275550000012
Calculating an information distribution factor omega of the maximum monomer voltage interaction model at the k momentk,maxAnd the information distribution factor omega of the minimum monomer voltage interaction model at the moment kk,min；
According to the formula
Figure FDA0002463275550000013
Obtaining the total state estimation value X of the battery pack at the moment kk,jAnd the total covariance P of the k time of the battery packk,j；
According to the formula SOCk,pack＝Xk,j×[1 0 0 0]TBattery pack overall SOC (State of Charge) for calculating k timek,pack；
Therein, SOCk,maxIs the SOC value, SOC at the k moment of the maximum monomer voltage interaction modelk,minFor the SOC value, SOC at the k moment of the minimum cell voltage interaction modelk,averageIs the SOC value, SOC, of the mean voltage interaction model at the k momentk,midIs SOCk,maxAnd SOCk,minAverage value of (1), SOCk-1,packThe SOC, X of the battery pack at the time k-1k,maxFor the maximum cell voltage interaction model kEstimate of the state of the moment, Xk,minIs the state estimation value, X, of the minimum cell voltage interaction model at the moment kk,averageIs a state estimation value, P, of the mean voltage interaction model at the moment kk,maxIs the covariance of the maximum cell voltage interaction model at the time k, Pk,minIs the covariance of the minimum cell voltage interaction model at the time k, Pk,averageAnd (4) the covariance of the k moment of the average voltage interaction model.
2. The method of claim 1, wherein the SOC of the maximum cell voltage interaction model, the SOC of the minimum cell voltage interaction model and the SOC of the average voltage interaction model are estimated by an IMM-EKF filter program, and the corresponding SOC is obtainedmax、SOCminAnd SOCaverageThe method comprises the following steps:
discretizing the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model to obtain a target state equation and an observation equation based on the IMM-EKF filtering program:
the state equation of the battery model is: xk+1,i＝A×Xk,i+B×(Ik-Ik,drf)；
The observation equation of the battery pack model is as follows: u shapek,i＝Uk,oc-V1k,i-V2k,i-ItrueR0,i；
Wherein i is max, min or average, corresponding to the maximum cell voltage interaction model, the minimum cell voltage interaction model or the average voltage interaction model, Xk+1,iFor the battery state at the moment corresponding to model k +1, Xk,iBattery state for the moment of the corresponding model k, IkFor the value of the collected current at time k, Ik,drfThe current drift amount at the moment k is shown, and A and B are system parameters; u shapek,iTo correspond to terminal voltage at model k time, Uk,ocV1 for the open circuit voltage at time corresponding to model kk,iPolarization capacitance C of battery pack for moment corresponding to model k1Polarization voltage of V2k,iFor time k of the modelPolarization capacitor C of battery pack2Polarization voltage of ItrueIs the true value of the current, and Itrue＝Ik-Ik,drf，R0,iThe internal resistance of the battery pack corresponding to the model.
3. The method of claim 2, comprising:
the battery state X at the moment corresponding to the model k +1k+1,iIn particular to
Figure FDA0002463275550000021
The battery state X corresponding to the model k momentk,iIn particular to
Figure FDA0002463275550000022
Respectively estimating the SOC of the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model by utilizing an IMM-EKF filtering program to correspondingly obtain the SOCmax、SOCminAnd SOCaverageThe method comprises the following steps:
according to the formula SOCk,i＝Xk,i×[1 0 0 0]TCalculating to obtain SOCk,max、SOCk,minAnd SOCk,average；
Therein, SOCk+1,iFor SOC values corresponding to model k +1 time, including SOCk+1,max、SOCk+1,minAnd SOCk+1,average，SOCk,iFor SOC values corresponding to model k times, including SOCk,max、SOCk,minAnd SOCk,average，Ik+1,drfThe amount of current drift at time k +1, V1k+1,iPolarization capacitance C of battery pack corresponding to time k +1 of model1Polarization voltage of V2k+1,iPolarization capacitance C of battery pack corresponding to time k +1 of model2The polarization voltage of (1).
4. The method of claim 3, wherein X isk,iIs calculated byThe method comprises the following steps:
according to the formula
Figure FDA0002463275550000031
Calculating a Kalman gain matrix K corresponding to a model K momentk,i；
According to formula Kk,i＝[KSOC,iKV1,iKV2,iKI]TThe Kalman gain values K corresponding to V1 and V2V1And KV2Multiplying by corresponding suppression coefficients K1 and K2 and adding Kk,iIs updated to Kk,i＝[KSOC,ik1×KV1,ik2×KV2,iKI]T；
According to the formula
Figure FDA0002463275550000032
Calculating to obtain the battery pack state X at the moment of the corresponding model kk,i；
Wherein, P- k,iError covariance estimation for the corresponding model, h (X)k,Itrue) An observation equation for the battery model, Ck,iAn observation equation h (X) for the battery modelk,Itrue) Partial derivative of, KSOC,iKalman gain values, K, corresponding to model SOC valuesV1,iIs a V1Corresponding Kalman gain value, KV2,iIs a V2Corresponding Kalman gain value, KIIs the current drift value IdrfCorresponding Kalman gain values, k1 and k2 are suppression coefficients,
Figure FDA0002463275550000033
for the state estimate at the moment of the corresponding model K, Kk,iKalman gain matrix, e, for the moment of model kk,iIs the information corresponding to the model k time.
5. The method of claim 4, wherein the equation is based on
Figure FDA0002463275550000034
Calculating to obtain the battery pack state X at the moment of the corresponding model kk,iThe method comprises the following steps:
according to the formula ek,i＝U(t,k)i-h(Xk,Itrue) Calculating information e of the battery pack modelk,i；
Judgment | ek,iWhether | is less than M;
if yes, let Kk,i＝0；
If not, according to formula Ek,i＝ek,i×ek-1,iCalculation of Ek,iAnd E is judgedk,iWhether greater than 0;
if Ek,iIf greater than 0, then K isk,iIs updated to Kk,i×Kstrength；
If Ek,iIf not greater than 0, then K isk,iIs updated to Kk,i×Kcontrol；
Wherein, h (X)k,Itrue) Is an observation equation of the battery model, U(t,k)iFor the voltage value, K, acquired at the moment of the corresponding model KstrengthAs gain enhancement factor, KcontrolIs the suppression factor.
6. The method according to any one of claims 2-5, comprising:
the system parameter A is specifically
Figure FDA0002463275550000035
The system parameter B is specifically
Figure FDA0002463275550000041
Where t is the sampling time, C1,iAnd C2,iFor the polarization capacitance, R, of the corresponding model battery1And R2Is the polarization internal resistance, R, of the battery in the second-order RC battery equivalent circuit model1,iAnd R2,iFor polarization internal resistance of corresponding model battery pack, η is coulombEfficiency, Ca is the maximum discharge capacity of the battery.
7. The method of claim 6, wherein R is1,i、C1,i、R2,i、C2,i、R0,iThe calculation process comprises the following steps:
establishing a battery pack parameter estimation model and setting a trigger condition;
according to the obtained R1,i、C1,i、R2,i、C2,i、R0,iDetermining model parameters in the battery parameter estimation model;
updating model parameters in the battery pack parameter estimation model when the trigger condition is triggered;
calculating the R using the updated battery parameter estimation model1,i、C1,i、R2,i、C2,i、R0,iThe current value of (a).
8. A system for battery SOC estimation based on IMM-EKF, comprising:
the modeling module is used for establishing a maximum cell voltage interaction model, a minimum cell voltage interaction model and an average voltage interaction model according to the voltage of the battery pack;
an estimation module for estimating the SOC of the maximum cell voltage interaction model, the minimum cell voltage interaction model and the average voltage interaction model by using an IMM-EKF filter program to obtain the SOC correspondinglymax、SOCminAnd SOCaverage；
The judging module is used for judging whether the terminal voltage at the moment of the maximum monomer voltage interaction model k is smaller than the maximum charging cut-off voltage or not and whether the terminal voltage at the moment of the minimum monomer voltage interaction model k is larger than the discharging cut-off voltage or not;
a first calculation module, configured to, when the terminal voltage at the time of the maximum cell voltage interaction model k is smaller than the maximum charge cut-off voltage, and the terminal voltage at the time of the minimum cell voltage interaction model k is larger than the discharge voltageAt electrical cut-off voltage according to the formula
Figure FDA0002463275550000042
Calculating an information distribution factor omega of the mean voltage interaction model at the k momentk,averageAccording to the formula
Figure FDA0002463275550000051
Calculating an information distribution factor omega of the maximum monomer voltage interaction model at the k momentk,maxAnd the information distribution factor omega of the minimum monomer voltage interaction model at the moment kk,min；
A second calculation module for calculating according to a formula
Figure FDA0002463275550000052
Obtaining the total state estimation value X of the battery pack at the moment kk,jAnd the total covariance P of the k time of the battery packk,j；
A third calculation module for calculating the formula SOCk,pack＝Xk,j×[1 0 0 0]TBattery pack overall SOC (State of Charge) for calculating k timek,pack；
Therein, SOCk,maxIs the SOC value, SOC at the k moment of the maximum monomer voltage interaction modelk,minFor the SOC value, SOC at the k moment of the minimum cell voltage interaction modelk,averageIs the SOC value, SOC, of the mean voltage interaction model at the k momentk,midIs SOCk,maxAnd SOCk,minAverage value of (1), SOCk-1,packThe SOC, X of the battery pack at the time k-1k,maxIs the state estimation value, X, of the maximum monomer voltage interaction model at the moment kk,minIs the state estimation value, X, of the minimum cell voltage interaction model at the moment kk,averageIs a state estimation value, P, of the mean voltage interaction model at the moment kk,maxIs the covariance of the maximum cell voltage interaction model at the time k, Pk,minIs the covariance of the minimum cell voltage interaction model at the time k, Pk,averageFor the time k of the mean voltage interaction modelThe covariance.
CN201711395497.1A 2017-12-21 2017-12-21 Battery pack SOC estimation method based on IMM-EKF Active CN108037464B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711395497.1A CN108037464B (en) 2017-12-21 2017-12-21 Battery pack SOC estimation method based on IMM-EKF

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711395497.1A CN108037464B (en) 2017-12-21 2017-12-21 Battery pack SOC estimation method based on IMM-EKF

Publications (2)

Publication Number Publication Date
CN108037464A CN108037464A (en) 2018-05-15
CN108037464B true CN108037464B (en) 2020-06-16

Family

ID=62100433

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711395497.1A Active CN108037464B (en) 2017-12-21 2017-12-21 Battery pack SOC estimation method based on IMM-EKF

Country Status (1)

Country Link
CN (1) CN108037464B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109120035A (en) * 2018-09-11 2019-01-01 深圳市科陆电子科技股份有限公司 SOC balance control method in battery box system
CN109188293B (en) * 2018-11-08 2020-01-07 武汉理工大学 EKF lithium ion battery SOC estimation method based on innovation covariance band fading factor
CN109633473B (en) * 2019-01-23 2021-03-09 刘平 Distributed battery pack state of charge estimation algorithm
CN109991543B (en) * 2019-03-29 2022-06-21 深圳猛犸电动科技有限公司 Method and device for acquiring residual electric quantity of battery pack and terminal equipment
CN110095728A (en) * 2019-05-23 2019-08-06 合肥工业大学智能制造技术研究院 Battery SOC, SOH combined estimation method based on interactive multi-model
CN110531269B (en) * 2019-09-05 2022-06-17 许继集团有限公司 SOC estimation method of series-parallel combined cell stack and cell management system
CN112485672A (en) * 2020-11-13 2021-03-12 上海电气集团股份有限公司 Battery state determination method and device
CN114035072B (en) * 2021-11-11 2023-10-03 重庆大学 Multi-state joint estimation method for battery pack based on Yun Bian cooperation

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104049213A (en) * 2013-03-12 2014-09-17 三星Sdi株式会社 Apparatus for predicting state of health of battery pack by using discrete wavelet transform
CN104849666A (en) * 2014-06-23 2015-08-19 北汽福田汽车股份有限公司 Battery pack SOC evaluation method and system
CN105652206A (en) * 2015-12-26 2016-06-08 深圳市沃特玛电池有限公司 Battery pack state-of-charge (Soc) evaluation method and battery pack Soc evaluation system
CN107037374A (en) * 2017-05-04 2017-08-11 广东电网有限责任公司电力科学研究院 A kind of SOC and SOH combined estimation methods of lithium ion battery
CN107390127A (en) * 2017-07-11 2017-11-24 欣旺达电动汽车电池有限公司 A kind of SOC estimation method
CN107402353A (en) * 2017-06-30 2017-11-28 中国电力科学研究院 A kind of state-of-charge to lithium ion battery is filtered the method and system of estimation
CN107422269A (en) * 2017-06-16 2017-12-01 上海交通大学 A kind of online SOC measuring methods of lithium battery

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104049213A (en) * 2013-03-12 2014-09-17 三星Sdi株式会社 Apparatus for predicting state of health of battery pack by using discrete wavelet transform
CN104849666A (en) * 2014-06-23 2015-08-19 北汽福田汽车股份有限公司 Battery pack SOC evaluation method and system
CN105652206A (en) * 2015-12-26 2016-06-08 深圳市沃特玛电池有限公司 Battery pack state-of-charge (Soc) evaluation method and battery pack Soc evaluation system
CN107037374A (en) * 2017-05-04 2017-08-11 广东电网有限责任公司电力科学研究院 A kind of SOC and SOH combined estimation methods of lithium ion battery
CN107422269A (en) * 2017-06-16 2017-12-01 上海交通大学 A kind of online SOC measuring methods of lithium battery
CN107402353A (en) * 2017-06-30 2017-11-28 中国电力科学研究院 A kind of state-of-charge to lithium ion battery is filtered the method and system of estimation
CN107390127A (en) * 2017-07-11 2017-11-24 欣旺达电动汽车电池有限公司 A kind of SOC estimation method

Also Published As

Publication number Publication date
CN108037464A (en) 2018-05-15

Similar Documents

Publication Publication Date Title
CN108037464B (en) Battery pack SOC estimation method based on IMM-EKF
Zhang et al. State-of-charge estimation of lithium-ion battery pack by using an adaptive extended Kalman filter for electric vehicles
CN111707951B (en) Battery pack consistency evaluation method and system
Xu et al. State of charge estimation for lithium-ion batteries based on adaptive dual Kalman filter
CN110441694B (en) Lithium battery state-of-charge estimation method based on multiple fading factors Kalman filtering
Duong et al. Online state of charge and model parameters estimation of the LiFePO4 battery in electric vehicles using multiple adaptive forgetting factors recursive least-squares
CN107402353B (en) Method and system for carrying out filtering estimation on state of charge of lithium ion battery
Xiong et al. A data-driven based adaptive state of charge estimator of lithium-ion polymer battery used in electric vehicles
Huang et al. A model-based state-of-charge estimation method for series-connected lithium-ion battery pack considering fast-varying cell temperature
Zhigang et al. State of charge estimation of power Li-ion batteries using a hybrid estimation algorithm based on UKF
US20150051853A1 (en) Apparatus for parameter estimation
Li et al. A novel parameter and state-of-charge determining method of lithium-ion battery for electric vehicles
US11143705B2 (en) Method and device for detecting battery cell states and battery cell parameters
JP7030191B2 (en) Methods and devices for monitoring the stable convergence behavior of the Kalman filter
CN116113837A (en) Method for estimating state of charge of battery
Jin et al. SOC estimation of lithium-ion battery considering the influence of discharge rate
CN113466725B (en) Method and device for determining state of charge of battery, storage medium and electronic equipment
CN111027203B (en) Super capacitor SOC calculation method
Miao et al. A novel online model parameters identification method with anti‐interference characteristics for lithium‐ion batteries
Dragičević et al. Battery state-of-charge and parameter estimation algorithm based on Kalman filter
CN112946481A (en) Based on federation H∞Filtering sliding-mode observer lithium ion battery SOC estimation method and battery management system
CN115656848A (en) Lithium battery SOC estimation method based on capacity correction
Manthopoulos et al. A review and comparison of lithium-ion battery SOC estimation methods for electric vehicles
Baba et al. State of charge estimation of lithium-ion battery using Kalman filters
CN114114019A (en) Lithium ion battery SOC estimation method

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