CN111060834A - Power battery state of health estimation method - Google Patents

Power battery state of health estimation method Download PDF

Info

Publication number
CN111060834A
CN111060834A CN201911319639.5A CN201911319639A CN111060834A CN 111060834 A CN111060834 A CN 111060834A CN 201911319639 A CN201911319639 A CN 201911319639A CN 111060834 A CN111060834 A CN 111060834A
Authority
CN
China
Prior art keywords
battery
state
internal resistance
covariance
power battery
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.)
Pending
Application number
CN201911319639.5A
Other languages
Chinese (zh)
Inventor
邓浩然
方锐
张亚军
邝男男
裴志豪
齐创
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.)
China Automotive Technology and Research Center Co Ltd
CATARC Tianjin Automotive Engineering Research Institute Co Ltd
Original Assignee
China Automotive Technology and Research Center Co Ltd
CATARC Tianjin Automotive Engineering Research Institute 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 China Automotive Technology and Research Center Co Ltd, CATARC Tianjin Automotive Engineering Research Institute Co Ltd filed Critical China Automotive Technology and Research Center Co Ltd
Priority to CN201911319639.5A priority Critical patent/CN111060834A/en
Publication of CN111060834A publication Critical patent/CN111060834A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/392—Determining battery ageing or deterioration, e.g. state of health
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/3644—Constructional arrangements
    • G01R31/3648—Constructional arrangements comprising digital calculation means, e.g. for performing an algorithm
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/389—Measuring internal impedance, internal conductance or related variables

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Tests Of Electric Status Of Batteries (AREA)
  • Secondary Cells (AREA)

Abstract

The invention provides a power battery state of health estimation method, which adopts a lithium ion battery second-order Thevenin equivalent circuit model and applies an Adaptive Unscented Kalman Filter (AUKF) algorithm to estimate the battery state in real time. And (3) establishing a loop iteration relation by combining the self-adaptive unscented Kalman filtering algorithm with the unscented Kalman filtering algorithm and the extended Kalman algorithm, knowing the battery parameter to estimate the battery state, then using the battery state as the known quantity estimation model parameter, and performing recursion operation by analogy to estimate the SOC and the ohmic internal resistance of the battery in real time. The battery SOH can be estimated in real time by using the function corresponding relation between the ohmic internal resistance and the battery SOH.

Description

Power battery state of health estimation method
Technical Field
The invention belongs to the technical field of new energy automobile battery management, and particularly relates to a power battery health state estimation method.
Background
With the increasing world energy consumption and the increasing air pollution, the development of new energy automobiles becomes an important task for the development of modern industry. Among them, electric vehicles are receiving much attention due to their characteristics such as high efficiency and low pollution. The lithium ion power battery pack is the only energy storage link in the electric automobile, and when the performance of the power battery pack of the electric automobile is reduced to 80% of the original performance, the lithium ion power battery pack is not suitable for being used in the electric automobile any more. Therefore, the health state of the power battery needs to be accurately estimated to evaluate the current capacity capability of the battery in real time, so that the preparation work of maintaining or replacing the battery can be timely made, unsafe factors of the battery can be effectively found and avoided, and the stability of the power battery is guaranteed.
The estimation of the state of charge of the battery needs to be based on accurate estimation of the state of charge of the battery, but since the lithium battery is a typical nonlinear system, the error of the traditional ampere-hour integration method is more and more accumulated along with the time, and therefore the ampere-hour integration method cannot be used alone for estimating the state of charge. In order to accurately estimate the state of charge, it is very important to design a closed-loop algorithm to make up for the shortcomings of the ampere-hour integration method. The most widely used algorithm is the extended Kalman filtering algorithm, but the capacitance in the model is approximate to an integer order, so that the estimation precision of the algorithm depends heavily on the accuracy of the noise variance, and the error of the extended Kalman filtering algorithm is relatively large.
Disclosure of Invention
In view of this, the present invention is directed to a method for estimating a state of health of a power battery, so as to achieve an accurate online estimation of the state of health of the power battery.
In order to achieve the purpose, the technical scheme of the invention is realized as follows:
a power battery state of health estimation method utilizes a self-adaptive unscented Kalman filtering algorithm to estimate the state of charge of a power battery, and utilizes a relative internal resistance method to estimate the SOH of the power battery in real time.
Further, the adaptive unscented kalman filter algorithm is to combine unscented kalman filtering of an estimated state with extended kalman filtering of an estimated battery ohmic internal resistance to use, establish a loop iteration relationship, know model parameters to estimate the battery state, then use the battery state as known quantity, identify the model parameters, perform recursive operation, estimate the battery SOC and ohmic internal resistance in real time, and estimate the battery SOH in real time by using a function corresponding relationship between the ohmic internal resistance and the battery SOH.
Further, the method for estimating the state of charge of the power battery by using the adaptive unscented kalman filter algorithm specifically comprises the following steps:
1) establishing a battery state space model and discretizing;
2) and (3) carrying out initialization operation on the state variables and the covariance thereof:
Figure BDA0002326790720000021
3) building a Sigma point set related to the state quantity and a corresponding weight omega by utilizing UT transformation, and setting an expanded state variable as
Figure BDA0002326790720000025
Covariance of PX,k＝diag{PxK, Q, R }, building a Sigma point set through the expanded state variables, and setting the estimation value of the previous step of circulation state quantity as
Figure BDA0002326790720000022
The method for selecting the Sigma point set comprises the following steps
Figure BDA0002326790720000023
Wherein the content of the first and second substances,
Figure BDA0002326790720000024
the ith column representing the square root of the matrix;
the number of the Sigma points is 2n +1, and the weight algorithm of the Sigma points is as follows:
Figure BDA0002326790720000031
wherein, ω is(m)Representing the weight, ω, corresponding to the mean estimate(c)Representing the weight corresponding to the covariance estimation, n being the dimension of the expanded state variable, α describing the degree of the Sigma point deviating from the estimated state value, satisfying 1e-4≤α＜1；λ＝α2(n + kappa) -n is a Sigma point scaling parameter, kappa is greater than or equal to 0, β is a quantity related to the Sigma point distribution, and when the Sigma points are in Gaussian distribution, β is 2;
4) sigma point set obtained by using Sigma point set selection method
Figure BDA0002326790720000032
5) One-step prediction for computing 2n +1 Sigma point sets
Figure BDA0002326790720000033
6) Utilizing weight omega in Sigma point weight algorithmiWeighting and summing the predicted values of the Sigma point set to obtain a one-step predicted value and a covariance matrix of the battery state quantity:
Figure BDA0002326790720000034
7) according to the one-step predicted value, UT transformation is utilized again to generate a new Sigma point set;
Figure BDA0002326790720000035
8) substituting the Sigma point set obtained in the step 6) into an observation equation to obtain one-step prediction of observed quantity
Figure BDA0002326790720000041
9) Weighting and summing the predicted values by one step to obtain the predicted value and covariance matrix of the observed quantity:
Figure BDA0002326790720000042
10) calculating a Kalman gain matrix:
Figure BDA0002326790720000043
11) calculating a state update and a covariance update of the battery:
Figure BDA0002326790720000044
12) mean and covariance update of process noise and measurement noise:
Figure BDA0002326790720000045
wherein b is a forgetting factor, 0<b<1，dk＝(1-b)/(1-bk) By using the formula, the noise parameter value can be updated online in real time, and the estimated value of the state variable SOC is corrected in real time;
13) calculating a state variable R using an extended Kalman filter algorithm0The measurement update gain of
Figure BDA0002326790720000046
14) Finally, the state variable R0The estimate and variance estimate are
Figure BDA0002326790720000051
And substituting the corresponding operation quantity in the state space expression into the self-adaptive unscented Kalman filtering algorithm by combining the discretized state space model and the state space model taking the ohmic internal resistance as the state variable, and estimating the SOC and the ohmic internal resistance of the battery in real time through cyclic iterative computation.
Further, the calculation of battery SOH using battery internal resistance is defined as
Figure BDA0002326790720000052
Wherein R isEOLIs the ohmic internal resistance value, R, at the end of the battery lifeNIs the ohmic resistance value, R, of the battery when it leaves the factorynowAnd substituting the ohmic internal resistance value of each single battery estimated by the self-adaptive unscented Kalman filtering algorithm into the formula to obtain the real-time SOH estimated value of each single battery.
Further, a second-order Thevenin equivalent circuit model is selected to establish a battery state space equation.
Further, the equivalent circuit model comprises a voltage source, ohmic internal resistance and two RC parallel circuits which are connected in series, and the Uoc represents the open-circuit voltage of the battery and is related to the SOC of the battery; r0Is the ohmic internal resistance of the cell; the RC parallel circuit describes the polarization characteristic of the battery, R1C1The circuit represents the electrochemical polarization process, R2C2The loop represents a concentration polarization process; u shapetFor the terminal voltage of the battery, the following equation can be established by a second-order Thevenin equivalent circuit model of the battery
Figure BDA0002326790720000053
The SOC of the battery is obtained by ampere-hour integration method, and the battery discharge time satisfies
Figure BDA0002326790720000061
Wherein C is the total ampere-hour capacity of the battery, η is the coulombic efficiency, also called charging efficiency, and refers to the ratio of the discharge capacity of the battery to the charging capacity in the circulation process;
the state equation and the output equation under the second-order equivalent circuit model of the battery can be obtained by the two formulas and are respectively as follows:
Figure BDA0002326790720000062
U(t)＝Uoc(SOC(t))-Up1(t)-Up2(t)-R0I(t)。
further, establishing a discretized battery state space model specifically comprises;
discretization form of second-order Thevenin equivalent circuit model:
Figure BDA0002326790720000063
Uk＝UOC(Soc,k)-ikR0-Up1,k-Up2,k+vk
wherein the state space variable is xk＝[Soc,k,Up1,k,Up2,k]T(ii) a The controlled variable is ik(ii) a Observed variable is yk＝Uk；wk＝[w1,k,w2,k,w3,k]TIs system noise, and the covariance is Q; v. ofkFor observation noise, the covariance is R; Δ t is the sampling interval; soc,k,Up1,k,Up2,kRespectively refers to the SOC of the battery at the kth sampling point moment and the terminal voltages of two RC parallel circuits at the moment, and the state space coefficient matrix is
Figure BDA0002326790720000071
Figure BDA0002326790720000072
Figure BDA0002326790720000073
Figure BDA0002326790720000074
In addition, the ohmic internal resistance R of the battery0As state space variables, the following discretized state space equations and output observation equations are obtained:
R0,k+1＝R0,k+rk
Uk＝UOC(Soc,k)-ikR0-Up1,k-Up2,k+ek
rkand ekRespectively, system noise and observation noise, with covariance Dr and De.
Another objective of the present invention is to provide a method and system for estimating a state of health of a power battery, including a battery state of charge calculation module, a battery parameter update module, a battery parameter identification module, and a battery state update module.
Another object of the present invention is to provide a storage medium, on which a computer program is stored, which when executed by a processor implements the above-mentioned power battery state of health estimation method.
Another object of the present invention is to provide an electronic device, comprising:
a processor; and the number of the first and second groups,
a memory for storing executable instructions of the processor; wherein the processor is configured to execute the above-described power battery state of health estimation method via execution of the executable instructions.
Compared with the prior art, the method for estimating the state of health of the power battery has the following advantages:
(1) the estimation algorithm of the SOC and the internal resistance of the battery designed by the invention is not limited by current working conditions, and the estimation precision is higher.
(2) The algorithm designed by the invention estimates the Kalman gain matrix by using two times of lossless transformation, increases the operation iteration times and improves the precision of the algorithm.
(3) The method provided by the invention adopts the adaptive unscented Kalman filtering algorithm to identify the ohmic internal resistance of the time-varying battery system, and then utilizes the functional relation between the internal resistance and the SOH of the battery to estimate the SOH of the battery in real time.
(4) The algorithm designed by the invention utilizes the measurement data to estimate the mean value and the variance of the noise on line, and the SOC value is continuously corrected according to the estimation result of the mean value and the variance of the noise of each sampling point, so that the influence of a tiny model error on the state of charge estimation precision can be corrected on line in real time. The algorithm has good adaptivity, and overcomes the defects of large noise error, low precision, poor stability and the like of the extended Kalman algorithm estimation.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate an embodiment of the invention and, together with the description, serve to explain the invention and not to limit the invention. In the drawings:
FIG. 1 is a schematic diagram illustrating a method for estimating a state of health of a power battery according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a second-order Thevenin equivalent circuit model of a lithium ion battery according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a voltage waveform of a battery terminal after pulse discharge according to an embodiment of the present invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The present invention will be described in detail below with reference to the embodiments with reference to the attached drawings.
The invention adopts a lithium ion battery second-order Thevenin equivalent circuit model and applies an Adaptive Unscented Kalman Filter (AUKF) algorithm to estimate the battery state in real time. And (3) establishing a loop iteration relation by combining the self-adaptive unscented Kalman filtering algorithm with the unscented Kalman filtering algorithm and the extended Kalman algorithm, knowing the battery parameter to estimate the battery state, then using the battery state as the known quantity estimation model parameter, and performing recursion operation by analogy to estimate the SOC and the ohmic internal resistance of the battery in real time. The battery SOH can be estimated in real time by using the function corresponding relation between the ohmic internal resistance and the battery SOH.
Establishment of battery state space model
A second-order Thevenin equivalent circuit model is selected to establish a state space equation of the battery, the model is formed by connecting a voltage source, ohmic internal resistance and two RC parallel circuits in series, and the circuit form is shown in figure 1. Uoc represents the battery open circuit voltage, related to the battery SOC; r0Is the ohmic internal resistance of the cell; the RC parallel circuit describes the polarization characteristic of the battery, R1C1The circuit represents the electrochemical polarization process, R2C2The loop represents a concentration polarization process; u shapetIs the terminal voltage of the battery.
From the second-order Thevenin equivalent circuit model of the battery, the following equation can be established
Figure BDA0002326790720000091
The SOC of the battery is obtained by ampere-hour integration method, and the battery discharge time satisfies
Figure BDA0002326790720000092
Where C is the total ampere-hour capacity of the cell and η is the coulombic efficiency, also called the charge efficiency, which refers to the ratio of the cell discharge capacity to the charge capacity during this cycle.
The state equation and the output equation under the second-order equivalent circuit model of the battery can be obtained by the formulas (1) and (2) and are respectively:
Figure BDA0002326790720000101
U(t)＝Uoc(SOC(t))-Up1(t)-Up2(t)-R0I(t) (4)
battery model parameter identification method
The invention utilizes the current pulse experimental data of the single battery to identify the electrical characteristic parameters of the battery.
(1) Ohmic internal resistance R0And (4) identifying.
Fig. 3 is a terminal voltage curve of primary pulse discharge of the lithium ion battery, wherein the current pulse starts at the time point a and ends at the time point C. The graph shows that the voltages of the AB section and the CD section are instantaneously dropped and risen. The instantaneous drop and the instantaneous rise are caused by the ohmic internal resistance of the battery, so the ohmic internal resistance identification formula is
Figure BDA0002326790720000102
(2) Identification of polarization resistance-capacitance parameters
In FIG. 2, the DE section RC network is zero input response, and the end voltage equation in the second-order equivalent circuit model of the battery meets the requirement
Figure BDA0002326790720000103
Up1(t0) For R in the equivalent circuit model of the battery1C1Initial value of network terminal voltage, Up2(t0) Is R2C2Initial value of network terminal voltage, Uoc (t)0) Is t0The open circuit voltage of the battery at that time. And fitting the terminal voltage response in the standing period after the pulse is ended by a matlab least square fitting method to obtain the time constants of the two RC networks at the moment and the open-circuit voltage of the battery.
And the battery starts pulse discharge after standing for a period of time, and at the moment, the RC network meets zero-state response. Terminal voltage of equivalent circuit model satisfies equation
Figure BDA0002326790720000111
And substituting the time constant obtained by fitting into a formula (7), and fitting the terminal voltage response of a period of time after the pulse starts by utilizing matlab least square fitting to respectively obtain the R value and the C value of the two RC networks.
And fitting to obtain the relation between the Open Circuit Voltage (OCV) and the state of charge (SOC) of the single battery by using the parameter identification result.
Adaptive unscented kalman filter algorithm to estimate battery state
The self-adaptive unscented Kalman filtering algorithm is used by combining unscented Kalman filtering in an estimation state with extended Kalman filtering for estimating the ohmic internal resistance of the battery, and a loop iteration relation is established. And estimating the battery state by using the known model parameters, and identifying the model parameters by using the battery state as a known quantity to perform recursive operation. The algorithm has good adaptivity.
(1) Establishing a discretized model
The discretization form of the second-order Thevenin equivalent circuit model can be obtained by the following equations (1) and (2):
Figure BDA0002326790720000112
Uk＝UOC(Soc,k)-ikR0-Up1,k-Up2,k+vk(9)
wherein the state space variable is xk＝[Soc,k,Up1,k,Up2,k]T(ii) a The controlled variable is ik(ii) a Observed variable is yk＝Uk；wk＝[w1,k,w2,k,w3,k]TIs system noise, and the covariance is Q; v. ofkFor observation noise, the covariance is R; Δ t is the sampling interval; soc,k,Up1,k,Up2,kRespectively referring to the SOC of the battery at the k-th sampling point moment and the terminal voltages of the two RC parallel circuits at the moment. The state space coefficient matrix is
Figure BDA0002326790720000121
Figure BDA0002326790720000122
Figure BDA0002326790720000123
Figure BDA0002326790720000124
In addition, the ohmic internal resistance R of the battery0As the state space variables, the following discretized state space equation and output observation equation can be obtained:
Figure BDA0002326790720000125
rkand ekRespectively, system noise and observation noise, with covariance Dr and De.
(2) AUKF estimates battery SOC and ohmic internal resistance
And carrying out lossless transformation (UT) near the estimation point when the nonlinear filtering is processed by the adaptive unscented Kalman filtering algorithm, matching the obtained mean value and covariance of the Sigma point sets with the original statistical characteristics, directly carrying out nonlinear mapping on the Sigma point sets to approximately obtain a state probability density function, and establishing a circular iteration relation.
The method for estimating the battery state by adopting the self-adaptive unscented Kalman filtering algorithm comprises the following steps:
1) and (3) carrying out initialization operation on the state variables and the covariance thereof:
Figure BDA0002326790720000131
2) and constructing a Sigma point set related to the state quantity and a corresponding weight omega by utilizing UT transformation. Let the expanded state variable be
Figure BDA0002326790720000137
Covariance of PX,k＝diag{Px,kQ, R, building a Sigma point set through the expanded state variables, and setting the estimation value of the previous step circulation state quantity as
Figure BDA0002326790720000132
The method for selecting the Sigma point set comprises the following steps
Figure BDA0002326790720000133
Wherein the content of the first and second substances,
Figure BDA0002326790720000134
the ith column representing the square root of the matrix.
The number of the Sigma points is 2n +1, and the weight algorithm of the Sigma points is as follows:
Figure BDA0002326790720000135
wherein, ω is(m)Representing the weight, ω, corresponding to the mean estimate(c)Representing the weight corresponding to the covariance estimation, n being the dimension of the expanded state variable, α describing the degree of the Sigma point deviating from the estimated state value, satisfying 1e-4≤α＜1；λ＝α2And (n + kappa) -n is a Sigma point scaling parameter, kappa is greater than or equal to 0, β is a quantity related to the Sigma point distribution, and when the Sigma points are in Gaussian distribution, β is equal to 2.
3) Sigma point set obtained by the formula (16)
Figure BDA0002326790720000136
4) One-step prediction for computing 2n +1 Sigma point sets
Figure BDA0002326790720000141
5) Using the weight ω in equation (17)iWeighting and summing the predicted values of the Sigma point set to obtain a one-step predicted value and a covariance matrix of the battery state quantity:
Figure BDA0002326790720000142
6) and generating a new Sigma point set by using UT transformation again according to the one-step predicted value.
Figure BDA0002326790720000143
7) Substituting the Sigma point set obtained in the step 6) into an observation equation to obtain one-step prediction of observed quantity
Figure BDA0002326790720000144
8) Weighting and summing the predicted values by one step to obtain the predicted value and covariance matrix of the observed quantity:
Figure BDA0002326790720000145
9) calculating a Kalman gain matrix:
Figure BDA0002326790720000146
10) calculating a state update and a covariance update of the battery:
Figure BDA0002326790720000147
12) mean and covariance update of process noise and measurement noise:
Figure BDA0002326790720000151
wherein b is a forgetting factor, 0<b<1，dk＝(1-b)/(1-bk). By using the formula, the noise parameter value can be updated online in real time, and the estimated value of the state variable SOC is corrected in real time.
13) Calculating a state variable R using an extended Kalman filter algorithm0The measurement update gain of
Figure BDA0002326790720000152
14) Finally, the state variable R0The estimate and variance estimate are
Figure BDA0002326790720000153
And substituting corresponding operation quantities in the state space expression into an AUKF algorithm by combining the state space models (8) and (9) and the state space model (14) taking the ohmic internal resistance as the state variable, and estimating the SOC and the ohmic internal resistance of the battery in real time through cyclic iterative calculation.
(3) Estimating battery SOH by internal resistance method
The definition of the calculation of the SOH of the battery by using the internal resistance of the battery is
Figure BDA0002326790720000154
Wherein R isEOLIs the ohmic internal resistance value, R, at the end of the battery lifeNIs the ohmic resistance value, R, of the battery when it leaves the factorynowThe ohmic internal resistance value of the battery under the current state is obtained. And substituting the ohmic internal resistance value of each single battery estimated by the AUKF algorithm into the formula (29) to obtain the real-time SOH estimated value of each single battery.
The invention provides an algorithm for estimating SOC and SOH of a power battery. The estimation result of the algorithm can be compared with the experimental result to verify the estimation precision. In practical application, the program of the algorithm is embedded into a battery management system background and is used for high-precision estimation of the SOC and the SOH of the power battery in the battery management system.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.

Claims (10)

1. A power battery state of health estimation method characterized by: estimating the state of charge of the power battery by using a self-adaptive unscented Kalman filtering algorithm, and estimating the SOH of the power battery in real time by using a relative internal resistance method.
2. The power battery state of health estimation method of claim 1, characterized in that: the self-adaptive unscented Kalman filtering algorithm is characterized in that unscented Kalman filtering in an estimation state and extended Kalman filtering for estimating ohmic internal resistance of a battery are combined for use, a loop iteration relation is established, model parameters are known to estimate the state of the battery, then the state of the battery is used as known quantity, the model parameters are identified to perform recursive operation, the SOC and the ohmic internal resistance of the battery are estimated in real time, and then the SOH of the battery is estimated in real time by utilizing the function corresponding relation of the ohmic internal resistance and the SOH of the battery.
3. The power battery state of health estimation method of claim 2, characterized in that: the method for estimating the state of charge of the power battery by using the adaptive unscented kalman filter algorithm specifically comprises the following steps:
1) establishing a battery state space model and discretizing;
2) and (3) carrying out initialization operation on the state variables and the covariance thereof:
Figure FDA0002326790710000011
3) building a Sigma point set related to the state quantity and a corresponding weight omega by utilizing UT transformation, and setting an expanded state variable as
Figure FDA0002326790710000012
Covariance of PX,k＝diag{Px,kQ, R, building a Sigma point set through the expanded state variables, and setting the estimation value of the previous step circulation state quantity as
Figure FDA0002326790710000013
The method for selecting the Sigma point set comprises the following steps
Figure FDA0002326790710000014
Wherein the content of the first and second substances,
Figure FDA0002326790710000021
Figure FDA0002326790710000022
the ith column representing the square root of the matrix;
the number of the Sigma points is 2n +1, and the weight algorithm of the Sigma points is as follows:
Figure FDA0002326790710000023
wherein, ω is(m)Representing the weight, ω, corresponding to the mean estimate(c)Representing the weight corresponding to the covariance estimation, n being the dimension of the expanded state variable, α describing the degree of the Sigma point deviating from the estimated state value, satisfying 1e-4≤α＜1；λ＝α2(n + kappa) -n is a Sigma point scaling parameter, kappa is greater than or equal to 0, β is a quantity related to the Sigma point distribution, and when the Sigma points are in Gaussian distribution, β is 2;
4) sigma point set obtained by using Sigma point set selection method
Figure FDA0002326790710000024
5) One-step prediction for computing 2n +1 Sigma point sets
Figure FDA0002326790710000025
6) Utilizing weight omega in Sigma point weight algorithmiWeighting and summing the predicted values of the Sigma point set to obtain a one-step predicted value and a covariance matrix of the battery state quantity:
Figure FDA0002326790710000026
7) according to the one-step predicted value, UT transformation is utilized again to generate a new Sigma point set;
Figure FDA0002326790710000031
8) substituting the Sigma point set obtained in the step 6) into an observation equation to obtain one-step prediction of observed quantity
Figure FDA0002326790710000032
9) Weighting and summing the predicted values by one step to obtain the predicted value and covariance matrix of the observed quantity:
Figure FDA0002326790710000033
10) calculating a Kalman gain matrix:
Figure FDA0002326790710000034
11) calculating a state update and a covariance update of the battery:
Figure FDA0002326790710000035
12) mean and covariance update of process noise and measurement noise:
Figure FDA0002326790710000036
wherein b is a forgetting factor, 0<b<1，dk＝(1-b)/(1-bk) By using the formula, the noise parameter value can be updated online in real time, and the estimated value of the state variable SOC is corrected in real time;
13) calculating a state variable R using an extended Kalman filter algorithm0The measurement update gain of
Figure FDA0002326790710000041
14) Finally, the state variable R0The estimate and variance estimate are
Figure FDA0002326790710000042
And substituting the corresponding operation quantity in the state space expression into the self-adaptive unscented Kalman filtering algorithm by combining the discretized state space model and the state space model taking the ohmic internal resistance as the state variable, and estimating the SOC and the ohmic internal resistance of the battery in real time through cyclic iterative computation.
4. The power battery state of health estimation method of claim 1, characterized in that: the definition of the calculation of the SOH of the battery by using the internal resistance of the battery is
Figure FDA0002326790710000043
Wherein R isEOLIs the ohmic internal resistance value, R, at the end of the battery lifeNIs the ohmic resistance value, R, of the battery when it leaves the factorynowAnd substituting the ohmic internal resistance value of each single battery estimated by the self-adaptive unscented Kalman filtering algorithm into the formula to obtain the real-time SOH estimated value of each single battery.
5. The power battery state of health estimation method of claim 3, characterized in that: and (4) selecting a second-order Thevenin equivalent circuit model to establish a battery state space equation.
6. The power battery state of health estimation method of claim 5, characterized in that: the equivalent circuit model comprises a voltage source, ohmic internal resistance and two RC parallel circuits which are connected in series, wherein the Uoc represents the open-circuit voltage of the battery and is related to the SOC of the battery; r0Is the ohmic internal resistance of the cell; the RC parallel circuit describes the polarization characteristic of the battery, R1C1The circuit represents the electrochemical polarization process, R2C2The loop represents a concentration polarization process; u shapetFor the terminal voltage of the battery, the following equation can be established by a second-order Thevenin equivalent circuit model of the battery
Figure FDA0002326790710000051
The SOC of the battery is obtained by ampere-hour integration method, and the battery discharge time satisfies
Figure FDA0002326790710000052
Wherein C is the total ampere-hour capacity of the battery, η is the coulombic efficiency, also called charging efficiency, and refers to the ratio of the discharge capacity of the battery to the charging capacity in the circulation process;
the state equation and the output equation under the second-order equivalent circuit model of the battery can be obtained by the two formulas and are respectively as follows:
Figure FDA0002326790710000053
U(t)＝Uoc(SOC(t))-Up1(t)-Up2(t)-R0I(t)。
7. the power battery state of health estimation method of claim 6, characterized in that: establishing a discretized battery state space model specifically comprises the following steps;
discretization form of second-order Thevenin equivalent circuit model:
Figure FDA0002326790710000061
Uk＝UOC(Soc,k)-ikR0-Up1,k-Up2,k+vk
wherein the state space variable is xk＝[Soc,k,Up1,k,Up2,k]T(ii) a The controlled variable is ik(ii) a Observed variable is yk＝Uk；wk＝[w1,k,w2,k,w3,k]TIs system noise, and the covariance is Q; v. ofkFor observation noise, the covariance is R; Δ t is the sampling interval; soc,k,Up1,k,Up2,kRespectively refers to the SOC of the battery at the kth sampling point moment and the terminal voltages of two RC parallel circuits at the moment, and the state space coefficient matrix is
Figure FDA0002326790710000062
Figure FDA0002326790710000063
Figure FDA0002326790710000064
Figure FDA0002326790710000065
In addition, the ohmic internal resistance R of the battery0As state space variables, the following discretized state space equations and output observation equations are obtained:
R0,k+1＝R0,k+rk
Uk＝UOC(Soc,k)-ikR0-Up1,k-Up2,k+ek
rkand ekRespectively, system noise and observation noise, with covariance Dr and De.
8. A power battery state of health estimation method system that employs a power battery state of health estimation method of claims 1-7, characterized by: the system comprises a battery state of charge calculation module, a battery parameter updating module, a battery parameter identification module and a battery state updating module.
9. A storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the power battery state of health estimation method of any one of claims 1 to 7.
10. An electronic device, comprising:
a processor; and the number of the first and second groups,
a memory for storing executable instructions of the processor; wherein the processor is configured to perform the power battery state of health estimation method of any one of claims 1 to 7 via execution of the executable instructions.
CN201911319639.5A 2019-12-19 2019-12-19 Power battery state of health estimation method Pending CN111060834A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911319639.5A CN111060834A (en) 2019-12-19 2019-12-19 Power battery state of health estimation method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911319639.5A CN111060834A (en) 2019-12-19 2019-12-19 Power battery state of health estimation method

Publications (1)

Publication Number Publication Date
CN111060834A true CN111060834A (en) 2020-04-24

Family

ID=70300788

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911319639.5A Pending CN111060834A (en) 2019-12-19 2019-12-19 Power battery state of health estimation method

Country Status (1)

Country Link
CN (1) CN111060834A (en)

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111426968A (en) * 2020-05-28 2020-07-17 山东交通学院 Method and system for estimating SOP of power battery of electric vehicle
CN111505506A (en) * 2020-05-15 2020-08-07 吉林大学 Battery SOC estimation method integrating multi-scale Kalman filtering and unscented Kalman filtering
CN111551869A (en) * 2020-05-15 2020-08-18 江苏科尚智能科技有限公司 Method and device for measuring low-frequency parameters of lithium battery, computer equipment and storage medium
CN111983476A (en) * 2020-08-24 2020-11-24 哈尔滨理工大学 Battery safety degree estimation method and device based on Kalman filtering method
CN112329352A (en) * 2021-01-06 2021-02-05 山东广域科技有限责任公司 Supercapacitor service life prediction method and device applied to power system
CN112379295A (en) * 2020-09-29 2021-02-19 合肥工业大学 Method, system and storage medium for predicting health state of power battery
CN112698231A (en) * 2020-11-20 2021-04-23 深圳供电局有限公司 Battery health state detection method and device, control equipment and storage medium
CN112710955A (en) * 2020-12-16 2021-04-27 上海理工大学 Algorithm for improving battery capacity estimation precision
CN112733411A (en) * 2020-12-18 2021-04-30 武汉大学 Lithium battery SOC estimation method and system based on second-order differential particle filter
CN112858928A (en) * 2021-03-08 2021-05-28 安徽理工大学 Lithium battery SOC estimation method based on online parameter identification
CN112858920A (en) * 2021-02-01 2021-05-28 山西国润储能科技有限公司 SOC estimation method of all-vanadium redox flow battery fusion model based on adaptive unscented Kalman filtering
CN113567861A (en) * 2021-07-06 2021-10-29 上海电机学院 Estimation method for health state of power battery
CN114239463A (en) * 2021-12-16 2022-03-25 杭州科工电子科技有限公司 Battery cluster state of charge correction method based on big data
CN115453376A (en) * 2022-10-26 2022-12-09 国网浙江省电力有限公司宁波市北仑区供电公司 State monitoring method and system for power battery of electric automobile and electric automobile
CN116500458A (en) * 2023-06-27 2023-07-28 中国第一汽车股份有限公司 Power battery capacity evaluation method and device, vehicle and electronic device
CN112327183B (en) * 2020-09-18 2023-11-28 国联汽车动力电池研究院有限责任公司 Lithium ion battery SOC estimation method and device
US20230393206A1 (en) * 2021-02-24 2023-12-07 Pmgrow Corporation Evaluation method and device according to purposes of reusable battery

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104698382A (en) * 2013-12-04 2015-06-10 东莞钜威新能源有限公司 Method for predicting the SOC and SOH of battery pack

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104698382A (en) * 2013-12-04 2015-06-10 东莞钜威新能源有限公司 Method for predicting the SOC and SOH of battery pack

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
颜湘武等: "基于自适应无迹卡尔曼滤波的动力电池健康状态检测及梯次利用研究", 《电工技术学报》 *

Cited By (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111505506A (en) * 2020-05-15 2020-08-07 吉林大学 Battery SOC estimation method integrating multi-scale Kalman filtering and unscented Kalman filtering
CN111551869A (en) * 2020-05-15 2020-08-18 江苏科尚智能科技有限公司 Method and device for measuring low-frequency parameters of lithium battery, computer equipment and storage medium
CN111426968A (en) * 2020-05-28 2020-07-17 山东交通学院 Method and system for estimating SOP of power battery of electric vehicle
CN111983476B (en) * 2020-08-24 2022-08-30 哈尔滨理工大学 Battery safety degree estimation method and device based on Kalman filtering method
CN111983476A (en) * 2020-08-24 2020-11-24 哈尔滨理工大学 Battery safety degree estimation method and device based on Kalman filtering method
CN112327183B (en) * 2020-09-18 2023-11-28 国联汽车动力电池研究院有限责任公司 Lithium ion battery SOC estimation method and device
CN112379295A (en) * 2020-09-29 2021-02-19 合肥工业大学 Method, system and storage medium for predicting health state of power battery
CN112379295B (en) * 2020-09-29 2023-08-04 合肥工业大学 Method, system and storage medium for predicting power battery health status
CN112698231A (en) * 2020-11-20 2021-04-23 深圳供电局有限公司 Battery health state detection method and device, control equipment and storage medium
CN112710955A (en) * 2020-12-16 2021-04-27 上海理工大学 Algorithm for improving battery capacity estimation precision
CN112710955B (en) * 2020-12-16 2022-11-01 上海理工大学 Algorithm for improving battery capacity estimation precision
CN112733411A (en) * 2020-12-18 2021-04-30 武汉大学 Lithium battery SOC estimation method and system based on second-order differential particle filter
CN112329352A (en) * 2021-01-06 2021-02-05 山东广域科技有限责任公司 Supercapacitor service life prediction method and device applied to power system
CN112858920A (en) * 2021-02-01 2021-05-28 山西国润储能科技有限公司 SOC estimation method of all-vanadium redox flow battery fusion model based on adaptive unscented Kalman filtering
US12007447B2 (en) * 2021-02-24 2024-06-11 Pmgrow Corporation Evaluation method and device according to purposes of reusable battery
US20230393206A1 (en) * 2021-02-24 2023-12-07 Pmgrow Corporation Evaluation method and device according to purposes of reusable battery
CN112858928B (en) * 2021-03-08 2024-02-06 安徽理工大学 Lithium battery SOC estimation method based on online parameter identification
CN112858928A (en) * 2021-03-08 2021-05-28 安徽理工大学 Lithium battery SOC estimation method based on online parameter identification
CN113567861A (en) * 2021-07-06 2021-10-29 上海电机学院 Estimation method for health state of power battery
CN114239463A (en) * 2021-12-16 2022-03-25 杭州科工电子科技有限公司 Battery cluster state of charge correction method based on big data
CN114239463B (en) * 2021-12-16 2022-11-01 杭州科工电子科技有限公司 Battery cluster state of charge correction method based on big data
CN115453376A (en) * 2022-10-26 2022-12-09 国网浙江省电力有限公司宁波市北仑区供电公司 State monitoring method and system for power battery of electric automobile and electric automobile
CN116500458A (en) * 2023-06-27 2023-07-28 中国第一汽车股份有限公司 Power battery capacity evaluation method and device, vehicle and electronic device
CN116500458B (en) * 2023-06-27 2023-09-22 中国第一汽车股份有限公司 Power battery capacity evaluation method and device, vehicle and electronic device

Similar Documents

Publication Publication Date Title
CN111060834A (en) Power battery state of health estimation method
CN110488194B (en) Lithium battery SOC estimation method and system based on electrochemical impedance model
CN111581904B (en) Lithium battery SOC and SOH collaborative estimation method considering cycle number influence
CN110286332B (en) Electric vehicle power battery SOC estimation method based on multiple innovation theory
CN110346734B (en) Machine learning-based lithium ion power battery health state estimation method
CN112034349B (en) Lithium battery health state online estimation method
CN111007400A (en) Lithium battery SOC estimation method based on self-adaptive double-extended Kalman filtering method
CN111505506A (en) Battery SOC estimation method integrating multi-scale Kalman filtering and unscented Kalman filtering
CN110824363B (en) Lithium battery SOC and SOE joint estimation method based on improved CKF
CN108445422B (en) Battery state of charge estimation method based on polarization voltage recovery characteristics
CN112269133B (en) SOC estimation method based on pre-charging circuit model parameter identification
CN111142025A (en) Battery SOC estimation method and device, storage medium and electric vehicle
CN112580284A (en) Hybrid capacitor equivalent circuit model and online parameter identification method
CN114660464A (en) Lithium ion battery state of charge estimation method
CN112528472A (en) Multi-innovation hybrid Kalman filtering and H-infinity filtering algorithm
CN111400935B (en) Lithium battery three-order model parameter identification method and system based on likelihood function
CN112946481A (en) Based on federation H∞Filtering sliding-mode observer lithium ion battery SOC estimation method and battery management system
CN112580289A (en) Hybrid capacitor power state online estimation method and system
CN115754741A (en) SOC estimation method and device of energy storage system, equipment and storage medium
CN114814591A (en) Lithium battery SOE estimation method, device and system
CN114779089A (en) Method for calculating battery state of charge based on energy storage lithium battery equivalent circuit model
CN112946480B (en) Lithium battery circuit model simplification method for improving SOC estimation real-time performance
CN112816873A (en) IEKF lithium battery SOC estimation method based on improved battery model
CN113125969A (en) Battery data processing method, device and medium based on AUKF
CN115436806A (en) SOC and SOH self-adaptive collaborative estimation method of lithium ion battery

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20200424