CN113702838B - Lithium ion battery state of charge estimation method based on disturbance observer - Google Patents

Lithium ion battery state of charge estimation method based on disturbance observer Download PDF

Info

Publication number
CN113702838B
CN113702838B CN202110995316.9A CN202110995316A CN113702838B CN 113702838 B CN113702838 B CN 113702838B CN 202110995316 A CN202110995316 A CN 202110995316A CN 113702838 B CN113702838 B CN 113702838B
Authority
CN
China
Prior art keywords
battery
soc
voltage
value
model
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
CN202110995316.9A
Other languages
Chinese (zh)
Other versions
CN113702838A (en
Inventor
刘思佳
向枫
范世军
代高强
彭诚
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.)
Sichuan Changhong Battery Co ltd
Original Assignee
Sichuan Changhong Battery 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 Sichuan Changhong Battery Co ltd filed Critical Sichuan Changhong Battery Co ltd
Priority to CN202110995316.9A priority Critical patent/CN113702838B/en
Publication of CN113702838A publication Critical patent/CN113702838A/en
Application granted granted Critical
Publication of CN113702838B publication Critical patent/CN113702838B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • Secondary Cells (AREA)
  • Tests Of Electric Status Of Batteries (AREA)

Abstract

The invention relates to the technical field of battery management systems, in the lithium ion battery state of charge estimation method based on a disturbance observer, a battery state estimation model is established through a first-order equivalent circuit model of the lithium ion battery, errors existing in the first-order equivalent circuit model are corrected in real time by using the disturbance observer, a battery voltage disturbance observer model is established, the battery state estimation model is combined with the battery voltage disturbance observer model to obtain a battery state equation and a battery observation equation, a Kalman gain matrix is obtained by adopting an extended Kalman filtering algorithm according to the battery state equation and the battery observation equation, and the battery state of charge of the lithium ion battery is estimated according to the battery state equation, the battery observation equation and the Kalman gain matrix, so that the problem that the accuracy of an SOC estimation result is reduced along with the increase of the service time and/or the cycle times of the battery is solved.

Description

Lithium ion battery state of charge estimation method based on disturbance observer
Technical Field
The invention relates to a lithium ion battery, in particular to a lithium ion battery state of charge estimation method based on a disturbance observer.
Background
The common estimation method of the State of Charge (SOC) of the lithium ion battery is an ampere-hour integration method based on SOC definition, but due to the open loop characteristic of the ampere-hour integration method, errors existing in initial assignment are difficult to eliminate, accumulated errors are easy to generate in the process of multiple estimation, so that in order to reduce the estimation errors of the State of Charge, the SOC estimation method of a battery model combined with a Kalman filtering algorithm is adopted, the accuracy requirement on parameters of the battery model is high, and the difference between the parameters of the battery model and actual parameters directly leads to the reduction of the accuracy of an estimation result.
The battery model parameters are generally obtained by battery standardized test data, a first-order model is more adopted, the complexity of the model is lower, the model can be realized by only needing fewer model parameters, engineering application is easy, but obvious dynamic errors exist in the simulation response effect on the battery polarization voltage, the hysteresis characteristics of the lithium ion battery are not considered by the battery model, namely, in the discharging and charging processes, the inconsistency exists between the corresponding relation between the open-circuit voltage and the SOC in the actual operation of the battery and the relation curve between the open-circuit voltage and the SOC obtained by experimental data, on the other hand, after the lithium battery is charged and discharged for a long time, the actual parameters of the battery can be greatly changed, the model parameter deviation in the SOC estimation algorithm is also increased, and the accuracy of the SOC estimation method can be reduced along with the increase of the service time and the cycle times of the battery.
Disclosure of Invention
The technical problems solved by the invention are as follows: the lithium ion battery state of charge estimation method based on the disturbance observer solves the problem that the accuracy of the SOC estimation result is reduced along with the increase of the battery service time and/or the cycle use times.
The invention solves the technical problems by adopting the technical scheme that: the lithium ion battery state of charge estimation method based on the disturbance observer comprises the following steps:
s01, establishing a battery voltage disturbance observer model according to a disturbance observer principle and an equivalent circuit mathematical model of the lithium ion battery, and obtaining a voltage disturbance variable estimated value;
s02, subtracting the voltage disturbance variable estimated value of the same period from the polarization voltage value in the battery voltage disturbance observer model, and discretizing to obtain a battery compensation voltage variable;
s03, discretizing a first-order equivalent circuit mathematical model formula of the lithium ion battery, and performing first-order Taylor expansion linearization to obtain a battery state estimation model;
s04, combining the battery state estimation model with the battery voltage disturbance observer model to obtain a battery state equation and a battery observation equation;
s05, acquiring a Kalman gain matrix by adopting an extended Kalman filtering algorithm according to a battery state equation and a battery observation equation;
s06, estimating the battery state of charge of the lithium ion battery according to a battery state equation, a battery observation equation and a Kalman gain matrix.
Further, in step S01, the equivalent mathematical model of the lithium ion battery is a first-order equivalent circuit model, and the formula of the first-order equivalent circuit model is:wherein U is p Is polarization voltage, +.>Is U p With respect to the derivative of time t, d is the differential operator, τ p Is the polarization time constant and τ p ＝R p C p ，R P Is a polarization resistance, C p Is a polarized capacitor, U oc Is the open circuit voltage value, U b Is the voltage value of the battery terminal, R o Is ohmic internal resistance value, I b Is a battery current value, the battery current I b The mathematical formula of (a) is: />
Further, in step S01, the battery voltage disturbance observer model includes a nominal model P n (s) is: p (P) n (s)＝(τ pm s+1)/R pm Wherein τ pm Is a polarization time constant model parameter, s is complex frequency of Laplace transformation, R pm Is a polarization internal resistance model parameter.
Further, in step S01, the estimated value U of the voltage disturbance variable e The calculation formula of (2) is as follows:wherein τ f ＝1/(2πf c )，f c Is the filter cut-off frequency.
Further, the steps ofIn step S02, the battery compensation voltage variable U c The calculation formula of (2) is as follows:wherein Z is a Z transformation operator, e is a natural constant, and Ts is a system sampling period.
Further, the battery state estimation model described in step S03:u in p,k For the battery polarization voltage variation of the kth period, U p,k-1 For the cell polarization voltage variation of the kth-1 cycle, I b,k For the battery current sample value of the kth period, I b,k-1 For the battery current sample value of the k-1 th period, U b,k Is the battery terminal voltage variable of the kth period, U oc,k Is the open-circuit voltage variable for the kth period.
Further, the battery state equation in step S04 is:u in c,k Compensating the voltage variation for the battery of the kth cycle, U c,k-1 Compensating voltage variation, SOC, for the battery of the kth-1 cycle e,k SOC for the k-th cycle is estimated as SOC k-1 For the SOC estimation result of the k-1 th period, η i For coulombic efficiency, C N Rated capacity of the lithium ion battery; the battery observation equation is: u (U) be,k ＝U oc (SOC e,k )-U c,k -R o I b,k Wherein U is oc (SOC e,k ) SOC for the k-th cycle is estimated e,k Corresponding open circuit voltage value, U be,k For the battery voltage terminal predictive value of the kth period, U c,k The voltage variation is compensated for the battery of the kth cycle.
Further, the Kalman gain matrix K described in step S05 k The method comprises the following steps:wherein T is the rotation of the matrixPut the operator, add>For error covariance estimate +.> The P is k-1 The error covariance calculated for the previous cycle is calculated by:wherein E is an identity matrix, and when k is 1, P k-1 Is a preset value.
Further, in step S06, the process of estimating the battery state of charge of the lithium ion battery includes the following steps:
s601, obtaining a current sampling value I of a kth period b,k And the current sampling value I of the k-1 th period b,k-1 Battery terminal voltage sampling value U of kth period bs,k Cell polarization voltage variable U of the kth-1 cycle p,k-1 SOC estimation result SOC of the (k-1) th cycle k-1 Error covariance P for the kth-1 period k-1 ；
S602, U p,k-1 And I b,k-1 Substitution intoCalculation of the (k-1) th period variable U c,k-1 ；
S603, will I b,k 、I b,k-1 、U c,k-1 And SOC (System on chip) k-1 Substituting into the battery state equationObtaining the predicted value U of the kth period c,k And SOC estimated value SOC e,k ；
S604, obtaining the state of charge SOC and open circuit voltage U of the battery according to the open circuit voltage method oc Is substituted into SOC e,k Obtaining U oc (SOC e,k )；
S605, will I b,k 、U c,k And U oc (SOC e,k ) Battery observation equation U substituted into SOC estimation method be,k ＝U oc (SOC ek ,)-U ck ,-R o I bk Obtaining a battery terminal voltage predicted value U be,k Battery terminal voltage sampling value U bs,k With battery terminal voltage predictive value U be,k The difference of (2) is the observed difference D k D is k ＝U bs,k -U be ,k；
S606, calculating a Kalman gain matrix of a kth period by using an extended Kalman filtering algorithm:in-> Will D k Multiplying by a coefficient K u,k Then is connected with U c,k Adding to obtain a battery polarization voltage variable U which needs to be input in the (k+1) th period p,k U, i.e. U p,k ＝K u,k D k +U c,k The method comprises the steps of carrying out a first treatment on the surface of the Will D k Multiplying by a coefficient K s,k And then is connected with SOC e,k Adding to obtain SOC estimation result SOC of the kth period k I.e. SOC k ＝K s,k D k +SOC e,k 。
The invention has the beneficial effects that: according to the lithium ion battery state of charge estimation method based on the disturbance observer, a battery state estimation model is built through a first-order equivalent circuit model of the lithium ion battery, errors existing in the first-order equivalent circuit model are corrected in real time by the disturbance observer, a battery voltage disturbance observer model is built, the battery state estimation model is combined with the battery voltage disturbance observer model to obtain a battery state equation and a battery observation equation, an extended Kalman filtering algorithm is adopted according to the battery state equation and the battery observation equation to obtain a Kalman gain matrix, and the battery state of charge of the lithium ion battery is estimated according to the battery state equation, the battery observation equation and the Kalman gain matrix, so that the problem that the accuracy of an SOC estimation result is reduced along with the increase of the service time and/or the cycle use times of the battery is solved.
Drawings
Fig. 1 is a schematic flow chart of a lithium ion battery state of charge estimation method based on a disturbance observer.
Fig. 2 is a first-order equivalent circuit model of a lithium ion battery in a method for estimating the state of charge of the lithium ion battery based on a disturbance observer.
Fig. 3 is a schematic diagram of a disturbance observer system in the disturbance observer-based lithium ion battery state of charge estimation method of the present invention.
Fig. 4 is a diagram of a model structure of a voltage disturbance observer in the disturbance observer-based lithium ion battery state of charge estimation method of the present invention.
FIG. 5 shows a battery hysteresis characteristic voltage U in the disturbance observer-based lithium ion battery state of charge estimation method of the present invention h And a corresponding relation diagram of the SOC.
Fig. 6 is a comparison chart of SOC estimation error values in the lithium battery discharging process calculated by the lithium ion battery state of charge estimation method based on the disturbance observer and the conventional EKF method.
Fig. 7 is a comparison chart of the SOC estimation error value of the lithium ion battery in the charging process calculated by the disturbance observer-based lithium ion battery state of charge estimation method and the conventional EKF method.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
The invention relates to a lithium ion battery state of charge estimation method based on a disturbance observer, which is shown in figure 1 and comprises the following steps:
and S01, establishing a battery voltage disturbance observer model according to a disturbance observer principle and an equivalent circuit mathematical model of the lithium ion battery, and obtaining a voltage disturbance variable estimated value.
Specifically, the first-order equivalent circuit model of the lithium ion battery shown in fig. 2 is adopted as the estimationThe battery mathematical model of the method can better represent the static and dynamic characteristics of the battery, has better balance between the accuracy and the complexity of the model, is easy for engineering realization, and is the mathematical model of the equivalent circuit:wherein U is p Is polarization voltage, +.>Is U p With respect to the derivative of time t, d is the differential operator, τ p Is the polarization time constant and τ p ＝R p C p ，R P Is a polarization resistance, C p Is a polarized capacitor, U oc Is the open circuit voltage value, U b Is the voltage value of the battery terminal, R o Is ohmic internal resistance value, I b Is the battery current value.
Compared with a higher-order model, the first-order equivalent circuit model has obvious dynamic errors on the simulation response effect of the polarized voltage, and meanwhile, the first-order model does not consider the hysteresis characteristic of the lithium ion battery, namely, in the discharging and charging processes, the correspondence between the open-circuit voltage and the SOC in the actual operation of the battery is inconsistent with the open-circuit voltage and the SOC obtained by experimental data, so that the problem is solved by adopting a disturbance observer, and a disturbance observer system schematic diagram is shown in figure 3, wherein P(s) and P(s) in the figure are shown in the disturbance observer system schematic diagram n (s) a single-input single-output actual system with time-varying property and a corresponding nominal model, signal d s Representing the external disturbance variable, delta is the estimated value of the disturbance variable of the system, Q(s) is a low-pass filter with unit gain, and the disturbance observer is used for generating a nominal model P of a mathematical model of the system n (s) the difference from the actual system P(s) under specific conditions is regarded as the internal disturbance of the system, when the system signal frequency is in the low frequency band of the filter Q(s) in FIG. 2, the input-output relationship of the system is related to the nominal model P of the system n (s) remain the same, the actual system P(s) parameter difference and the external disturbance variable d are counteracted in principle s The influence on the accuracy of the system model is that a disturbance observer is adopted to reduce the lithium ion batteryThe effect of polarization voltage dynamic error and hysteresis characteristics on charge estimation.
According to the disturbance observer principle, a battery voltage disturbance observer model is established, real-time correction of lithium ion battery voltage dynamic errors is achieved, the structure of the voltage disturbance observer model is shown in fig. 4, and the model is formed by an actual system P(s) and a nominal model P n The difference between(s) is defined as battery polarization voltage error, and hysteresis characteristic voltage U is used as the internal disturbance quantity of the system h As an external disturbance quantity of the voltage disturbance observer model. Obtaining a mathematical formula about battery current according to a mathematical model of a first-order equivalent circuit of the battery:laplace transformation is carried out on the formula to obtain the following components: />Wherein s is complex frequency of Laplace transformation, and input parameters of a battery voltage disturbance observer model are set as polarization voltage U according to the formula p The output parameter is the battery current I b And gets a nominal model P of the battery voltage disturbance observer n (s) is: p (P) n (s)＝(τ pm s+1)/R pm Wherein τ pm Is a polarization time constant model parameter, R pm Is a polarization internal resistance model parameter; the actual system P(s) corresponding thereto is: p(s) = (τ) pr s+1)/R pr Wherein τ pr Is the actual value of the polarization time constant, R pr Is the actual value of the polarization internal resistance, and the tau pr And R is pr Has time-varying properties; the low-pass filter Q(s) of the voltage disturbance observer model adopts a first-order filter, Q(s) =1/(τ) f s+1), with a time constant τ f ＝1/(2πf c )，f c Is the filter cut-off frequency; the system disturbance variable estimation value of the battery voltage disturbance observer model is defined as a voltage disturbance variable estimation value U e Obtaining U according to the disturbance observer system structure e And polarization voltage U p Battery current I b The relation of (2) is: />In->Is P n Inverse model of(s), ->Will->Substituting the formula of Q(s) into the relation to obtain the estimated value U of the voltage disturbance variable e The mathematical formula of (a) is: />
S02, subtracting the voltage disturbance variable estimated value of the same period from the polarization voltage value in the battery voltage disturbance observer model, and discretizing to obtain a battery compensation voltage variable;
specifically, the battery compensation voltage variable U c The calculation formula of (2) is as follows:wherein Z is a Z transform operator, e is a natural constant, T S Is the system sampling period.
S03, discretizing a first-order equivalent circuit mathematical model formula of the lithium ion battery, and performing first-order Taylor expansion linearization to obtain a battery state estimation model;
specifically, the battery state estimation model:u in p,k For the battery polarization voltage variation of the kth period, U p,k-1 For the cell polarization voltage variation of the kth-1 cycle, I b,k For the battery current sample value of the kth period, I b,k-1 For the battery current sample value of the k-1 th period, U b,k Is the battery terminal voltage variable of the kth period, U oc,k Is the open-circuit voltage variable for the kth period.
S04, combining the battery state estimation model with the battery voltage disturbance observer model to obtain a battery state equation and a battery observation equation.
Specifically, the voltage variable U is compensated for the battery c The calculation formula of (2) carries out inverse Z transformation and substitutes I b,k-1 And U p,k-1 Calculating to obtain a battery compensation voltage variable U of the kth-1 period c,k-1 The method comprises the steps of carrying out a first treatment on the surface of the Battery compensation voltage variable U for comparing the kth period with the kth-1 period c,k And U c,k-1 Respectively replace polarization voltage variable U in battery state estimation model p,k And U p,k-1 The method comprises the steps of carrying out a first treatment on the surface of the And then, combining an ampere-hour integration method and an open-circuit voltage method of SOC estimation, and establishing a battery state equation and a battery observation equation required by the SOC estimation method, wherein the battery state equation is as follows:u in c,k Compensating the voltage variation for the battery of the kth cycle, U c,k-1 Compensating voltage variation, SOC, for the battery of the kth-1 cycle e,k SOC for the k-th cycle is estimated as SOC k-1 For the SOC estimation result of the k-1 th period, η i For coulombic efficiency, C N Rated capacity of the lithium ion battery; the battery observation equation is: u (U) be,k ＝U oc (SOC e,k )-U c,k -R o I b,k Wherein U is oc (SOC e,k ) SOC for the k-th cycle is estimated e,k Corresponding open circuit voltage value, U be,k For the battery voltage terminal predictive value of the kth period, U c,k The voltage variation is compensated for the battery of the kth cycle.
S05, obtaining a Kalman gain matrix by adopting an extended Kalman filtering algorithm according to a battery state equation and a battery observation equation.
Specifically, an Extended Kalman Filter (EKF) algorithm is adopted, and a Kalman gain matrix is calculated synchronously according to a battery state equation and a battery observation equation:where T is the transpose operator of the matrix,for error covariance estimate +.>The P is k-1 The error covariance calculated for the previous cycle is calculated by: />Wherein E is an identity matrix, and when k is 1, P k-1 Is a preset value.
S06, estimating the battery state of charge of the lithium ion battery according to a battery state equation, a battery observation equation and a Kalman gain matrix.
Specifically, the specific calculation process of each operation period is as follows:
s601, obtaining a current sampling value I of a kth period b,k And the current sampling value I of the k-1 th period b,k-1 Battery terminal voltage sampling value U of kth period bs,k Cell polarization voltage variable U of the kth-1 cycle p,k-1 SOC estimation result SOC of the (k-1) th cycle k-1 Error covariance P for the kth-1 period k-1 ；
S602, U p,k-1 And I b,k-1 Substitution intoCalculation of the (k-1) th period variable U c,k-1 ；
S603, will I b,k 、I b,k-1 、U c,k-1 And SOC (System on chip) k-1 Substituting into the battery state equationObtaining the predicted value U of the kth period c,k And SOC estimated value SOC e,k ；
S604, obtaining the state of charge SOC and open circuit voltage U of the battery according to the open circuit voltage method oc Is substituted into SOC e,k Obtaining U oc (SOC e,k )；
S605, will I b,k 、U c,k And U oc (SOC e,k ) Battery observation equation U substituted into SOC estimation method be,k ＝U oc (SOC ek ,)-U ck ,-R o I bk Obtaining a battery terminal voltage predicted value U be,k Battery terminal voltage sampling value U bs,k With battery terminal voltage predictive value U be,k The difference of (2) is the observed difference D k D is k ＝U bs,k -U be,k ；
S606, calculating a Kalman gain matrix of a kth period by using an extended Kalman filtering algorithm:in-> Will D k Multiplying by a coefficient K u,k Then is connected with U c,k Adding to obtain a battery polarization voltage variable U which needs to be input in the (k+1) th period p,k U, i.e. U p,k ＝K u,k D k +U c,k The method comprises the steps of carrying out a first treatment on the surface of the Will D k Multiplying by a coefficient K s,k And then is connected with SOC e,k Adding to obtain SOC estimation result SOC of the kth period k I.e. SOC k ＝K s,k D k +SOC e,k 。
The lithium ion battery charge state estimation method based on the disturbance observer is used for carrying out a lithium ion battery charge and discharge simulation experiment, and the estimation accuracy is compared with that of a conventional EKF method. Taking a lithium iron phosphate battery with 3.2V/50Ah monomer as a test object, adopting an HPPC standard test method to obtain battery experimental data at normal temperature, building a battery monomer simulation module according to the experimental data, and increasing hysteresis characteristic voltage U at a battery end according to the hysteresis characteristic of the lithium iron phosphate battery h The correspondence relationship with the battery SOC is shown in fig. 5.
A battery discharge simulation experiment is carried out by adopting a current with 0.5C multiplying power, the charge state of the lithium ion battery is discharged from 100% to 0%, the initial SOC values of the two SOC estimation methods are both set to 95%, and the SOC estimation error value pairs of the two methods in the discharging process are shown in FIG. 6. As can be seen from fig. 6, both methods in the discharge test have low dependence on the estimated initial value, and can quickly correct the error of 5% between the initial value and the SOC actual value, but since the lithium ion battery has a significant hysteresis voltage in the low SOC region in the late stage of discharge, the estimated value of the conventional EKF method has a large dynamic error, the maximum value of which is 2.5%; the invention effectively reduces the influence of hysteresis characteristic on the estimation effect, and the SOC estimation error value is stabilized at about 0.3%.
A battery charging simulation experiment is carried out by adopting a current with 0.5C multiplying power, the charge state of the lithium ion battery is charged from 0% to 100%, the initial SOC values of the two SOC estimation methods are set to be 5%, and the SOC estimation error value pairs of the two methods in the charging process are shown in FIG. 7. As can be seen from fig. 7, the conventional EKF method cannot quickly correct 5% of initial error due to the hysteresis characteristic voltage in the initial charge period, and slowly decreases after the SOC estimation error increases to 6.7%, the hysteresis characteristic voltage decreases in the late charge period, the estimation effect of the method is improved, and the final error value decreases to 0.9%; the estimated error value in the initial stage of charging is 5.2% at maximum and rapidly drops, the absolute value of the error in the middle and later stages of charging is kept at about 0.2%, the initial error can be rapidly corrected without being influenced by hysteresis characteristic voltage in the whole charging process, and the estimated error value is kept low.
By combining the simulation results, the invention can be seen that the estimation error is reduced in a voltage disturbance compensation mode, and the estimation effect in the discharging and charging processes is better than that of the conventional EKF method.

Claims (8)

1. The lithium ion battery state of charge estimation method based on the disturbance observer is characterized by comprising the following steps of:
s01, establishing a battery voltage disturbance observer model according to a disturbance observer principle and an equivalent circuit mathematical model of the lithium ion battery, and obtaining a polarization voltage disturbance variable estimated value;
s02, subtracting the polarized voltage disturbance variable estimated value of the same period from the polarized voltage value in the battery voltage disturbance observer model, and discretizing to obtain a battery compensation voltage variable;
s03, discretizing a first-order equivalent circuit mathematical model formula of the lithium ion battery, and performing first-order Taylor expansion linearization to obtain a battery state estimation model;
s04, combining the battery state estimation model with the battery voltage disturbance observer model to obtain a battery state equation and a battery observation equation, wherein the battery state equation is as follows:u in c,k Compensating the voltage variation for the battery of the kth cycle, U c,k-1 Compensating the voltage variation for the battery of the (k-1) th period, e being a natural constant, τ pm Is a polarization time constant model parameter, T S For sampling period, R pm Is a polarized internal resistance model parameter, I b,k-1 For the current sampling value of the k-1 th period, SOC e,k SOC for the k-th cycle is estimated as SOC k-1 For the SOC estimation result of the k-1 th period, η i For coulombic efficiency, C N Is the rated capacity of the lithium ion battery, I b,k A current sampling value for the kth period; the battery observation equation is: u (U) be,k ＝U oc (SOC e,k )-U c,k -R o I b,k Wherein U is oc (SOC e,k ) SOC for the k-th cycle is estimated e,k Corresponding open circuit voltage value, U be,k For the battery voltage terminal predictive value of the kth period, U c,k Compensating the voltage variation, R, for the battery of the kth cycle o Is an ohmic internal resistance value;
s05, acquiring a Kalman gain matrix by adopting an extended Kalman filtering algorithm according to a battery state equation and a battery observation equation;
s06, estimating the battery state of charge of the lithium ion battery according to a battery state equation, a battery observation equation and a Kalman gain matrix.
2. The method for estimating a state of charge of a lithium ion battery based on a disturbance observer according to claim 1, wherein in step S01, the equivalent mathematical model of the lithium ion battery is a first-order equivalent circuit model, and the formula of the first-order equivalent circuit model is:wherein U is p Is polarization voltage, +.>Is U p With respect to the derivative of time t, d is the differential operator, τ p Is the polarization time constant and τ p ＝R p C p ，R P Is a polarization resistance, C p Is a polarized capacitor, U oc Is the open circuit voltage value, U b Is the voltage value of the battery terminal, R o Is ohmic internal resistance value, I b Is a battery current value, the battery current value I b The mathematical formula of (a) is:
3. the disturbance observer-based lithium ion battery state of charge estimation method according to claim 1 or 2, wherein in step S01, the battery voltage disturbance observer model comprises a nominal model, the nominal model P n (s) is: p (P) n (s)＝(τ pm s+1)/R pm Wherein τ pm Is a polarization time constant model parameter, s is complex frequency of Laplace transformation, R pm Is a polarization internal resistance model parameter.
4. The method for estimating the state of charge of a lithium ion battery based on a disturbance observer according to claim 1 or 2, wherein in step S01, the estimated value U of the disturbance variable of the polarization voltage e The calculation formula of (2) is：Wherein τ f Is a time constant τ f ＝1/(2πf c )，f c Is the filter cut-off frequency τ pm Is a polarization time constant model parameter, s is complex frequency of Laplace transformation, R pm Is a polarized internal resistance model parameter, I b Is the battery current value, U p Is the polarization voltage.
5. The method of estimating the state of charge of a lithium-ion battery based on a disturbance observer according to claim 1 or 2, wherein in step S02, the battery compensation voltage variable U c The calculation formula of (2) is as follows:wherein Z is a Z transformation operator, e is a natural constant, ts is a sampling period, τ f Is a time constant, U p Is the polarization voltage, R pm Is the parameter of the polarized internal resistance model, τ pm Is a polarization time constant model parameter, I b Is the battery current value.
6. The method of estimating state of charge of a lithium-ion battery based on a disturbance observer according to claim 1 or 2, wherein the battery state estimation model in step S03:u in p,k The battery polarization voltage variable in the kth period, e is a natural constant, τ p Is the polarization time constant and τ p ＝R p C p ，R P Is a polarization resistance, C p Is a polarized capacitor, T S For sampling period, U p,k-1 For the cell polarization voltage variation of the kth-1 cycle, I b,k For the battery current sample value of the kth period, I b,k-1 For the battery current sample value of the k-1 th period, U b,k Is the battery terminal voltage variable of the kth period, U oc,k R is the open-circuit voltage variable of the kth period o Is the ohmic internal resistance value.
7. The method for estimating the state of charge of a lithium ion battery based on a disturbance observer according to claim 1 or 2, wherein the kalman gain matrix K in step S05 k The method comprises the following steps:wherein T is the transpose operator of the matrix, +.>U oc (SOC e,k ) SOC for the k-th cycle is estimated e,k The corresponding open-circuit voltage value, d is the differential operator,>for error covariance estimate +.>e is a natural constant, ts is a sampling period, τ pm Is a polarization time constant model parameter, the P k-1 The error covariance calculated for the previous cycle is calculated by: />Wherein E is an identity matrix, and when k is 1, P k-1 Is a preset value.
8. The method for estimating the state of charge of a lithium ion battery based on a disturbance observer according to claim 1 or 2, wherein in step S06, the process of estimating the state of charge of the lithium ion battery comprises the steps of:
s601, obtaining a current sampling value I of a kth period b,k And the current sampling value I of the k-1 th period b,k-1 Battery terminal voltage sampling value U of kth period bs,k Kth-1Periodic battery polarization voltage variation U p,k-1 SOC estimation result SOC of the (k-1) th cycle k-1 Error covariance P for the kth-1 period k-1 ；
S602, U p,k-1 And I b,k-1 Substitution intoCalculation of the (k-1) th period variable U c,k-1 ，τ f Is a time constant, Z is a Z transformation operator, U p Is the polarization voltage, R pm Is the parameter of the polarized internal resistance model, τ pm Is a model parameter of polarization time constant, e is a natural constant, T S For sampling period, I b Is a battery current value;
s603, will I b,k 、I b,k-1 、U c,k-1 And SOC (System on chip) k-1 Substituting into the battery state equationObtaining the predicted value U of the kth period c,k And SOC estimated value SOC e,k ，U c,k-1 Compensating voltage variation, SOC, for the battery of the kth-1 cycle k-1 For the SOC estimation result of the k-1 th period, η i For coulombic efficiency, C N Rated capacity of the lithium ion battery;
s604, obtaining the state of charge SOC and open circuit voltage U of the battery according to the open circuit voltage method oc Is substituted into SOC e,k Obtaining U oc (SOC e,k )，U oc (SOC e,k ) SOC for the k-th cycle is estimated e,k The corresponding open circuit voltage value;
s605, will I b,k 、U c,k And U oc (SOC e,k ) Battery observation equation U substituted into SOC estimation method be,k ＝U oc (SOC e,k )-U c,k -R o I b,k Obtaining a battery terminal voltage predicted value U be,k Battery terminal voltage sampling value U bs,k With battery terminal voltage predictive value U be,k The difference of (2) is the observed difference D k D is k ＝U bs,k -U be,k ，R o Is an ohmic internal resistance value;
s606, calculating a Kalman gain matrix of a kth period by using an extended Kalman filtering algorithm:in-> Will D k Multiplying by a coefficient K u,k Then is connected with U c,k Adding to obtain a battery polarization voltage variable U which needs to be input in the (k+1) th period p,k U, i.e. U p,k ＝K u,k D k +U c,k The method comprises the steps of carrying out a first treatment on the surface of the Will D k Multiplying by a coefficient K s,k And then is connected with SOC e,k Adding to obtain SOC estimation result SOC of the kth period k I.e. SOC k ＝K s,k D k +SOC e,k 。
CN202110995316.9A 2021-08-27 2021-08-27 Lithium ion battery state of charge estimation method based on disturbance observer Active CN113702838B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110995316.9A CN113702838B (en) 2021-08-27 2021-08-27 Lithium ion battery state of charge estimation method based on disturbance observer

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110995316.9A CN113702838B (en) 2021-08-27 2021-08-27 Lithium ion battery state of charge estimation method based on disturbance observer

Publications (2)

Publication Number Publication Date
CN113702838A CN113702838A (en) 2021-11-26
CN113702838B true CN113702838B (en) 2024-03-12

Family

ID=78655941

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110995316.9A Active CN113702838B (en) 2021-08-27 2021-08-27 Lithium ion battery state of charge estimation method based on disturbance observer

Country Status (1)

Country Link
CN (1) CN113702838B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116224099B (en) * 2023-05-06 2023-07-21 力高(山东)新能源技术股份有限公司 Method for dynamically and adaptively estimating battery SOC
CN118068198B (en) * 2024-04-18 2024-08-02 北京智芯微电子科技有限公司 Battery voltage sampling method and device based on multiple correction Kalman filtering

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103293485A (en) * 2013-06-10 2013-09-11 北京工业大学 Model-based storage battery SOC (state of charge) estimating method
WO2015056964A1 (en) * 2013-10-14 2015-04-23 주식회사 엘지화학 Apparatus for estimating state of hybrid secondary battery and method therefor
CN106324523A (en) * 2016-09-26 2017-01-11 合肥工业大学 Discrete variable structure observer-based lithium battery SOC (state of charge) estimation method
CN111366855A (en) * 2020-03-19 2020-07-03 北京理工大学 Battery equivalent circuit model disturbance-resistant parameterization method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103293485A (en) * 2013-06-10 2013-09-11 北京工业大学 Model-based storage battery SOC (state of charge) estimating method
WO2015056964A1 (en) * 2013-10-14 2015-04-23 주식회사 엘지화학 Apparatus for estimating state of hybrid secondary battery and method therefor
CN106324523A (en) * 2016-09-26 2017-01-11 合肥工业大学 Discrete variable structure observer-based lithium battery SOC (state of charge) estimation method
CN111366855A (en) * 2020-03-19 2020-07-03 北京理工大学 Battery equivalent circuit model disturbance-resistant parameterization method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
"Disturbance observer-based state-of-charge estimation for Li-ion battery used in light electric vehicles";Pascal Messier等;《Journal of Energy Storage》;第27卷;101144 *
"基于扰动观测器的锂电池荷电状态估算方法";刘思佳等;《电源技术》;第45卷(第10期);1256-1259 *

Also Published As

Publication number Publication date
CN113702838A (en) 2021-11-26

Similar Documents

Publication Publication Date Title
Li et al. Enhanced online model identification and state of charge estimation for lithium-ion battery under noise corrupted measurements by bias compensation recursive least squares
Yang et al. Online parameters identification and state of charge estimation for lithium-ion capacitor based on improved Cubature Kalman filter
CN109061520B (en) Power battery health and power state online estimation method and system
CN110398691B (en) Lithium ion power battery SoC estimation method based on improved self-adaptive double unscented Kalman filter
CN106324523B (en) Lithium battery SOC estimation method based on discrete-time variable structure observer
CN113702838B (en) Lithium ion battery state of charge estimation method based on disturbance observer
CN105425153B (en) A kind of method of the state-of-charge for the electrokinetic cell for estimating electric vehicle
CN109633479B (en) Lithium battery SOC online estimation method based on embedded type volume Kalman filtering
CN110907835B (en) Battery model parameter identification and SOC estimation method with noise immunity characteristic
CN112379270B (en) Rolling time domain estimation method for state of charge of power battery of electric automobile
CN111985154B (en) Self-adaptive fuzzy Kalman estimation SOC algorithm
CN115932591A (en) Lithium battery SOC estimation method, system, medium and electronic equipment based on PID-EKF
CN113219344A (en) Lead-acid storage battery SOC estimation method
CN111060823A (en) DP model-based battery SOP online estimation method in low-temperature environment
KR20110118246A (en) System for calculating parameter and charge rate of battery
CN113671377A (en) Lithium battery SOC estimation method based on improved EKF algorithm
CN115327415A (en) Lithium battery SOC estimation method based on limited memory recursive least square algorithm
CN112946480B (en) Lithium battery circuit model simplification method for improving SOC estimation real-time performance
CN113109725B (en) Parallel battery state-of-charge estimation method based on state noise matrix self-adjustment
CN114740385A (en) Self-adaptive lithium ion battery state of charge estimation method
CN117074947A (en) Battery parameter estimation method, apparatus, device, storage medium, and program product
CN112114255B (en) Lithium battery state-of-charge estimation method based on combined set Kalman filtering
CN116165545A (en) Battery SOC state estimation method based on unscented Kalman filtering and neural network
CN114859235A (en) Lithium battery SOC estimation method based on extended Kalman filtering
CN115407204A (en) Battery state of charge estimation method and system based on battery parameter compensation

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