LU503873B1 - Online self-adaptive method for estimating battery internal temperature - Google Patents
Online self-adaptive method for estimating battery internal temperature Download PDFInfo
- Publication number
- LU503873B1 LU503873B1 LU503873A LU503873A LU503873B1 LU 503873 B1 LU503873 B1 LU 503873B1 LU 503873 A LU503873 A LU 503873A LU 503873 A LU503873 A LU 503873A LU 503873 B1 LU503873 B1 LU 503873B1
- Authority
- LU
- Luxembourg
- Prior art keywords
- battery
- model
- state
- internal temperature
- temperature
- Prior art date
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
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K7/00—Measuring temperature based on the use of electric or magnetic elements directly sensitive to heat ; Power supply therefor, e.g. using thermoelectric elements
- G01K7/42—Circuits effecting compensation of thermal inertia; Circuits for predicting the stationary value of a temperature
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K7/00—Measuring temperature based on the use of electric or magnetic elements directly sensitive to heat ; Power supply therefor, e.g. using thermoelectric elements
- G01K7/42—Circuits effecting compensation of thermal inertia; Circuits for predicting the stationary value of a temperature
- G01K7/427—Temperature calculation based on spatial modeling, e.g. spatial inter- or extrapolation
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M10/00—Secondary cells; Manufacture thereof
- H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
- H01M10/48—Accumulators combined with arrangements for measuring, testing or indicating the condition of cells, e.g. the level or density of the electrolyte
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Chemical & Material Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Electrochemistry (AREA)
- General Chemical & Material Sciences (AREA)
- Secondary Cells (AREA)
Abstract
The invention discloses an online self-adaptive method for estimating the battery internal temperature, which comprises the following steps: establishing a lumped thermal characteristic equivalent circuit model of the battery based on the modeling similarity between physical systems of the same order; establishing a state space model of the lumped thermal characteristic equivalent circuit model in a time domain state based on Kirchhoff's circuit laws; acquiring a battery internal temperature state model based on the state space model in the time domain state; obtaining thermal characteristic model parameters to be identified in the battery internal temperature state model based on the battery internal temperature state model by adopting the forgetting factor least square method; jointly estimating the parameters of the thermal characteristic model and the battery internal temperature by using a joint Kalman filter method.
Description
DESCRIPTION LU503873
ONLINE SELF-ADAPTIVE METHOD FOR ESTIMATING BATTERY INTERNAL
TEMPERATURE
The invention belongs to the technical field of power batteries, and in particular relates to an online self-adaptive method for estimating the battery internal temperature.
In recent years, benefiting from the explosive development of global new energy automobile industry, lithium-ion batteries have quickly become the mainstream power source due to their high energy density and long cycle life. Compared with its own defects such as aging, self-discharge and capacity decline, the safety problems caused by the difficulty in real-time prediction of battery temperature seriously restrict the large-scale development of new energy vehicles. Therefore, the thermal characteristics modeling and high-precision prediction of internal temperature of power battery are very important to prevent thermal runaway of battery and ensure safe and long-life operation.
In addition, for the thermal management system of battery, different thermal management systems may have different temperature control modes, which makes different thermal management systems of battery have different external heat transfer coefficients. Using the same external heat transfer coefficient in different working conditions will increase the error of battery internal temperature estimation. Therefore, the adaptive working condition prediction method of battery external heat transfer coefficients is very important for high-accuracy thermal characteristic modeling and efficient thermal management system construction.
The complex internal structure and changeable application environment of power battery make it easy to produce uneven internal temperature distribution and high internal temperature during its use, which increases the difficulty of estimating the battery internal temperature. At present, the commonly used methods for measuring the)503873 battery internal temperature includes the internal temperature measurement method based on experiment, the internal temperature prediction method based on off-line and the internal temperature estimation method based on online estimation. The internal temperature measurement method based on experiment is usually to embed thermocouple in the battery, which is only suitable for testing the relevant thermal characteristics of the battery under experimental environmental conditions. In the practical application of the battery, the feasibility of this method is low and there are huge hidden dangers in the long-term use of the battery, which cannot guarantee the safety of the battery in use. The internal temperature prediction method based on off-line is based on the finite element numerical calculation method, and the internal temperature estimation of the battery is obtained by establishing the thermal model of the battery and carrying out off-line simulation. This method is usually used in the packaging design and module design of battery cells, and the finite element numerical calculation method has high computational complexity, which is not suitable for the practical application and thermal management of batteries.
The above-mentioned researches on the battery internal temperature are based on the laboratory or off-line, and the development of an application-grade vehicle-mounted battery thermal management system needs to meet the online estimation of the battery internal temperature. The internal temperature prediction method based on online estimation often takes practical application as the starting point to realize the thermal management of batteries. The prediction methods based on online estimation of battery internal temperature can be roughly divided into DC internal resistance method, electrochemical impedance spectroscopy (EIS) measurement analysis method and temperature difference transfer function method. Some researchers have studied the DC internal resistance method, using the functional relationship between DC internal resistance and the internal and external temperature of the battery to estimate the battery internal temperature by looking up the table or functional calculation. However, different types of batteries, even different batteries of the same type, have different functional relationships between DC internal resistance and internal and external temperatures of batteries, which makes the DC internal resistance method not universal.
In addition, the research on EIS measurement analysis method by scholars in the field 19503873 not rare, and the principle of this method is similar to that of DC internal resistance method. The battery internal temperature is estimated by looking up the table or calculating the function according to the corresponding functional relationship between the average amplitude and phase angle of the AC impedance of the battery and the battery internal temperature. However, EIS measuring instrument is not suitable for the practical application of automobiles because of its demanding working environment and extremely high price.
In view of the obvious disadvantages of the above methods, in order to improve the reliability of battery thermal modeling, the temperature difference transfer function method based on thermal network film modeling is applied. This method uses the transfer function between the battery internal temperature and the heating power of the system to calculate the battery internal temperature through the input current of the system. This kind of method is commonly used in the current battery internal temperature estimation, but this method belongs to open-loop estimation, and it cannot update the parameters of the transfer function adaptively according to the changes of the external environment, so the accuracy is low. From the above discussion, it can be found that the existing research on battery thermal characteristics modeling is often accompanied by high computational complexity. At the same time, the harsh environmental conditions also make most of the research stay in the laboratory stage, which cannot realize the actual vehicle application of the system. In addition, the uncertainty of external thermal resistance and the inability to estimate the internal text adaptively and accurately in changeable environment have not been paid enough attention.
SUMMARY LU503873
In order to solve the above technical problems, the invention provides an online self-adaptive method for estimating the battery internal temperature, which realizes the self-adaptive collaborative estimation of the parameters of a thermal characteristic model and the battery internal temperature, and solves the problem that the battery internal temperature is difficult to adaptively and accurately estimate online in the prior art.
In order to achieve the above purpose, the present invention provides an online self-adaptive method for estimating the battery internal temperature, which comprises the following steps: establishing a lumped thermal characteristic equivalent circuit model of the battery based on the modeling similarity between physical systems of the same order; establishing a state space model of the lumped thermal characteristic equivalent circuit model in a time domain state based on Kirchhoff's circuit laws; acquiring a battery internal temperature state model based on the state space model in the time domain state; obtaining thermal characteristic model parameters to be identified in the battery internal temperature state model based on the battery internal temperature state model by adopting the forgetting factor least square method; jointly estimating the parameters of the thermal characteristic model and the battery internal temperature by using a joint Kalman filter method.
Optionally, the step of establishing a state space model of the lumped thermal characteristic equivalent circuit model in a time domain state based on Kirchhoff's circuit laws includes:
Processing a heating power model based on the lumped thermal characteristic equivalent circuit model to obtain a new heating power model of the lumped thermal characteristic equivalent circuit model; obtaining the state models of the highest temperature node in the center of the battery and the temperature node in the surface area of the battery, based on the new heating power model and the Kirchhoff circuit law; obtaining an observation model of Tss based on the difference between the highest temperature node in the inner center of the battery and the ambient temperature wherg,503873 the battery is located: acquiring a state space model of the lumped thermal characteristic equivalent circuit model based on the state model and the observation model.
Optionally, the new heating power model is: (> ff SR. + IF {AREY dT SE IR, = IR,
Where Q is the heating power, Rs is the DC internal resistance of the battery, / is the charging and discharging current of the battery, T is the battery temperature, E is the balanced electromotive force inside the battery, and U; is the terminal voltage of the battery; the state model is: ar BE i= RE = Oo {is = Toma wherein, Q is the heating power, T; is the highest temperature in the center of the battery, Ts is the surface temperature of the battery, Tam is the ambient temperature of the battery, Riis the internal thermal resistance of the battery, Ro is the external thermal resistance of the battery, Ci is the equivalent thermal capacity of the battery, Cs is the equivalent hot melting of the battery shell, Tis is the difference between the highest temperature 7; of the internal central temperature of the battery and the ambient temperature Tamb Of the battery, and Tss is the difference between the surface temperature Ts of the battery and the ambient temperature Tamb Of the battery;
The observation model is: =o ifr oe se Er SS Be ns where, y: is the system output in time domain, Ct is the output matrix in time domain, xt is the system state variable in continuous state, D: is the branch transfer matrix in time domain, and u is the system input in continuous state.
Optionally, the step of acquiring a battery internal temperature state model based on the state space model in the time domain state includes:
Acquiring a transfer function of the state space model based on the new heating, 503873 power model and the internal temperature state of the battery, wherein the internal temperature state of the battery is the difference between the highest point of the internal central temperature of the battery and the ambient temperature where the battery is located: carrying out integrated deformation on the transfer function by using Z transform to obtain the internal temperature state model of the battery.
Optionally, the battery internal temperature state model is:
Finer ™ Oper + MaTion + Bi Quay + Bali where Tis, x is the temperature value of the highest temperature point in the battery at time k, Tis, k+1 is the temperature value of the highest temperature point in the battery at time k+1, Tis k+2 is the temperature value of the highest temperature point in the battery at time k+2, Q« is the heating power of the battery at time k, Qx+1 is the heating power of the battery at time k+1, ay, az, B1 and B2 are the parameters to be identified of the difference equation of the internal temperature state model.
Optionally, the step of obtaining thermal characteristic model parameters to be identified in the battery internal temperature state model based on the battery internal temperature state model by adopting the forgetting factor least square method comprises: acquiring an exogenous autoregressive model based on the battery internal temperature state model; solving the thermal characteristic model parameters that need to be identified in the battery internal temperature state model by the least method based on the exogenous autoregressive model.
Optionally, the exogenous autoregressive model is as follows: 8 = {ay daB Bi y = #78 € | ÿ 7 too. Bu u, dp ke Pr-2 3 ; lo = [Tropes Tinie Oren Pel’
Where Y is the output matrix of the exogenous autoregressive model, ¢ is the data matrix, 6 is the coefficient matrix, and @x is the system data matrix at time k;
the parameters of the thermal characteristic model parameters are Ci, Ri and Cs; LU503873 the thermal characteristic model parameter equation is: le, = CESR, + ata, BE SPF,
Where Ci is the equivalent heat capacity inside the battery, Riis the internal thermal resistance of the battery, Cs is the equivalent hot melt of the battery shell, andAt is the system sampling time.
Optionally, the step of jointly estimating the parameters of the thermal characteristic model and the battery internal temperature by using a joint Kalman filter method comprises: acquiring the state space model in the discrete state of the lumped thermal characteristic equivalent circuit model based on the state space model in the time domain state; constructing the JKF model of self-adaptive cooperative estimation of the external equivalent thermal resistance and the internal temperature state of the battery by using the joint Kalman filtering method and based on the state space equation in the discrete state: realizing the joint estimation of the parameters of the thermal characteristic model and the internal temperature based on the JKF model.
Optionally, the state space model in the discrete state is:
PT open St | ( at ét 3 Tas je 1Qu + wi
Lise ky nn Foe | i de me LE ask qi ve fes RS, ! (RS Rp doo ing UE ot ~ * Fax 4,
Tosser = {0 1] eo 4 101 [0]
Where xq, k+1 are the state variables of the system at time k+1 in discrete state, xq, « are the state variables of the system at time k in discrete state, uk is the input variable of the system at time k, wk is system noise, yx is the output variable of the system at time k in discrete state, Au is the system matrix in discrete state, Ba is the control matrix in discrete state, Ca is the observation matrix in discrete state and Du is the direct transfer matrix in discrete state.
Optionally, the JKF model is: LU503873
Le Flap Up) + we wo Ag X e+ Rata + wy,
Fe = ALG pe Med PUR 0px; + Dre + vy where xy is the 3-dimensional system state matrix at time k, xyx is the system state matrix engraved at time k+1, ux is the input variable of the system at time k, yk is the output variable of the system at time k in discrete state, Ay is the 3x3 -dimensional system matrix, By is the 3x1 -dimensional control matrix, Cy is the 1x3 -dimensional output matrix, Dy is the direct matrix, wk is the system noise and vx is the observation noise.
Compared with the prior art, the invention has the following advantages and technical effects:
According to the invention, a battery lumped thermal characteristic model is established, and the discrete state space equation expression of the thermal characteristic model is realized by combining a control theory. Considering the uncertainty of external thermal resistance, the error can be controlled within 0.5°C by combining the forgetting factor least square algorithm with the joint Kalman filter algorithm, and the initial values of different external equivalent thermal resistances will converge to the same value with the iteration, thus realizing the adaptive collaborative estimation of the parameters of thermal characteristic model and the battery internal temperature, and solving the problem that it is difficult to estimate the battery internal temperature adaptively and accurately in the prior art.
The accompanying drawings, which constitute a part of this application, are used to provide a further understanding of this application. The illustrative embodiments of this application and their descriptions are used to explain this application, and do not constitute an improper limitation of this application. In the attached drawings:
FIG. 1 is a flowchart of an embodiment of the present invention;
FIG. 2 is a schematic diagram of a lumped thermal characteristic model according to an embodiment of the present invention.
DESCRIPTION OF THE INVENTION LU503873
It should be noted that the embodiments in this application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail with reference to the attached drawings and examples.
It should be noted that the steps shown in the flowchart of the accompanying drawings may be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order from here.
The invention provides an online self-adaptive method for estimating the battery internal temperature, and the specific process is shown in FIG. 1, which comprises the following steps: establishing the lumped thermal characteristic equivalent circuit model of the battery based on the modeling similarity between physical systems of the same order; establishing a state space model of the lumped thermal characteristic equivalent circuit model in a time domain state based on Kirchhoff's circuit laws; acquiring a battery internal temperature state model based on the state space model in the time domain state; obtaining thermal characteristic model parameters to be identified in the battery internal temperature state model based on the battery internal temperature state model by adopting the forgetting factor least square method; jointly estimating the parameters of the thermal characteristic model and the battery internal temperature by using a joint Kalman filter method.
Further, the step of establishing a state space model of the lumped thermal characteristic equivalent circuit model in a time domain state based on Kirchhoff's circuit laws includes:
Processing a heating power model based on the lumped thermal characteristic equivalent circuit model to obtain a new heating power model of the lumped thermal characteristic equivalent circuit model; obtaining the state models of the highest temperature node in the center of the battery and the temperature node in the surface area of the battery, based on the new heating power model and the Kirchhoff circuit law; LU503873 obtaining an observation model of Tss based on the difference between the highest temperature node in the inner center of the battery and the ambient temperature where the battery is located: acquiring a state space model of the lumped thermal characteristic equivalent circuit model based on the state model and the observation model.
Further, the step of acquiring a battery internal temperature state model based on the state space model in the time domain state includes:
Acquiring a transfer function of the state space model based on the new heating power model and the internal temperature state of the battery, wherein the internal temperature state of the battery is the difference between the highest point of the internal central temperature of the battery and the ambient temperature where the battery is located; carrying out integrated deformation on the transfer function by using Z transform to obtain the internal temperature state model of the battery.
Further, the step of obtaining thermal characteristic model parameters to be identified in the battery internal temperature state model based on the battery internal temperature state model by adopting the forgetting factor least square method comprises: acquiring an exogenous autoregressive model based on the battery internal temperature state model; solving the thermal characteristic model parameters that need to be identified in the battery internal temperature state model by the least method based on the exogenous autoregressive model.
Further, the step of jointly estimating the parameters of the thermal characteristic model and the battery internal temperature by using a joint Kalman filter method comprises: acquiring the state space model in the discrete state of the lumped thermal characteristic equivalent circuit model based on the state space model in the time domain state; constructing the JKF model of self-adaptive cooperative estimation of the external equivalent thermal resistance and the internal temperature state of the battery by Using 503873 the joint Kalman filtering method and based on the state space equation in the discrete state; realizing the joint estimation of the parameters of the thermal characteristic model and the internal temperature based on the JKF model.
The specific flow of this embodiment is shown in FIG. 1, and the implementation steps are as follows:
S1, establishing a lumped thermal characteristic equivalent circuit model of a lithium ion battery by using the modeling similarity characteristics between physical systems of the same order. As shown in FIG. 2, the composition of the model is based on the similarity between lumped electrical characteristic modeling and thermal characteristic modeling of the battery, and the thermal characteristic network of the battery is described by an equivalent electrical characteristic network. Among them, the current source, capacitance and resistance of the electrical characteristic network are equivalent to the heating power, heat capacity and thermal resistance of the thermal characteristic network respectively. Q is the total heating power of the battery, and there are two ways to conduct it: one is to heat the battery, and the other is to conduct it through thermal resistance. 7; is the highest temperature in the center of the battery, Ts is the surface temperature of the battery. Tams is the ambient temperature of the battery. To is the model reference temperature point, and in general, Tams is set as the external reference temperature point, that is, Tam» = To. Ri is the internal thermal resistance of the battery, which is related to the introduction coefficient of the battery. Ro is the external thermal resistance of the battery, which is related to the heat transfer coefficient outside the battery. Ci is the equivalent thermal capacity of the battery. Cs is the equivalent hot melting of the battery shell. It is worth noting that the cylindrical battery sample used in the embodiment of the invention is an aluminum shell with excellent thermal conductivity, and the heat capacity of the shell is extremely small. The invention realizes the modeling and solving of the thermal characteristics of the battery based on FIG. 2.
S2, constructing a state space equation of the lumped thermal characteristic equivalent circuit model based on Kirchhoff's circuit law.
S21, in order to reduce the computational complexity of the battery heat source,
further simplifying the Bernardi's heating power equation according to Kirchhoff's law 503873 and obtaining a simplified expression for calculating the heating power of the battery lumped parameter thermal model: 9x FR, +IT{dE)/dT « E —U, = IR,
Where Q is the heating power, Rs is the DC internal resistance of the battery, / is the charging and discharging current of the battery, 7 is the battery temperature, E is the balanced electromotive force inside the battery, and U; is the terminal voltage of the battery:
S22, using the similarity of system modeling to express the thermal characteristic model in time domain state space equation. Obtaining the differential equations of 7; node and Ts node based on Bernardi's simplified equation and Kirchhoff's circuit law in circuit science:
In the formula, Q is the heating power, Ti is the highest temperature in the center of the battery, Ts is the surface temperature of the battery, Tamb is the ambient temperature of the battery, Ri is the internal thermal resistance of the battery, Ro is the external thermal resistance of the battery, Ci is the equivalent thermal capacity of the battery, Cs is the equivalent hot melting of the battery shell, Tis is the difference between the highest temperature 7; of the internal central temperature of the battery and the ambient temperature Tamb Of the battery, and Tss is the difference between the surface temperature 7s of the battery and the ambient temperature Tamb of the battery.
In the optimal control of engineering practice, state variables are often required as feedback quantities. In the embedded system application of lithium-ion battery, the value of Tss is easier to be measured safely than that of Tis, therefore, the embodiment of the invention selects the output vector of the thermal characteristic model as Tss, and the observation equation in the continuous time state is obtained:
Te (011 [7 | + 0)
Where y: is the system output in time domain, Ct is the output matrix in time domain, 503873 xt is the system state variable in continuous state, D: is the branch transfer matrix in time domain, and u is the system input in continuous state.
S3, calculating the difference equation of Tis according to the state space equation of the equivalent circuit model, where Tis is the difference between the highest internal temperature Ti; of the lithium-ion battery and the ambient temperature Tamb where the battery is located.
S31, taking the heating power Q as the system input and the highest internal temperature Tis as the system output, obtaining the system transfer function based on
LTCM:
S32, integrating and deforming the transfer function of the lumped thermal characteristic model by Z transform to obtain the difference equation of the thermal characteristic model, that is, the difference equation of Tis:
Tiss = Tiger + aise + Pa Qrei + 52 where Tis, x is the temperature value of the highest temperature point in the battery at time k, Tis, k+1 is the temperature value of the highest temperature point in the battery at time k+1, Tis k+2 is the temperature value of the highest temperature point in the battery at time k+2, Q« is the heating power of the battery at time k, Qx+1 is the heating power of the battery at time k+1, ay, az, B1 and B2 are the parameters to be identified of the difference equation of the internal temperature state model.
In the formula, a, az, B1 and B2 are the parameters to be identified, and their specific expansion forms are as follows: (a = — AIRC AERO — SRC, +2 (ta = (1 ~ ASR CO{AE SRC, + AE /R, GC — 1}
Le, cs ST AC: Pa = USER + ARC ~ 1)
Considering the uncertainty characteristics of the external equivalent thermal resistance Ro value of the battery, when identifying the parameters in LTCM, the invention couples Ro with the state variables of the thermal characteristic model, so as to realize the independent identification of Ro and the collaborative estimation of the internal temperature, and further, the calculation equations of the thermal parameters C;,
Ri and Cs of the battery can be obtained: LU503873
Where Ci is the equivalent heat capacity inside the battery, Riis the internal thermal resistance of the battery, Cs is the equivalent hot melt of the battery shell, andAt is the system sampling time.
S4, adopting the forgetting factor least square method to solve the parameters to be identified in the difference equation of the highest temperature point Tis in the lithium-ion battery.
S41, further deforming the obtained difference equation form to obtain an exogenous autoregressive model suitable for online iteration form: 8= {ed Be PT
Y= To Tasse Disa Tass Diem] " = {Trenet Piste Puen Gil”
Where Y is the output matrix of the exogenous autoregressive model, ¢ is the data matrix, 6 is the coefficient matrix, and @x is the system data matrix at time k;
S42. Based on the exogenous autoregressive model, the parameters of LTCM (Lumped Thermal Characteristic Model) are solved, where the parameters of the thermal characteristic model are Ci, Riand Cs. The main steps of identifying LTCM parameters by
FFRLS algorithm are as follows: 1): Assigning an initial value to k and initializing a coefficient matrix and an error covariance matrix Po: {k = 0,8, = Ef] {Ps = EL(85 — 80)(8 — 80)" 2) reading the data matrix «+1 at the time k+1 and calculating the gain matrix Gr+1;
Gist = Putroas/ A + Grows Pras) 3) updating the error covariance matrix Px+1 at time k+1;
Pir = (E = GpsrPpai I Pr fA 4): calculating the coefficient matrix at time k+1:
(Os = 8, + Be Be LU503873
Léger © Pres 7 Pros By 5): using the calculation equations of Ci, Ri and Cs to realize the identification of
LTCM parameters in a cyclic iteration until the end of the complete time series.
In the formula of the above steps, Pı1 is the error covariance matrix of the coefficient matrix at time k+1, Gz+1 is the gain matrix of FFRLS algorithm at time k+1, and
Ex +1 is the innovation value at time k+1, which indicates the error value between the measured value and the estimated value of the highest temperature inside the battery. A is the forgetting factor of the algorithm. In this embodiment, the value of the forgetting factor A is 0.98. Using the system input data matrix , the values of the coefficient matrix are obtained under the iterative computation of FFRLS. Further, the parameter values of LTCM are obtained through the calculation equations of Ci, Ri, and Cs.
S5, adopting the joint Kalman filter method to realize the joint estimation of the external equivalent thermal resistance Ro and the internal maximum temperature point
Tis of the lithium ion battery.
S51, from the point of view of engineering practice, the state equation description of the thermal circuit model in time domain is not suitable for online temperature estimation.
Therefore, the state space equation in the time domain is discretized to obtain the state space equation in the discrete state of the system based on LTCM: {vers x à Ba x In the formula, xa, x and xq, k+1 are the state variables of the system at time k and k+1 in discrete state, ya is the system output vector in discrete state, ux is the system input variable at time k, Au is the system matrix in discrete state, Ba is the control matrix in discrete state, Ca is the observation matrix in discrete state and Du is the direct transfer matrix in discrete state, since the direct transfer of the input vector is not considered in the discrete state space equation of the thermal characteristic model, that is, Da=0, Wi3 503873 and vk are respectively the system noise and the observation noise.
S52, realizing the adaptive common estimation strategy of external equivalent thermal resistance Ro and internal temperature state Tis based on JKF (Joint Kalman filter) and Kalman filtering algorithm. JKF algorithm takes Ro as one of the position state variables of the system, and realizes the adaptive identification of Ro through online state estimation algorithm. Constructing a JKF algorithm for adaptive collaborative estimation of external equivalent thermal resistance and internal temperature of battery. The state space equation of JKF algorithm is as follows: por er 7 fled Hwy Apt a + Bru, + We
Va = BR Wp) PS Org + Bag, + Va
Where x, is the 3-dimensional system state matrix, and the matrix is expressed as
Fe = her Teen Rol uk is the input variable of the system at time k, the matrix is expressed as uk= Qk. yk is the output variable of the system at time k in discrete state, the matrix is expressed as yı= ss, k. Au is the 3x3 -dimensional system matrix, By is the 3x1 -dimensional control matrix, Cu is the 1x3 -dimensional output matrix, Du is the direct matrix, the specific expansion form of the matrix of Ay, By, Cs and Du is as follows: (Ay [A 1] 8, = [4t/C; 0 of lc, =10 1 ©] D, = 101
Where C3 is calculated as follows:
REX pr Un) DR Un tp) dry, dT At
Finally, based on the state space equation of LTCM system in discrete state, the recursive iterative process of JKF algorithm suitable for online embedded application is obtained. The main steps are as follows: 1): assigning an initial value to k, and initializing a coefficient matrix and an error covariance matrix; {k= 0,2 = Exo] £80 = Elle = £5) (xg — 20] 2): updating the state variables in time:
peer = Ap + Bru LU503873 3): updating error covariance matrix in time: 4): calculating the Kalman gain: 5): updating the state variable measurement:
Tiger = EN 6): updating the error covariance matrix measurement: 7). lterating the equivalent thermal resistance Ro of the calculation part and the internal temperature state Tis circularly until the end of the complete time series.
The above implementation of the present invention only takes the commercial 18650 battery as an example to realize the modeling of the lumped thermal characteristics of the battery and the adaptive collaborative estimation of the model parameters and the battery internal temperature, but it can be understood that those skilled in the art can make any changes and changes without departing from the spirit and scope of the present invention.
The above is only the preferred embodiment of this application, but the protection scope of this application is not limited to this. Any change or replacement that can be easily thought of by a person familiar with this technical field within the technical scope disclosed in this application should be included in the protection scope of this application.
Therefore, the protection scope of this application should be based on the protection scope of the claims.
Claims (10)
1. An online self-adaptive method for estimating the battery internal temperature, characterized by comprising the following steps: establishing a lumped thermal characteristic equivalent circuit model of the battery based on the modeling similarity between physical systems of the same order; establishing a state space model of the lumped thermal characteristic equivalent circuit model in a time domain state based on Kirchhoff's circuit laws; acquiring a battery internal temperature state model based on the state space model in the time domain state; obtaining thermal characteristic model parameters to be identified in the battery internal temperature state model based on the battery internal temperature state model by adopting the forgetting factor least square method; jointly estimating the parameters of the thermal characteristic model and the battery internal temperature by using a joint Kalman filter method.
2. The online self-adaptive method for estimating the battery internal temperature according to claim 1, characterized in that the step of establishing a state space model of the lumped thermal characteristic equivalent circuit model in a time domain state based on Kirchhoff's circuit laws comprises: processing a heating power model based on the lumped thermal characteristic equivalent circuit model to obtain a new heating power model of the lumped thermal characteristic equivalent circuit model; obtaining the state models of the highest temperature node in the center of the battery and the temperature node in the surface area of the battery, based on the new heating power model and the Kirchhoff circuit law; obtaining an observation model of Tss based on the difference between the highest temperature node in the inner center of the battery and the ambient temperature where the battery is located; acquiring a state space model of the lumped thermal characteristic equivalent circuit model based on the state model and the observation model.
3. The online self-adaptive method for estimating the battery internal temperature 503873 according to claim 2, characterized in that the new heating power model is:
Q x PROAITIRY/dT & FE —U, = IR,
where Q is the heating power, Rs is the DC internal resistance of the battery, / is the charging and discharging current of the battery, 7 is the battery temperature, E is the balanced electromotive force inside the battery, and U; is the terminal voltage of the battery:
the state model is as follows:
where, Q is the heating power, 7; is the highest temperature in the center of the battery, Ts is the surface temperature of the battery, Tamb is the ambient temperature of the battery, Riis the internal thermal resistance of the battery, Ro is the external thermal resistance of the battery, C; is the equivalent thermal capacity of the battery, Cs is the equivalent hot melting of the battery shell, Tis is the difference between the highest temperature 7; of the internal central temperature of the battery and the ambient temperature Tamb Of the battery, and Tss is the difference between the surface temperature Ts of the battery and the ambient temperature Tamb Of the battery;
the observation model is:
1-10 ul] +09)
Ft £3 Lu By ug where, y: is the system output in time domain, Ct is the output matrix in time domain, xt is the system state variable in continuous state, D: is the branch transfer matrix in time domain, and u is the system input in continuous state.
4. The online self-adaptive method for estimating the battery internal temperature 503873 according to claim 2, characterized in that the step of acquiring a battery internal temperature state model based on the state space model in the time domain state comprises: acquiring a transfer function of the state space model based on the new heating power model and the internal temperature state of the battery, wherein the internal temperature state of the battery is the difference between the highest point of the internal central temperature of the battery and the ambient temperature where the battery is located: carrying out integrated deformation on the transfer function by using Z transform to obtain the internal temperature state model of the battery.
5. The online self-adaptive method for estimating the battery internal temperature according to claim 4, characterized in that the difference equation of the battery internal temperature state model is as follows: Tara = Ma Tiger + olin + Bi Queer + He where Tis, x is the temperature value of the highest temperature point in the battery at time k, Tis, k+1 is the temperature value of the highest temperature point in the battery at time k+1, Tis k+2 is the temperature value of the highest temperature point in the battery at time k+2, Q« is the heating power of the battery at time k, Qx+1 is the heating power of the battery at time k+1, ay, az, B1 and B2 are the parameters to be identified of the difference equation of the internal temperature state model.
6. The online self-adaptive method for estimating the battery internal temperature 503873 according to claim 1, characterized in that, the step of obtaining thermal characteristic model parameters to be identified in the battery internal temperature state model based on the battery internal temperature state model by adopting the forgetting factor least square method comprises: acquiring an exogenous autoregressive model based on the battery internal temperature state model; solving the thermal characteristic model parameters that need to be identified in the battery internal temperature state model by the least method based on the exogenous autoregressive model.
7. The online self-adaptive method for estimating the battery internal temperature according to claim 6, characterized in that the exogenous autoregressive model is as follows: (8 = {oy da, By, PT Y= #78 | ? = io, Pu Po Ps Pa} _ Y= {Teo Tes Toe Teper Trent lo = Tispen Tor Queer OT" where Y is the output matrix of the exogenous autoregressive model, @ is the data matrix, 6 is the coefficient matrix, and @x is the system data matrix at time k; the parameters of the thermal characteristic model parameters are Ci, Ri and Cs; the thermal characteristic model parameter equation is as follows: where C; is the equivalent heat capacity inside the battery, Ri is the internal thermal resistance of the battery, Cs is the equivalent hot melt of the battery shell, andAt is the system sampling time.
8. The online self-adaptive method for estimating the battery internal temperature 503873 according to claim 1, characterized in that the step of jointly estimating the parameters of the thermal characteristic model and the battery internal temperature by using a joint Kalman filter method comprises: acquiring the state space model in the discrete state of the lumped thermal characteristic equivalent circuit model based on the state space model in the time domain state; constructing the JKF model of self-adaptive cooperative estimation of the external equivalent thermal resistance and the internal temperature state of the battery by using the joint Kalman filtering method and based on the state space equation in the discrete state; realizing the joint estimation of the parameters of the thermal characteristic model and the internal temperature based on the JKF model.
9. The online self-adaptive method for estimating the battery internal temperature according to claim 8, characterized in that the state space model in the discrete state is: re {ow 3 A 1 Ar od x Ag $ where Xu, k+1 are the state variables of the system at time k+1 in discrete state, xq, « are the state variables of the system at time Æ in discrete state, ux is the input variable of the system at time k, wk is system noise, yx is the output variable of the system at time k in discrete state, Au is the system matrix in discrete state, Ba is the control matrix in discrete state, Ca is the observation matrix in discrete state and Du is the direct transfer matrix in discrete state.
10. The online self-adaptive method for estimating the battery internal temperature 503873 according to claim 8, characterized in that the JKF model is as follows:
(pa = Fla ed + we SS App + Brg Hwy iv, = REX pe Ua) + y € CX + Brig + uy where xy is the 3-dimensional system state matrix at time k, xyx is the system state matrix engraved at time k+1, ux is the input variable of the system at time k, yk is the output variable of the system at time k in discrete state, Ay is the 3x3 -dimensional system matrix, By is the 3x1 -dimensional control matrix, Cy is the 1x3 -dimensional output matrix, Dy is the direct matrix, wk is the system noise and vx is the observation noise.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| LU503873A LU503873B1 (en) | 2023-04-07 | 2023-04-07 | Online self-adaptive method for estimating battery internal temperature |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| LU503873A LU503873B1 (en) | 2023-04-07 | 2023-04-07 | Online self-adaptive method for estimating battery internal temperature |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| LU503873B1 true LU503873B1 (en) | 2023-10-09 |
Family
ID=88289566
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| LU503873A LU503873B1 (en) | 2023-04-07 | 2023-04-07 | Online self-adaptive method for estimating battery internal temperature |
Country Status (1)
| Country | Link |
|---|---|
| LU (1) | LU503873B1 (en) |
-
2023
- 2023-04-07 LU LU503873A patent/LU503873B1/en active IP Right Grant
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shi et al. | A novel lumped thermal characteristic modeling strategy for the online adaptive temperature and parameter co-estimation of vehicle lithium-ion batteries | |
| Xie et al. | An improved resistance-based thermal model for a pouch lithium-ion battery considering heat generation of posts | |
| CN110068772B (en) | State-of-charge estimation method for lithium-ion batteries based on improved fractional-order model | |
| CN104462847B (en) | A kind of internal temperature of battery real-time predicting method | |
| CN104849675B (en) | Acquisition method of electrochemical and thermal coupling model of lithium-ion battery | |
| CN107576919A (en) | Power battery charged state estimating system and method based on ARMAX models | |
| CN106597308B (en) | A kind of power battery method for estimating remaining capacity | |
| CN104991980A (en) | Electrochemical mechanism modeling method for lithium-ion battery | |
| CN111914503A (en) | Lithium ion battery power input electric heating coupling model building method | |
| CN105116343A (en) | LS-SVM power cell SOC estimation method and system | |
| CN105572596B (en) | Lithium battery SOC estimation method and system | |
| CN114545262A (en) | A lithium-ion battery parameter identification and SOC estimation method for loss data | |
| CN108427079B (en) | A kind of power battery method for estimating remaining capacity | |
| CN110795851A (en) | Lithium ion battery modeling method considering environmental temperature influence | |
| Wang et al. | An integrated online adaptive state of charge estimation approach of high-power lithium-ion battery packs | |
| CN110146823A (en) | A SOC Estimation Method for Series Battery Packs Based on Information Fusion | |
| CN114498882B (en) | A hybrid energy storage system with low-temperature self-heating function and a self-heating method | |
| CN108414947B (en) | A joint estimation method of spatial lithium-ion battery state based on multiple time scales | |
| CN113839442A (en) | A bidirectional power converter-based filter state estimation method for chemical synthesis | |
| Zhuang et al. | Research on estimation of state of charge of li-ion battery based on cubature kalman filter | |
| CN108732509A (en) | A kind of On-line Estimation method of the charge states of lithium ion battery of space-oriented application | |
| CN118095032A (en) | Modeling and simulation analysis method for lithium iron phosphate battery based on digital twin | |
| CN107167741A (en) | A kind of lithium battery SOC observation procedures based on neutral net | |
| CN113671384A (en) | A kind of battery temperature prediction method and system | |
| Zeng et al. | Improved fractional‐order hysteresis‐equivalent circuit modeling for the online adaptive high‐precision state of charge prediction of urban‐electric‐bus lithium‐ion batteries |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| FG | Patent granted |
Effective date: 20231009 |