1. A parameter identification method of a battery equivalent circuit model is characterized by comprising the following steps:

establishing an N-order RC equivalent circuit model and an N-1-order PNGV model, and acquiring a first model parameter of the N-order RC equivalent circuit model to be estimated and a second model parameter of the N-1-order PNGV model;

a first state equation of the N-order RC equivalent circuit model is simultaneously established according to the first model parameter, and a second state equation of the N-1-order PNGV model is simultaneously established according to the second model parameter;

converting the first state equation and the second state equation into a difference equation;

estimating the first model parameter and the second model parameter by a parameter estimation algorithm based on the difference equation.

2. The method of claim 1, wherein the step of simultaneously establishing the first state equation of the N-th order RC equivalent circuit model according to the first model parameters and the second state equation of the N-1 th order PNGV model according to the second model parameters comprises the steps of:

the charging current is regulated to be positive;

acquiring terminal voltage, open-circuit voltage, polarization capacitance and ohmic drop of the N-order RC equivalent circuit model based on the first model parameter, and acquiring terminal voltage, open-circuit voltage, polarization capacitance, equivalent capacitance and ohmic drop of the N-1-order PNGV model based on the second model parameter;

the first state equation and the second state equation are simultaneous.

3. The method of claim 1, wherein the step of estimating the first model parameter and the second model parameter by a parameter estimation algorithm based on the difference equation comprises the steps of:

acquiring a difference coefficient and a difference matrix of the difference equation based on the difference equation;

establishing a conversion relation based on the difference coefficient, the first model parameter and the second model parameter;

acquiring a first relation between the differential coefficient and the first model parameter and a second relation between the differential coefficient and the second model parameter according to the conversion relation;

estimating the difference coefficient through a parameter estimation algorithm to obtain an estimation result;

estimating the first model parameter from the conversion relation and the first relation, and estimating the second model parameter from the conversion relation and the second relation, based on the estimation result.

4. The method for identifying parameters of a battery equivalent circuit model according to claim 3, wherein the step of estimating the difference coefficients by a parameter estimation algorithm comprises:

sampling the voltage and the current of the battery for the ith time;

determining initial values of the difference coefficient and the covariance matrix;

calculating a gain matrix based on the difference matrix and the covariance matrix;

updating a covariance matrix;

and estimating the difference coefficient according to the difference matrix, the difference equation and the gain matrix to obtain an estimation result of the difference coefficient.

5. The method for identifying parameters of a battery equivalent circuit model according to claim 1, wherein the step of estimating the difference coefficients by a parameter estimation algorithm comprises:

sampling the voltage and the current of the battery for the ith time;

determining initial values of the difference coefficient and the covariance matrix;

determining a forgetting factor;

calculating a gain matrix based on the difference matrix, the covariance matrix, and the forgetting factor;

estimating the difference coefficients from the difference matrix, the difference equation and the gain matrix;

updating the covariance matrix;

sampling the voltage and the current of the battery for the ith (i + 1) time;

and repeatedly calculating the difference coefficient, the covariance matrix and the gain matrix until N times of observation are finished to obtain the estimation result of the difference coefficient.

6. The method for identifying parameters of a battery equivalent circuit model according to claim 4 or 5, wherein the step of determining the initial values of the difference coefficients and covariance matrix comprises the steps of:

selecting the previous n observations;

calculating initial values of the difference coefficient and the covariance matrix through the difference equation and the difference matrix;

and (5) performing the (n + 1) th observation, and estimating the difference coefficient through the parameter estimation algorithm.

7. The method for identifying parameters of a battery equivalent circuit model according to claim 4 or 5, wherein in the step of determining the initial values of the difference coefficients and covariance matrix, the initial value of the difference coefficient is defined as θ₀＝[0 0 … 0 0 0]^TAnd the initial value of the covariance matrix is P₀＝σ²E, where E is an NxN identity matrix, σ²≥10⁶。

8. A parameter identification system of a battery equivalent circuit model is characterized by comprising: the system comprises an establishing module, an equation module, a conversion module and an estimation module;

the establishing module is used for establishing an N-order RC equivalent circuit model and an N-1-order PNGV model, and acquiring a first model parameter of the N-order RC equivalent circuit model to be estimated and a second model parameter of the N-1-order PNGV model;

the equation module is used for simultaneously establishing a first state equation of the N-order RC equivalent circuit model according to the first model parameter and establishing a second state equation of the N-1-order PNGV model according to the second model parameter;

the conversion module is used for converting the first state equation and the second state equation into a difference equation;

the estimation module is configured to estimate the first model parameter and the second model parameter by a parameter estimation algorithm based on the difference equation.

9. A terminal device, comprising: a memory and a processor:

the memory for storing a computer program;

the processor configured to execute the computer program stored in the memory to cause the terminal device to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium comprising a program or instructions for performing the method of any one of claims 1 to 7 when the program or instructions are run on a computer.

[ background of the invention ]

Today, accurate and reliable battery models are not only a prerequisite for implementing battery characteristic simulations, but also a basis for battery state estimation. The equivalent circuit model is widely applied to battery characteristic simulation and state estimation due to the advantages of simple structure, clear physical significance, high precision and the like. The equivalent circuit model of the battery includes an RC equivalent circuit model and a PNGV model, and although there are many methods for identifying parameters of these equivalent circuit models, these equivalent circuit models do not have the characteristics of real-time and online estimation. According to the current research results, the parameter identification of the RC equivalent circuit model and the PNGV model needs to be separately identified, but a unified method capable of identifying the two model parameters at the same time is not formed yet.

[ summary of the invention ]

In view of this, embodiments of the present invention provide a method and a system for identifying parameters of a battery equivalent circuit model, so as to solve the technical problem that the parameter identification of an RC equivalent circuit model and a PNGV model in the prior art cannot form uniform and simultaneous identification.

In a first aspect, an embodiment of the present invention provides a method for identifying parameters of a battery equivalent circuit model, including the following steps:

establishing an N-order RC equivalent circuit model and an N-1-order PNGV model, and acquiring a first model parameter of the N-order RC equivalent circuit model to be estimated and a second model parameter of the N-1-order PNGV model;

a first state equation of the N-order RC equivalent circuit model is simultaneously established according to the first model parameter, and a second state equation of the N-1-order PNGV model is simultaneously established according to the second model parameter;

converting the first state equation and the second state equation into a difference equation;

estimating the first model parameter and the second model parameter by a parameter estimation algorithm based on the difference equation.

Through the scheme provided by the embodiment, the model parameter identification of the N-order RC equivalent circuit model and the N-1-order PNGV model can be processed together, two model parameters can be calculated simultaneously only through one difference equation, the battery state can be evaluated quickly, the battery performance can be evaluated, and the use safety of equipment for normal work of the battery is guaranteed.

In a preferred embodiment, the step of simultaneously establishing a first state equation of the RC equivalent circuit model of order N according to the first model parameters and a second state equation of the PNGV model of order N-1 according to the second model parameters comprises the steps of:

the charging current is regulated to be positive;

acquiring terminal voltage, open-circuit voltage, polarization capacitance and ohmic drop of the N-order RC equivalent circuit model based on the first model parameter, and acquiring terminal voltage, open-circuit voltage, polarization capacitance, equivalent capacitance and ohmic drop of the N-1-order PNGV model based on the second model parameter;

the first state equation and the second state equation are simultaneous.

By the scheme provided by the embodiment, different state equations are respectively connected aiming at different model parameters in the N-order RC equivalent circuit model and the N-1-order PNGV model, so that after the difference coefficient is calculated by the difference equation, the estimation results of the model parameters can be respectively obtained through the corresponding state equations.

In a preferred embodiment, in the step of estimating the first model parameter and the second model parameter by a parameter estimation algorithm based on the difference equation, the method comprises the steps of:

acquiring a difference coefficient and a difference matrix of the difference equation based on the difference equation;

establishing a conversion relation based on the difference coefficient, the first model parameter and the second model parameter;

acquiring a first relation between the differential coefficient and the first model parameter and a second relation between the differential coefficient and the second model parameter according to the conversion relation;

estimating the difference coefficient through a parameter estimation algorithm to obtain an estimation result;

estimating the first model parameter from the conversion relation and the first relation, and estimating the second model parameter from the conversion relation and the second relation, based on the estimation result.

By the scheme provided by the embodiment, for the N-order RC equivalent circuit model and the N-1-order PNGV model, the difference coefficient and the first model parameter and the second model parameter respectively establish a mathematical relationship, so that the model parameters corresponding to the N-order RC equivalent circuit model and the N-1-order PNGV model can be rapidly calculated after the estimation result of the difference coefficient is obtained.

In a preferred embodiment, in the step of estimating the difference coefficient by a parameter estimation algorithm, the method comprises:

sampling the voltage and the current of the battery for the ith time;

determining initial values of the difference coefficient and the covariance matrix;

calculating a gain matrix based on the difference matrix and the covariance matrix;

estimating the difference coefficients from the difference matrix, the difference equation and the gain matrix;

updating a covariance matrix;

sampling the voltage and the current of the battery for the ith (i + 1) time;

and repeatedly calculating the difference coefficient, the covariance matrix and the gain matrix until L times of observation are finished to obtain the estimation result of the difference coefficient.

By the scheme provided by the embodiment, the difference coefficient is estimated by using a recursive least square algorithm, so that the operation speed and the accuracy of the estimation result are improved.

In a preferred embodiment, in the step of estimating the difference coefficient by a parameter estimation algorithm, the method comprises:

sampling the voltage and the current of the battery for the ith time;

determining initial values of the difference coefficient and the covariance matrix;

determining a forgetting factor;

calculating a gain matrix based on the difference matrix, the covariance matrix, and the forgetting factor;

estimating the difference coefficients from the difference matrix, the difference equation and the gain matrix;

updating the covariance matrix;

sampling the voltage and the current of the battery for the ith (i + 1) time;

and repeatedly calculating the difference coefficient, the covariance matrix and the gain matrix until L times of observation are finished to obtain the estimation result of the difference coefficient.

By the scheme provided by the embodiment, the differential coefficient is estimated by using the recursive least square algorithm with forgetting factors, so that the information quantity provided by new voltage and current data obtained by sampling can be enhanced, old data can be weakened gradually, data saturation can be prevented, and the operation speed and the accuracy of the estimation result can be improved.

In a preferred embodiment, the step of determining the initial values of the difference coefficients and covariance matrix comprises the steps of:

selecting the previous n observations;

calculating initial values of the difference coefficient and the covariance matrix through the difference equation and the difference matrix;

and (5) performing the (n + 1) th observation, and estimating the difference coefficient through the parameter estimation algorithm.

By the scheme provided by the embodiment, the initial values of the difference coefficient and the covariance matrix are calculated by using a recursion algorithm, and the calculation speed is improved.

In a preferred embodiment, in the step of determining the initial values of the difference coefficients and covariance matrix, the initial value of the difference coefficient is defined as θ₀＝[00…000]^TAnd the initial value of the covariance matrix is P₀＝σ²E, where E is an NxN identity matrix, σ²≥10⁶。

By the scheme provided by the embodiment, the method for defining the initial value is utilized, and the computing resource for computing the initial value is saved.

In a second aspect, an embodiment of the present invention provides a parameter identification system for a battery equivalent circuit model, including: the system comprises an establishing module, an equation module, a conversion module and an estimation module;

the establishing module is used for establishing an N-order RC equivalent circuit model and an N-1-order PNGV model, and acquiring a first model parameter of the N-order RC equivalent circuit model to be estimated and a second model parameter of the N-1-order PNGV model;

the equation module is used for simultaneously establishing a first state equation of the N-order RC equivalent circuit model according to the first model parameter and establishing a second state equation of the N-1-order PNGV model according to the second model parameter;

the conversion module is used for converting the first state equation and the second state equation into a difference equation;

the estimation module is configured to estimate the first model parameter and the second model parameter by a parameter estimation algorithm based on the difference equation.

According to the scheme provided by the embodiment, the model parameter identification of the N-order RC equivalent circuit model and the N-1-order PNGV model is processed by the four modules together, two model parameters can be calculated simultaneously only through one difference equation, the battery state can be evaluated quickly, the battery performance can be evaluated, and the use safety of equipment for normal work of the battery is guaranteed.

In a third aspect, an embodiment of the present invention provides a terminal device, including: a memory and a processor:

the memory for storing a computer program;

the processor is configured to execute the computer program stored in the memory to cause the terminal device to perform the method according to the first aspect.

In a fourth aspect, embodiments of the present invention provide a computer-readable storage medium, comprising a program or instructions, which when run on a computer, performs the method according to the first aspect.

Compared with the prior art, the technical scheme at least has the following beneficial effects:

according to the parameter identification method and system of the battery equivalent circuit model disclosed by the embodiment of the invention, the respective state equations of the N-order RC equivalent circuit model and the N-1-order PNGV model are connected, the state equations are converted into the same differential equation, the model parameters of the N-order RC equivalent circuit model and the N-1-order PNGV model can be estimated by estimating the differential coefficient of the differential equation and the mathematical relationship between the differential coefficient and the respective model parameters of the N-order RC equivalent circuit model and the N-1-order PNGV model, and the effect of identifying the two model parameters of the N-order RC equivalent circuit model and the N-1-order PNGV model at the same time is realized.

[ description of the drawings ]

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1 is a flowchart illustrating a method for identifying parameters of a battery equivalent circuit model according to embodiment 1 of the present invention;

fig. 2 is a flowchart of Step200 in the method for identifying parameters of a battery equivalent circuit model according to embodiment 1 of the present invention;

fig. 3 is a flowchart of Step400 in the method for identifying parameters of a battery equivalent circuit model according to embodiment 1 of the present invention;

fig. 4 is a flowchart of steps when the recursive least square algorithm is adopted in Step440 in the method for identifying parameters of a battery equivalent circuit model according to embodiment 1 of the present invention;

fig. 5 is a flowchart of a Step when Step440 adopts a recursive least square algorithm with a forgetting factor in the parameter identification method of the battery equivalent circuit model according to embodiment 1 of the present invention;

fig. 6 is a block diagram of a parameter identification system of a battery equivalent circuit model according to embodiment 2 of the present invention.

Reference numerals:

1-establishing a module; 2-equation module; 3-a transformation module; 4-estimation module.

[ detailed description ] embodiments

For better understanding of the technical solutions of the present invention, the following detailed descriptions of the embodiments of the present invention are provided with reference to the accompanying drawings.

It should be understood that the described embodiments are only some embodiments of the invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Example 1

As shown in fig. 1, an embodiment 1 of the present invention discloses a parameter identification method for a battery equivalent circuit model, including the following steps:

step 100: an N-order RC equivalent circuit model and an N-1-order PNGV model are established, and a first model parameter of the N-order RC equivalent circuit model to be estimated and a second model parameter of the N-1-order PNGV model are obtained.

Step 200: and simultaneously establishing a first state equation of an N-order RC equivalent circuit model according to the first model parameter and establishing a second state equation of an N-1-order PNGV model according to the second model parameter.

Step 300: the first state equation and the second state equation are converted into a difference equation.

Step 400: the first model parameters and the second model parameters are estimated by a parameter estimation algorithm based on a difference equation.

The parameter identification method of embodiment 1 can process the model parameter identification of the N-order RC equivalent circuit model and the N-1-order PNGV model together, and can calculate two model parameters at the same time only through one difference equation, which is beneficial to rapidly evaluating the battery state, further evaluating the battery performance, and providing guarantee for the use safety of the equipment in normal operation of the battery.

As shown in fig. 2, in the parameter identification method of the embodiment 1, in Step 200: the method comprises the following steps of simultaneously establishing a first state equation of an N-order RC equivalent circuit model according to a first model parameter and establishing a second state equation of an N-1-order PNGV model according to a second model parameter:

step 210: the charging current I is defined to be positive.

Step 220: acquiring terminal voltage U and open-circuit voltage U of N-order RC equivalent circuit model based on first model parameters_oPolarization voltage U_pAnd a polarization capacitor C_pAnd the ohmic drop in the voltage RI,acquiring terminal voltage U and open-circuit voltage U of N-1-order PNGV model based on second model parameters_oPolarization voltage U_pAnd a polarization capacitor C_pAnd an equivalent capacitance C_bAnd ohmic drop RI.

Step 230: the first state equation and the second state equation are simultaneously established.

In the parameter identification method of this embodiment 1, different state equations are respectively associated with different model parameters in the N-order RC equivalent circuit model and the N-1-order PNGV model, so that after the difference coefficients are calculated by the difference equations, estimation results of the respective model parameters can be respectively obtained by the respective corresponding state equations.

As shown in fig. 3, in the parameter identification method of the embodiment 1, in Step 400: based on the difference equation, the method for estimating the first model parameter and the second model parameter through the parameter estimation algorithm comprises the following steps:

step 410: obtaining a difference coefficient theta and a difference matrix of a difference equation based on the difference equation

Step 420: and establishing a conversion relation based on the difference coefficient theta, the first model parameter and the second model parameter.

Step 430: and acquiring a first relation between the difference coefficient theta and the first model parameter and a second relation between the difference coefficient theta and the second model parameter according to the conversion relation.

Step 440: and estimating the difference coefficient theta through a parameter estimation algorithm, and obtaining an estimation result.

Step 450: based on the estimation result, the first model parameter is estimated from the conversion relation and the first relation, and the second model parameter is estimated from the conversion relation and the second relation.

In Step400, Step420, Step430 and Step440 are not in strict sequential Step order, and difference equation u (k) adopts difference coefficient theta and difference matrixAfter the representation, Step420 and Step430 are to establish a transformation relationship, a first relationship and a second relationship between the difference coefficient θ and the first model parameter and the second model parameter, Step440 is to estimate the difference coefficient θ, both of which can be performed simultaneously, and finally, the Step450 is executed to estimate the first model parameter and the second model parameter by combining them.

In the parameter identification method of embodiment 1, for the N-order RC equivalent circuit model and the N-1-order PNGV model, mathematical relationships are respectively established between the difference coefficient θ and the first model parameter and the second model parameter, so that after an estimation result of the difference coefficient θ is obtained, the model parameters corresponding to the N-order RC equivalent circuit model and the N-1-order PNGV model can be quickly calculated.

As shown in fig. 4, in the parameter identification method of embodiment 1, in Step 440: in the differential coefficient estimation through the parameter estimation algorithm, when the parameter estimation algorithm adopts a recursive least square algorithm, the method comprises the following steps:

step 441: the voltage and current of the battery are sampled for the ith time.

Step 442: determining difference coefficients theta and covariance matrix P_NThe initial value of (c).

Step 443: based on a difference matrixSum covariance matrix P_NA gain matrix is calculated.

Step 444: according to a difference matrixDifference equation y (k) and gain matrix G_iThe difference coefficient theta is estimated.

Step 445: updating the covariance matrix P_N。

Step 446: the voltage and current of the battery are sampled for the ith (i + 1) times.

Step 447: repeatedly calculating difference coefficient theta and covariance matrix P_NAnd a gain matrix G_iAnd obtaining the estimation result of the difference coefficient theta until L times of observation are finished.

In the parameter identification method of embodiment 1, the difference coefficient θ is estimated by using a recursive least square algorithm, so that the operation speed and the accuracy of the estimation result are improved.

As shown in fig. 5, in the parameter identification method of embodiment 1, in Step 440: in the differential coefficient theta is estimated through a parameter estimation algorithm, when the parameter estimation algorithm adopts a recursive least square algorithm of a factor p to be forgotten, the method comprises the following steps:

step 441': the voltage and current of the battery are sampled for the ith time.

Step 442': determining difference coefficients theta and covariance matrix P_NThe initial value of (c).

Step 443': a forgetting factor p is determined.

Step 444': based on a difference matrixCovariance matrix P_NAnd a forgetting factor p to calculate a gain matrix G_i。

Step 445': according to a difference matrixDifference equation y (k) and gain matrix G_iThe difference coefficient theta is estimated.

Step 446': updating the covariance matrix P_N。

Step 447': the voltage and current of the battery are sampled for the ith (i + 1) times.

Step 448': repeatedly calculating difference coefficient theta and covariance matrix P_NAnd a gain matrix G_iAnd obtaining the estimation result of the difference coefficient theta until L times of observation are finished.

In the parameter identification method of embodiment 1, the recursive least square algorithm with the forgetting factor ρ is used to estimate the difference coefficient θ, so that the information amount provided by the new voltage and current data obtained by sampling can be enhanced, old data can be gradually weakened, data saturation can be prevented, and the operation speed and the accuracy of the estimation result can be improved.

In the parameter identification method of the present embodiment 1, the difference coefficient θ and the covariance matrix P are determined_NThe initial value of (a) includes the steps of:

selecting the previous n observations;

by difference equations y (k) and difference matricesCalculating difference coefficient theta and covariance matrix P_NAn initial value of (d);

and (5) performing the (n + 1) th observation, and estimating a difference coefficient theta through a parameter estimation algorithm.

In the parameter identification method of embodiment 1, the difference coefficient θ and the covariance matrix P are calculated by using a recursive algorithm_NThe initial value of (2) and the calculation speed are improved.

In the parameter identification method of the present embodiment 1, the difference coefficient θ and the covariance matrix P are determined_NIn the step of defining the initial value of the difference coefficient theta as theta₀＝[0 0 … 0 0 0]^TAnd a covariance matrix P_NHas an initial value of P₀＝σ²E, where E is an NxN identity matrix, σ²≥10⁶。

The parameter identification method of embodiment 1 utilizes a method of defining an initial value, thereby saving the calculation resources for calculating the initial value.

Specifically, in the method for identifying parameters of the battery equivalent circuit model in embodiment 1, Step100 is adopted to establish an N-order RC equivalent circuit model and an N-1-order PNGV model, a first model parameter of the N-order RC equivalent circuit model to be estimated and a second model parameter of the N-1-order PNGV model are obtained, Step210 is adopted to specify that the charging current I is positive, and Step220 is adopted to obtain the terminal voltage U and the open-circuit voltage U of the N-order RC equivalent circuit model based on the first model parameter_oPolarization voltage U_pAnd a polarization capacitor C_pAnd ohmic drop RI, using Step230 to establish a first equation of state, which can be expressed as:

likewise, Step220 is employed based on the first modelParameter acquisition terminal voltage U and open-circuit voltage U of N-1-order PNGV model_oPolarization voltage U_pAnd a polarization capacitor C_pAnd an equivalent capacitance C_bAnd ohmic drop RI, using Step230 to establish a second equation of state, which can be expressed as:

wherein:

τ_p,i＝R_p,iC_p,iequation (3).

After the equations (1) and (2) are converted into the difference equation by the Step300, they have the same structure and form:

U(k)＝α₀+α₁U(k-1)+α₂U(k-2)+…+α_NU(k-N)+β₀I(k)+β₁I(k-1)+β₂I(k-2)+…+β_Ni (k-N) formula (4).

Equation (4) can be written as follows:

wherein: difference equation y (k) u (k), and difference coefficient θ [ α ═ α₀ α₁ α₂ … β₀ β₁ β₂ … β_N]Differential matrix

Step420 is adopted to establish a conversion relation between the difference coefficient theta of the difference equation y (k) in the formula (4) and the first model parameter and the second model parameter:

then, Step430 is adopted to establish the difference coefficient θ ═ α in equation (6)₀ α₁ α₂ … β₀ β₁ β₂ … β_N]And a first relation between the first model parameter of the N-order RC equivalent circuit model and a second relation between the second model parameter of the N-1-order PNGV model, wherein the first relation is formula (7) and the second relation is formula (8).

In particular, when N is 0, a_i＝b_iWhen the Open Circuit Voltage (OCV) is equal to 0, the first model parameters of the zero-order RC equivalent circuit model including the Open Circuit Voltage (OCV) can be calculated by equations (6) and (7). When N is 1, the equivalent circuit model is a first-order RC model, and the first model parameters of the first-order RC equivalent circuit model including the open-circuit voltage can be calculated by using the formula (6) and the formula (7). If N is present>1, the first model parameter of the N-order RC equivalent circuit model including the open circuit voltage can be calculated by using the formula (6) and the formula (7), and the second model parameter of the N-1-order PNGV model including the open circuit voltage can be calculated by combining the formula (6) and the formula (8).

The Step440 may be used to estimate the difference coefficient θ in the formula (5) by using a parameter estimation algorithm, which includes, but is not limited to, Least Square-like algorithms such as Least Square (LS), Recursive Least Square (RLS), Recursive augmented Least Square (RELS), Recursive Least Square with forgetting factor (RFF), and so on. The following is an exemplary presentation of a recursive least squares algorithm and a recursive least squares algorithm with forgetting factor.

When the step of estimating the parameters of the battery equivalent circuit model by using the recursive least square algorithm is as follows:

step 441: the voltage and current of the battery are sampled i times.

Step 442: determining difference coefficients theta and covariance matrix P_NThe initial value of (c).

Step 443: based on a difference matrixSum covariance matrix P_NComputing a gain matrix G_i：

Step 444: according to a difference matrixDifference equation y (k) and gain matrix G_iEstimating a difference coefficient θ:

and after calculating theta, if N is less than or equal to 1, obtaining a first model parameter of the N-order RC equivalent circuit model by utilizing the relation between the formula (6) and the formula (7). If N >1, the first model parameter of the N-order RC equivalent circuit model can be obtained by using the relationship between the formula (6) and the formula (7), and the second model parameter of the N-1-order PNGV model can be obtained by using the relationship between the formula (6) and the formula (8).

Step 445: updating the covariance matrix P_N：

Step 446: the battery voltage and current are sampled i +1 times.

Step 447: repeating the steps from Step443 to Step447, and calculating the difference coefficient theta and the covariance matrix P_NAnd a gain matrix G_iAnd obtaining the estimation result of the difference coefficient theta until the L times of observation are finished.

When the recursive least square algorithm with forgetting factors is used for estimating the parameters of the battery equivalent circuit model, the method comprises the following steps:

step 441': the voltage and current of the battery are sampled i times.

Step 442': determining difference coefficients theta and covariance matrix P_NThe initial value of (c).

Step 443': determining a forgetting factor ρ: 0< rho is less than or equal to 1.

Step 444': based on a difference matrixSum covariance matrix P_NComputing a gain matrix G_i：

Step 445': according to a difference matrixDifference equation y (k) and gain matrix G_iEstimating a difference coefficient θ:

and after calculating theta, if N is less than or equal to 1, obtaining a first model parameter of the N-order RC equivalent circuit model by utilizing the relation between the formula (6) and the formula (7). If N >1, the first model parameter of the N-order RC equivalent circuit model can be obtained by using the relationship between the formula (6) and the formula (7), and the second model parameter of the N-1-order PNGV model can be obtained by using the relationship between the formula (6) and the formula (8).

Step 446': updating the covariance matrix P_N：

Step 447': the voltage and current of the battery are sampled for the ith (i + 1) times.

Step 448': repetition ofStep444 'to Step 448', calculating the difference coefficient theta and the covariance matrix P_NAnd a gain matrix G_iAnd obtaining the estimation result of the difference coefficient theta until L times of observation are finished.

In estimating battery equivalent circuit model parameters using a recursive least squares algorithm and a recursive least squares algorithm with a forgetting factor, steps 442 and 442' both involve determining a difference coefficient θ and a covariance matrix P_NInitial value of (a), difference coefficient theta and covariance matrix P_NThe initial value of (2) is determined in the following two ways.

The method comprises the following steps:

the first n observations (n < L) were chosen.

By difference equations y (k) and difference matricesThe difference coefficient theta and the covariance matrix P are calculated using the formula (15) and the formula (16)_NThe initial value of (c).

θ(n)＝(Φ^T(n)Φ(n))^-1Φ^T(n) Y (n) formula (15);

P(n)＝(Φ^T(n)Φ(n))^-1equation (16).

And (4) performing the (n + 1) th observation, and estimating a difference coefficient theta by using a recursive least square algorithm or a recursive least square algorithm with a forgetting factor.

The method 2 comprises the following steps: let theta₀＝[0 0 … 0 0 0]^T，P₀＝σ²E, E is an N × N identity matrix, σ²≥10⁶。

Example 2

As shown in fig. 6, an embodiment 2 of the present invention provides a parameter identification system for a battery equivalent circuit model, including: the model parameter identification and calculation system comprises an establishing module 1, an equation module 2, a conversion module 3 and an estimation module 4, wherein the establishing module 1, the equation module 2, the conversion module 3 and the estimation module 4 are in communication connection with each other, and model parameters can be identified and calculated together.

The establishing module 1 is used for establishing an N-order RC equivalent circuit model and an N-1-order PNGV model, and acquiring a first model parameter of the N-order RC equivalent circuit model to be estimated and a second model parameter of the N-1-order PNGV model;

the equation module 2 is used for simultaneously establishing a first state equation of an N-order RC equivalent circuit model according to the first model parameter and establishing a second state equation of an N-1-order PNGV model according to the second model parameter;

the conversion module 3 is used for converting the first state equation and the second state equation into a difference equation;

the estimation module 4 is configured to estimate the first model parameter and the second model parameter by a parameter estimation algorithm based on a difference equation.

In the parameter identification system of the battery equivalent circuit model in this embodiment 2, the four modules are used to process the model parameter identification of the N-order RC equivalent circuit model and the N-1-order PNGV model together, and two model parameters can be calculated simultaneously only by using one differential equation, so that the battery state can be evaluated quickly, the battery performance can be evaluated, and the use safety of the normally-working equipment of the battery can be guaranteed.

Example 3

Embodiment 3 of the present invention provides a terminal device, including: a memory and a processor; a memory for storing a computer program; a processor for executing the computer program stored in the memory to make the terminal device execute the method disclosed in embodiment 1 of the present invention.

Example 4

Embodiment 4 of the present invention provides a computer-readable storage medium including a program or instructions, where when the program or instructions are run on a computer, the method disclosed in embodiment 1 of the present invention is executed.

According to the parameter identification method and system of the battery equivalent circuit model disclosed by the embodiment of the invention, the respective state equations of the N-order RC equivalent circuit model and the N-1-order PNGV model are connected, the state equations are converted into the same differential equation, the model parameters of the N-order RC equivalent circuit model and the N-1-order PNGV model can be estimated by estimating the differential coefficient of the differential equation and the mathematical relationship between the differential coefficient and the respective model parameters of the N-order RC equivalent circuit model and the N-1-order PNGV model, and the effect of identifying the two model parameters of the N-order RC equivalent circuit model and the N-1-order PNGV model at the same time is realized.

The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.