CN111477981A - Lithium ion battery interval optimization charging method - Google Patents

Abstract

The invention relates to a lithium ion battery interval optimization charging method, which comprises the steps of establishing a capacity decline rate model and an energy consumption model by analyzing characteristic parameter change characteristics of a lithium ion battery in different multiplying powers and different SOC intervals, providing a multi-objective function taking the capacity decline rate and the energy consumption as penalty terms comprehensively, taking the capacity decline rate and the energy consumption as optimization targets, taking SOC as a state variable, and obtaining a group of optimized current sequences by utilizing an optimization algorithm to calculate under the constraint of average charging multiplying power, charging and discharging voltage, maximum charging multiplying power and total charging electric quantity; meanwhile, the control of multi-target optimized charging of the battery is realized based on a model predictive control theory, and an optimized current sequence is obtained by adopting a multi-step predictive control method. Compared with the traditional charging method, the invention reduces the energy consumption in the charging process, has the function of slowing down the decline of the battery capacity and prolongs the service life.

Description

Lithium ion battery interval optimization charging method

Technical Field

The invention belongs to the technical application field of power batteries, and mainly relates to a lithium ion battery interval optimization charging method.

Background

The existing charging methods include constant-current constant-voltage charging, current intermittent charging, step variable-current charging, alternating-current charging and the like. The constant-current and constant-voltage charging is most widely applied, but the constant-voltage stage consumes long time, and the charging efficiency is influenced by the polarization effect generated in the charging process. The current intermittent charging is divided into constant current intermittent charging and variable current intermittent charging, in the constant current intermittent charging process, the current amplitude is constant, the duty ratio is constant, the determination of the current amplitude and the determination of the duty ratio are not solved all the time, and the contradiction between the charging time and the charging polarization cannot be balanced; the current amplitude of the variable-current intermittent charging is not unique, the purpose of improving the charging efficiency can be achieved, but the control in the variable-current intermittent charging process is complex, the service life of the battery is often influenced if the current is too large, and the charging time is increased if the current is too small. The step variable current charging is the most widely researched charging method at present, the whole charging process is divided into a plurality of stages, and the optimized charging current sequence is obtained by using different intelligent algorithms.

Disclosure of Invention

In order to overcome the defects that the existing quick charging method has adverse effect on the service life of a battery and the loss of the charging process is large, the invention provides an optimized charging method based on a model integrating the capacity decline rate of the battery and the energy consumption. The technical scheme adopted by the invention for solving the technical problems is as follows: an optimized charging method based on a lithium ion battery capacity fading rate and an energy consumption model. The method mainly comprises two parts, namely model establishment and algorithm realization.

In order to achieve the above purposes, the technical scheme adopted by the invention is as follows:

a lithium ion battery partition optimizing charging method comprises the following steps:

s1 according to battery capacity decline rate Q_lossCarrying out differential processing on the relation with the equivalent cycle times x to obtain the relation between the battery capacity fading rate DS and x;

s2, according to the capacity fading rate model of the lithium ion battery under the full SOC circulation intervals with different charging multiplying powers and different aging degrees, establishing the battery capacity fading rate DS, the SOC interval and the charging multiplying power I_cAnd a relationship model between aging degrees;

s3, establishing a primary energy consumption model according to the first-order equivalent circuit model;

s4, establishing DC internal resistance function R of the battery by a DC internal resistance incremental method_internalSelecting a DC internal resistance curve R with the multiplying power of 8C_b[SOC,I_c]The method includes the steps that as a reference curve of the direct current internal resistance, the multiplying power is set to be an 8C reference curve, and direct current internal resistance curves under other multiplying powers are different from the reference value to obtain a direct current internal resistance increment curve;

s5, fitting the direct current internal resistance increment curve obtained in the S4 by utilizing a fifth-order function to obtain a relational expression of internal resistance increments under different SOCs and different multiplying powers, and obtaining a final energy consumption model;

s6, determining an index function and a constraint condition according to the capacity fading rate model obtained in the S2 and the final energy consumption model of the S5, then selecting a state variable, the number of charging stages and a decision variable, and calculating by using an optimization algorithm to obtain a group of optimized charging current sequences;

and S7, selecting battery SOC change as a prediction model based on a model prediction control theory, selecting current multiplying power in different SOC intervals as a constraint condition by taking energy consumption as an optimization target, selecting multi-step prediction with variable step length, and obtaining an optimized current sequence by adopting an optimized charging method of a dynamic programming algorithm.

On the basis of the above scheme, S1 specifically includes the following steps:

s11 rate of decline of battery capacity Q_lossApproximately presents a power exponential relation with the equivalent cycle number x, and is shown in formula (1):

Q_loss＝β×x^α(1)

the equivalent cycle times are calculated by dividing the actual cycle times by the interval number 5, and α and β are constant terms and exponential terms respectively;

s12 rate of decline of battery capacity Q_losDerivation to give formula (2):

DS＝d(Q_loss)/dx＝α×β×x^α-1(2)

where DS represents the battery capacity fade rate.

On the basis of the above scheme, S2 specifically includes the following steps:

s21 lithium ion battery charging rate I_cAnd a battery capacity fading rate model under the full SOC circulation intervals with different aging degrees, as shown in formula (3):

DS(I_c，Q_loss)＝a(Q_loss)×I_cb(Q_loss)+c(Q_loss) (3)

wherein a, b and c are parameters related to the aging degree of the battery;

s22, coupling the SOC cycle interval by the battery capacity decline rate model, as shown in formula (4):

DS(soc，I_c，Q_loss)＝ΔSOC_k×_g[DS(I_c，Q_loss)](4)

wherein, Δ SOC_kIndicating the amount of charge in the kth stage,

ΔSOC_kas shown in formula (5):

where Q is the battery capacity, η is the charge efficiency, Δ t_kIs the charging time of the kth stage, I_oIs the charging current of the battery, g [ DS (I)_c，Q_loss)]The relation between the battery capacity decline rate under the cycle condition of each SOC subarea and the whole subarea is indicated;

S23:g[DS(I_c，Q_loss)]the establishment process is as follows:

fitting with formula (1) to obtain corresponding α, β,

substituting α and β obtained in the step S1 into the formula (2) to obtain the change curves of the battery capacity fading rate along with the cycle times under different subsection SOC cycle intervals and full SOC cycle intervals;

the capacity fading rate in the segmented SOC circulation interval and the capacity fading rate in the full SOC circulation interval are approximately in a quadratic function relation as shown in the formula (6);

g[DS(I_c，Q_loss)]＝M+N×DS(I_c，Q_loss)+P×[DS(I_c，Q_loss)]²(6)

m, N and P are relation parameter values between the capacity fading rate of the segmented SOC circulation interval and the full SOC circulation interval;

arbitrary charge phase k, battery capacity fade rate L_DS(k) Different charging multiplying power I in different SOC circulation intervals_cAnd the calculation model under different aging degrees is shown as the formula (7):

in the formula (7), DS (I)_c，Q_loss) For full SOC cycle interval, different charging multiplying power I_cAnd rate of capacity fade (as a known amount) for different degrees of aging of the battery, L_DS(k) For different SOC cycle intervals and different charging multiplying power I_cAnd the rate of capacity fade for different degrees of battery aging.

On the basis of the above scheme, the preliminary energy consumption model establishing process described in S3 is as follows:

s31 energy consumption based on first-order equivalent circuit modelPhase performance indicator function L_Eloss(k) As shown in formula (8):

L_Eloss(k)＝{I_o ²(k)×R_o[SOC(k),I_c(k)]+I_p ²(k)×R_p[SOC(k),I_c(k)]}×△t_k(8)

wherein R is_o[SOC(k),I_c(k)]And R_p[SOC(k),I_c(k)]Respectively with different charging multiplying power I_cOhmic and polarization internal resistances, I, at the lower SOC points_oCharging current for the battery, I_pFor the current flowing through the polarized internal resistance, I_cAs the charge rate of the battery, Δ t_kCharging time of each kth stage;

s32, discovering the polarized capacitance C in the equivalent circuit model by analyzing the identified parameter values_pThe value is large, and the current I on the polarized internal resistance is considered in the constant current charging process_pApproximate charging current I of the battery_oEqual, so the phase performance indicator function of energy consumption is simplified to equation (9):

will ohm the internal resistance R_o[SOC(k),I_c(k)]And polarization internal resistance R_p[SOC(k),I_c(k)]The sum is defined as the DC internal resistance R_internalTherefore, equation (9) is simplified to equation (10), and a preliminary energy consumption model is obtained:

L_Eloss(k)＝I_o ²(k)×R_internal[SOC(k),I_c(k)]×△tk (10)。

on the basis of the above scheme, S4 specifically includes the following steps:

s41, according to the change relation of the direct current internal resistance along with the SOC interval under different charging multiplying factors: the direct current internal resistance is smaller along with the increase of the charging multiplying power, and a direct current internal resistance function R of the battery is established by using an internal resistance incremental method_internal[SOC(k),I_c(k)]And the DC internal resistance under 8C is used as a reference value R_b[SOC,I_c]，

The incremental internal resistance is shown in formula (11):

ΔR(SOC，I_c)＝R(SOC，I_c)-R_b(SOC，I_c) (11)

and (4) obtaining the direct current internal resistance increment curves under different charging multiplying powers through the formula (11).

On the basis of the above scheme, S5 specifically includes the following steps:

s51, fitting the internal resistance incremental curve by adopting a fifth-order function, wherein the equation is shown as the following formula (12):

ΔR(SOC，I_c)＝r(I_c)SOC⁵+s(I_c)SOC⁴+t(I_c)SOC³+u(I_c)SOC²+v(I_c)SOC+w(I_c)

(12)

wherein r, s, t, u, v and w are all equal to the charging multiplying power I_cA coefficient of correlation;

s52, in order to obtain the relation between the direct current internal resistance increment equation coefficient and the multiplying factor, 6 coefficients r, S, t, u, v and w in the formula (12) and the charging multiplying factor I_cFitting the relation curve between the two.

Because the relationship between the internal resistance increment coefficient and the multiplying power is in a quadratic function relationship, fitting is carried out according to the formula (13):

x(I_c)＝H×I_c ²+Y×I_c+Z (13)

wherein, x (I)_c) Is the relationship between the direct current internal resistance incremental equation coefficient and the charging multiplying power, and H, Y, Z is a fitting coefficient;

in order to verify the accuracy of the direct current internal resistance function model, the charging multiplying power 1C and the charging multiplying power 5C are used for verification to obtain a final energy consumption model of the lithium ion battery, and the final energy consumption model is shown as a formula (14):

L_Eloss(k)＝I_o ²(k)×{R_b[SOC(k)，I_c(k)]+ΔR[SOC(k)，I_c(k)]}×Δt_k(14)。

on the basis of the above scheme, S6 specifically includes the following steps:

s61, stage division:

selecting 6C as the average charging multiplying power in the optimized charging process, and comprehensively considering the charging and discharging process to be divided into 15 charging stages;

s62 selection of state variables:

selecting SOC as a state variable;

s63, determination of decision variables and column writing of state transition equations:

selecting charging multiplying power I_cAs decision variables:

the state transition equation is written according to the relation between the battery state variables SOC as shown in equation (15):

wherein Q is the battery capacity;

and S64, determining an index function and a constraint condition:

the indicator function comprehensively considers the capacity fading rate L_DS(k) And energy consumption L_Eloss(k) Objective function f as penalty term_k[SOC(k)]As shown in formula (16):

in order to obtain the optimal charging current sequence, a corresponding constraint condition needs to be added, wherein the constraint condition is shown as a formula (17):

s65 programming

As the dynamic programming algorithm follows the algorithm principle of inverse solution, the optimal index function f (k) of the stage is determined according to the formula (18), so that the decision current I of the stage is determined_L(k)；

min_f(k+1)+L(k)<f(k)

f(k)＝min_f(k+1)+L(k) (18)

And designing an optimized charging method based on a dynamic programming algorithm to obtain an optimized charging current sequence.

On the basis of the above scheme, S7 specifically includes the following steps:

s71 selection of prediction model

Selecting a first-order equivalent circuit model as a controlled object, selecting a battery SOC change model as a prediction model, and expressing the following formula (19):

wherein, T_sA charging time for each charging phase;

s72 determination of the phase index function:

only energy consumption is taken as an optimization target, wherein the stage index function is shown as a formula (20):

wherein the value of the DC internal resistance R is a function related to the SOC and the charging current value I_kIs the charging current for the kth charging phase.

S73 determination of constraint conditions

Related constraint conditions are added to the charging rate in different SOC intervals, and the constraint conditions are shown as a formula (21):

s74 selection of prediction step size

The prediction step size is shown in equation (22):

s75 selection of an optimization algorithm

And selecting a dynamic programming algorithm as an optimization algorithm.

The invention has the following beneficial effects:

the invention has the advantages that the cycle service life of the battery is obviously prolonged under the condition of optimizing the charging method, and the energy consumption in the charging process is lower than that of the traditional charging method.

Drawings

The invention has the following drawings:

FIG. 1 is a graph of capacity fade rate with cycle number for different SOC cycle intervals

FIG. 2 fitting graph of capacity fade rate

FIG. 3 is a graph of capacity fade rates for different SOC cycle intervals

FIG. 4 is a graph of the relationship between the rate of capacity fade between partial regions and full regions

FIG. 5 is a DC internal resistance variation graph

FIG. 6 is a graph showing incremental changes in internal resistance

FIG. 7 is a graph showing the increase coefficient of internal resistance with the change of magnification

FIG. 8 is a comparison graph of DC internal resistance experiment and calculated value

FIG. 9 dynamic programming flow chart

FIG. 10 optimized Current-Voltage sequence diagram

FIG. 11 is a control flow diagram based on MPC charging method

FIG. 12 optimized Current sequence diagram

FIG. 13 energy consumption vs. graph

FIG. 14 graph of capacity fade rate versus cycle number

Detailed Description

The technical solution in the embodiment of the present invention will be clearly and completely described below with reference to fig. 1 to 14 of the embodiment of the present invention.

A lithium ion battery partition optimizing charging method comprises the following steps:

s1 according to the battery capacity decline rate Q_lossAnd (3) carrying out differential processing on the relation with the equivalent cycle times x to obtain the relation between the battery capacity fading rate DS and x:

s1 specifically includes the following steps:

s11 rate of decline of battery capacity Q_lossThe equivalent cycle times x are approximately in a power exponential relationship, as shown in formula (1)：

Q_loss＝β×x^α(1)

The equivalent cycle times are calculated by dividing the actual cycle times by the interval number 5, and α and β are constant terms and exponential terms respectively;

s12 rate of decline of battery capacity Q_losDerivation to give formula (2):

DS＝d(Q_loss)/dx＝α×β×x^α-1(2)

where DS represents the battery capacity fade rate.

S2: according to a battery capacity decline rate model of the lithium ion battery under full SOC circulation intervals with different charging multiplying powers and different aging degrees, a battery capacity decline rate DS, an SOC interval and a charging multiplying power I are established_cAnd a relationship model between aging degrees;

s2 specifically includes the following steps:

s21 lithium ion battery charging rate I_cAnd a battery capacity fading rate model under the full SOC circulation intervals with different aging degrees, as shown in formula (3):

DS(I_c,Q_loss)＝a(Q_loss)×I_cb(Q_loss)+c(Q_loss) (3)

wherein a, b and c are parameters related to the aging degree of the battery;

s22, coupling the SOC cycle interval by the battery capacity decline rate model, as shown in formula (4):

DS(soc,I_c,Q_loss)＝ΔSOC_k×g[DS(I_c,Q_loss)](4)

wherein, Δ SOC_kIndicating the amount of charge in the kth stage,

ΔSOC_kas shown in formula (5):

wherein Q is the battery capacity, η is the charging efficiency, △ t_kIs the charging time of the kth stage, I_oBeing batteriesCharging current, g [ DS (I)_c,Q_loss)]The relation between the battery capacity decline rate under the cycle condition of each SOC subarea and the whole subarea is indicated;

S23:g[DS(I_c,Q_loss)]the establishment process is as follows:

fitting with formula (1) to obtain corresponding α, β,

α and β obtained in S1 are substituted into formula (2), and a change curve of the battery capacity fading rate with the cycle number in each of the different segment SOC cycle intervals and the full SOC cycle interval is obtained, as shown in fig. 3.

The relation between the battery capacity fading rate in the segmented SOC cycle section and the full-section capacity fading rate can be obtained as shown in fig. 4 by taking the capacity fading rate in the full-section SOC cycle section as the abscissa.

The capacity fading rate in the segmented SOC circulation interval and the capacity fading rate in the full SOC circulation interval are approximately in a quadratic function relation as shown in the formula (6), and the parameter value of M, N, P is obtained through the fitting of the graph 4;

g[DS(I_c,Q_loss)]＝M+N×DS(I_c,Q_loss)+P×[DS(I_c,Q_loss)]²(6)

wherein M, N and P are relation parameter values between the capacity fading rate of the segmented SOC circulation interval and the full SOC circulation interval.

In summary, the battery capacity degradation rate L for any charging phase k_DS(k) At different SOC cycle intervals and different charging multiplying power I_cAnd the calculation model under different aging degrees is shown as the formula (7):

in the formula (7), DS (I)_c，Q_loss) For full SOC cycle interval, different charging multiplying power I_cAnd rate of capacity fade (as a known amount) for different degrees of aging of the battery, L_DS(k) For different SOC cycle intervals and different charging multiplying power I_cAnd the rate of capacity fade for different degrees of battery aging.

S3, establishing a preliminary energy consumption model according to the first-order equivalent circuit model,

the preliminary energy consumption model establishing process of S3 is as follows:

s31 phase performance index function L of energy consumption based on first-order equivalent circuit model_Eloss(k) As shown in formula (8):

L_Eloss(k)＝{I_o ²(k)×R_o[SOC(k),I_c(k)]+I_p ²(k)×R_p[SOC(k),I_c(k)]}×△t_k(8)

wherein R is_o[SOC(k),I_c(k)]And R_p[SOC(k),I_c(k)]Respectively with different charging multiplying power I_cOhmic and polarization internal resistances, I, at the lower SOC points_oCharging current for the battery, I_pFor the current flowing through the polarized internal resistance, I_cAs the charge rate of the battery, Δ t_kCharging time of each kth stage;

s32, the polarized capacitor C in the equivalent circuit model can be found by analyzing the identified parameter values_pThe value is large, and the current I on the polarized internal resistance can be considered in the constant current charging process_pApproximate charging current I of the battery_oEqual, the phase performance indicator function of energy consumption can be simplified to equation (9):

will ohm the internal resistance R_o[SOC(k),I_c(k)]And polarization internal resistance R_p[SOC(k),I_c(k)]The sum is defined as the DC internal resistance R_internalTherefore, the formula (9) is simplified to the formula (10), and a preliminary energy consumption model is obtained. :

L_Eloss(k)＝I_o ²(k)×R_internal[SOC(k),I_c(k)]×△t_k(10)。

s4, establishing DC internal resistance function R of the battery by a DC internal resistance incremental method_internalSelecting a DC internal resistance curve R with a charging multiplying power of 8C_b[SOC,I_c]As a reference curve of DC internal resistance, other charging timesAnd (3) making a difference value between the direct current internal resistance curve and the reference value under the rate to obtain a direct current internal resistance increment curve:

s4 specifically includes the following steps:

and S41, according to the change relationship of the direct current internal resistance along with the SOC interval under different charging multiplying factors, the direct current internal resistance is shown in the figure 5. As the charging rate increases, the dc internal resistance decreases. Therefore, the direct current internal resistance function R of the battery is established by using the internal resistance increment method_internal[SOC(k),I_c(k)]And the DC internal resistance under 8C is used as a reference value R_b[SOC,I_c]。

The incremental internal resistance is shown in formula (11):

△R(SOC,I_c)＝R(SOC,I_c)-R_b(SOC,I_c) (11)

the direct current internal resistance increment curves at different charging rates are obtained by equation (11), as shown in fig. 6.

And S5, fitting the direct current internal resistance increment curve obtained in the step S4 by utilizing a fifth-order function to obtain a relational expression of the internal resistance increment under different SOC and different charging multiplying factors, and obtaining a final energy consumption model.

S5 specifically includes the following steps:

s51, fitting the internal resistance incremental curve by adopting a fifth-order function, wherein the equation is shown as the following formula (12):

ΔR(SOC，I_c)＝r(I_c)SOC⁵+s(I_c)SOC⁴+t(I_c)SOC³+u(I_c)SOC²+v(I_c)SOC+w(I_c)

(12)

wherein r, s, t, u, v and w are all equal to the charging multiplying power I_cThe coefficient of correlation.

S52, in order to obtain the relation between the direct current internal resistance increment equation coefficient and the multiplying factor, 6 coefficients r, S, t, u, v and w in the formula (12) and the charging multiplying factor I_cThe relationship between them was fitted as shown in fig. 7.

Because the relationship between the internal resistance increment coefficient and the multiplying power is in a quadratic function relationship, fitting is carried out according to the formula (13):

x(I_c)＝H×I_c ²+Y×I_c+Z (13)

wherein, x (I)_c) Is the relation between the direct current internal resistance increment equation coefficient and the charging multiplying factor, and H, Y, Z is a fitting coefficient.

In order to verify the accuracy of the dc internal resistance function model, the charging magnifications 1C and 5C are used for verification, as shown in fig. 8.

Obtaining a final energy consumption model of the lithium ion battery, wherein the final energy consumption model is shown as a formula (14):

L_Eloss(k)＝I_o ²(k)×{R_b[SOC(k),I_c(k)]+△R[SOC(k),I_c(k)]}×△t_k(14)。

and S6, determining an index function and a constraint condition according to the battery capacity fading rate model of S2 and the final energy consumption model of S5, then selecting a state variable, the number of charging stages and a decision variable, and obtaining a group of optimized current sequences by using an optimized charging method of a dynamic programming algorithm:

s6 specifically includes the following steps:

s61, stage division:

and 6C is selected as the average charging rate in the optimized charging process, and the charging and discharging process is divided into 15 charging stages by comprehensive consideration.

S62 selection of state variables:

the SOC is selected as a state variable.

S63, determination of decision variables and column writing of state transition equations:

selecting charging multiplying power I_cAs a decision variable.

The state transition equation is written according to the relation between the battery state variables SOC as shown in equation (15):

wherein Q is the battery capacity;

and S64, determining an index function and a constraint condition:

the index function comprehensively considers the capacityVolume decay Rate L_DS(k) And energy consumption L_Eloss(k) Objective function f as penalty term_k[SOC(k)]As shown in formula (16):

in order to obtain the optimal charging current sequence, a corresponding constraint condition needs to be added, wherein the constraint condition is shown as a formula (17):

s65 programming

The dynamic programming algorithm flow is obtained from fig. 9.

As the dynamic programming algorithm follows the algorithm principle of inverse solution, the initial value of the input stage number k is 15, and the optimal index function f (k) of the stage is determined according to the formula (18), so that the decision current I of the stage is determined_L(k)。

min_f(k+1)+L(k)<f(k)

f(k)＝min_f(k+1)+L(k) (18)

The optimal charging current sequence obtained by the design of the optimal charging method based on the dynamic programming algorithm with the balance coefficient M of 100 is shown in FIG. 10.

S7, based on the model prediction control theory, selecting the battery SOC variation as a prediction model, taking energy consumption as an optimization target, taking current multiplying power in different SOC intervals as a constraint condition, selecting multi-step prediction with variable step length, and obtaining an optimized current sequence by adopting a dynamic programming algorithm:

s7 specifically includes the following steps:

s71 selection of prediction model

Selecting a first-order equivalent circuit model as a controlled object, selecting a battery SOC change model as a prediction model, and expressing the following formula (19):

wherein, T_sA charging time for each charging phase;

s72 determination of phase index function

Only energy consumption is taken as an optimization target, wherein the stage index function is shown as a formula (20):

wherein the value of the DC internal resistance R is a function related to the SOC and the charging current value I_kIs the charging current for the kth charging phase.

S73 determination of constraint conditions

Related constraint conditions are added to the charging rate in different SOC intervals, and the constraint conditions are shown as a formula (21):

s74 selection of prediction step size

The first 10 charging stages adopt a prediction mode with a fixed step length of 5 steps, only the first element in a control sequence is used as the charging current of the stage in each optimization stage, and the last 5 charging stages adopt a multi-step prediction mode with a variable step length, wherein the prediction step length is shown as a formula (22):

s75 selection of an optimization algorithm

The design of the model predictive control process is basically completed, the last step is optimization algorithm selection, a dynamic programming algorithm is selected as an optimization algorithm, and the specific implementation process is shown in fig. 11.

Fig. 12 shows a current sequence obtained in each charging stage by using a model predictive control algorithm based on the simulation model under the condition that the average charging rate is 6C. Energy consumption change comparison curve continuous charging stage energy consumption accumulation under conventional charging method and optimized charging method with same average multiplying powerEvaluation of E_lossThe comparison curve of (a) is shown in fig. 13.

Under different cycle times of the lithium ion battery, by using a constant-current constant-voltage charge-discharge capacity test experiment under a 1C multiplying power, the relationship between the change rate of the battery capacity with the cycle times under the optimized charging method and the traditional charging method is respectively obtained and is shown in FIG. 14.

It should be understood that the above-mentioned embodiments of the present invention are only examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention, and it will be obvious to those skilled in the art that other variations and modifications can be made on the basis of the above description, and all embodiments cannot be exhaustive, and obvious variations and modifications of the present invention are included in the protection scope of the present invention.

Those not described in detail in this specification are within the skill of the art.

Claims (8)

1. A lithium ion battery partition optimization charging method is characterized by comprising the following steps:

s1 according to battery capacity decline rate Q_lossCarrying out differential processing on the relation with the equivalent cycle times x to obtain the relation between the battery capacity fading rate DS and x;

s2, according to the capacity fading rate model of the lithium ion battery under the full SOC circulation intervals with different charging multiplying powers and different aging degrees, establishing the battery capacity fading rate DS, the SOC interval and the charging multiplying power I_cAnd a relationship model between aging degrees;

s3, establishing a primary energy consumption model according to the first-order equivalent circuit model;

s4, establishing DC internal resistance function R of the battery by a DC internal resistance incremental method_internalSelecting a DC internal resistance curve R with the multiplying power of 8C_b[SOC,I_c]Taking the direct current internal resistance curve as a reference curve of the direct current internal resistance, and making a difference value between the direct current internal resistance curve under other multiplying powers and the reference value to obtain a direct current internal resistance incremental curve;

s5, fitting the direct current internal resistance increment curve obtained in the S4 by utilizing a fifth-order function to obtain a relational expression of internal resistance increments under different SOCs and different multiplying powers, and obtaining a final energy consumption model;

s6, determining an index function and a constraint condition according to the capacity fading rate model obtained in the S2 and the final energy consumption model of the S5, then selecting a state variable, the number of charging stages and a decision variable, and calculating by using an optimization algorithm to obtain a group of optimized charging current sequences;

and S7, selecting battery SOC change as a prediction model based on a model prediction control theory, selecting current multiplying power in different SOC intervals as a constraint condition by taking energy consumption as an optimization target, selecting multi-step prediction with variable step length, and obtaining an optimized current sequence by adopting an optimized charging method of a dynamic programming algorithm.

2. The method for optimizing charging of lithium ion batteries according to claim 1, wherein S1 specifically includes the following steps:

s11 rate of decline of battery capacity Q_lossApproximately presents a power exponential relation with the equivalent cycle number x, and is shown in formula (1):

Q_loss＝β×x^α(1)

the equivalent cycle times are calculated by dividing the actual cycle times by the interval number 5, and α and β are constant terms and exponential terms respectively;

s12 rate of decline of battery capacity Q_losDerivation to give formula (2):

DS＝d(Q_loss)/dx＝α×β×x^α-1(2)

where DS represents the battery capacity fade rate.

3. The method for optimizing charging of lithium ion batteries according to claim 2, wherein S2 specifically includes the following steps:

s21 lithium ion battery charging rate I_cAnd a battery capacity fading rate model under the full SOC circulation intervals with different aging degrees, as shown in formula (3):

DS(I_c,Q_loss)＝a(Q_loss)×I_cb(Q_loss)+c(Q_loss) (3)

wherein a, b and c are parameters related to the aging degree of the battery;

s22, coupling the SOC cycle interval by the battery capacity decline rate model, as shown in formula (4):

DS(soc,I_c,Q_loss)＝ΔSOC_k×g[DS(I_c,Q_loss)](4)

wherein, Δ SOC_kIndicating the amount of charge in the kth stage,

ΔSOC_kas shown in formula (5):

wherein Q is the battery capacity, η is the charging efficiency, △ t_kIs the charging time of the kth stage, I_oIs the charging current of the battery, g [ DS (I)_c,Q_loss)]The relation between the battery capacity decline rate under the cycle condition of each SOC subarea and the whole subarea is indicated;

S23:g[DS(I_c,Q_loss)]the establishment process is as follows:

fitting with formula (1) to obtain corresponding α, β,

substituting α and β obtained in the step S1 into the formula (2) to obtain the change curves of the battery capacity fading rate along with the cycle times under different subsection SOC cycle intervals and full SOC cycle intervals;

the capacity fading rate in the segmented SOC circulation interval and the capacity fading rate in the full SOC circulation interval are approximately in a quadratic function relation as shown in the formula (6);

g[DS(I_c，Q_loss)]＝M+N×DS(I_c，Q_loss)+P×[DS(I_c，Q_loss)]²(6)

m, N and P are relation parameter values between the capacity fading rate of the segmented SOC circulation interval and the full SOC circulation interval;

arbitrary charge phase k, battery capacity fade rate L_DS(k) Different charging multiplying power I in different SOC circulation intervals_cAnd the calculation model under different aging degrees is shown as the formula (7):

in the formula (7), DS (I)_c，Q_loss) For full SOC cycle interval, different charging multiplying power I_cAnd rate of capacity fade for different degrees of battery aging, L_DS(k) For different SOC cycle intervals and different charging multiplying power I_cAnd the rate of capacity fade for different degrees of battery aging.

4. The method according to claim 3, wherein the preliminary energy consumption model building process of S3 is as follows:

s31 phase performance index function L of energy consumption based on first-order equivalent circuit model_Eloss(k) As shown in formula (8):

L_Eloss(k)＝{I_o ²(k)×R_o[SOC(k),I_c(k)]+I_p ²(k)×R_p[SOC(k),I_c(k)]}×△t_k(8)

wherein R is_o[SOC(k),I_c(k)]And R_p[SOC(k),I_c(k)]Respectively with different charging multiplying power I_cOhmic and polarization internal resistances, I, at the lower SOC points_oCharging current for the battery, I_pFor the current flowing through the polarized internal resistance, I_cAs the charge rate of the battery, Δ t_kCharging time of each kth stage;

s32, discovering the polarized capacitance C in the equivalent circuit model by analyzing the identified parameter values_pThe value is large, and the current I on the polarized internal resistance is considered in the constant current charging process_pApproximate charging current I of the battery_oEqual, so the phase performance indicator function of energy consumption is simplified to equation (9):

will ohm the internal resistance R_o[SOC(k),I_c(k)]And polarization internal resistance R_p[SOC(k),I_c(k)]The sum is defined as the DC internal resistance R_internalTherefore, equation (9) is simplified to equation (10), and a preliminary energy consumption model is obtained:

L_Eloss(k)＝I_o ²(k)×R_internal[SOC(k),I_c(k)]×△t_k(10)。

5. the method for optimizing charging of lithium ion batteries according to claim 4, wherein S4 specifically comprises the following steps:

s41, according to the change relation of the direct current internal resistance along with the SOC interval under different charging multiplying factors: the direct current internal resistance is smaller along with the increase of the charging multiplying power, and a direct current internal resistance function R of the battery is established by using an internal resistance incremental method_internal[SOC(k)，I_c(k)]And the DC internal resistance under 8C is used as a reference value R_b[SOC，I_c]，

The incremental internal resistance is shown in formula (11):

ΔR(SOC，I_c)＝R(SOC，I_c)-R_b(SOC，I_c) And (11) obtaining the direct current internal resistance increment curves under different charging multiplying powers through the formula (11).

6. The method for optimizing charging of lithium ion batteries according to claim 5, wherein S5 specifically comprises the following steps:

s51, fitting the internal resistance incremental curve by adopting a fifth-order function, wherein the equation is shown as the following formula (12):

ΔR(SOC，I_c)＝r(I_c)SOC⁵+s(I_c)SOC⁴+t(I_c)SOC³+u(I_c)SOC²+v(I_c)SOC+w(I_c)

(12)

wherein r, s, t, u, v and w are all equal to the charging multiplying power I_cA coefficient of correlation;

s52, 6 coefficients r, S and t in the equation (12) are paired to obtain the relation between the direct current internal resistance incremental equation coefficient and the multiplying powerU, v and w and charging rate I_cFitting the relation curve;

because the relationship between the internal resistance increment coefficient and the multiplying power is in a quadratic function relationship, fitting is carried out according to the formula (13):

x(I_c)＝H×I_c ²+Y×I_c+Z (13)

wherein, x (I)_c) Is the relationship between the direct current internal resistance incremental equation coefficient and the charging multiplying power, and H, Y, Z is a fitting coefficient;

in order to verify the accuracy of the direct current internal resistance function model, the charging multiplying power 1C and the charging multiplying power 5C are used for verification to obtain a final energy consumption model of the lithium ion battery, and the final energy consumption model is shown as a formula (14):

L_Eloss(k)＝I_o ²(k)×{R_b[SOC(k)，I_c(k)]+ΔR[SOC(k)，I_c(k)]}×Δt_k(14)。

7. the method for optimizing charging of lithium ion batteries according to claim 6, wherein the step S6 specifically includes the following steps:

s61, stage division:

selecting 6C as the average charging multiplying power in the optimized charging process, and comprehensively considering the charging and discharging process to be divided into 15 charging stages;

s62 selection of state variables:

selecting SOC as a state variable;

s63, determination of decision variables and column writing of state transition equations:

selecting charging multiplying power I_cAs decision variables:

the state transition equation is written according to the relation between the battery state variables SOC as shown in equation (15):

wherein Q is the battery capacity;

and S64, determining an index function and a constraint condition:

the indicator function comprehensively considers the capacity fading rate L_DS(k) And energy consumption L_Eloss(k) Objective function f as penalty term_k[SOC(k)]As shown in formula (16):

in order to obtain the optimal charging current sequence, a corresponding constraint condition needs to be added, wherein the constraint condition is shown as a formula (17):

s65 programming

As the dynamic programming algorithm follows the algorithm principle of inverse solution, the optimal index function f (k) of the stage is determined according to the formula (18), so that the decision current I of the stage is determined_L(k)；

min_f(k+1)+L(k)<f(k)

f(k)＝min_f(k+1)+L(k) (18)

And designing an optimized charging method based on a dynamic programming algorithm to obtain an optimized charging current sequence.

8. The method for optimizing charging of lithium ion batteries according to claim 7, wherein S7 specifically includes the following steps: s71 selection of prediction model:

selecting a first-order equivalent circuit model as a controlled object, selecting a battery SOC change model as a prediction model, and expressing the following formula (19):

wherein, T_sA charging time for each charging phase;

s72 determination of the phase index function:

only energy consumption is taken as an optimization target, wherein the stage index function is shown as a formula (20):

wherein the value of the DC internal resistance R is a function related to the SOC and the charging current value I_kIs the charging current of the kth charging phase;

s73 determination of constraint conditions

Related constraint conditions are added to the charging rate in different SOC intervals, and the constraint conditions are shown as a formula (21):

s74 selection of prediction step size

The prediction step size is shown in equation (22):

s75 selection of an optimization algorithm

And selecting a dynamic programming algorithm as an optimization algorithm.

Images