A battery equivalent circuit model parameter self-learning method and system

Abstract

The present invention discloses a battery equivalent circuit model parameter self-learning method, comprising the following steps: S1, taking a lithium iron phosphate battery, performing charge and discharge tests on the battery, and obtaining test data at time intervals of dt seconds, wherein the test data includes a battery terminal voltage measurement value, a current, and a battery surface temperature; S2, based on a third-order RC equivalent circuit model, according to the least squares principle, performing offline identification of battery parameters based on the test data, and obtaining the battery's ohmic internal resistance, polarization internal resistance, and time constant; S3, performing self-learning on the battery's ohmic internal resistance; the present application performs offline identification of the battery's internal resistance, and online compensation of the battery's internal resistance according to the third-order RC equivalent circuit model during the battery's operation, thereby completing battery internal resistance self-learning for SOC, SOP, and SOH state estimation, thereby accurately identifying the parameters.

Description

Technical Field

[0001] The present invention belongs to the technical field of battery equivalent circuit models, and in particular relates to a battery parameter self-learning method and system. Background Art

[0002] In battery management systems, the equivalent circuit model of the battery is often used for offline parameter identification, and the identified parameters are used to estimate the battery operating status. The model used for offline battery parameter identification is as follows: Figure 1 Since the battery internal resistance is difficult to identify online, and the battery internal resistance will change due to factors such as current, temperature, and SOH, the battery internal resistance estimation is inaccurate during the operation process, and the SOC, SOP, and SOH state estimation have large deviations, which creates battery safety hazards. Summary of the Invention

[0003] The purpose of the present invention is to address the problems of the existing technology. This application proposes a battery equivalent circuit model parameter self-learning method and system. By offline identifying the battery internal resistance, the battery internal resistance is compensated online according to the third-order RC equivalent circuit model during the battery operation process, and the battery internal resistance self-learning is completed for SOC, SOP, and SOH state estimation, and the battery internal resistance parameters are accurately identified.

[0004] The technical solution of this application is as follows:

[0005] A battery equivalent circuit model parameter self-learning method includes the following steps:

[0006] S1, take a lithium iron phosphate battery, perform charge and discharge tests on the battery, and obtain test data at time intervals of dt seconds. The test data includes the battery terminal voltage measurement value, current and battery surface temperature;

[0007] S2, based on the third-order RC equivalent circuit model and the principle of least squares method, the battery parameters are identified offline based on the test data of step S1 to obtain the battery's ohmic internal resistance, polarization internal resistance and time constant;

[0008] S3, self-learning of the battery's ohmic internal resistance.

[0009] Step S1 specifically includes the following steps:

[0010] S11, discharge the fully charged battery based on the adjusted current, collect the battery terminal voltage measurement value, battery surface temperature and current at time intervals of dt seconds, and record the battery terminal voltage measurement value U1(k) and the battery surface temperature T(k) and current I(k) every time the battery SOC (state of charge) decreases by a factor of s until the battery SOC = 0; U1(k) represents the battery terminal voltage measurement value at the kth sampling, T(k) represents the battery surface temperature at the kth sampling, I(k) represents the current at the kth sampling; OCV1(k) represents the battery open circuit voltage measurement value at the kth sampling; SOC(k) represents the battery SOC value at the kth sampling;

[0011] S12, charging the battery with SOC=0 based on the adjusted current, synchronously sampling at time intervals of dt seconds, recording the terminal voltage value U1(j) of the battery, and recording the battery surface temperature T(j) and current I(j) every time the battery SOC (state of charge) increases by s times, until the battery SOC=100%; U1(j) represents the terminal voltage value of the battery at the kth sampling, T(j) represents the battery surface temperature at the kth sampling, I(j) represents the current at the jth sampling; OCV1(j) represents the battery open circuit voltage measurement value at the jth sampling; SOC(j) represents the battery SOC value at the jth sampling;

[0012] k=1,2……n; n represents the number of sampling times from when the battery is fully charged to when SOC=0;

[0013] j=n+1,n+2…n+m; m represents the number of sampling times from SOC=0 to full charge of the battery;

[0014] Let i = 1, 2...n+m; OCV1(i) represents the battery open circuit voltage measurement value of the i-th sampling during the test data acquisition process, U1(i) represents the battery terminal voltage measurement value of the i-th sampling; I(i) represents the battery current measurement value of the i-th sampling; SOC(i) represents the battery SOC at the i-th sampling.

[0015] S2 specifically includes the following steps:

[0016] S21, calculate the battery ohmic internal resistance R0(i) based on formula (1):

[0017] R0(i)=[U1(i)-U1(i+1)] / [I(i)-I(i+1)](1)U1(i) represents the battery terminal voltage measurement value of the i-th sampling, U1(i+1) represents the battery terminal voltage measurement value of the i+1-th sampling, I(i) represents the battery current measurement value of the i-th sampling, I(i+1) represents the battery current measurement value of the i+1-th sampling, and R0(i) represents the ohmic internal resistance corresponding to SOC(i);

[0018] S22, identify the time constant and polarization internal resistance.

[0019] The identification of the time constant and polarization internal resistance specifically includes the following steps:

[0020] S2201, according to the characteristics of different batteries, set the range of the polarization internal resistance and time constant of the battery. When identifying the polarization internal resistance and time constant of the battery within the setting range, set the polarization internal resistance range [lb1, ub1] and the time constant value range [lb2, ub2];

[0021] S2202, setting the initial value of the polarization internal resistance and the initial value of the time constant of the battery. The design value of the initial value of the polarization internal resistance is between the lower boundary lb1 of the polarization internal resistance and the upper boundary ub1 of the polarization internal resistance. The setting value of the initial value of the time constant is between the lower boundary lb2 of the time constant and the upper boundary ub2 of the time constant.

[0022] S2203, the terminal voltage measurement value of the battery sampled for the i-th time is U1(i), and the terminal voltage calculated value of the battery calculated based on the test data is U2′(i). The initial value of the polarization internal resistance and the initial value of the time constant of the battery are substituted into formulas (2) to (7) to calculate the terminal voltage calculated value U2′(i) of the battery based on the test data.

[0023] U2'(i)=OCV(i)+Vp(i)+Vr(i)(2)

[0024] Vr(i)=I(i)×R0(i)(3)

[0025] Vp(i)=Vp1(i)+Vp2(i)+Vp3(i)(4)

[0026]

[0027] U2'(i) is the calculated value of the battery terminal voltage at the i-th sampling, OCV1(i) is the measured value of the battery open circuit voltage at the i-th sampling, Vp(i) is the polarization voltage of the battery at the i-th sampling, and the polarization voltage includes the first-order polarization voltage Vp1(i), the second-order polarization voltage Vp2(i), and the third-order polarization voltage Vp3(i); Vr(i) is the ohmic voltage of the battery, I(i) is the current of the battery, and R0(i) is the ohmic internal resistance of the battery ; The time constants of the battery model are divided into the first-order time constant τ1, the second-order time constant τ2 and the third-order time constant τ3; the polarization internal resistance of the battery includes the first-order polarization internal resistance Rp1(i), the second-order polarization internal resistance Rp2(i) and the third-order polarization internal resistance Rp3(i); dt is the battery data sampling time interval, Vp1(0), Vp2(0) and Vp3(0) are the initial values ​​of the first-order polarization voltage, the second-order polarization voltage and the third-order polarization voltage respectively;

[0028] S2204, setting the objective function Fcn;

[0029] Fcn=U1(i)-U2'(i) (8) U1(i) is the measured value of the battery terminal voltage at the i-th sampling, and U2'(i) is the calculated value of the battery terminal voltage obtained based on the test data of the i-th sampling;

[0030] S2205, using the least squares method, the objective function Fcn is minimized; when the objective function Fcn takes the minimum value, the polarization internal resistance and time constant of the corresponding battery are the parameters (polarization internal resistance and time constant) for offline identification.

[0031]

[0032] I(i) is the current value of the battery sampled for the i-th time, Rp is the polarization internal resistance of the battery, and the value of F(τ,Rp,I(i)) is equal to U2'(i), which represents the calculated value of the battery terminal voltage under the given time constant τ, polarization internal resistance Rp and measurement current U1(i).

[0033] By iterating the battery polarization internal resistance and time constant, the solution value of Fcn is updated until the solution value of Fcn reaches the minimum value, and the identification of the polarization internal resistance Rp and time constant τ is completed;

[0034] Step S3 specifically includes the following steps:

[0035] S31, the battery ohmic internal resistance R0(i) performs the first self-learning step, and the first self-learning output value is recorded as R0_SL1;

[0036] Set the upper and lower limits of the SOC interval USOC and LSOC, and the SOC(i) interval corresponding to the self-learning of the battery ohmic internal resistance R0(i) is [LSOC(i), USOC(i)]. Within the SOC(i) interval [LSOC(i), USOC(i)], the battery ohmic internal resistance corresponding to SOC(i) is R0(i); calculate the first self-learning output value of the battery ohmic internal resistance (average ohmic internal resistance value) R0_SL1(i) based on formula (11); when the error between the self-learning average ohmic internal resistance R0_SL1(i+1) and the previous self-learning average ohmic internal resistance R0_SL1(i) is less than or equal to Tol, the first self-learning calculation of the battery ohmic internal resistance is completed, and the last self-learning value of the battery is taken as the first self-learning output value of the battery ohmic internal resistance R0_SL1:

[0037]

[0038] When the condition |R0_SL1(i+1)-R0_SL1(i)|≤Tol is satisfied, R0_SL1=R0_SL1(i+1);

[0039] Tol is configurable. The value of Tol is adjusted according to different batteries and is in milliohms. SOC(i) is any value between 0% and 100%. b represents the number of R0(i) corresponding to SOC(i) in the interval [LSOC(i), USOC(i)]. SOC(i)∈[LSOC(i), USOC(i)];

[0040] S32, performing a second self-learning step based on the first self-learning output value R0_SL1 of the battery ohmic internal resistance, where the second self-learning output value is R0_SL2;

[0041] S33, performing a third self-learning step based on the second self-learning output value R0_SL2, and the third self-learning step output value is R0_SL3.

[0042] S32 specifically includes the following steps:

[0043] S3201, after one hour of open circuit, the battery is charged and discharged at a constant current. During the real-time operation of the battery, the terminal voltage of the battery is collected at intervals of dt seconds and recorded as U0(c). U0(c) is recorded in sequence as U0(1), U0(2), ...U0(c), U0(c+1), U0(c+2) ...U0(A). The collected current is recorded as I0(1), I0(2), ...I0(c), I0(c+1), I0(c+2) ...I0(A). The collected temperature is recorded as T 0(1), T0(2), ... T0(c), T0(c+1), T0(c+2) ... T0(A); U0(c), I0(c), T0(c) are the battery terminal voltage, current and battery surface temperature collected for the cth time when the voltage enters the constant current charge and discharge condition from the open circuit state for one hour, SOC(c) is the SOC value collected for the cth time when the voltage enters the constant current charge and discharge condition from the open circuit state for one hour; the real-time dynamic ohmic internal resistance R0 of the battery is calculated based on formula (12) d :When the battery SOC (c) ∈ ​​[LSOC, USOC] and the voltage enters the constant current charge and discharge condition from the open circuit state for one hour, calculate the battery dynamic ohmic internal resistance R0 d :

[0044]

[0045] In order to avoid the problems of asynchronous battery current and voltage acquisition, zero drift in current sampling, and inaccurate battery terminal voltage and current measurement values ​​due to the incomplete disappearance of the battery polarization voltage, the average value of the voltage matrix [U0(1), U0(2), … U0(c-1)] is taken as the open circuit voltage before the battery enters the constant current charge and discharge condition, and the average value of the current matrix [I0(c), I0(c+1), I0(c+2)] is taken as the current after the battery enters the constant current charge and discharge condition.

[0046] S3202, based on the battery capacity, calculate the ohmic internal resistance R0_in after being corrected for the rated capacity, formula (13):

[0047] R0_in=CN0 / CN×R0_SL1(13)CN0 is the rated capacity of the test battery a, CN is the rated capacity of a certain type of battery a; R0_in is the ohmic internal resistance corrected by the rated capacity;

[0048] S3203, calculating the battery ohmic internal resistance second self-learning output value R0_SL2;

[0049] R0_SL2=R0_in±|R0 d -R0_in|(14)

[0050] Based on |R0 d -R0_in| self-learning calculation R0_SL2 value, the calculation formula is formula (14). d -R0_in|, because |R0 d -R0_in| is related to the battery temperature, current, and battery health status SOH. It is necessary to calculate R0 at the same temperature and current. d And R0_in is subtracted. When R0 d When -R0_in≥0, ​​formula (14) takes the plus sign, when R0 d When -R0_in<0, the minus sign is used in formula (14).

[0051] Step S33 specifically includes the following steps:

[0052] S3301, during the real-time operation of the battery, the battery SOC value SOC(t) at time t and the battery offline identification parameters time constant τ and polarization internal resistance Rp in step S2 are used to linearly interpolate and calculate the battery Rp(t) at time t and the battery time constants τ1(t), τ2(t), and τ3(t). Rp(t) and τ1(t), τ2(t), and τ3(t) are substituted into formulas (15)(16)(17)(18) to calculate the battery polarization voltages Vp1(t), Vp2(t), and Vp3(t) at time t.

[0053] Vp(t)=Vp1(t)+Vp2(t)+Vp3(t)(15)

[0054]

[0055] Using the battery SOC(t) at time t and the actual value of the battery open circuit voltage OCV1(t) at time t, linear interpolation is used to calculate the battery open circuit voltage OCV2(t) at time t. The battery ohmic voltage Vr(t) at time t is calculated using the second self-learning output value R0_SL2, and the calculated value of the battery terminal voltage at time t, U2(t), is calculated as follows:

[0056] OCV1(t)=U1(t)-VP1(t)-VR1(t) (19)

[0057] OCV2(t)=U2(t)-VP2(t)-VR2(t) (20)U1(t)-VP1(t)-VR1(t)≈U2(t)-VP2(t)-VR2(t)(21)

[0058] U1(t)-U2(t)≈[VP1(t)+VR1(t)]-[VP2(t)+VR2(t)](22)

[0059] Vr(t)=I(t)×R0_SL2(23)The polarization internal resistance Rp(t) of the battery at time t includes the first-order polarization internal resistance Rp1(t), the second-order polarization internal resistance Rp2(t) and the third-order polarization internal resistance Rp3(t); τ1(t), τ2(t) and τ3(t) are the first-order time constant, second-order time constant and third-order time constant at time t respectively; Vp1(t), Vp2(t) and Vp3(t) are the first-order polarization voltage, second-order polarization voltage and third-order polarization voltage at time t respectively, Vr(t), OCV2(t) and U2(t) are the calculated values ​​of the battery at time t during actual operation; I(t) is the current acquisition value at time t; t represents the real-time acquisition time.

[0060] S3302, using the battery during actual operation, based on the battery terminal voltage U1(t) collected at time t and the value U2(t) calculated at time t in step S3301, calculate the battery voltage difference DV(t) at time t, and use DV(t) at time t to self-learn R0_SL3.

[0061] Before calculating the battery voltage differential DV(t) at time t, calibrate the battery's SOC. SOC calibration is performed after the battery is fully charged or discharged. Fully charged or discharged means charging or discharging the battery with an adjusted current. SOC calibration is complete when the voltage reaches the maximum charge cutoff voltage or the discharge cutoff voltage.

[0062] DV(t)=U1(t)-U2(t)(24)

[0063] DR0(t)=DV(t) / I(t)(25)

[0064]

[0065] Among them, OCV1(t) is the actual value of the open-circuit voltage of the battery at time t, OCV2(t) is the calculated value of the open-circuit voltage of the battery at time t, VP1(t) is the actual value of the polarization voltage of the battery at time t under zero-state response, VP2(t) is the calculated value of the polarization voltage of the battery at time t, VP2(t) = Vp1(t) + Vp2(t) + Vp3(t), VR1(t) is the actual value of the ohmic voltage of the battery at time t under zero-state response, VR2(t) is the calculated value of the ohmic voltage of the battery at time t, and VR2(t) is equal to Vr(t). When calculating Vr(t), R0_SL2 is substituted into formula (25) as the ohmic internal resistance learning value to calculate the ohmic voltage Vr(t) at time t. DR0(t) represents the ratio of the battery voltage difference DV(t) at time t to the battery current value I(t) collected at time t, representing the ohmic internal resistance compensation value calculated at time t. After the battery is fully charged or discharged, the SOC error is eliminated and OCV1(t) is equal to OCV2(t);

[0066] Since the voltage sensor cannot collect the polarization voltage and ohmic voltage during the actual operating conditions of the battery, when using the third-order RC equivalent circuit model to calculate the errors of the polarization voltage and ohmic voltage of the battery, under the premise that the open circuit voltage of the battery is negligible, the difference DV(t) between U1(t) and U2(t) is used to represent the error between the sum of the measured values ​​and the sum of the calculated values ​​of the polarization voltage and ohmic voltage of the battery.

[0067] The value range of t is [1, A]. The process of t changing from 1 to A corresponds to a period of SOC change length that can be set according to different battery types, such as 5%, 10%, 20%, etc. The number of calculations is A. Considering the error caused by the single compensation value DR0(t) of the ohmic internal resistance, a period of data from the process of t changing from 1 to A is selected for ohmic internal resistance R0 compensation.

[0068] When the battery OCV is in the platform area or the SOC is in the interval corresponding to the OCV platform area, calculate DR0(t). When the SOC change length reaches the set threshold, the ohmic internal resistance compensation is completed. Since the calculated value of DV(t) is different each time, the ohmic internal resistance compensation value is taken as the average value within the SOC change length. The error DV(t) between the measured and calculated values ​​of the battery polarization voltage and ohmic voltage is iteratively calculated to iteratively compensate the ohmic internal resistance, R0_SL3 = [DR0(1)+DR0(2)+…DR0(t)…+DR0(A-1)+DR0(A)] / A.

[0069] A battery equivalent circuit model parameter self-learning system includes a data acquisition unit, an offline identification unit and a self-learning unit;

[0070] The data acquisition unit takes a lithium iron phosphate battery, performs charge and discharge tests on the battery, and obtains test data at intervals of dt seconds. The test data includes the battery terminal voltage measurement value, current, and battery surface temperature;

[0071] The offline identification unit is based on a third-order RC equivalent circuit model and the least squares method. It performs offline identification of battery parameters based on test data to obtain the battery's ohmic internal resistance, polarization internal resistance, and time constant.

[0072] The self-learning unit performs self-learning on the battery's ohmic internal resistance.

[0073] The working process of the data acquisition unit specifically includes the following steps:

[0074] S11, discharge the fully charged battery based on the adjusted current, collect the battery terminal voltage measurement value, battery surface temperature and current at time intervals of dt seconds, and record the battery terminal voltage measurement value U1(k) and the battery surface temperature T(k) and current I(k) every time the battery SOC decreases by a factor of s until the battery SOC = 0; U1(k) represents the battery terminal voltage measurement value at the kth sampling, T(k) represents the battery surface temperature at the kth sampling, I(k) represents the current at the kth sampling; OCV1(k) represents the battery open circuit voltage measurement value at the kth sampling; SOC(k) represents the battery SOC value at the kth sampling;

[0075] S12, charging the battery with SOC=0 based on the adjusted current, synchronously sampling at time intervals of dt seconds, recording the terminal voltage value U1(j) of the battery, and recording the battery surface temperature T(j) and current I(j) every time the battery SOC (state of charge) increases by s times, until the battery SOC=100%; U1(j) represents the terminal voltage value of the battery at the kth sampling, T(j) represents the battery surface temperature at the kth sampling, I(j) represents the current at the jth sampling; OCV1(j) represents the battery open circuit voltage measurement value at the jth sampling; SOC(j) represents the battery SOC value at the jth sampling;

[0076] k=1,2……n; n represents the number of sampling times from when the battery is fully charged to when SOC=0;

[0077] j=n+1,n+2…n+m; m represents the number of sampling times from SOC=0 to full charge of the battery;

[0078] Let i = 1, 2...n+m; OCV1(i) represents the battery open circuit voltage measurement value at the i-th sampling during the test data acquisition process, U1(i) represents the battery terminal voltage measurement value at the i-th sampling; I(i) represents the battery current measurement value at the i-th sampling; SOC(i) represents the battery SOC at the i-th sampling;

[0079] The working process of the offline identification unit specifically includes the following steps:

[0080] S21, calculate the battery ohmic internal resistance R0(i):

[0081] R0(i)=[U1(i)-U1(i+1)] / [I(i)-I(i+1)](1)U1(i) represents the battery terminal voltage measurement value of the i-th sampling, U1(i+1) represents the battery terminal voltage measurement value of the i+1-th sampling, I(i) represents the battery current measurement value of the i-th sampling, I(i+1) represents the battery current measurement value of the i+1-th sampling, and R0(i) represents the ohmic internal resistance corresponding to SOC(i);

[0082] S22, identification of time constant and polarization internal resistance;

[0083] The identification of the time constant and polarization internal resistance specifically includes the following steps:

[0084] S2201, setting a range for the polarization internal resistance and time constant of the battery. When identifying the polarization internal resistance and time constant of the battery within the set range, set the polarization internal resistance range [lb1, ub1] and the time constant value range [lb2, ub2].

[0085] S2202, setting the initial value of the polarization internal resistance and the initial value of the time constant of the battery. The design value of the initial value of the polarization internal resistance is between the lower boundary lb1 of the polarization internal resistance and the upper boundary ub1 of the polarization internal resistance. The setting value of the initial value of the time constant is between the lower boundary lb2 of the time constant and the upper boundary ub2 of the time constant.

[0086] S2203, the terminal voltage measurement value of the battery sampled at the i-th time is U1(i), and the terminal voltage calculated value of the battery calculated based on the test data is U2′(i). The initial value of the polarization internal resistance and the initial value of the time constant of the battery are used to calculate the terminal voltage calculated value of the battery based on the test data, U2′(i):

[0087] U2'(i)=OCV(i)+Vp(i)+Vr(i)(2)

[0088] Vr(i)=I(i)×R0(i)(3)

[0089] Vp(i)=Vp1(i)+Vp2(i)+Vp3(i)(4)

[0090]

[0091] U2'(i) is the calculated value of the battery terminal voltage at the i-th sampling, OCV1(i) is the measured value of the battery open circuit voltage at the i-th sampling, Vp(i) is the polarization voltage of the battery at the i-th sampling, and the polarization voltage includes the first-order polarization voltage Vp1(i), the second-order polarization voltage Vp2(i), and the third-order polarization voltage Vp3(i); Vr(i) is the ohmic voltage of the battery, I(i) is the current of the battery, and R0(i) is the ohmic internal resistance of the battery ; The time constants of the battery model are divided into the first-order time constant τ1, the second-order time constant τ2 and the third-order time constant τ3; the polarization internal resistance of the battery includes the first-order polarization internal resistance Rp1(i), the second-order polarization internal resistance Rp2(i) and the third-order polarization internal resistance Rp3(i); dt is the battery data sampling time interval, Vp1(0), Vp2(0) and Vp3(0) are the initial values ​​of the first-order polarization voltage, the second-order polarization voltage and the third-order polarization voltage respectively;

[0092] S2204, setting the objective function Fcn;

[0093] Fcn=U1(i)-U2'(i) (8) U1(i) is the measured value of the battery terminal voltage at the i-th sampling, and U2'(i) is the calculated value of the battery terminal voltage obtained based on the test data of the i-th sampling;

[0094] S2205, using the least squares method, the objective function Fcn is minimized; when the objective function Fcn is minimized, the polarization internal resistance and time constant of the corresponding battery are the polarization internal resistance and time constant identified offline:

[0095]

[0096] I(i) is the current value of the battery sampled at the i-th time, Rp is the polarization internal resistance of the battery, and the value of F(τ,Rp,I(i)) is equal to U2'(i), which represents the calculated value of the battery terminal voltage under the given time constant τ, polarization internal resistance Rp and measurement current U1(i);

[0097] By iterating the battery polarization internal resistance and time constant, the solution value of Fcn is updated until the solution value of Fcn reaches the minimum value, and the identification of the polarization internal resistance Rp and time constant τ is completed.

[0098] The working process of the self-learning unit specifically includes the following steps:

[0099] S31, the battery ohmic internal resistance R0(i) performs the first self-learning step, and the first self-learning output value is recorded as R0_SL1;

[0100] Set the upper and lower limits of the SOC interval USOC and LSOC, and the SOC(i) interval corresponding to the self-learning of the battery ohmic internal resistance R0(i) is [LSOC(i), USOC(i)]. Within the SOC(i) interval [LSOC(i), USOC(i)], the battery ohmic internal resistance corresponding to SOC(i) is R0(i); calculate the first self-learning output value of the battery ohmic internal resistance (average ohmic internal resistance value) R0_SL1(i) based on formula (11); when the error between the self-learning average ohmic internal resistance R0_SL1(i+1) and the previous self-learning average ohmic internal resistance R0_SL1(i) is less than or equal to Tol, the first self-learning calculation of the battery ohmic internal resistance is completed, and the last self-learning value of the battery is taken as the first self-learning output value of the battery ohmic internal resistance R0_SL1:

[0101]

[0102] When the condition |R0_SL1(i+1)-R0_SL1(i)|≤Tol is satisfied, R0_SL1=R0_SL1(i+1);

[0103] Tol is the set threshold; b represents the number of R0(i) corresponding to SOC(i) in the interval [LSOC(i), USOC(i)], SOC(i)∈[LSOC(i), USOC(i)];

[0104] S32, performing a second self-learning step based on the first self-learning output value R0_SL1 of the battery ohmic internal resistance, where the second self-learning output value is R0_SL2;

[0105] S33, performing a third self-learning step based on the second self-learning output value R0_SL2, and the third self-learning step output value is R0_SL3;

[0106] S32 specifically includes the following steps:

[0107] S3201, after one hour of open circuit, the battery is charged and discharged at a constant current. During the real-time operation of the battery, the terminal voltage of the battery is collected at intervals of dt seconds and recorded as U0(c). U0(c) is recorded in sequence as U0(1), U0(2), ...U0(c), U0(c+1), U0(c+2) ...U0(A). The collected current is recorded as I0(1), I0(2), ...I0(c), I0(c+1), I0(c+2) ...I0(A). The collected temperature is recorded as T 0(1), T0(2), ... T0(c), T0(c+1), T0(c+2) ... T0(A); U0(c), I0(c), T0(c) are the battery terminal voltage, current and battery surface temperature collected for the cth time when the voltage enters the constant current charge and discharge condition from the open circuit state for one hour, SOC(c) is the SOC value collected for the cth time when the voltage enters the constant current charge and discharge condition from the open circuit state for one hour; the real-time dynamic ohmic internal resistance R0 of the battery is calculated based on formula (12) d :When the battery SOC (c) ∈ ​​[LSOC, USOC] and the voltage enters the constant current charge and discharge condition from the open circuit state for one hour, calculate the battery dynamic ohmic internal resistance R0 d :

[0108]

[0109] The average value of the voltage matrix [U0(1), U0(2), ... U0(c-1)] is taken as the open circuit voltage before the battery enters the constant current charge and discharge condition, and the average value of the current matrix [I0(c), I0(c+1), I0(c+2)] is taken as the current after the battery enters the constant current charge and discharge condition;

[0110] S3202, based on the battery capacity, calculate the ohmic internal resistance R0_in after being corrected for the rated capacity, formula (13):

[0111] R0_in=CN0 / CN×R0_SL1(13)CN0 is the rated capacity of the test battery, CN is the rated capacity of a certain type of battery under test; R0_in is the ohmic internal resistance corrected by the rated capacity;

[0112] S3203, calculating the battery ohmic internal resistance second self-learning output value R0_SL2;

[0113] R0_SL2=R0_in±|R0 d -R0_in|(14)

[0114] Based on |R0 d -R0_in|Self-learning calculation of R0_SL2 value, in the calculation|R0 d -R0_in|, because |R0 d-R0_in| is related to the battery temperature, current, and battery health status. d and R0_in; when R0 d When -R0_in≥0, ​​formula (14) takes the plus sign, when R0 d When -R0_in<0, the minus sign is used in formula (14);

[0115] Step S33 specifically includes the following steps:

[0116] S3301, during the real-time operation of the battery, the battery SOC value SOC(t) at time t and the battery offline identification parameters time constant τ and polarization internal resistance Rp in step S2 are used to linearly interpolate and calculate the battery Rp(t) at time t and the battery time constants τ1(t), τ2(t), and τ3(t). Rp(t) and τ1(t), τ2(t), and τ3(t) are substituted into formulas (15)(16)(17)(18) to calculate the battery polarization voltages Vp1(t), Vp2(t), and Vp3(t) at time t.

[0117] Vp(t)=Vp1(t)+Vp2(t)+Vp3(t)(15)

[0118]

[0119] Using the battery SOC(t) at time t and the actual value of the battery open circuit voltage OCV1(t) at time t, linear interpolation is used to calculate the battery open circuit voltage OCV2(t) at time t; the battery ohmic voltage Vr(t) at time t is calculated using the second self-learning output value R0_SL2, and the calculated value of the battery terminal voltage at time t, U2(t), is obtained:

[0120] OCV1(t)=U1(t)-VP1(t)-VR1(t) (19)

[0121] OCV2(t)=U2(t)-VP2(t)-VR2(t) (20)U1(t)-VP1(t)-VR1(t)≈U2(t)-VP2(t)-VR2(t)(21)

[0122] U1(t)-U2(t)≈[VP1(t)+VR1(t)]-[VP2(t)+VR2(t)](22)

[0123] Vr(t)=I(t)×R0_SL2(23) The polarization internal resistance Rp(t) of the battery at time t includes the first-order polarization internal resistance Rp1(t), the second-order polarization internal resistance Rp2(t) and the third-order polarization internal resistance Rp3(t); τ1(t), τ2(t) and τ3(t) are the first-order time constant, the second-order time constant and the third-order time constant at time t respectively; Vp1(t), Vp2(t) and Vp3(t) are the first-order polarization voltage, the second-order polarization voltage and the third-order polarization voltage at time t respectively; I(t) is the current acquisition value at time t; t represents the real-time acquisition time;

[0124] S3302, using the battery terminal voltage U1(t) collected at time t and the value U2(t) calculated at time t in step S3301 during actual operation, calculate the battery voltage difference DV(t) at time t, and use DV(t) at time t to self-learn R0_SL3;

[0125] Battery SOC calibration refers to the calibration of the battery SOC after the battery is fully charged or discharged, and the calculation of the battery voltage difference DV(t) at time t:

[0126] DV(t)=U1(t)-U2(t)(24)

[0127] DR0(t)=DV(t) / I(t)(25)

[0128]

[0129] Among them, OCV1(t) is the actual value of the open circuit voltage of the battery at time t, OCV2(t) is the calculated value of the open circuit voltage of the battery at time t, VP1(t) is the actual value of the polarization voltage of the battery at time t under zero-state response, VP2(t) is the calculated value of the polarization voltage of the battery at time t, VP2(t)=Vp1(t)+Vp2(t)+Vp3(t), VR1(t) is the actual value of the ohmic voltage of the battery at time t under zero-state response, VR2(t) is the calculated value of the ohmic voltage of the battery at time t, and VR2(t) is equal to Vr(t); DR0(t) represents the ratio of the voltage difference DV(t) of the battery at time t to the current collected value I(t) of the battery at time t, and represents the ohmic internal resistance compensation value calculated at time t; after the battery is fully charged or discharged, the SOC error is eliminated, and OCV1(t) is equal to OCV2(t);

[0130] The value range of t is [1, A]. The SOC change length is set according to different types of batteries during the process of t changing from 1 to A, and the number of calculations is A. Considering the error caused by the single compensation value DR0(t) of the ohmic internal resistance, the data during the process of t changing from 1 to A is selected to compensate the ohmic internal resistance R0.

[0131] The error DV(t) between the measured and calculated values ​​of the battery polarization voltage and ohmic voltage is iteratively calculated, and the ohmic internal resistance is iteratively compensated, R0_SL3 = [DR0(1)+DR0(2)+…DR0(t)…+DR0(A-1)+DR0(A)] / A.

[0132] Compared with the prior art, the present invention has the following beneficial effects:

[0133] The present invention discloses a battery equivalent circuit model parameter self-learning method. The method identifies the battery internal resistance offline, compensates the battery internal resistance online according to a third-order RC equivalent circuit model during the battery operation process, completes the battery internal resistance self-learning, and provides SOC, SOP, and SOH state estimation, thereby accurately identifying the parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0134] Other features, objects and advantages of the present application will become more apparent upon reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0135] Figure 1 Third-order RC equivalent circuit model diagram;

[0136] Figure 2 Parameter offline identification process diagram;

[0137] Figure 3 Offline identification of ohmic internal resistance based on SOC interval learning process diagram;

[0138] Figure 4 Self-learning ohmic internal resistance R0 d Learning process diagram;

[0139] Figure 5 Voltage difference compensation diagram during charging process. DETAILED DESCRIPTION

[0140] The following will clearly and completely describe the technical solution of the present invention in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0141] A battery equivalent circuit model parameter self-learning method includes the following steps:

[0142] S1: Take a lithium iron phosphate battery and perform charge and discharge tests on the battery at room temperature (25±2°C). The test data is obtained at a time interval of 0.1 seconds. The test data includes the battery terminal voltage measurement value, current, and battery surface temperature. The test battery is recorded as battery a.

[0143] Step S1 specifically includes the following steps:

[0144] S11, discharge the fully charged battery based on the adjusted current, collect the battery terminal voltage measurement value, battery surface temperature and current at 0.1 second time intervals, and record the battery terminal voltage measurement value U1(k) and the battery surface temperature T(k) and current I(k) every time the battery SOC (state of charge) decreases by s times; until the battery SOC = 0; U1(k) represents the battery terminal voltage measurement value at the kth sampling, T(k) represents the battery surface temperature at the kth sampling, I(k) represents the current at the kth sampling; OCV1(k) represents the battery open circuit voltage measurement value at the kth sampling; SOC(k) represents the battery SOC value at the kth sampling;

[0145] S12, charging the battery with SOC=0 based on the adjusted current, synchronously sampling at a time interval of 0.1 seconds, recording the terminal voltage value U1(j) of the battery, and recording the battery surface temperature T(j) and current I(j) every time the battery SOC (state of charge) increases by s times, until the battery SOC=100%; U1(j) represents the terminal voltage value of the battery at the kth sampling, T(j) represents the battery surface temperature at the kth sampling, I(j) represents the current at the jth sampling; OCV1(j) represents the battery open circuit voltage measurement value at the jth sampling; SOC(j) represents the battery SOC value at the jth sampling;

[0146] k=1,2……n; n represents the number of sampling times from when the battery is fully charged to when SOC=0;

[0147] j=n+1,n+2…n+m; m represents the number of sampling times from SOC=0 to full charge of the battery;

[0148] Let i = 1, 2...n+m; OCV1(i) represents the battery open circuit voltage measurement value at the i-th sampling during the test data acquisition process, U1(i) represents the battery terminal voltage measurement value at the i-th sampling; I(i) represents the battery current measurement value at the i-th sampling; SOC(i) represents the battery SOC at the i-th sampling;

[0149] In step S1 of this embodiment, a lithium iron phosphate battery is taken and the battery terminal voltage, current, and temperature are synchronously sampled at 0.1 second intervals at room temperature (25±2°C). The battery test experimental steps are as follows:

[0150] In step S101, fully charge the battery using a 1.0C current and let it rest for one hour. The battery terminal voltage at the last collected point is recorded as the open-circuit voltage at which the SOC equals 100%. C represents the current rate symbol, and SOC represents the battery state of charge. C and SOC are standard symbols in this field.

[0151] The current I is adjusted to a current range of 0.1C to 1.0C. Using 1.0C for the adjusted current is more in line with national standards.

[0152] S102: Discharge the battery with a current of 0.5C for 60 seconds, let it stand for 1 hour, then discharge the battery with a current of 1.0C to 95% SOC and let it stand for 1 hour. The battery terminal voltage value at the last acquisition point is recorded as the open circuit voltage when the SOC is equal to 95%.

[0153] S103 , performing a discharge test on the battery according to the test method of S102 until the SOC reaches 5%.

[0154] S104 , the battery is completely discharged using a current of 1.0 C and then left to stand for 1 hour. The battery terminal voltage at the last acquisition point is recorded as the open circuit voltage when the SOC is equal to 0%.

[0155] S15: Charge with 0.5C current for 60 seconds, let it stand for 1 hour, then charge with 1.0C current to 5% SOC and let it stand for 1 hour. The battery terminal voltage at the last collection point is recorded as the open circuit voltage when SOC equals 5%.

[0156] S106 , performing a charging test on the battery according to the test method of S105 until the SOC reaches 95%.

[0157] S107 , fully charge the battery using a 1.0C current and let it sit for 1 hour. The battery terminal voltage at the last acquisition point is recorded as the open circuit voltage when the SOC is equal to 100%.

[0158] S108, through the test steps S101 and S107, obtain the test data corresponding to each 5% interval of the SOC in the range of 0% to 100%, and the obtained test data includes current, voltage terminal voltage and battery surface temperature.

[0159] S2, based on the third-order RC equivalent circuit model and the principle of least squares method, the battery parameters are identified offline based on the test data of step S1 to obtain the battery's ohmic internal resistance, polarization internal resistance and time constant;

[0160] like Figure 2 As shown, S2 specifically includes the following steps:

[0161] S21, calculate the battery ohmic internal resistance R0(i) based on formula (1).

[0162] R0(i)=[U1(i)-U1(i+1)] / [I(i)-I(i+1)](1)U1(i) represents the battery terminal voltage measurement value of the i-th sampling, U1(i+1) represents the battery terminal voltage measurement value of the i+1-th sampling, I(i) represents the battery current measurement value of the i-th sampling, I(i+1) represents the battery current measurement value of the i+1-th sampling, and R0(i) represents the ohmic internal resistance corresponding to SOC(i).

[0163] S22, Identification of time constant and polarization internal resistance.

[0164] S2201 determines the polarization internal resistance and time constant of the battery based on its characteristics. When identifying the polarization internal resistance and time constant, set the polarization internal resistance range [lb1, ub1] and the time constant range [lb2, ub2], with lb1 and lb2 as the lower limits and ub1 and ub2 as the upper limits. The lb and ub values ​​are adjusted to suit the battery's characteristics.

[0165] S2202: Set the initial value of the battery's polarization internal resistance and time constant. The initial value of the polarization internal resistance is designed to be between the lower boundary lb1 and the upper boundary ub1, and the initial value of the time constant is set to be between the lower boundary lb2 and the upper boundary ub2. In this embodiment, the time constant is set to 100 seconds and the polarization internal resistance is set to 0.5 ohms.

[0166] S2203, the terminal voltage measurement value of the battery sampled for the i-th time is U1(i), and the terminal voltage calculated value of the battery calculated based on the test data is U2′(i). The initial value of the polarization internal resistance and the initial value of the time constant of the battery are substituted into formulas (2) to (7) to calculate the terminal voltage calculated value U2′(i) of the battery based on the test data.

[0167] U2'(i)=OCV(i)+Vp(i)+Vr(i)(2)

[0168] Vr(i)=I(i)×R0(i)(3)

[0169] Vp(i)=Vp1(i)+Vp2(i)+Vp3(i)(4)

[0170]

[0171] U2'(i) is the calculated value of the battery terminal voltage at the i-th sampling, OCV1(i) is the measured value of the battery open circuit voltage at the i-th sampling, Vp(i) is the polarization voltage of the battery at the i-th sampling, and the polarization voltage includes the first-order polarization voltage Vp1(i), the second-order polarization voltage Vp2(i), and the third-order polarization voltage Vp3(i); Vr(i) is the ohmic voltage of the battery, I(i) is the current of the battery, and R0(i) is the ohmic internal resistance of the battery ; The time constants of the battery model are divided into the first-order time constant τ1, the second-order time constant τ2 and the third-order time constant τ3; the polarization internal resistance of the battery includes the first-order polarization internal resistance Rp1(i), the second-order polarization internal resistance Rp2(i) and the third-order polarization internal resistance Rp3(i); dt is the battery data sampling time interval, Vp1(0), Vp2(0) and Vp3(0) are the initial values ​​of the first-order polarization voltage, the second-order polarization voltage and the third-order polarization voltage respectively;

[0172] S2204, setting the objective function Fcn;

[0173] Fcn=U1(i)-U2'(i) (8) U1(i) is the measured value of the battery terminal voltage at the i-th sampling, and U2'(i) is the calculated value of the battery terminal voltage obtained based on the test data of the i-th sampling;

[0174] S2205, using the least squares method, the objective function Fcn is minimized; when the objective function Fcn takes the minimum value, the polarization internal resistance and time constant of the corresponding battery are the parameters (polarization internal resistance and time constant) for offline identification.

[0175]

[0176] I(i) is the current value of the battery sampled for the i-th time, Rp is the polarization internal resistance of the battery, and the value of F(τ,Rp,I(i)) is equal to U2'(i), which represents the calculated value of the battery terminal voltage under the given time constant τ, polarization internal resistance Rp and measurement current U1(i).

[0177] When minimizing the objective function Fcn, the initial values ​​of the battery's polarization internal resistance and time constant are substituted into formulas (2) to 7) to calculate U2'(i), completing the first solution to Fcn. Based on the solution value of Fcn, the polarization internal resistance and time constant are updated using the least squares method. The updated polarization internal resistance and time constant are used to continue calculating U2'(i), completing the second solution to Fcn. By continuously updating the battery's polarization internal resistance and time constant, the solution value of Fcn is updated until the solution value of Fcn reaches the minimum value, completing the identification of the polarization internal resistance Rp and the time constant τ.

[0178] S3, self-learning of the battery's ohmic internal resistance.

[0179] Step S3 specifically includes the following steps:

[0180] S31, the battery ohmic internal resistance R0(i) performs the first self-learning step, and the first self-learning output value is recorded as R0_SL1;

[0181] Set the upper and lower limits of the SOC range, USOC and LSOC, and set LSOC∈[0,30] and USOC∈[70,100]. The SOC(i) range corresponding to the self-learning of the battery ohmic internal resistance R0(i) is [LSOC(i),USOC(i)]. Within the SOC(i) range [LSOC(i),USOC(i)], the battery ohmic internal resistance corresponding to SOC(i) is R0(i). Based on formula (11), calculate the first self-learning output value (average value of the ohmic internal resistance) R0_SL1(i) of the battery ohmic internal resistance. The first calculation is USOC(1) equal to 100, LSOC(1) equal to 0, the second calculation is USOC(2) equal to 95, LSOC(2) equal to 5, and so on. When the error between the self-learning average value of the ohmic internal resistance R0_SL1(i+1) and the previous self-learning average value of the ohmic internal resistance R0_SL1(i) is less than or equal to Tol, the first self-learning calculation of the battery's ohmic internal resistance is completed, and the last self-learning value of the battery is taken as the first self-learning output value of the battery's ohmic internal resistance R0_SL1.

[0182]

[0183] When the condition |R0_SL1(i+1)-R0_SL1(i)|≤Tol is satisfied, R0_SL1=R0_SL1(i+1);

[0184] Tol can be set. The value of Tol is adjusted according to different batteries and is in milliohms. SOC(i) is any value between 0% and 100%. b represents the number of R0(i) corresponding to SOC(i) in the interval [LSOC(i), USOC(i)]. SOC(i)∈[LSOC(i), USOC(i)]. USOC(i) and LSOC(i) are expressed as Figure 3 express.

[0185] S32, performing a second self-learning step based on the first self-learning output value R0_SL1 of the battery ohmic internal resistance, where the second self-learning output value is R0_SL2;

[0186] S32 specifically includes the following steps:

[0187] S3201, after one hour of open circuit, the battery is charged and discharged at a constant current. During the real-time operation of the battery, the terminal voltage of the battery is collected at intervals of dt seconds and recorded as U0(c). U0(c) is recorded in sequence as U0(1), U0(2), ... U0(c), U0(c+1), U0(c+2) ... U0(A). The collected current is recorded as I0(1), I0(2), ... I0(c), I0(c+1), I0(c+2) ... I0(A). The collected temperature is recorded as T0(1), T0(2), ... T0(c), T0(c+1), T0(c+2)…T0(A); U0(c), I0(c), T0(c) are the battery terminal voltage, current and battery surface temperature collected for the cth time when the voltage enters the constant current charge and discharge condition from the open circuit state for one hour; SOC(c) is the SOC value collected for the cth time when the voltage enters the constant current charge and discharge condition from the open circuit state for one hour; A represents the number of samples in the real-time operation process, c=1, 2…A; The real-time dynamic ohmic internal resistance R0 of the battery is calculated based on formula (12): d :

[0188] When the battery SOC (c) ∈ ​​[LSOC, USOC] and the voltage enters the constant current charge and discharge condition from the open circuit state for one hour, calculate the battery dynamic ohmic internal resistance R0 d :

[0189]

[0190] In order to avoid the problem of asynchronous acquisition of battery current and voltage, zero drift in current sampling, and inaccurate battery terminal voltage and current measurement due to incomplete disappearance of the battery polarization voltage, the average value of the voltage matrix [U0(1), U0(2), ... U0(c-1)] is taken as the open circuit voltage before the battery enters the constant current charge and discharge condition, and the average value of the current matrix [I0(c), I0(c+1), I0(c+2)] is taken as the current after the battery enters the constant current charge and discharge condition. Figure 4 shown.

[0191] S3202, based on the battery capacity, calculate the ohmic internal resistance R0_in after being corrected for the rated capacity, formula (13):

[0192] R0_in=CN0 / CN×R0_SL1(13)CN0 is the rated capacity of the test battery a, CN is the rated capacity of a certain type of battery a; R0_in is the ohmic internal resistance corrected by the rated capacity;

[0193] S3203, calculating the battery ohmic internal resistance second self-learning output value R0_SL2;

[0194] R0_SL2=R0_in±|R0 d -R0_in|(14)

[0195] Based on |R0 d -R0_in|Self-learning calculation of R0_SL2 value, see formula (14). d -R0_in|, because |R0 d -R0_in| is related to the battery temperature, current, and battery health status SOH. It is necessary to calculate R0 at the same temperature and current. d And R0_in is subtracted. When R0 d When -R0_in≥0, ​​formula (14) takes the plus sign, when R0 d When -R0_in<0, the minus sign is used in formula (14).

[0196] S33, performing a third self-learning step based on the second self-learning output value R0_SL2, and the third self-learning step output value is R0_SL3.

[0197] S3301, during the real-time operation of the battery, the battery SOC value SOC(t) at time t and the battery offline identification parameters time constant τ and polarization internal resistance Rp in step S2 are used to linearly interpolate and calculate the battery Rp(t) at time t and the battery time constants τ1(t), τ2(t), τ3(t). Rp(t) and τ1(t), τ2(t), τ3(t) are substituted into formulas (15)(16)(17)(18) to calculate the battery polarization voltage Vp1(t), Vp2(t), Vp3(t) at time t; the battery open circuit voltage OCV2(t) at a certain time is calculated by linear interpolation using the battery SOC′(t) at time t and the actual value of the battery open circuit voltage OCV1(t) at time t. The battery ohmic voltage Vr(t) at time t is calculated using the second self-learning output value R0_SL2, and the calculated value of the battery terminal voltage U2(t) at time t is obtained:

[0198] Vp(t)=Vp1(t)+Vp2(t)+Vp3(t)(15)

[0199]

[0200] Using the battery SOC(t) at time t and the actual value of the battery open circuit voltage OCV1(t) at time t, linear interpolation is used to calculate the battery open circuit voltage OCV2(t) at time t. The battery ohmic voltage Vr(t) at time t is calculated using the second self-learning output value R0_SL2, and the calculated value of the battery terminal voltage at time t, U2(t), is calculated as follows:

[0201] OCV1(t)=U1(t)-VP1(t)-VR1(t)(19)

[0202] OCV2(t)=U2(t)-VP2(t)-VR2(t)(20)U1(t)-VP1(t)-VR1(t)≈U2(t)-VP2(t)-VR2(t)(21)

[0203] U1(t)-U2(t)≈[VP1(t)+VR1(t)]-[VP2(t)+VR2(t)](22)

[0204] Vr(t)=I(t)×R0_SL2(23)Rp(t), τ1(t), τ2(t), τ3(t), Vp1(t), Vp2(t), Vp3(t), Vr(t), OCV2(t), and U2(t) are the calculated values ​​of the battery at time t during actual operation; I(t) is the current value collected at time t; and t represents the real-time collection time. The polarization internal resistance Rp(t) of the battery at time t includes the first-order polarization internal resistance Rp1(t), the second-order polarization internal resistance Rp2(t) and the third-order polarization internal resistance Rp3(t); τ1(t), τ2(t) and τ3(t) are the first-order time constant, second-order time constant and third-order time constant at time t, respectively; Vp1(t), Vp2(t) and Vp3(t) are the first-order polarization voltage, second-order polarization voltage and third-order polarization voltage at time t, respectively; Vr(t), OCV2(t) and U2(t) are the calculated values ​​of the battery at time t during actual operation; I(t) is the current collected value at time t; t represents the real-time collection time.

[0205] In step S3302, the voltage U1(t) collected at time t during actual battery operation and the value U2(t) calculated at time t in step S3301 are used to calculate the battery voltage difference DV(t) at time t. This value DV(t) is used to self-learn R0_SL3, as shown in equations (20) to (25).

[0206] Before calculating the battery voltage differential DV(t) at time t, calibrate the battery's SOC. SOC calibration is performed after the battery is fully charged or discharged. The full or empty standard is to charge or discharge the battery at an adjusted current (such as 0.5C or 1C). SOC calibration is completed when the battery voltage reaches the maximum charge cutoff voltage (such as 3.6V or 4.2V) or the discharge cutoff voltage (such as 2.0V or 2.7V).

[0207] DV(t)=U1(t)-U2(t)(24)

[0208] DR0(t)=DV(t) / I(t)(25)

[0209]

[0210] Among them, OCV1(t) is the actual value of the battery's open-circuit voltage at time t, OCV2(t) is the calculated value of the battery's open-circuit voltage at time t, VP1(t) is the actual value of the battery's polarization voltage at time t under zero-state response. Although VP1(t) cannot be directly measured, it can be directly calculated from the measured value U1(t) under the condition of zero-state response of the battery current. VP2(t) is the calculated value of the battery's polarization voltage at time t, VP2(t) = Vp1(t) + Vp2(t) + Vp3(t). VR1(t) is the actual value of the battery's ohmic voltage at time t under zero-state response, VR2(t) is the calculated value of the battery's ohmic voltage at time t, and VR2(t) is the calculated value of the battery's ohmic voltage at time t. VR2(t) is equal to Vr(t). When calculating Vr(t), substitute R0_SL2 as the ohmic internal resistance learning value into formula (19) to calculate the ohmic voltage Vr(t) at time t. DR0(t) represents the ratio of the voltage difference DV(t) of the battery at time t to the current value I(t) collected at time t, and represents the ohmic internal resistance compensation value calculated at time t. After the battery is fully charged or discharged, the SOC error is eliminated, and OCV1(t) is approximately equal to OCV2(t), that is, formula (22). Formula (23)(24) can be derived from formula (22). Under a certain constant current condition, the true value in Table 1 is the voltage sensor measurement value with a sampling accuracy of one thousandth, and the polarization voltage and ohmic voltage are directly calculated from the voltage sensor measurement value. The calculated values ​​in Table 1 are the calculated values ​​obtained by the third-order RC equivalent circuit model using offline identification parameters. Taking the battery charging process as an example, formulas (20) to (24) are illustrated as shown in Table 1.

[0211] Table 1 Examples of formulas (16)-(20)

[0212] variable name Value (unit: mv) variable name Value (unit: mv) OCV1(t) 3300 VP1(t) 100 OCV2(t) 3299 VP2(t) 90 U1(t) 3450 VR1(t) 50 U2(t) 3444 VR2(t) 55

[0213] Since the voltage sensor cannot collect the polarization voltage and ohmic voltage during the actual operating conditions of the battery, when using the third-order RC equivalent circuit model to calculate the errors of the polarization voltage and ohmic voltage of the battery, under the premise that the open circuit voltage of the battery is negligible, the difference DV(t) between U1(t) and U2(t) is used to represent the error between the sum of the measured values ​​and the sum of the calculated values ​​of the polarization voltage and ohmic voltage of the battery.

[0214] As in this embodiment, OCV1(t)≈OCV2(t), U1(t)-U2(t)=6(mv), VP1(t)+VR1(t)=150(mv), VP2(t)+VR2(t)=145(mv). After the battery eliminates the SOC error, the battery OCV1(t) is approximately equal to OCV2(t). The error between the actual value and the calculated value of the polarization voltage and the ohmic voltage in the open circuit state is 5(mv) in total. DV(t) is equal to 6(mv), and DV(t) is approximately equal to the sum of the errors of the polarization voltage and the ohmic voltage.

[0215] The value range of t is [1, A]. When t changes from 1 to A, it corresponds to a period of SOC change length that can be set according to different types of batteries, such as 5%, 10%, 20%, etc. The number of calculations is A. Considering the error caused by the single compensation value DR0(t) of the ohmic internal resistance, a period of data in the process of t changing from 1 to A is selected for ohmic internal resistance R0 compensation. Taking the battery charging process as an example, assuming the operating current is 100 (A), the example is as follows Figure 5 and as shown in Table 2.

[0216] Table 2 Example of calculation for ohmic internal resistance compensation

[0217]

[0218]

[0219] Calculate DR0(t) when the battery's OCV is in the plateau region or when the SOC is in the corresponding interval of the OCV plateau region. When the SOC variation reaches the set threshold, ohmic internal resistance compensation is completed. Because DV(t) calculations vary, the ohmic internal resistance compensation value is the average value over the SOC variation period. Figure 5 For the example of the battery OCV being in the voltage plateau region, the error DV(t) between the measured and calculated values ​​of the battery polarization voltage and ohmic voltage is iteratively calculated, and the ohmic internal resistance is iteratively compensated, R0_SL3 = [DR0(1)+DR0(2)+…DR0(t)…+DR0A-1+DR0A / A].

[0220] R0_SL2 is learned based on R0_SL1, and R0_SL3 is learned based on R0_SL2. R0_SL1 is the average value of the internal resistance in ohms. The learning process of R0_SL1, R0_SL2, and R0_SL3 progresses gradually.

[0221] A battery equivalent circuit model parameter self-learning system includes a data acquisition unit, an offline identification unit and a self-learning unit;

[0222] The data acquisition unit takes a lithium iron phosphate battery, performs charge and discharge tests on the battery, and obtains test data at intervals of dt seconds. The test data includes the battery terminal voltage measurement value, current, and battery surface temperature;

[0223] The offline identification unit is based on a third-order RC equivalent circuit model and the least squares method. It performs offline identification of battery parameters based on test data to obtain the battery's ohmic internal resistance, polarization internal resistance, and time constant.

[0224] The self-learning unit performs self-learning on the battery's ohmic internal resistance.

[0225] The working process of the data acquisition unit specifically includes the following steps:

[0226] S11, discharge the fully charged battery based on the adjusted current, collect the battery terminal voltage measurement value, battery surface temperature and current at time intervals of dt seconds, and record the battery terminal voltage measurement value U1(k) and the battery surface temperature T(k) and current I(k) every time the battery SOC decreases by a factor of s until the battery SOC = 0; U1(k) represents the battery terminal voltage measurement value at the kth sampling, T(k) represents the battery surface temperature at the kth sampling, I(k) represents the current at the kth sampling; OCV1(k) represents the battery open circuit voltage measurement value at the kth sampling; SOC(k) represents the battery SOC value at the kth sampling;

[0227] S12, charging the battery with SOC=0 based on the adjusted current, synchronously sampling at time intervals of dt seconds, recording the terminal voltage value U1(j) of the battery, and recording the battery surface temperature T(j) and current I(j) every time the battery SOC (state of charge) increases by s times, until the battery SOC=100%; U1(j) represents the terminal voltage value of the battery at the kth sampling, T(j) represents the battery surface temperature at the kth sampling, I(j) represents the current at the jth sampling; OCV1(j) represents the battery open circuit voltage measurement value at the jth sampling; SOC(j) represents the battery SOC value at the jth sampling;

[0228] k=1,2……n; n represents the number of sampling times from when the battery is fully charged to when SOC=0;

[0229] j=n+1,n+2…n+m; m represents the number of sampling times from SOC=0 to full charge of the battery;

[0230] Let i = 1, 2...n+m; OCV1(i) represents the battery open circuit voltage measurement value at the i-th sampling during the test data acquisition process, U1(i) represents the battery terminal voltage measurement value at the i-th sampling; I(i) represents the battery current measurement value at the i-th sampling; SOC(i) represents the battery SOC at the i-th sampling;

[0231] The working process of the offline identification unit specifically includes the following steps:

[0232] S21, calculate the battery ohmic internal resistance R0(i):

[0233] R0(i)=[U1(i)-U1(i+1)] / [I(i)-I(i+1)](1)U1(i) represents the battery terminal voltage measurement value of the i-th sampling, U1(i+1) represents the battery terminal voltage measurement value of the i+1-th sampling, I(i) represents the battery current measurement value of the i-th sampling, I(i+1) represents the battery current measurement value of the i+1-th sampling, and R0(i) represents the ohmic internal resistance corresponding to SOC(i);

[0234] S22, identification of time constant and polarization internal resistance;

[0235] The identification of the time constant and polarization internal resistance specifically includes the following steps:

[0236] S2201, setting a range for the polarization internal resistance and time constant of the battery. When identifying the polarization internal resistance and time constant of the battery within the set range, set the polarization internal resistance range [lb1, ub1] and the time constant value range [lb2, ub2].

[0237] S2202, setting the initial value of the polarization internal resistance and the initial value of the time constant of the battery. The design value of the initial value of the polarization internal resistance is between the lower boundary lb1 of the polarization internal resistance and the upper boundary ub1 of the polarization internal resistance. The setting value of the initial value of the time constant is between the lower boundary lb2 of the time constant and the upper boundary ub2 of the time constant.

[0238] S2203, the terminal voltage measurement value of the battery sampled at the i-th time is U1(i), and the terminal voltage calculated value of the battery calculated based on the test data is U2′(i). The initial value of the polarization internal resistance and the initial value of the time constant of the battery are used to calculate the terminal voltage calculated value of the battery based on the test data, U2′(i):

[0239] U2'(i)=OCV(i)+Vp(i)+Vr(i)(2)

[0240] Vr(i)=I(i)×R0(i)(3)

[0241] Vp(i)=Vp1(i)+Vp2(i)+Vp3(i)(4)

[0242]

[0243] U2'(i) is the calculated value of the battery terminal voltage at the i-th sampling, OCV1(i) is the measured value of the battery open circuit voltage at the i-th sampling, Vp(i) is the polarization voltage of the battery at the i-th sampling, and the polarization voltage includes the first-order polarization voltage Vp1(i), the second-order polarization voltage Vp2(i), and the third-order polarization voltage Vp3(i); Vr(i) is the ohmic voltage of the battery, I(i) is the current of the battery, and R0(i) is the ohmic internal resistance of the battery ; The time constants of the battery model are divided into the first-order time constant τ1, the second-order time constant τ2 and the third-order time constant τ3; the polarization internal resistance of the battery includes the first-order polarization internal resistance Rp1(i), the second-order polarization internal resistance Rp2(i) and the third-order polarization internal resistance Rp3(i); dt is the battery data sampling time interval, Vp1(0), Vp2(0) and Vp3(0) are the initial values ​​of the first-order polarization voltage, the second-order polarization voltage and the third-order polarization voltage respectively;

[0244] S2204, setting the objective function Fcn;

[0245] Fcn=U1(i)-U2'(i) (8) U1(i) is the measured value of the battery terminal voltage at the i-th sampling, and U2'(i) is the calculated value of the battery terminal voltage obtained based on the test data of the i-th sampling;

[0246] S2205, using the least squares method, the objective function Fcn is minimized; when the objective function Fcn is minimized, the polarization internal resistance and time constant of the corresponding battery are the polarization internal resistance and time constant identified offline:

[0247]

[0248] I(i) is the current value of the battery sampled at the i-th time, Rp is the polarization internal resistance of the battery, and the value of F(τ,Rp,I(i)) is equal to U2'(i), which represents the calculated value of the battery terminal voltage under the given time constant τ, polarization internal resistance Rp and measurement current U1(i);

[0249] By iterating the battery polarization internal resistance and time constant, the solution value of Fcn is updated until the solution value of Fcn reaches the minimum value, and the identification of the polarization internal resistance Rp and time constant τ is completed.

[0250] The working process of the self-learning unit specifically includes the following steps:

[0251] S31, the battery ohmic internal resistance R0(i) performs the first self-learning step, and the first self-learning output value is recorded as R0_SL1;

[0252] Set the upper and lower limits of the SOC interval USOC and LSOC, and the SOC(i) interval corresponding to the self-learning of the battery ohmic internal resistance R0(i) is [LSOC(i), USOC(i)]. Within the SOC(i) interval [LSOC(i), USOC(i)], the battery ohmic internal resistance corresponding to SOC(i) is R0(i); calculate the first self-learning output value of the battery ohmic internal resistance (average ohmic internal resistance value) R0_SL1(i) based on formula (11); when the error between the self-learning average ohmic internal resistance R0_SL1(i+1) and the previous self-learning average ohmic internal resistance R0_SL1(i) is less than or equal to Tol, the first self-learning calculation of the battery ohmic internal resistance is completed, and the last self-learning value of the battery is taken as the first self-learning output value of the battery ohmic internal resistance R0_SL1:

[0253]

[0254] When the condition |R0_SL1(i+1)-R0_SL1(i)|≤Tol is satisfied, R0_SL1=R0_SL1(i+1);

[0255] Tol is the set threshold; b represents the number of R0(i) corresponding to SOC(i) in the interval [LSOC(i), USOC(i)], SOC(i)∈[LSOC(i), USOC(i)];

[0256] S32, performing a second self-learning step based on the first self-learning output value R0_SL1 of the battery ohmic internal resistance, where the second self-learning output value is R0_SL2;

[0257] S33, performing a third self-learning step based on the second self-learning output value R0_SL2, and the third self-learning step output value is R0_SL3;

[0258] S32 specifically includes the following steps:

[0259] S3201, after one hour of open circuit, the battery is charged and discharged at a constant current. During the real-time operation of the battery, the terminal voltage of the battery is collected at intervals of dt seconds and recorded as U0(c). U0(c) is recorded in sequence as U0(1), U0(2), ...U0(c), U0(c+1), U0(c+2) ...U0(A). The collected current is recorded as I0(1), I0(2), ...I0(c), I0(c+1), I0(c+2) ...I0(A). The collected temperature is recorded as T 0(1), T0(2), ... T0(c), T0(c+1), T0(c+2) ... T0(A); U0(c), I0(c), T0(c) are the battery terminal voltage, current and battery surface temperature collected for the cth time when the voltage enters the constant current charge and discharge condition from the open circuit state for one hour, SOC(c) is the SOC value collected for the cth time when the voltage enters the constant current charge and discharge condition from the open circuit state for one hour; the real-time dynamic ohmic internal resistance R0 of the battery is calculated based on formula (12) d :When the battery SOC (c) ∈ ​​[LSOC, USOC] and the voltage enters the constant current charge and discharge condition from the open circuit state for one hour, calculate the battery dynamic ohmic internal resistance R0 d :

[0260]

[0261] The average value of the voltage matrix [U0(1), U0(2), ... U0(c-1)] is taken as the open circuit voltage before the battery enters the constant current charge and discharge condition, and the average value of the current matrix [I0(c), I0(c+1), I0(c+2)] is taken as the current after the battery enters the constant current charge and discharge condition;

[0262] S3202, based on the battery capacity, calculate the ohmic internal resistance R0_in after being corrected for the rated capacity, formula (13):

[0263] R0_in=CN0 / CN×R0_SL1(13)CN0 is the rated capacity of the test battery, CN is the rated capacity of a certain type of battery under test; R0_in is the ohmic internal resistance corrected by the rated capacity;

[0264] S3203, calculating the battery ohmic internal resistance second self-learning output value R0_SL2;

[0265] R0_SL2=R0_in±|R0 d -R0_in|(14)

[0266] Based on |R0 d -R0_in|Self-learning calculation of R0_SL2 value, in the calculation|R0 d -R0_in|, because |R0 d -R0_in| is related to the battery temperature, current, and battery health status. d and R0_in; when R0 d When -R0_in≥0, ​​formula (14) takes the plus sign, when R0 d When -R0_in<0, the minus sign is used in formula (14);

[0267] Step S33 specifically includes the following steps:

[0268] S3301, during the real-time operation of the battery, the battery SOC value SOC(t) at time t and the battery offline identification parameters time constant τ and polarization internal resistance Rp in step S2 are used to linearly interpolate and calculate the battery Rp(t) at time t and the battery time constants τ1(t), τ2(t), and τ3(t). Rp(t) and τ1(t), τ2(t), and τ3(t) are substituted into formulas (15)(16)(17)(18) to calculate the battery polarization voltages Vp1(t), Vp2(t), and Vp3(t) at time t.

[0269] Vp(t)=Vp1(t)+Vp2(t)+Vp3(t)(15)

[0270]

[0271] Using the battery SOC(t) at time t and the actual value of the battery open circuit voltage OCV1(t) at time t, linear interpolation is used to calculate the battery open circuit voltage OCV2(t) at time t; the battery ohmic voltage Vr(t) at time t is calculated using the second self-learning output value R0_SL2, and the calculated value of the battery terminal voltage at time t, U2(t), is obtained:

[0272] OCV1(t)=U1(t)-VP1(t)-VR1(t) (19)

[0273] OCV2(t)=U2(t)-VP2(t)-VR2(t) (20) U1(t)-VP1(t)-VR1(t)≈U2(t)-VP2(t)-VR2(t)(21)

[0274] U1(t)-U2(t)≈[VP1(t)+VR1(t)]-[VP2(t)+VR2(t)](22)

[0275] Vr(t)=I(t)×R0_SL2(23) The polarization internal resistance Rp(t) of the battery at time t includes the first-order polarization internal resistance Rp1(t), the second-order polarization internal resistance Rp2(t) and the third-order polarization internal resistance Rp3(t); τ1(t), τ2(t) and τ3(t) are the first-order time constant, the second-order time constant and the third-order time constant at time t respectively; Vp1(t), Vp2(t) and Vp3(t) are the first-order polarization voltage, the second-order polarization voltage and the third-order polarization voltage at time t respectively; I(t) is the current acquisition value at time t; t represents the real-time acquisition time;

[0276] S3302, using the battery terminal voltage U1(t) collected at time t and the value U2(t) calculated at time t in step S3301 during actual operation, calculate the battery voltage difference DV(t) at time t, and use DV(t) at time t to self-learn R0_SL3;

[0277] Battery SOC calibration refers to the calibration of the battery SOC after the battery is fully charged or discharged, and the calculation of the battery voltage difference DV(t) at time t:

[0278] DV(t)=U1(t)-U2(t)(24)

[0279] DR0(t)=DV(t) / I(t)(25)

[0280]

[0281] Among them, OCV1(t) is the actual value of the open circuit voltage of the battery at time t, OCV2(t) is the calculated value of the open circuit voltage of the battery at time t, VP1(t) is the actual value of the polarization voltage of the battery at time t under zero-state response, VP2(t) is the calculated value of the polarization voltage of the battery at time t, VP2(t)=Vp1(t)+Vp2(t)+Vp3(t), VR1(t) is the actual value of the ohmic voltage of the battery at time t under zero-state response, VR2(t) is the calculated value of the ohmic voltage of the battery at time t, and VR2(t) is equal to Vr(t); DR0(t) represents the ratio of the voltage difference DV(t) of the battery at time t to the current collected value I(t) of the battery at time t, and represents the ohmic internal resistance compensation value calculated at time t; after the battery is fully charged or discharged, the SOC error is eliminated, and OCV1(t) is equal to OCV2(t);

[0282] The value range of t is [1, A]. The SOC change length is set according to different types of batteries during the process of t changing from 1 to A, and the number of calculations is A. Considering the error caused by the single compensation value DR0(t) of the ohmic internal resistance, the data during the process of t changing from 1 to A is selected to compensate the ohmic internal resistance R0.

[0283] The error DV(t) between the measured and calculated values ​​of the battery polarization voltage and ohmic voltage is iteratively calculated, and the ohmic internal resistance is iteratively compensated, R0_SL3 = [DR0(1)+DR0(2)+…DR0(t)…+DR0(A-1)+DR0(A)] / A.

[0284] Thus far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present invention.

[0285] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures, and techniques are not shown in detail so as not to obscure the understanding of this description.

[0286] Similarly, it should be understood that in order to streamline the present disclosure and aid understanding of one or more of the various inventive aspects, in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the claims, inventive aspects lie in less than all the features of the individual embodiments disclosed above. Accordingly, the claims that follow the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of the invention.

[0287] Those skilled in the art will appreciate that the modules, units, or groups of devices in the examples disclosed herein may be arranged in the device described in the embodiment, or alternatively may be located in one or more devices different from the devices in the examples. The modules in the aforementioned examples may be combined into one module or further divided into multiple submodules.

[0288] It will be appreciated by those skilled in the art that the modules in the devices of the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or groups in the embodiments may be combined into one module or unit or group, and further may be divided into a plurality of submodules or subunits or subgroups. All features disclosed in this specification (including the accompanying claims, abstract and drawings) and all processes or units of any method or device so disclosed may be combined in any combination, except that at least some of such features and / or processes or units are mutually exclusive. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0289] Furthermore, those skilled in the art will appreciate that although some embodiments described herein include certain features included in other embodiments but not other features, combinations of features from different embodiments are intended to be within the scope of the present invention and to form different embodiments. For example, in the claims below, any of the claimed embodiments may be used in any combination.

[0290] In addition, some of the embodiments are described herein as methods or combinations of method elements that can be implemented by a processor of a computer system or by other devices that perform the functions described. Thus, a processor having the necessary instructions for implementing the method or method element forms a device for implementing the method or method element. Furthermore, the elements described herein of the device embodiments are examples of devices for implementing the functions performed by the elements for the purpose of implementing the invention.

[0291] The various techniques described herein may be implemented in conjunction with hardware or software, or a combination thereof. Thus, the methods and apparatus of the present invention, or certain aspects or portions of the methods and apparatus of the present invention, may take the form of program code (i.e., instructions) embedded in a tangible medium, such as a floppy disk, CD-ROM, hard drive, or any other machine-readable storage medium, wherein when the program is loaded into a machine such as a computer and executed by the machine, the machine becomes an apparatus for practicing the present invention.

[0292] When the program code is executed on a programmable computer, the computing device generally includes a processor, a storage medium readable by the processor (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device. The memory is configured to store the program code; and the processor is configured to execute the method of the present invention according to the instructions in the program code stored in the memory.

[0293] By way of example and not limitation, computer-readable media include computer storage media and communication media. Computer-readable media include computer storage media and communication media. Computer storage media stores information such as computer-readable instructions, data structures, program modules, or other data. Communication media generally embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and includes any information delivery media. Combinations of any of the above are also included within the scope of computer-readable media.

[0294] As used herein, unless otherwise specified, the use of ordinal numbers "first," "second," "third," etc. to describe common objects merely indicates that different instances of similar objects are involved and are not intended to imply that the objects so described must have a given order in time, space, ranking, or in any other manner.

[0295] Although the present invention has been described with respect to a limited number of embodiments, it will be apparent to those skilled in the art, having benefit of the foregoing description, that other embodiments are contemplated within the scope of the invention thus described. Furthermore, it should be noted that the language used in this specification has been selected primarily for readability and didactic purposes, rather than for the purpose of explaining or limiting the subject matter of the present invention. Consequently, many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the appended claims. The disclosure of the present invention is intended to be illustrative rather than restrictive of the scope of the invention, which is defined by the appended claims.

Claims

1. A battery equivalent circuit model parameter self-learning method, characterized in that: The following steps are included: S1, take a lithium iron phosphate battery, perform charge and discharge tests on the battery, and obtain test data at time intervals of dt seconds. The test data includes the battery terminal voltage measurement value, current and battery surface temperature; S2, based on the third-order RC equivalent circuit model, performs offline identification of the battery's ohmic internal resistance based on Ohm's law and test data. It also performs offline identification of the battery's polarization internal resistance and time constant based on test data using the least squares method. S3, self-learning of the battery's ohmic internal resistance; The step S3 specifically includes the following steps: S31, performing the first self-learning of the battery ohmic internal resistance R0(i), and the output value of the first self-learning is recorded as R0_SL1; Set the upper and lower limits of the SOC interval USOC and LSOC, and the SOC(i) interval corresponding to the self-learning of the battery ohmic internal resistance R0(i) is [LSOC(i), USOC(i)]. Within the SOC(i) interval [LSOC(i), USOC(i)], the battery ohmic internal resistance corresponding to SOC(i) is R0(i); i represents the i-th sampling, and the first self-learning output value of the battery ohmic internal resistance is calculated based on formula (11), and the first self-learning output value of the battery ohmic internal resistance is the average value of the ohmic internal resistance R0_SL1(i); when the error between the self-learning average value of the ohmic internal resistance R0_SL1(i+1) and the previous self-learning average value of the ohmic internal resistance R0_SL1(i) is less than or equal to Tol, the first self-learning calculation of the battery ohmic internal resistance is completed, and the last self-learning value of the battery is taken as the first self-learning output value of the battery ohmic internal resistance R0_SL1: When the condition |R0_SL1(i+1)-R0_SL1(i)|≤Tol is satisfied, R0_SL1=R0_SL1(i+1); Tol is the set threshold; b represents the number of R0(i) corresponding to SOC(i) in the interval [LSOC(i), USOC(i)], SOC(i)∈[LSOC(i), USOC(i)]; S32, performing a second self-learning step based on the first self-learning output value R0_SL1 of the battery ohmic internal resistance, where the second self-learning output value is R0_SL2; S33, performing a third self-learning step based on the second self-learning output value R0_SL2, and the third self-learning step output value is R0_SL3.

2. A battery equivalent circuit model parameter self-learning method according to claim 1, characterized in that: S2 specifically includes the following steps: S21, calculate the battery ohmic internal resistance R0(i): R0(i)=[U1(i)-U1(i+1)] / [I(i)-I(i+1)](1) U1(i) represents the battery terminal voltage measurement value of the i-th sampling, U1(i+1) represents the battery terminal voltage measurement value of the i+1-th sampling, I(i) represents the battery current measurement value of the i-th sampling, I(i+1) represents the battery current measurement value of the i+1-th sampling, and R0(i) represents the ohmic internal resistance corresponding to SOC(i); S22, identify the time constant and polarization internal resistance.

3. A battery equivalent circuit model parameter self-learning method according to claim 2, characterized in that: The identification of the time constant and polarization internal resistance specifically includes the following steps: S2201, setting a range for the polarization internal resistance and time constant of the battery. When identifying the polarization internal resistance and time constant of the battery within the set range, set the polarization internal resistance range [lb1, ub1] and the time constant value range [lb2, ub2]. S2202, setting the initial value of the polarization internal resistance and the initial value of the time constant of the battery. The design value of the initial value of the polarization internal resistance is between the lower boundary lb1 of the polarization internal resistance and the upper boundary ub1 of the polarization internal resistance. The setting value of the initial value of the time constant is between the lower boundary lb2 of the time constant and the upper boundary ub2 of the time constant. S2203, the terminal voltage measurement value of the battery sampled at the i-th time is U1(i), and the terminal voltage calculated value of the battery based on the test data is U2'(i). The battery's initial polarization internal resistance value and the initial time constant value are used to calculate the battery terminal voltage calculated value U2'(i) based on the test data: U2'(i)=OCV(i)+Vp(i)+Vr(i)(2) Vr(i)=I(i)×R0(i)(3) Vp(i)=Vp1(i)+Vp2(i)+Vp3(i)(4) U2'(i) is the calculated value of the battery terminal voltage at the i-th sampling, OCV1(i) is the measured value of the battery open circuit voltage at the i-th sampling, Vp(i) is the polarization voltage of the battery at the i-th sampling, and the polarization voltage includes the first-order polarization voltage Vp1(i), the second-order polarization voltage Vp2(i), and the third-order polarization voltage Vp3(i); Vr(i) is the ohmic voltage of the battery, I(i) is the current of the battery, and R0(i) is the ohmic internal resistance of the battery; the time constant of the battery model is divided into the first-order time constant τ1, the second-order time constant τ2, and the third-order time constant τ3; the polarization internal resistance of the battery includes the first-order polarization internal resistance Rp1(i), the second-order polarization internal resistance Rp2(i), and the third-order polarization internal resistance Rp3(i); dt is the battery data sampling time interval; S2204, setting the objective function Fcn; Fcn=U1(i)-U2'(i)(8) U1(i) is the measured value of the battery terminal voltage at the i-th sampling, and U2'(i) is the calculated value of the battery terminal voltage obtained based on the test data of the i-th sampling; S2205, using the least squares method, the objective function Fcn is minimized; when the objective function Fcn is minimized, the polarization internal resistance and time constant of the corresponding battery are the polarization internal resistance and time constant identified offline: I(i) is the current value of the battery sampled at the i-th time, Rp is the polarization internal resistance of the battery, and the value of F(τ,Rp,I(i)) is equal to U2'(i), which represents the calculated value of the battery terminal voltage under the given time constant τ, polarization internal resistance Rp and measurement current I(i); By iterating the battery polarization internal resistance and time constant, the solution value of Fcn is updated until the solution value of Fcn reaches the minimum value, and the identification of the polarization internal resistance Rp and time constant τ is completed.

4. A battery equivalent circuit model parameter self-learning method according to claim 1, characterized in that: S32 specifically includes the following steps: S3201, after one hour of open circuit, the battery is charged and discharged at a constant current. During the real-time operation of the battery, the terminal voltage of the battery is collected at intervals of dt seconds and recorded as U0(c). U0(c) is recorded in sequence as U0(1), U0(2), ...U0(c), U0(c+1), U0(c+2) ...U0(A). The collected current is recorded as I0(1), I0(2), ...I0(c), I0(c+1), I0(c+2) ...I0(A). The collected temperature is recorded as T 0(1), T0(2), ... T0(c), T0(c+1), T0(c+2) ... T0(A); U0(c), I0(c), T0(c) are the battery terminal voltage, current and battery surface temperature collected for the cth time when the voltage enters the constant current charge and discharge condition from the open circuit state for one hour, SOC(c) is the SOC value collected for the cth time when the voltage enters the constant current charge and discharge condition from the open circuit state for one hour; the real-time dynamic ohmic internal resistance R0 of the battery is calculated based on formula (12) d :When the battery SOC (c) ∈ ​​[LSOC, USOC] and the voltage enters the constant current charge and discharge condition from the open circuit state for one hour, calculate the battery dynamic ohmic internal resistance R0 d : The average value of the voltage matrix [U0(1), U0(2), ... U0(c-1)] is taken as the open circuit voltage before the battery enters the constant current charge and discharge condition, and the average value of the current matrix [I0(c), I0(c+1), I0(c+2)] is taken as the current after the battery enters the constant current charge and discharge condition; S3202, based on the battery capacity, calculate the ohmic internal resistance R0_in after being corrected for the rated capacity, formula (13): R0_in=CN0 / CN×R0_SL1(13) CN0 is the rated capacity of the test battery, CN is the rated capacity of a certain type of battery under test; R0_in is the ohmic internal resistance corrected by the rated capacity; S3203, calculating the battery ohmic internal resistance second self-learning output value R0_SL2; R0_SL2=R0_in±|R0 d -R0_in|(14) Based on |R0 d -R0_in|Self-learning calculation of R0_SL2 value, in the calculation|R0 d -R0_in|, because |R0 d -R0_in| is related to the battery temperature, current, and battery health status. d and R0_in; when R0 d When -R0_in≥0, ​​formula (14) takes the plus sign, when R0 d When -R0_in<0, the minus sign is used in formula (14).

5. A battery equivalent circuit model parameter self-learning method according to claim 4, characterized in that: Step S33 specifically includes the following steps: S3301, during real-time operation of the battery, the battery SOC value SOC(t) at time t and the battery offline identification parameters time constant τ and polarization internal resistance Rp in step S2 are used to linearly interpolate and calculate the battery Rp(t) at time t and the battery time constants τ1(t), τ2(t), and τ3(t) at time t; Rp(t) and τ1(t), τ2(t), and τ3(t) are substituted into formulas (15), (16), (17), and (18) to calculate the battery polarization voltages Vp1(t), Vp2(t), and Vp3(t) at time t; Vp(t)=Vp1(t)+Vp2(t)+Vp3(t)(15) Using the battery SOC(t) at time t and the actual value of the battery open circuit voltage OCV1(t) at time t, linear interpolation is used to calculate the battery open circuit voltage OCV2(t) at time t; the battery ohmic voltage Vr(t) at time t is calculated using the second self-learning output value R0_SL2, and the calculated value of the battery terminal voltage at time t, U2(t), is obtained: OCV1(t)=U1(t)-VP1(t)-VR1(t) (19) OCV2(t)=U2(t)-VP2(t)-VR2(t) (20) U1(t)-VP1(t)-VR1(t)≈U2(t)-VP2(t)-VR2(t)(21) U1(t)-U2(t)≈[VP1(t)+VR1(t)]-[VP2(t)+VR2(t)](22) Vr(t)=I(t)×R0_SL2(23) The polarization internal resistance Rp(t) of the battery at time t includes the first-order polarization internal resistance Rp1(t), the second-order polarization internal resistance Rp2(t) and the third-order polarization internal resistance Rp3(t); τ1(t), τ2(t) and τ3(t) are the first-order time constant, second-order time constant and third-order time constant at time t, respectively; Vp1(t), Vp2(t) and Vp3(t) are the first-order polarization voltage, second-order polarization voltage and third-order polarization voltage at time t, respectively; I(t) is the current acquisition value at time t; t represents the real-time acquisition time; S3302, using the battery terminal voltage U1(t) collected at time t and the value U2(t) calculated at time t in step S3301 during actual operation, calculate the battery voltage difference DV(t) at time t, and use DV(t) at time t to self-learn R0_SL3; Battery SOC calibration refers to the calibration of the battery SOC after the battery is fully charged or discharged, and the calculation of the battery voltage difference DV(t) at time t: DV(t)=U1(t)-U2(t)(24) DR0(t)=DV(t) / I(t)(25) Among them, OCV1(t) is the actual value of the open circuit voltage of the battery at time t, OCV2(t) is the calculated value of the open circuit voltage of the battery at time t, VP1(t) is the actual value of the polarization voltage of the battery at time t under zero-state response, VP2(t) is the calculated value of the polarization voltage of the battery at time t, VP2(t)=Vp1(t)+Vp2(t)+Vp3(t), VR1(t) is the actual value of the ohmic voltage of the battery at time t under zero-state response, VR2(t) is the calculated value of the ohmic voltage of the battery at time t, and VR2(t) is equal to Vr(t); DR0(t) represents the ratio of the voltage difference DV(t) of the battery at time t to the current collected value I(t) of the battery at time t, and represents the ohmic internal resistance compensation value calculated at time t; after the battery is fully charged or discharged, the SOC error is eliminated, and OCV1(t) is equal to OCV2(t); The value range of t is [1, A]. The SOC change length is set according to different types of batteries during the process of t changing from 1 to A, and the number of calculations is A. Considering the error caused by the single compensation value DR0(t) of the ohmic internal resistance, the data during the process of t changing from 1 to A is selected to compensate the ohmic internal resistance R0. The error DV(t) between the measured and calculated values ​​of the battery polarization voltage and ohmic voltage is iteratively calculated, and the ohmic internal resistance is iteratively compensated, R0_SL3 = [DR0(1)+DR0(2)+…DR0(t)…+DR0(A-1)+DR0(A)] / A.

6. A battery equivalent circuit model parameter self-learning system, characterized in that: It includes data acquisition unit, offline identification unit and self-learning unit; The data acquisition unit takes a lithium iron phosphate battery, performs charge and discharge tests on the battery, and obtains test data at intervals of dt seconds. The test data includes the battery terminal voltage measurement value, current, and battery surface temperature; The offline identification unit uses a third-order RC equivalent circuit model to perform offline identification of the battery's ohmic internal resistance based on Ohm's law and test data. It also uses the least squares method to perform offline identification of the battery's polarization internal resistance and time constant based on test data. The self-learning unit performs self-learning on the battery's ohmic internal resistance; The working process of the self-learning unit specifically includes the following steps: S31, performing the first self-learning of the battery ohmic internal resistance R0(i), and the output value of the first self-learning is recorded as R0_SL1; The upper and lower limits of the SOC interval USOC and LSOC are set, and the SOC(i) interval corresponding to the self-learning of the battery ohmic internal resistance R0(i) is [LSOC(i), USOC(i)]. In the SOC(i) interval [LSOC(i), USOC(i)], the battery ohmic internal resistance corresponding to SOC(i) is R0(i); the first self-learning output value of the battery ohmic internal resistance is calculated based on formula (11), and the calculated first self-learning output value of the battery ohmic internal resistance is the average value of the ohmic internal resistance R0_SL1(i); when the error between the self-learning average value of the ohmic internal resistance R0_SL1(i+1) and the previous self-learning average value of the ohmic internal resistance R0_SL1(i) is less than or equal to Tol, the first self-learning calculation of the battery ohmic internal resistance is completed, and the last self-learning value of the battery is taken as the first self-learning output value of the battery ohmic internal resistance R0_SL1: When the condition |R0_SL1(i+1)-R0_SL1(i)|≤Tol is satisfied, R0_SL1=R0_SL1(i+1); Tol is the set threshold; b represents the number of R0(i) corresponding to SOC(i) in the interval [LSOC(i), USOC(i)], SOC(i)∈[LSOC(i), USOC(i)]; S32, performing a second self-learning step based on the first self-learning output value R0_SL1 of the battery ohmic internal resistance, where the second self-learning output value is R0_SL2; S33, performing a third self-learning step based on the second self-learning output value R0_SL2, and the third self-learning step output value is R0_SL3.

7. A battery equivalent circuit model parameter self-learning system according to claim 6, characterized in that: The working process of the offline identification unit specifically includes the following steps: S21, calculate the battery ohmic internal resistance R0(i): R0(i)=[U1(i)-U1(i+1)] / [I(i)-I(i+1)](1) U1(i) represents the battery terminal voltage measurement value of the i-th sampling, U1(i+1) represents the battery terminal voltage measurement value of the i+1-th sampling, I(i) represents the battery current measurement value of the i-th sampling, I(i+1) represents the battery current measurement value of the i+1-th sampling, and R0(i) represents the ohmic internal resistance corresponding to SOC(i); S22, identification of time constant and polarization internal resistance; The identification of the time constant and polarization internal resistance specifically includes the following steps: S2201, setting a range for the polarization internal resistance and time constant of the battery. When identifying the polarization internal resistance and time constant of the battery within the set range, set the polarization internal resistance range [lb1, ub1] and the time constant value range [lb2, ub2]. S2202, setting the initial value of the polarization internal resistance and the initial value of the time constant of the battery. The design value of the initial value of the polarization internal resistance is between the lower boundary lb1 of the polarization internal resistance and the upper boundary ub1 of the polarization internal resistance. The setting value of the initial value of the time constant is between the lower boundary lb2 of the time constant and the upper boundary ub2 of the time constant. S2203, the terminal voltage measurement value of the battery sampled at the i-th time is U1(i), and the terminal voltage calculated value of the battery based on the test data is U2'(i). The battery's initial polarization internal resistance value and the initial time constant value are used to calculate the battery terminal voltage calculated value U2'(i) based on the test data: U2'(i)=OCV(i)+Vp(i)+Vr(i)(2) Vr(i)=I(i)×R0(i)(3) Vp(i)=Vp1(i)+Vp2(i)+Vp3(i)(4) U2'(i) is the calculated value of the battery terminal voltage at the i-th sampling, OCV1(i) is the measured value of the battery open circuit voltage at the i-th sampling, Vp(i) is the polarization voltage of the battery at the i-th sampling, and the polarization voltage includes the first-order polarization voltage Vp1(i), the second-order polarization voltage Vp2(i), and the third-order polarization voltage Vp3(i); Vr(i) is the ohmic voltage of the battery, I(i) is the current of the battery, and R0(i) is the ohmic internal resistance of the battery; the time constant of the battery model is divided into the first-order time constant τ1, the second-order time constant τ2, and the third-order time constant τ3; the polarization internal resistance of the battery includes the first-order polarization internal resistance Rp1(i), the second-order polarization internal resistance Rp2(i), and the third-order polarization internal resistance Rp3(i); dt is the battery data sampling time interval; S2204, setting the objective function Fcn; Fcn=U1(i)-U2'(i)(8) U1(i) is the measured value of the battery terminal voltage at the i-th sampling, and U2'(i) is the calculated value of the battery terminal voltage obtained based on the test data of the i-th sampling; S2205, using the least squares method, the objective function Fcn is minimized; when the objective function Fcn is minimized, the polarization internal resistance and time constant of the corresponding battery are the polarization internal resistance and time constant identified offline: I(i) is the current value of the battery sampled at the i-th time, Rp is the polarization internal resistance of the battery, and the value of F(τ,Rp,I(i)) is equal to U2'(i), which represents the calculated value of the battery terminal voltage under the given time constant τ, polarization internal resistance Rp and measurement current I(i); By iterating the battery polarization internal resistance and time constant, the solution value of Fcn is updated until the solution value of Fcn reaches the minimum value, and the identification of the polarization internal resistance Rp and time constant τ is completed.

8. A battery equivalent circuit model parameter self-learning system according to claim 7, characterized in that: S32 specifically includes the following steps: S3201, after one hour of open circuit, the battery is charged and discharged at a constant current. During the real-time operation of the battery, the terminal voltage of the battery is collected at intervals of dt seconds and recorded as U0(c). U0(c) is recorded in sequence as U0(1), U0(2), ...U0(c), U0(c+1), U0(c+2) ...U0(A). The collected current is recorded as I0(1), I0(2), ...I0(c), I0(c+1), I0(c+2) ...I0(A). The collected temperature is recorded as T 0(1), T0(2), ... T0(c), T0(c+1), T0(c+2) ... T0(A); U0(c), I0(c), T0(c) are the battery terminal voltage, current and battery surface temperature collected for the cth time when the voltage enters the constant current charge and discharge condition from the open circuit state for one hour, SOC(c) is the SOC value collected for the cth time when the voltage enters the constant current charge and discharge condition from the open circuit state for one hour; the real-time dynamic ohmic internal resistance R0 of the battery is calculated based on formula (12) d :When the battery SOC (c) ∈ ​​[LSOC, USOC] and the voltage enters the constant current charge and discharge condition from the open circuit state for one hour, calculate the battery dynamic ohmic internal resistance R0 d : The average value of the voltage matrix [U0(1), U0(2), ... U0(c-1)] is taken as the open circuit voltage before the battery enters the constant current charge and discharge condition, and the average value of the current matrix [I0(c), I0(c+1), I0(c+2)] is taken as the current after the battery enters the constant current charge and discharge condition; S3202, based on the battery capacity, calculate the ohmic internal resistance R0_in after being corrected for the rated capacity, formula (13): R0_in=CN0 / CN×R0_SL1(13) CN0 is the rated capacity of the test battery, CN is the rated capacity of a certain type of battery under test; R0_in is the ohmic internal resistance corrected by the rated capacity; S3203, calculating the battery ohmic internal resistance second self-learning output value R0_SL2; R0_SL2=R0_in±|R0 d -R0_in|(14) Based on |R0 d -R0_in|Self-learning calculation of R0_SL2 value, in the calculation|R0 d -R0_in|, because |R0 d -R0_in| is related to the battery temperature, current, and battery health status. d and R0_in; when R0 d When -R0_in≥0, ​​formula (14) takes the plus sign, when R0 d When -R0_in<0, the minus sign is used in formula (14); Step S33 specifically includes the following steps: S3301, during real-time operation of the battery, the battery SOC value SOC(t) at time t and the battery offline identification parameters time constant τ and polarization internal resistance Rp in step S2 are used to linearly interpolate and calculate the battery Rp(t) at time t and the battery time constants τ1(t), τ2(t), and τ3(t) at time t; Rp(t) and τ1(t), τ2(t), and τ3(t) are substituted into formulas (15), (16), (17), and (18) to calculate the battery polarization voltages Vp1(t), Vp2(t), and Vp3(t) at time t; Vp(t)=Vp1(t)+Vp2(t)+Vp3(t)(15) Using the battery SOC(t) at time t and the actual value of the battery open circuit voltage OCV1(t) at time t, linear interpolation is used to calculate the battery open circuit voltage OCV2(t) at time t; the battery ohmic voltage Vr(t) at time t is calculated using the second self-learning output value R0_SL2, and the calculated value of the battery terminal voltage at time t, U2(t), is obtained: OCV1(t)=U1(t)-VP1(t)-VR1(t) (19) OCV2(t)=U2(t)-VP2(t)-VR2(t) (20) U1(t)-VP1(t)-VR1(t)≈U2(t)-VP2(t)-VR2(t)(21) U1(t)-U2(t)≈[VP1(t)+VR1(t)]-[VP2(t)+VR2(t)](22) Vr(t)=I(t)×R0_SL2(23) The polarization internal resistance Rp(t) of the battery at time t includes the first-order polarization internal resistance Rp1(t), the second-order polarization internal resistance Rp2(t) and the third-order polarization internal resistance Rp3(t); τ1(t), τ2(t) and τ3(t) are the first-order time constant, second-order time constant and third-order time constant at time t, respectively; Vp1(t), Vp2(t) and Vp3(t) are the first-order polarization voltage, second-order polarization voltage and third-order polarization voltage at time t, respectively; I(t) is the current acquisition value at time t; t represents the real-time acquisition time; S3302, using the battery terminal voltage U1(t) collected at time t and the value U2(t) calculated at time t in step S3301 during actual operation, calculate the battery voltage difference DV(t) at time t, and use DV(t) at time t to self-learn R0_SL3; Battery SOC calibration refers to the calibration of the battery SOC after the battery is fully charged or discharged, and the calculation of the battery voltage difference DV(t) at time t: DV(t)=U1(t)-U2(t)(24) DR0(t)=DV(t) / I(t)(25) Among them, OCV1(t) is the actual value of the open circuit voltage of the battery at time t, OCV2(t) is the calculated value of the open circuit voltage of the battery at time t, VP1(t) is the actual value of the polarization voltage of the battery at time t under zero-state response, VP2(t) is the calculated value of the polarization voltage of the battery at time t, VP2(t)=Vp1(t)+Vp2(t)+Vp3(t), VR1(t) is the actual value of the ohmic voltage of the battery at time t under zero-state response, VR2(t) is the calculated value of the ohmic voltage of the battery at time t, and VR2(t) is equal to Vr(t); DR0(t) represents the ratio of the voltage difference DV(t) of the battery at time t to the current collected value I(t) of the battery at time t, and represents the ohmic internal resistance compensation value calculated at time t; after the battery is fully charged or discharged, the SOC error is eliminated, and OCV1(t) is equal to OCV2(t); The value range of t is [1, A]. The SOC change length is set according to different types of batteries during the process of t changing from 1 to A, and the number of calculations is A. Considering the error caused by the single compensation value DR0(t) of the ohmic internal resistance, the data during the process of t changing from 1 to A is selected to compensate the ohmic internal resistance R0. The error DV(t) between the measured and calculated values ​​of the battery polarization voltage and ohmic voltage is iteratively calculated, and the ohmic internal resistance is iteratively compensated, R0_SL3 = [DR0(1)+DR0(2)+…DR0(t)…+DR0(A-1)+DR0(A)] / A.

Images