Offline measuring method for available capacity of power battery, device, and storage medium

Abstract

The present invention relates to the technical field of power battery measurement. Disclosed are an offline measuring method for available capacity of a power battery, a device, and a storage medium. A circuit initial model is established by means of pulse working conditions, actual charging data is used to train the initial model, data of the battery in a constant-current charging state is used as negative feedback to correct battery model parameters, then a battery end voltage is simulated by means of the corrected model, and calculation is performed to obtain the actual available capacity of the battery and the state of health of the battery. According to the present invention, on the basis of existing voltages and Coulomb counting algorithms, a dynamic learning mechanism of battery parameters is added, reducing model prediction errors, enabling a predicted voltage to be closest to an actual voltage, solving the problem of the impedance of a battery changing over time, rapidly measuring the available capacity of the battery, and obtaining the state of health of the battery.

Description

A method, device and storage medium for offline detection of available capacity of a power battery Technical Field

[0001] The present invention relates to the technical field of power battery detection, and in particular to an offline detection method, device and storage medium for the available capacity of a power battery. Background Art

[0002] The degradation of power batteries during use is affected by factors such as material aging, driving habits, and operating conditions. Therefore, even batteries of the same brand / model may experience different degradation conditions. The traditional battery capacity full charge and discharge test method is simple to implement, but has a high time cost and does not meet the current requirements for rapid testing. The method based on vehicle operation big data is still limited by data quality issues and cannot be applied to the battery residual value assessment of all vehicles.

[0003] Existing technology uses voltage and coulomb counting algorithms to quickly detect the available capacity of power batteries. This detection method is widely used in handheld devices such as mobile phones. Usually, after a period of use, the battery impedance will change, affecting the detection accuracy. The formula shows that the battery voltage: V BAT =V ocv -I*R BAT , where V ocv is the battery open circuit voltage, R BAT is the battery's internal DC impedance. As shown in Figure 1, the voltage of an aged battery is lower than that of a new one, causing the system to shut down earlier. The coulomb counting method, on the other hand, continuously integrates the coulombs to calculate the consumed charge and the state of charge (SOC). Since the total capacity is known, the remaining capacity can be determined. However, this method struggles to accurately quantify (model) self-discharge. Furthermore, because it lacks periodic full-cycle calibration, measurement errors increase over time.

[0004] Summary of the Invention

[0005] In order to solve the problem in the prior art that it is difficult to accurately quantify the self-discharge amount and the measurement error is large when detecting the available capacity of a power battery, the present invention provides an offline detection method, device and storage medium for the available capacity of a power battery.

[0006] The technical solutions of the present invention are as follows:

[0007] The present invention provides a method for offline detection of available capacity of a power battery, comprising the following steps:

[0008] S1: obtain the battery's initial charging state SOCstart, the battery's data in the pulse charging state, and the battery's data in the constant current charging state;

[0009] S2: After parameter identification based on the data of the battery in the pulse charging state, a battery model is obtained;

[0010] S3: The battery terminal voltage is obtained through the battery model prediction, and the data of the battery under constant current charging state is used as negative feedback to correct the battery model parameters;

[0011] S4: Simulate the battery model to obtain the actual available capacity and health status of the battery.

[0012] Furthermore, when the battery is pulse charged, the pulse operating condition is as follows:

[0013] Charge at 1 / 3C constant current for 10-60 seconds, then let it rest for 2-10 minutes.

[0014] Charge at 1C constant current for 10-60 seconds, then let it rest for 2-10 minutes.

[0015] Where C represents the charge and discharge rate;

[0016] And obtain the actual charging current value, voltage value and internal resistance of the battery during the pulse working condition.

[0017] Furthermore, when the battery is charged at a constant current, the constant current working condition is as follows:

[0018] Charge at a specific constant current for 5 to 30 minutes, and let it rest for 5 to 10 minutes;

[0019] And obtain the actual charging current value, voltage value and internal resistance of the battery under constant current working conditions.

[0020] Furthermore, the battery model predicts the battery terminal voltage, including:

[0021] I. Determine the SOC at the next moment using formula (1): SOC next =SOC start +fIΔt / Q design (1);

[0022] SOC next is the estimated battery SOC value at the next moment, fIΔt is the integrated capacity at adjacent moments, Q design It is the nominal capacity or design capacity of the battery;

[0023] II. Obtain the battery terminal voltage at the next moment through formula (2): U est =U oc (SOC next ,T)+IR0(SOC next ,T) (2);

[0024] U oc (SOCnext ,T) is the battery open circuit voltage at the next moment, which has a functional relationship with the battery SOC and temperature T. R0(SOC next ,T) is the battery internal resistance at the next moment, which has a functional relationship with the battery SOC and temperature T, U est is the predicted value of the battery terminal voltage at the next moment.

[0025] Furthermore, the voltage data of the battery under constant current charging is obtained, and the voltage data is used as negative feedback to correct the battery model parameters. The negative feedback link is: △U=U est -U real (3);

[0026] U real The voltage data actually detected when the battery is in a constant current charging state;

[0027] Determine the size of the predicted battery terminal voltage and adjust R0 or Q on the battery model based on the judgment result. design After correction, obtain the available battery capacity.

[0028] Furthermore, the battery model parameters are corrected according to the judgment result, including:

[0029] 1) When the predicted voltage is too high and greater than a certain threshold ErrUlimit, reduce the battery internal resistance R0, and the reduction is proportional to the predicted voltage error △U. When the predicted voltage is too low and less than a certain threshold ErrUlimit, increase the battery internal resistance R0, and the increase is proportional to the predicted voltage error △U.

[0030] 2) When the predicted voltage is too high and between [0 and ErrUlimit], increase the battery capacity value Q design , the capacity increase is proportional to the voltage prediction error △U. When the predicted voltage is too small and between [-ErrUlimit~0], reduce the battery capacity value Q design , the magnitude of capacity reduction is proportional to the voltage prediction error △U;

[0031] 3) Update the battery internal resistance or capacity according to step 1 or step 2 to obtain the corrected battery electromotive force OCV correct , according to OCV correct Calibrate SOC to get SOC correct , when the SOC change reaches the target value SOC required (0~1), calculate the battery available capacity Q available : Q available =fIΔt / (SOC correct -SOC start ) (4).

[0032] Furthermore, when performing simulation calculation on the battery model to obtain the actual available capacity and battery health status of the battery, the following steps are included:

[0033] The simulation starting state of charge SOC is determined by formula (5) sim0 :SOC sim0 =SOC start +fIΔt / Q available (5);

[0034] The battery terminal voltage is simulated by incrementing the SOC interval ΔSOC: U = U oc (SOC sim0 +ΔSOC)+I R0 (SOC sim0 +ΔSOC) (6);

[0035] When U<=U term When the battery charge end SOC is obtained, it is recorded as EOC (End of charge), and the remaining charging time Δt', U term Charging cut-off voltage for the battery system;

[0036] Calculate the remaining battery capacity Q by formula (7) rem : Q rem =fIΔt' (7);

[0037] By correcting the initial charge battery capacity: Q start =SOC start *Q available (8);

[0038] Calculate the actual available capacity of the battery: Q1=Q start +fIΔt+Q rem (9); Q max =CapCoff2*Q1+(1-CapCoff2)*Q2 (11);

[0039] Among them, CapCoff1 and CapCoff2 are weighted coefficients with values ​​of (0 to 1), fIΔt' is the integral capacity during the simulation process, and Q max is the actual available capacity of the battery after weighting.

[0040] That is, the battery health status is estimated as:

[0041] The present invention further provides an electronic device, comprising:

[0042] memory for storing computer programs;

[0043] A processor is configured to implement the steps of any of the above-mentioned methods for offline detection of the available capacity of a power battery when executing the computer program.

[0044] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the offline detection method for the available capacity of a power battery as described in any one of the above items are implemented.

[0045] The beneficial effects of the present invention include at least: the solution provided by the present invention adds a dynamic learning mechanism for battery parameters on the basis of the existing voltage and coulomb counting algorithm, predicts the battery terminal voltage by establishing a battery model, and then uses the actual voltage data of the battery under constant current charging state as negative feedback to correct the battery model parameters, thereby reducing the model prediction error, making the predicted voltage closest to the actual voltage, solving the problem of battery impedance changing over time, quickly detecting the available capacity of the battery, and obtaining the health status of the battery. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a schematic diagram of battery cycle discharge characteristics.

[0047] FIG2 is a schematic flow chart of an offline detection method for the available capacity of a power battery provided by the present invention.

[0048] FIG3 is a schematic diagram of a battery model provided by the present invention. DETAILED DESCRIPTION

[0049] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0050] 2-3 , the present invention provides a method for offline detection of available capacity of a power battery, comprising the following steps:

[0051] S1: obtain the battery's initial charging state SOCstart, the battery's data in the pulse charging state, and the battery's data in the constant current charging state;

[0052] S2: After parameter identification based on the data of the battery in the pulse charging state, a battery model is obtained;

[0053] S3: The battery terminal voltage is obtained through the battery model prediction, and the data of the battery under constant current charging state is used as negative feedback to correct the battery model parameters;

[0054] S4: Simulate the battery model to obtain the actual available capacity and health status of the battery.

[0055] Furthermore, when the battery is pulse charged, the pulse operating condition is as follows:

[0056] Charge at 1 / 3C constant current for 10-60 seconds, then let it rest for 2-10 minutes.

[0057] Charge at 1C constant current for 10-60 seconds, then let it rest for 2-10 minutes.

[0058] Where C represents the charge and discharge rate;

[0059] And obtain the actual charging current value, voltage value and internal resistance of the battery during the pulse working condition.

[0060] When SOCstart is less than 25% of the battery's full charge, the battery is pulse charged 1 to 2 times repeatedly.

[0061] Furthermore, when the battery is charged at a constant current, the constant current working condition is as follows:

[0062] Charge at a specific constant current for 5 to 30 minutes, and let it rest for 5 to 10 minutes;

[0063] And obtain the actual charging current value, voltage value and internal resistance of the battery under constant current working conditions.

[0064] Furthermore, the battery model predicts the battery terminal voltage, including:

[0065] According to the battery circuit principle, the battery terminal voltage of the battery is predicted as: U = Uoc + IR0;

[0066] Cut-off condition: U<=U term , U term The charging cut-off voltage for the battery system is as follows:

[0067] I. Determine the SOC at the next moment using formula (1): SOC next =SOC start +fIΔt / Q design (1);

[0068] SOC next is the estimated battery SOC value at the next moment, fIΔt is the integrated capacity at adjacent moments, Q design It is the nominal capacity or design capacity of the battery;

[0069] II. Obtain the battery terminal voltage at the next moment through formula (2): U est =U oc (SOC next ,T)+IR0(SOC next ,T) (2);

[0070] U oc (SOC next ,T) is the battery open circuit voltage at the next moment, which has a functional relationship with the battery SOC and temperature T. R0(SOC next ,T) is the battery internal resistance at the next moment, which has a functional relationship with the battery SOC and temperature T, U est is the predicted value of the battery terminal voltage at the next moment.

[0071] Furthermore, the voltage data of the battery under constant current charging is obtained, and the voltage data is used as negative feedback to correct the battery model parameters. The negative feedback link is: △U=U est -U real (3);

[0072] U real The voltage data actually detected when the battery is in a constant current charging state;

[0073] Determine the size of the predicted battery terminal voltage and adjust R0 or Q on the battery model based on the judgment result. design After correction, obtain the available battery capacity.

[0074] Furthermore, the battery model parameters are corrected according to the judgment result, including:

[0075] 1) When the predicted voltage is too high and greater than a certain threshold ErrUlimit, reduce the battery internal resistance R0, and the reduction is proportional to the predicted voltage error △U. When the predicted voltage is too low and less than a certain threshold ErrUlimit, increase the battery internal resistance R0, and the increase is proportional to the predicted voltage error △U.

[0076] 2) When the predicted voltage is too high and between [0 and ErrUlimit], increase the battery capacity value Q design , the capacity increase is proportional to the voltage prediction error △U. When the predicted voltage is too small and between [-ErrUlimit~0], reduce the battery capacity value Q design , the magnitude of capacity reduction is proportional to the voltage prediction error △U;

[0077] 3) Update the battery internal resistance or capacity according to step 1 or step 2 to obtain the corrected battery electromotive force OCV correct , according to OCV correct Calibrate SOC to get SOCcorrect , when the SOC change reaches the target value SOC required (0~1), calculate the battery available capacity Q available : Q available =fIΔt / (SOC correct -SOC start ) (4).

[0078] Among them: when the predicted voltage is too large and greater than a certain threshold ErrUlimit, the battery internal resistance R0 is reduced, and the reduction range is proportional to the predicted voltage error △U; when the predicted voltage is too small and less than a certain threshold ErrUlimit, the battery internal resistance R0 is increased, and the increase range is proportional to the predicted voltage error △U. The correction is completed by the following formula: R O '=R O +a△U; a=a+a0;

[0079] a0 is set in stages according to the error △U. The adjustment coefficient a is a parameter that is strongly related to the charging current I, T, and SOC. a=function(I,T,SOC) can be obtained when the battery is in a constant current charging state.

[0080] When the predicted voltage is too high and between [0 and ErrUlimit], increase the battery capacity value Q design , the capacity increase is proportional to the voltage prediction error △U. When the predicted voltage is too small and between [-ErrUlimit~0], reduce the battery capacity value Q design When the capacity reduction is proportional to the voltage prediction error △U, the correction is completed by the following formula: Q design '=Q design +b△U; b=b+b0;

[0081] b0 is set in stages according to the error △U. The adjustment coefficient b is a parameter that is strongly related to the charging current I, T, and SOC. b=function(I,T,SOC) can be obtained when the battery is in a constant current charging state.

[0082] Furthermore, when performing simulation calculation on the battery model to obtain the actual available capacity and battery health status of the battery, the following steps are included:

[0083] The simulation starting state of charge SOC is determined by formula (5) sim0 :SOC sim0 =SOC start +fIΔt / Q available (5);

[0084] The battery terminal voltage is simulated by incrementing the SOC interval ΔSOC: U = U oc (SOCsim0 +ΔSOC)+I R0 (SOC sim0 +ΔSOC) (6);

[0085] When U<=U term When the battery charge end SOC is obtained, it is recorded as EOC (End of charge), and the remaining charging time Δt', U term Charging cut-off voltage for the battery system;

[0086] Calculate the remaining battery capacity Q by formula (7) rem : Q rem =fIΔt' (7);

[0087] By correcting the initial charge battery capacity: Q start =SOC start *Q available (8);

[0088] Calculate the actual available capacity of the battery: Q1=Q start +fIΔt+Q rem (9); Q max =CapCoff2*Q1+(1-CapCoff2)*Q2 (11);

[0089] Among them, Q1 / Q2 / Q max The actual available capacity of the battery is calculated by weighting Q1 / Q2 as the final predicted value. CapCoff1 and CapCoff2 are weighted coefficients with values ​​of (0 to 1). fIΔt' is the integral capacity during the simulation. Q max is the actual available capacity of the battery after weighting.

[0090] That is, the battery health status is estimated as:

[0091] The present invention establishes an initial circuit model through pulse working conditions, uses actual charging data to train the initial model, uses the data of the battery under constant current charging state as negative feedback to correct the battery model parameters, and then simulates the battery terminal voltage through the corrected model to calculate the actual available capacity and battery health status of the battery.

[0092] The present invention further provides an electronic device, comprising:

[0093] memory for storing computer programs;

[0094] A processor is configured to implement the steps of any of the above-mentioned methods for offline detection of the available capacity of a power battery when executing the computer program.

[0095] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the offline detection method for the available capacity of a power battery as described in any one of the above items are implemented.

[0096] The solution provided by the present invention adds a dynamic learning mechanism for battery parameters to the existing voltage- and coulomb-counting algorithm. By establishing a battery model to predict the battery terminal voltage, the actual voltage data of the battery under constant current charging is used as negative feedback to correct the battery model parameters, thereby reducing the model prediction error and making the predicted voltage closest to the actual voltage. This solves the problem of battery impedance changing over time, quickly detects the battery's available capacity, and obtains the battery's health status.

[0097] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention's description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for offline detection of available capacity of a power battery, characterized in that: The following steps are involved: S1: Obtain the battery's initial charging state SOCstart, the battery's data under pulse charging state, and the battery's data under constant current charging state; S2: After parameter identification is performed through the data of the battery under pulse charging state, a battery model is obtained; S3: The battery terminal voltage is obtained through the battery model prediction, and the data of the battery under constant current charging state is used as negative feedback to correct the battery model parameters; S4: Perform simulation calculation on the battery model to obtain the actual available capacity and health status of the battery.

2. The method for offline detection of available capacity of a power battery according to claim 1, characterized in that: When the battery is pulse charged, the pulse working condition is as follows: Charge at 1 / 3C constant current for 10 to 60 seconds, and let stand for 2 to 10 minutes; Charge at 1C constant current for 10-60 seconds, and let stand for 2-10 minutes; Wherein, C represents the charge and discharge rate; And obtain the actual charging current value, voltage value and internal resistance of the battery during the pulse working condition.

3. The method for offline detection of available capacity of a power battery according to claim 1, characterized in that: When the battery is charged with constant current, the constant current working condition is as follows: Charge at a specific constant current for 5 to 30 minutes, and let stand for 5 to 10 minutes; And obtain the actual charging current value, voltage value and internal resistance of the battery under constant current working conditions.

4. The method for offline detection of available capacity of a power battery according to claim 3, characterized in that: The battery model predicts the battery terminal voltage, including: I. Determine the SOC at the next moment by formula (1): SOC next =SOC start +fIΔt / Q design (1); SOC next is the estimated battery SOC value at the next moment, fIΔt is the integrated capacity at adjacent moments, Q design It is the nominal capacity or design capacity of the battery; II. Obtain the battery terminal voltage at the next moment through formula (2): The est =U oc (SOC next ,T)+IR0(SOC next ,T) (2); U oc (SOC next ,T) is the battery open circuit voltage at the next moment, which has a functional relationship with the battery SOC and temperature T. R0(SOC next ,T) is the battery internal resistance at the next moment, which has a functional relationship with the battery SOC and temperature T. est It is the predicted value of the battery terminal voltage at the next moment.

5. The method for offline detection of available capacity of a power battery according to claim 4, characterized in that: The voltage data of the battery under constant current charging is obtained, and the voltage data is used as negative feedback to correct the battery model parameters. The negative feedback link is: △U=U est -U real (3); U real The voltage data actually detected by the battery under constant current charging state; Determine the predicted battery terminal voltage and adjust R0 or Q on the battery model based on the result of the determination. design After correction, obtain the available battery capacity.

6. The method for offline detection of available capacity of a power battery according to claim 5, characterized in that: Correcting the battery model parameters according to the judgment result includes: 1) When the predicted voltage is too large and greater than a certain threshold ErrUlimit, the battery internal resistance R0 is reduced, and the reduction is proportional to the predicted voltage error △U. When the predicted voltage is too small and less than a certain threshold ErrUlimit, the battery internal resistance R0 is increased, and the increase is proportional to the predicted voltage error △U. 2) When the predicted voltage is too high and between [0 and ErrUlimit], increase the battery capacity value Q design The capacity increase is proportional to the voltage prediction error △U. When the predicted voltage is too small and between [-ErrUlimit~0], reduce the battery capacity value Q design , the magnitude of capacity reduction is proportional to the voltage prediction error △U; 3) Update the battery internal resistance or capacity according to step 1 or step 2 to obtain the corrected battery electromotive force OCV correct , according to OCV correct Calibrate SOC to get SOC correct , when the SOC change reaches the target value SOC required (0~1), calculate the available battery capacity Q available : Q available =fIΔt / (SOC correct -SOC start ) (4)。 7. The method for offline detection of available capacity of a power battery according to claim 6, characterized in that: When the battery model is simulated and calculated to obtain the actual available capacity and health status of the battery, the following steps are included: The simulation starting charging state SOC is determined by formula (5) sim0 : SOC sim0 =SOC start +fIΔt / Q available (5); The battery terminal voltage is simulated by incrementing the SOC interval ΔSOC: U=U oc (SOC sim0 +ΔSOC)+I R0 (SOC sim0 +ΔSOC) (6); When U<=U term When the battery charge cut-off SOC is obtained, recorded as EOC (End of charge), and the remaining charging time Δt', U term Charging cut-off voltage for battery system; Calculate the remaining battery capacity Q by formula (7) rem : Q rem =fIΔt′ (7); By correcting the initial charge battery capacity: Q start =SOC start *Q available (8); Calculate the actual available battery capacity: Q1=Q start +fIΔt+Q rem (9); Q max =CapCoff2*Q1+(1-CapCoff2)*Q2 (11); Among them, CapCoff1 and CapCoff2 are weighted coefficients with values ​​of (0 to 1), fIΔt' is the integral capacity during the simulation, and Q max It is the actual available capacity of the weighted battery.

8. The method for offline detection of available capacity of a power battery according to claim 7, characterized in that: You can also get the battery health status as:

9. An electronic device, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of the method for offline detection of available capacity of a power battery as claimed in any one of claims 1 to 8 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the offline detection method for the available capacity of a power battery according to any one of claims 1 to 8 are implemented.

Images