CN114325445A - A rapid assessment method of lithium-ion battery state of health based on regional frequency - Google Patents

Abstract

The invention discloses a method for rapidly assessing the health state of a lithium ion battery based on a regional frequency. Count the DL of the entire charge-discharge voltage curve; Step 2, convert the charge-discharge voltage data into PDF curve P(x), and search for the voltage V _peak corresponding to the maximum peak of P(x); Step 3, select the regional voltage according to the voltage V _peak ΔV _reg , and calculate the probability P of the regional voltage ΔV _reg ; Step 4, multiply the probability P and DL to obtain the regional frequency F; Step 5, take the regional frequency F as the independent variable and the battery SOH as the dependent variable, establish a linear regression Equation; Step 6, select another lithium-ion sample battery to be tested, repeat steps 1 to 4, substitute the frequency F' of the area to be measured into the linear regression equation, calculate the SOH value of the battery corresponding to the frequency F' of the area to be measured, and realize the measurement Battery SOH evaluation of Li-ion sample batteries.

Description

A rapid assessment method of lithium-ion battery state of health based on regional frequency

technical field

The invention relates to the technical field of battery state of health assessment, in particular to a method for rapidly assessing the state of health of lithium ion batteries based on regional frequencies.

Background technique

China is now the world's largest carbon emitter. By the end of 2020, China's total carbon emissions were 9.899 billion tons, accounting for 30.7% of the global total. Among them, transportation accounts for 15% of China's end-use carbon emissions, which has grown at an average annual rate of about 5% over the past nine years. In order to achieve the goal of "carbon peaking by 2030 and carbon neutrality by 2060", the Chinese government vigorously promotes transportation electrification, which is considered to be an effective way to significantly reduce carbon emissions. As of the end of June 2021, China has 6.03 million new energy vehicles, including 4.93 million pure electric vehicles. The power source of electric vehicles is usually high-performance lithium-ion batteries. However, the lifespan of Li-ion batteries is not infinite due to the gradual aging of the battery during charge-discharge cycles. The irreversible and unavoidable internal reactions during cycling degrade battery performance, such as capacity reduction and resistance increase. If the decline of the battery state of health (SOH) cannot be detected in time, the safety and reliability of the entire battery system will not be guaranteed.

There are many evaluation methods for battery SOH. The traditional method is based on capacity calibration and pulse step internal resistance measurement. Although this method is accurate, it is time-consuming, and requires the normal commercial operation of lithium-ion batteries to be stopped, and the capacity calibration and pulse step internal resistance measurement are specially performed, which will inevitably affect the use efficiency of lithium-ion batteries. If the battery parameters collected by the battery management system (BMS) during the normal operation of the battery can be used, such as temperature, voltage, current, etc., and the health factor reflecting the battery SOH can be extracted through their changes, the health status of the lithium battery can be realized. Online assessment. At present, there are some reports on the online assessment of lithium battery health status.

Chinese patent CN 111458649 A (a rapid detection method for the health of a battery module) discloses a rapid evaluation method for the actual capacity of a battery module, by acquiring the platform voltage data during the charging and discharging process of a battery module sample with known available capacity , convert it into a probability density function (PDF) curve, calculate the peak area of the set voltage range, and fit the regression equation between the SOH and the peak area of the battery module's state of health. The detection process only needs to obtain the platform voltage data during the charging and discharging process of the battery module to be detected, calculate the peak area according to the same steps, and substitute the peak area into the regression equation to obtain the SOH of the battery module to be detected. However, this method is sensitive to the sampling frequency, and often requires a sampling frequency of more than 1 Hz to ensure that the model evaluation is not distorted, but it is difficult to guarantee such a high sampling frequency in actual practical conditions.

Chinese patent CN 111948546 A (a method and system for evaluating the state of health of a lithium battery) discloses a method for evaluating the state of health of a lithium battery, which finds that the maximum peak height of the incremental capacity analysis (ICA) curve decreases during the charging and discharging process of the battery There is a linear relationship between the degree ΔH _max and the state of health of the battery. According to the maximum peak height drop degree ΔH _max corresponding to the sample battery under the state of health value, the ΔH _max -SOH fitting curve is drawn, and the voltage curve during the charging and discharging process of the battery to be tested is collected to calculate ICA The maximum peak height drop in the curve is the ΔH _max value, and the SOH value of the battery to be tested can be obtained according to the ΔH _max -SOH fitting curve. However, this approach also has certain limitations. When voltage plateaus appear in charge-discharge tests, the resolution of the voltage measurement may not be sufficient to distinguish voltage differences. This may lead to dV=0, especially for lithium iron phosphate batteries, which have a flatter voltage curve. Even if the measurement accuracy is acceptable, the noise can be reduced, but this results in a higher cost of the sampling module of the BMS.

Chinese patent CN 108490366 A (rapid assessment method for the health status of retired battery modules of electric vehicles) discloses a rapid assessment method for the health status of retired battery modules of electric vehicles. In contrast, the correlation between SOH and the Lorenz dispersion of the voltage during the discharge process is analyzed. Since the working voltage during the battery charge and discharge process can be collected in real time according to the battery management system, no additional collection is required, and the workload does not increase; the collected working voltage is in The corresponding state of charge (SOC) interval is sufficient, and is not limited to a certain SOC value, which is more convenient; the calculation of the voltage Lorenz dispersion is based on the result of voltage averaging in the SOC interval, and the result is more accurate. It is only necessary to calculate the Lorenz dispersion of the voltage during the discharge process of the decommissioned battery module to be tested to realize the rapid evaluation of the SOH of the decommissioned battery of electric vehicles, so as to conduct consistent and rapid sorting, so as to achieve the simple, convenient and low-cost reuse of the decommissioned battery. Target. However, the working voltage of the battery sample in the model and the working voltage of the battery to be tested need to be in the same SOC range. Since the SOC value is a dynamically changing value, as the environment and operating conditions change, the SOC value estimated by the battery management system may deviate from the real value, which brings uncertainty to the rapid assessment of the battery module's state of health.

SUMMARY OF THE INVENTION

The present invention is made to solve the above problems, and the purpose is to provide a method for rapidly assessing the state of health of a lithium ion battery based on a regional frequency.

The present invention provides a method for rapidly assessing the health state of a lithium-ion battery based on a regional frequency, which has the following characteristics and includes the following steps: Step 1: Charge and discharge the lithium-ion battery at a certain rate, collect battery charge and discharge voltage data, and obtain Working voltage curve, and count the number of voltage sampling points of the entire charge-discharge voltage curve; Step 2, convert the charge-discharge voltage data into a PDF curve P(x), and search for the voltage V _peak corresponding to the maximum peak of the PDF curve P(x); Step 3, select the regional voltage ΔV _reg according to the voltage V _peak , and calculate the probability P of the regional voltage ΔV _reg ; Step 4, multiply the probability P by the number of voltage sampling points to obtain the regional frequency F; Step 5, take the regional frequency F as The independent variable, taking the battery SOH as the dependent variable, establishes a linear regression equation; step 6, select another lithium-ion sample battery to be tested, repeat steps 1 to 4, and collect the charge and discharge of the lithium-ion sample battery to be tested at the same rate Voltage data, obtain the frequency F' of the area to be measured corresponding to the area voltage ΔV _reg of the lithium-ion sample battery to be measured, substitute the frequency F' of the area to be measured as the health factor index into the linear regression equation, and calculate the frequency corresponding to the frequency of the area to be measured F' The battery SOH value, realizes the battery SOH evaluation of the lithium-ion sample battery to be tested.

The method for rapidly assessing the state of health of lithium ion batteries based on regional frequency provided by the present invention may also have the following feature: wherein, in step 1, the method of collecting battery charge and discharge voltage data is automatic collection by the battery management system.

In the method for quickly assessing the health state of lithium-ion batteries based on the regional frequency provided by the present invention, it may also have the following characteristics: wherein, in step 2, the charge and discharge voltage data is converted into a PDF curve P by using the ksdensity function in the Matlab statistical toolbox. (x), the specific process is as follows:

x=[], in the formula, import the charging and discharging voltage data in [];

[f, x _i ]=ksdensity(x), in the formula, [f, x _i ] is the result form calculated and displayed by MATLAB, f corresponds to the value of the probability density, and x _i is the corresponding abscissa (voltage value).

In the method for quickly assessing the state of health of a lithium-ion battery based on the regional frequency provided by the present invention, it may also have the following characteristics: wherein, in step 3, the specific process of calculating the probability P of the regional voltage _ΔVreg includes the following sub-steps: step 3-1, take the voltage V _peak as the center in the PDF curve P(x), and select a certain value on the left and right sides to obtain the regional voltage ΔV _reg ; Step 3-2, put the PDF curve P(x) in the regional voltage ΔV _reg . Carry out definite integration inside to obtain the definite integral value of the regional voltage ΔV _reg ; step 3-3, take the ratio of the definite integral value of the regional voltage ΔV _reg to the definite integral value of the entire PDF curve P(x), which is the regional voltage ΔV _reg the probability P.

The role and effect of the invention

According to the method for rapidly assessing the state of health of lithium ion batteries based on regional frequency, the lithium ion battery is firstly charged and discharged at a certain rate, the battery charge and discharge voltage data are collected, the working voltage curve is obtained, and the entire charge and discharge voltage curve is counted The number of _{voltage sampling points} of The probability P of the regional voltage ΔV _reg ; then multiply the probability P by the number of voltage sampling points to obtain the regional frequency F; then use the regional frequency F as the independent variable and the battery SOH as the dependent variable to establish a linear regression equation; Measure the lithium-ion sample battery, repeat the above steps, collect the charge and discharge voltage data of the lithium-ion sample battery to be tested at the same rate, and obtain the area frequency F' to be tested corresponding to the area voltage ΔV _reg of the lithium-ion sample battery to be tested, The frequency F' of the area to be measured is substituted into the linear regression equation as a health factor index, and the SOH value of the battery corresponding to the frequency F' of the area to be measured is calculated to realize the battery SOH evaluation of the lithium-ion sample battery to be measured.

In the above process of this embodiment, the working voltage during the charging and discharging process of the battery can be collected in real time by the battery management system, and there is no need for offline experiments and additional collection of the lithium battery, which reduces the workload. And compared with the traditional PDF method that directly uses the PDF peak height as the health factor, this method is not sensitive to the sampling frequency, and can still obtain high accuracy at low sampling frequency (such as 1/60Hz, that is, sampling once a minute). , and the calculation is simple, no filtering is required, and it is suitable for real-time online SOH evaluation.

Description of drawings

FIG. 1 is a flowchart of a method for assessing battery state of health based on regional frequency in an embodiment of the present invention;

Fig. 2 is the charge-discharge curve of lithium iron phosphate module under 1/3C capacity calibration condition in the embodiment of the present invention;

FIG. 3 is a result diagram of the battery state of health assessment method based on regional frequency in an embodiment of the present invention.

Detailed ways

In order to make the technical means, creative features, goals and effects realized by the present invention easy to understand, the following embodiments specifically describe a method for quickly assessing the health status of lithium ion batteries based on regional frequency of the present invention with reference to the accompanying drawings.

In this embodiment, a method for rapidly assessing the state of health of a lithium-ion battery based on a regional frequency is provided.

FIG. 1 is a flowchart of a method for evaluating the state of health of a battery based on regional frequency in this embodiment.

As shown in FIG. 1 , the method for quickly assessing the state of health of a lithium-ion battery based on a regional frequency involved in this embodiment includes the following steps:

Step S1, charge and discharge the lithium ion battery at a certain rate, collect battery charge and discharge voltage data, obtain a working voltage curve, and count the number of voltage sampling points of the entire charge and discharge voltage curve.

In this example, 8 lithium iron phosphate battery modules (15P4S, 15 parallel and 4 strings) retired from the Chery S18B electric vehicle are used, with a nominal capacity of 40Ah, consisting of 4 15P1S battery cells connected in series. The rated voltage of the 15P1S battery unit is 3.2V, and the rated voltage of the entire module is 12.8V.

Table 1 is the charging and discharging test steps of the lithium iron phosphate battery module in this embodiment.

Table 1

本实施例中使用的电池测试系统电压采样频率为1/60Hz。As shown in Table 1, the lithium iron phosphate module is rapidly aged at a rate of 2C (80A). During cycling, a 30-min rest time was set between charge and discharge. The battery was cycled to age until its capacity dropped below 60% SOH. After every 50 cycles of aging, perform capacity calibration at 1/3C (13.3A) to obtain the remaining capacity and charge-discharge voltage curve, and calculate its SOH,

The voltage sampling frequency of the battery testing system used in this embodiment is 1/60 Hz.

FIG. 2 is the charge-discharge curve of the lithium iron phosphate module in the present embodiment under the 1/3C capacity calibration condition.

As shown in Figure 2, as the aging period increases, the charge and discharge platform of the module becomes shorter, and the SOH of the display module becomes smaller. The plateau potential of the charging curve increases with the increase of the aging cycle, and the plateau potential of the discharge curve decreases with the increase of the aging cycle, indicating that the aging leads to an increase in the internal resistance of the battery.

Table 2 shows the available capacity and SOH value of the lithium iron phosphate module in different cycle cycles under the 1/3C calibration condition in this example.

As shown in Table 2, the lithium iron phosphate battery module was aged in the 2C cycling protocol of 0-100% SOC, completed 400 charge-discharge cycles, and the SOH decayed from 96.7% to 55.73%.

FIG. 3 is a result diagram of the battery state of health assessment method based on regional frequency in this embodiment.

FIG. 3( a ) is a charging curve diagram of the battery in this embodiment.

As shown in FIG. 3( a ), the data length DL of the charging curve is 176 .

In step S2, the charge-discharge voltage data is converted into a PDF curve P(x), and a voltage V _peak corresponding to the maximum peak of the PDF curve P(x) is searched.

FIG. 3(b) is a PDF graph in this embodiment.

As shown in Fig. 3(b), the searched maximum PDF peak coordinates are (13.49, 2.326).

In step S3, the regional voltage ΔV _reg is selected according to the voltage V _peak , and the probability P of the regional voltage Δ V _reg is calculated. The specific implementation is divided into the following sub-steps:

FIG. 3( c ) is a schematic diagram of the selected region voltage ΔV _reg in this embodiment.

Step S3-1, taking the voltage V _peak as the center in the PDF curve P(x), and selecting a certain value on the left and right sides to obtain the regional voltage ΔV _reg .

As shown in Figure 3(c), taking the regional voltage ΔV _reg as 200mV, taking V _peak =13.49 V as the center, and taking 100 mV to the left and right respectively, the starting voltage of the regional voltage V ₀ =13.39 V, and the ending voltage V _t =13.59 V.

In step S3-2, the PDF curve P(x) is definitely integrated within the region voltage _ΔVreg to obtain the definite integral value of the region voltage _ΔVreg .

Step S3-3, take the ratio of the definite integral value of the regional voltage ΔV _reg to the definite integral value of the entire PDF curve P(x), which is the probability P of the regional voltage ΔV _reg .

As shown in Fig. 3(c), the definite integral of the termination voltage V _end to the initial voltage V _start is 1 due to the entire PDF curve. In fact, then the definite integral of the PDF curve P(x) in the region voltage is just the probability P that the voltage falls within the region voltage. Figure 3(c) shows that the probability P of the area voltage is 0.4036.

Step S4, multiply the probability P by the number of voltage sampling points to obtain the regional frequency F;

FIG. 3(d) is a schematic diagram of the calculation area frequency F in this embodiment.

As shown in Fig. 3(d), the area frequency F is 71.03.

Step S5 , establishing a rapid evaluation model, using the regional frequency F of the charging and discharging process of the sample battery in 8 cycles as an independent variable, and the SOH value of the sample battery in 8 cycles as a dependent variable, and establishing a linear regression equation.

FIG. 3(e) is the fitting curve of the area frequency and the SOH when the area voltage is 200mV during the charging and discharging process in this embodiment. The left side is the charging process, and the right side is the discharging process.

As shown in Fig. ³ (e), the fitted R2 for the charging process is 0.9254. The fitted R2 of the discharge process was ^0.9560 . Regardless of the charging and discharging process, it shows a good fitting accuracy.

Step S6, select another lithium ion sample battery to be tested, repeat steps 1 to 4, as long as the charge and discharge voltage data of the lithium ion sample battery to be tested under the same magnification are collected to obtain the regional voltage of the lithium ion sample battery to be tested ΔV _reg corresponds to the frequency F' of the area to be measured, and substitute the frequency F' of the area to be measured as the health factor index into the linear regression equation to calculate the SOH value of the battery corresponding to the frequency of the area to be measured F', to realize the battery of the lithium-ion sample battery to be measured. SOH assessment.

Action and effect of the embodiment

According to the method for rapidly assessing the health status of lithium-ion batteries based on the regional frequency involved in this embodiment, the lithium-ion battery is firstly charged and discharged at a certain rate, the battery charge and discharge voltage data are collected, the working voltage curve is obtained, and the entire charge and discharge voltage is counted The number of voltage sampling points of the curve; then convert the charge and discharge voltage data into the PDF curve P(x), and search for the voltage V _peak corresponding to the maximum peak of the PDF curve P(x); then select the regional voltage ΔV _reg according to the voltage V _peak , and Calculate the probability P of the regional voltage ΔV _reg ; then multiply the probability P by the number of voltage sampling points to obtain the regional frequency F; then use the regional frequency F as the independent variable and the battery state of health as the dependent variable to establish a linear regression equation; Select the lithium-ion sample battery to be tested, repeat the above steps, collect the charge and discharge voltage data of the lithium-ion sample battery to be tested at the same rate, and obtain the test area frequency F corresponding to the area voltage ΔV _reg of the lithium-ion sample battery to be tested ', Substitute the frequency F' of the area to be tested as a health factor index into the linear regression equation, calculate the battery state of health value corresponding to the frequency of the area to be measured F', and realize the battery state of health evaluation of the lithium-ion sample battery to be tested.

In the above process of this embodiment, the working voltage during the charging and discharging process of the battery can be collected in real time by the battery management system, and there is no need to perform offline experiments on the lithium battery, no additional collection is required, and no workload is increased. And compared with the traditional PDF method that directly uses the PDF peak height as the health factor, this method is not sensitive to the sampling frequency, and can still obtain high accuracy at low sampling frequency (such as 1/60Hz, that is, sampling once a minute). , and the calculation is simple, no filtering is required, and it is suitable for real-time online SOH evaluation.

The above embodiments are preferred cases of the present invention, and are not intended to limit the protection scope of the present invention.

Claims (4)

1. A lithium ion battery health state rapid evaluation method based on region frequency is characterized by comprising the following steps:

step 1, charging and discharging a lithium ion battery under a certain multiplying power, collecting battery charging and discharging voltage data, acquiring a working voltage curve, and counting the number of voltage sampling points of the whole charging and discharging voltage curve;

step 2, converting the charge and discharge voltage data into a probability density function curve P (x), and searching a voltage V corresponding to the maximum peak of the probability density function curve P (x)_peak；

Step 3, according to the voltage V_peakSelected area voltage DeltaV_regAnd calculating the area voltage DeltaV_regThe probability P of (A);

step 4, multiplying the probability P and the frequency of the voltage sampling points to obtain an area frequency F;

step 5, establishing a linear regression equation by taking the region frequency F as an independent variable and the battery health state as a dependent variable;

step 6, replacing the lithium ions to be detectedAnd (4) repeating the steps 1 to 4, collecting the charge-discharge voltage data of the lithium ion sample battery to be tested under the same multiplying power, and obtaining the voltage delta V of the lithium ion sample battery to be tested in the region_regAnd substituting the corresponding to-be-detected region frequency F 'as a health factor index into the linear regression equation, and calculating a battery health state value corresponding to the to-be-detected region frequency F', so as to realize the battery health state evaluation of the to-be-detected lithium ion sample battery.

2. The method for rapidly evaluating the health status of a lithium ion battery according to claim 1, wherein the method comprises the following steps:

in the step 1, the battery charging and discharging voltage data is acquired automatically by a battery management system.

3. The method for rapidly evaluating the health status of a lithium ion battery according to claim 1, wherein the method comprises the following steps:

in the step 2, a ksDensity function in a Matlab statistical tool box is used for converting the charging and discharging voltage data into a probability density function curve P (x), and the specific process is as follows:

x＝[]，

wherein the charge-discharge voltage data is introduced into [ ];

[f,x_i]＝ksdensity(x)，

wherein [ f, x [ ]_i]Is the result form displayed by MATLAB calculation, and f is the numerical value of probability density, x_iIs the corresponding abscissa (voltage value).

4. The method for rapidly evaluating the health status of a lithium ion battery according to claim 1, wherein the method comprises the following steps:

wherein, in step 3, the area voltage Δ V is calculated_regThe specific process of probability P of (a) includes the following substeps:

step 3-1, using the voltage V in the probability density function curve P (x)_peakSelecting a certain value from the left and right sides to obtain a region voltage delta V_reg；

Step 3-2, the voltage delta V of the probability density function curve P (x) in the region_regPerforming internal fixed integration to obtain area voltage delta V_regA fixed integral value of (2);

step 3-3, taking the area voltage delta V_regThe ratio of the constant integral value of (a) to the constant integral value of the entire probability density function curve P (x) is the area voltage DeltaV_regThe probability P of (a).

Images