An SOC calibration algorithm

Abstract

An SOC calibration algorithm includes the following steps. Step S1: Obtain a Coulomb efficiency table based on the test data of the battery cell materials at different temperatures and different currents, and use the Coulomb efficiency table to obtain the total battery capacity value. Step S2: Obtain the battery state. When the battery is in the charging state, compare the SOC value with the magnitude of M. If 0 < SOC value < M, calculate the SOC value using the ampere-hour integration algorithm. If the SOC value = M, enter the pre-calibration. If M < SOC value < 100%, use the real-time voltage calibration method. When the battery is in the discharging state, compare the SOC value with the magnitude of N. If the SOC value > N, calculate the SOC value using the ampere-hour integration algorithm. If the SOC value = N, enter the pre-calibration. If 0 < SOC value < N, use the real-time voltage calibration method. The SOC value of the present invention has a high accuracy, and can enter the voltage calibration state in advance, avoiding calibration at the end of charging or discharging when the SOC deviation reaches, reducing the deviation rate of SOC, and greatly reducing the amount of data.

Description

Technical Field

[0001] The present invention relates to the evaluation of the remaining battery power, and specifically to a SOC calibration algorithm. Background Art

[0002] State of Charge (SOC) refers to the ratio of the remaining capacity of a storage battery after being used for a period of time or left unused for a long time to its fully charged state capacity, which is usually expressed as a percentage and can be represented by the following formula: SOC is an important parameter in the Battery Management System (BMS), and many functions in the BMS are based on SOC. For batteries with the same capacity, a battery with a higher SOC accuracy can have a longer cruising range. Without accurate SOC, adding more protection functions cannot make the BMS work properly, which will put the battery in a protected state and is not conducive to extending the battery life. Therefore, high-precision SOC estimation can effectively reduce battery costs.

[0003] There are several common SOC algorithms: algorithms based on ampere-hour integration, algorithms based on open-circuit voltage (OCV) calibration, etc. However, in the actual test process, even a very small error integration (due to the accuracy of sensors and limited sampling intervals) gradually makes the SOC inaccurate. When the battery is fully charged or discharged and left standing for a period of time, calibration is performed according to the OCV voltage. However, in the actual use process of the battery, such conditions are rarely met, so it is very difficult to perform SOC calibration. Therefore, it is very necessary to obtain an algorithm for accurately calculating the SOC value to reduce battery maintenance costs. Summary of the Invention

[0004] The purpose of the present invention is to overcome the defects of the existing evaluation system and provide a SOC calibration algorithm.

[0005] To achieve the above purpose, the present invention adopts the following technical solutions:

[0006] A SOC calibration algorithm includes the following steps

[0007] Step S1: Obtain a Coulomb efficiency table based on the test data of the battery cell material at different temperatures and different currents, obtain the total capacity value of the battery cell using the rated total capacity value of the battery cell and the Coulomb efficiency table, and acquire and store the SOC-voltage table of the battery cell;

[0008] Step S2: Obtain the battery state;

[0009] When the battery is in the charging state, go to step S3;

[0010] When the battery is in the discharging state, go to step S4;

[0011] When the battery is in a standby state, keep the SOC value unchanged;

[0012] Step S3: Set a fixed boundary M, and compare the size of the SOC value with M. M is a fixed value, and 70% < M < 100%;

[0013] If 0 < SOC value < M, still use the ampere-hour integration algorithm to calculate the SOC value;

[0014] If the SOC value = M, enter the pre-calibration;

[0015] If M < SOC value < 100%, use the real-time voltage calibration method to calculate the SOC value;

[0016] Step S4: Set a fixed boundary N, and compare the size of the SOC value with N. N is a fixed value, and 0% < N < M < 100%;

[0017] If N < SOC value < 100%, still use the ampere-hour integration algorithm to calculate the SOC value;

[0018] If the SOC value = N, enter the pre-calibration;

[0019] If 0 < SOC value < N, use the real-time voltage calibration method to calculate the SOC value.

[0020] Preferably, the pre-calibration in step S3 includes the following steps:

[0021] Poll the current temperature and current charging current of all battery cells, select the SOC-voltmeter applicable to the current temperature and current charging current of the battery cells, and query the voltage Y when SOC = M as the calibration voltage value;

[0022] Compare whether the current voltage of the battery cell reaches Y. If it reaches Y, use the real-time voltage calibration to calculate the SOC value and quickly fit and display the SOC value as M to complete the calibration.

[0023] Preferably, when the voltage value of the highest single battery cell among all battery cells is less than the current calibration voltage value Y, charging still needs to be carried out until the calibration voltage value Y is reached.

[0024] Preferably, the pre-calibration in step S4 includes the following steps:

[0025] Poll the current temperature and current discharge current of all battery cells, select the SOC-voltmeter applicable to the current temperature and current discharge current of the battery cells, and query the voltage Z when SOC = N as the calibration voltage value; compare whether the current voltage of the battery cell reaches Z. If it reaches Z, use the real-time voltage calibration to calculate the SOC value and quickly fit and display the SOC value as N to complete the calibration.

[0026] Preferably, when the voltage value of the lowest single cell among all cells is greater than Z times the current calibration voltage value, it is still necessary to discharge until it reaches Z times the calibration voltage value.

[0027] Preferably, M = 90% in step S3 and N = 10% in step S4.

[0028] Preferably, in steps S3 and S4, based on the battery current and temperature within the time period t, the Coulomb efficiency coefficient k is obtained from the Coulomb efficiency table, and the ampere-hour integration algorithm is calculated based on the Coulomb efficiency coefficient k. The formula for calculating the SOC value by the ampere-hour integration algorithm is:

[0029]

[0030] where SOC 0 is the initial SOC value, C N is the total capacity of the cell, I is the battery current; η is the discharge efficiency, k is the Coulomb efficiency coefficient, and t is the time period.

[0031] Preferably, in step S3, the method of real-time voltage calibration is to poll the current temperature and current charging current of the cell, select the SOC-voltmeter applicable to the current temperature and current charging current of the cell, and obtain the SOC value through the SOC corresponding to the current voltage;

[0032] In step S4, the method of real-time voltage calibration is to poll the current temperature and current discharge current of the cell, select the SOC-voltmeter applicable to the current temperature and current discharge current of the cell, and obtain the SOC value through the SOC corresponding to the current voltage.

[0033] Preferably, in step S3, when the SOC value calculated by the method of real-time voltage calibration is 100%, stop charging;

[0034] In step S4, when the SOC value calculated by the method of real-time voltage calibration is 0%, stop discharging.

[0035] An SOC calibration algorithm of the present invention obtains the Coulomb efficiency table according to the test data of the cell materials forming the battery at different temperatures and different currents, and obtains the total battery capacity according to the rated total capacity of the cell and the Coulomb efficiency table. And after comparing the SOC value with the set fixed limit value, it enters the voltage calibration state in advance, avoiding calibrating when the SOC deviation reaches the charging end or the discharging end, and reducing the SOC deviation rate in advance; at the same time, during the later calibration, the calibration voltage used is obtained according to the original data under different temperatures and different current conditions during the charging or discharging process, rather than using the open circuit voltage (OCV) value, ensuring the accuracy of voltage calibration during the charging process and the discharging process. Compared with the existing full-process voltage calibration method, the data volume of the present invention is greatly reduced, and it is easier to maintain the stability and safety of the system.

[0036] In addition, compared with the existing algorithms, when there is a difference between the results obtained by the ampere-hour integration method and voltage calibration, sudden changes in SOC are likely to occur. However, this algorithm has pre-calibration, which can make the displayed value of the system SOC smoothly approach the value of real-time voltage calibration, ensuring accurate parameters and improving the user experience of customers at the same time.

[0037] In addition, the accuracy of a SOC calibration algorithm of the present invention is relatively high, the SOC estimation error is less than 5%, reducing the maintenance cost of the battery and meeting the usage requirements of the BMS. Description of the Drawings

[0038] Figure 1 is a flowchart of a SOC calibration algorithm of the present invention;

[0039] Figure 2 is a corresponding relationship diagram of the SOC curve calculated by the present invention and the SOC curve of the original data (25°C, 0.3C, charging);

[0040] Figure 3 is a corresponding relationship diagram of the SOC curve calculated by the present invention and the SOC curve of the original data (25°C, 0.3C, discharging). Detailed Embodiments

[0041] The following further describes the detailed embodiments of a SOC calibration algorithm of the present invention in conjunction with the Figure 1 embodiments given. The SOC calibration algorithm of the present invention is not limited to the descriptions of the following embodiments.

[0042] A SOC calibration algorithm includes the following steps.

[0043] Step S1: Obtain a Coulomb efficiency table based on the test data of the battery cell material at different temperatures and different currents, obtain the total capacity value of the battery cell by using the rated total capacity of the battery cell and the Coulomb efficiency table, and acquire and store the SOC-voltmeter of the battery cell.

[0044] Step S2: Obtain the battery state.

[0045] When the battery is in the charging state, go to Step S3.

[0046] When the battery is in the discharging state, go to Step S4.

[0047] When the battery is in the standby state, keep the SOC value unchanged.

[0048] Step S3: Set a fixed limit M, M is a fixed value, and 70% < M < 100%, and compare the size of the SOC value with M.

[0049] If 0 < SOC value < M, the Ampere-hour integration algorithm is still used to calculate the SOC value;

[0050] When the SOC value = M, enter the pre-calibration;

[0051] If M < SOC value < 100%, the real-time voltage calibration method is used to calculate the SOC value;

[0052] Step S4: Set a fixed limit N. N is a fixed value, and 0% < N < M < 100%. Compare the SOC value with N;

[0053] If N < SOC value < 100%, the Ampere-hour integration algorithm is still used to calculate the SOC value;

[0054] When the SOC value = N, enter the pre-calibration;

[0055] If 0 < SOC value < N, the real-time voltage calibration method is used to calculate the SOC value.

[0056] A SOC calibration algorithm of the present invention obtains a Coulomb efficiency table according to the test data of the battery cell materials forming the battery at different temperatures and different currents, and obtains the total battery cell capacity according to the rated total capacity of the battery cell and the Coulomb efficiency table. And when comparing the SOC value with the set fixed limit value, it can enter the voltage calibration state in advance, avoiding calibrating when the SOC deviation reaches the end of charging or discharging, and reducing the SOC deviation rate in advance; at the same time, during the later calibration, the calibration voltage used is obtained from the original data under different temperatures and different current conditions during the charging or discharging process, rather than using the open-circuit voltage (OCV) value, ensuring the accuracy of voltage calibration during the charging process and the discharging process. Compared with the existing full-process voltage calibration method, the data volume of the present invention is greatly reduced, and it is easier to maintain the stability and safety of the system.

[0057] In addition, compared with the existing algorithms, when there is a difference between the results obtained by the Ampere-hour integration method and the voltage calibration, SOC mutation is likely to occur. However, this algorithm has pre-calibration, which can make the SOC display value of the system smoothly approach the value of the real-time voltage calibration, ensuring the accuracy of the parameters and improving the user experience of customers at the same time.

[0058] Combined with the attached Figure 1 A SOC calibration algorithm is introduced in detail, including the following steps,

[0059] Step S1: Obtain a Coulomb efficiency table with corresponding Coulomb efficiency coefficients k based on the test data of the battery cell at different temperatures and different currents. Use the Coulomb efficiency table to obtain the total capacity value of a single battery cell based on the rated total capacity Q of the battery cell and the Coulomb efficiency coefficient k. The Coulomb efficiency table is shown in Table 1, where C is the multiple of the current magnitude during battery charging and discharging. For example, for a 63 Ah battery with a unit of ampere, a 1C current is 63A. When discharging at a 0.3C multiple current, the 0.3C current is 18.9A (63 * 0.3 = 18.9):

[0060] Magnification / Temperature 0.3C 0.6C 1C 2C 0℃ 85.92％ 86.18％ 85.51％ 84.87％ 10℃ 92.30％ 89.94％ 88.73％ 88.42％ 25℃ 97.11％ 95.47％ 94.53％ 93.63％ 35℃ 98.94％ 98.07％ 96.94％ 96.83％ 45℃ 98.85％ 99.37％ 96.94％ 98.13％

[0061] Table 1. Coulomb efficiency table

[0062] Therefore, in this algorithm, the total capacity value of the battery refers to the actual available capacity value of the fully charged battery at different temperatures and currents. The product of the rated total capacity of the battery cell and the Coulomb efficiency value in Table 1 is the total capacity value of the battery cell at the corresponding temperature and current. The SOC value is calculated using the formula SOC = remaining capacity of the battery cell / total capacity of the battery cell.

[0063] Step S2: Obtain the battery state;

[0064] When the battery is in the charging state, go to Step S3;

[0065] When the battery is in the discharging state, go to Step S4;

[0066] When the battery is in the standby state, keep the SOC value unchanged; the SOC value remains the same as the previously stored or calculated SOC 0 value.

[0067] Step S3: First, determine an SOC limit M for entering voltage calibration. M is a fixed value and 70% < M < 100%, preferably 90%. Compare the size of the SOC value with M;

[0068] If 0 < SOC value < M, calculate the SOC value using the ampere-hour integration algorithm;

[0069] If the SOC value = M, enter the pre-calibration;

[0070] If M < SOC value < 100%, calculate the SOC value using the real-time voltage calibration method.

[0071] The ampere-hour integration algorithm for the SOC value is:

[0072] Set a period t, combine the battery current I during charging within a period t and the accumulation of charge capacity within a period t, or combine the battery current I during discharging within a period t and the subtraction of charge capacity within a period t. Based on the battery current and temperature during battery charging and discharging within the time period t, obtain the Coulomb efficiency coefficient k from the Coulomb efficiency table, and perform the ampere-hour integration algorithm calculation based on the Coulomb efficiency coefficient k. SOC value = current capacity value / total cell capacity. For example, the calculation formula for the current SOC value during discharging is as follows:

[0073]

[0074] where SOC 0 is the initial SOC value, C N is the rated capacity of the cell, I is the battery current; η is the discharge efficiency, k is the Coulomb efficiency coefficient, and t is the time period.

[0075] Preferably, in the charging state, when the SOC value = M, poll the current temperature and current charging current of all cells (hereinafter referred to as the operating conditions), select the SOC-voltmeter applicable to this operating condition, and query the voltage Y when SOC = M as the calibration voltage value. Compare whether the current voltage of the cell reaches the calibration voltage value Y. If it reaches Y, use the real-time voltage calibration to calculate the SOC value and quickly fit and display the SOC value as M to complete the calibration. If it does not reach, continue to calculate the SOC by the ampere-hour integration method. Thereafter, when M ≤ SOC value ≤ 100%, the SOC value changes using the real-time voltage calibration method. The SOC-calibration voltmeter under each operating condition is obtained through prior experiments.

[0076] It should be particularly noted that due to the error of the ampere-hour integration method, when the SOC value obtained by integration reaches 90%, the voltage of the highest single cell may not reach the calibration voltage value. To ensure the reliability of the SOC, it is necessary to wait until the voltage value of the highest single cell reaches the calibration voltage value to determine that the SOC of the cell truly reaches 90%. Therefore, when the voltage value of the highest single cell among all cells is less than the calibration voltage value Y, charging still needs to be continued until the voltage value of the highest single cell is equal to the calibration voltage value Y to complete the calibration. When the SOC value is calculated as 100% using the real-time voltage calibration method, stop charging.

[0077] When M < SOC value < 100%, the method of real-time voltage calibration is: poll the current temperature and current charging current of the cell, select the SOC-voltmeter applicable to this operating condition, and obtain the SOC value through the SOC corresponding to the current voltage. The SOC-voltmeter is the test data of the battery and is known data.

[0078] Step S4: First, determine a SOC limit N for entering voltage calibration. N is a fixed value and 0% < N < M < 100%, preferably 10%. Compare the size of the SOC value with N, and 0% < N < M < 100%.

[0079] If N < SOC value < 100%, still use the ampere-hour integration algorithm to calculate the SOC value;

[0080] If the SOC value = N, enter early calibration;

[0081] If 0 < SOC value < N, use the real-time voltage calibration method to calculate the SOC value.

[0082] Preferably, in the charging state, when the SOC value = N, poll the current temperature and current charging current of all cells (hereinafter referred to as the working condition), select the SOC-voltmeter applicable to this working condition, query the voltage Z when SOC = N as the calibration voltage value. Compare whether the current voltage of the cell reaches Z. If the current voltage of the cell reaches Z, then use the real-time voltage calibration to calculate the SOC value and quickly fit and display the SOC value as N to complete the calibration. If not, continue to calculate the SOC by the ampere-hour integration method. Thereafter, when 0% ≤ SOC value ≤ N, the SOC value changes using the real-time voltage calibration method.

[0083] It should be noted that when the voltage value of the lowest single cell among all cells is greater than the current calibration voltage value Z, it is still necessary to discharge until the voltage of the lowest single cell is equal to the current calibration voltage value Z to complete the calibration. When the SOC value calculated by the real-time voltage calibration method is 0%, stop discharging.

[0084] When 0 < SOC value < N, the real-time voltage calibration method is: poll the current temperature and current discharge current of the cell, select the SOC-voltmeter applicable to this working condition, and obtain the SOC value through the SOC corresponding to the current voltage. The SOC-voltmeter is the test data of the battery.

[0085] The SOC-voltmeter used in the real-time voltage calibration method refers to the original data obtained according to the battery voltage values corresponding to the SOC values under different temperatures and different current conditions during the charging or discharging process.

[0086] The SOC accuracy verification method of the present invention:

[0087] Taking the calculation of the SOC accuracy under the charging and discharging conditions of the battery at 25°C and 0.3C as an example:

[0088] The rated total capacity of the battery cell is Q. The battery is charged with a certain current value, and the SOC value Q1 at time t1 is recorded. At the same time, the charged capacity is calculated by ampere-hour integration. Continue charging until time t2 and record the SOC value Q2 at time t2. At the same time, the capacity charged until time t2 is calculated by ampere-hour integration. k is the Coulomb efficiency coefficient, taking 97.11%. The SOC error is calculated using the following formula (the SOC value corresponding to time t0 is 0%):

[0089]

[0090]

[0091]

[0092] The test result is that the SOC estimation error ≤ 5%.

[0093] SOC curve: Under the charging condition of 25°C and 0.3C, the corresponding relationship between the calculated SOC curve and the original data SOC curve is as Figure 2 shown.

[0094] SOC curve: Under the discharging condition of 25°C and 0.3C, the corresponding relationship between the calculated SOC curve and the original data SOC curve is as Figure 3 shown.

[0095] Combined with the figure corresponding to the SOC curve calculated and the original data SOC curve, it can be seen that the deviation between the SOC curve using this algorithm and the original data SOC curve is within 5%, which can meet the usage requirements of the BMS.

[0096] Provide an embodiment of charging the battery at room temperature of 25°C. The battery in this embodiment is first discharged until the SOC value shows 0%, and then the following steps are carried out:

[0097] Step 1: Charge the battery at a current of 0.3C (18.9A). The SOC value changes using the ampere-hour integration method within the range of 0% to 90%.

[0098] Step 2: Poll the current current and current temperature of all battery cells, query the SOC-voltmeter, and obtain the corresponding calibration voltage Y at 90%. When the voltage of the highest single battery cell reaches the calibration voltage Y, the SOC value is fitted and displayed as 90%, and the calibration is completed;

[0099] Step 3: When the SOC value is between 90% - 100%, the SOC value changes using the real-time voltage calibration method, and when the SOC value calculated using the real-time voltage calibration method is 100%, stop charging.

[0100] An embodiment of discharging the battery at room temperature of 25°C is provided. In this embodiment, the battery is first charged until the displayed SOC value reaches 100%, and then the following steps are carried out:

[0101] Step 1: The battery discharges at a current of 0.3C (18.9), and the SOC value changes using the ampere-hour integration method within the range of 100% to 10%.

[0102] Step 2: Poll the current current and current temperature of all battery cells, query the SOC-voltmeter, and obtain the corresponding calibrated voltage Z at 10%. When the voltage of the lowest single battery cell reaches the calibrated voltage Z, the SOC value is fitted and displayed as 10%, and the calibration is completed.

[0103] Step 3: When the SOC value is between 10% and 0%, the SOC value changes using the real-time voltage calibration method, and when the SOC value calculated using the real-time voltage calibration method is 0%, the discharge stops.

[0104] The SOC calibration method of the present invention can be used in the battery management system of various electrical energy storage systems with battery cells, stored in the memory, and executed by the microprocessor MCU of the electrical energy storage system. For example, it can be used in home energy storage batteries for solar energy, electric vehicles, etc.

[0105] The above content is a further detailed description of the present invention in combination with specific preferred embodiments. It cannot be determined that the specific implementation of the present invention is only limited to these descriptions. For those of ordinary skill in the technical field to which the present invention belongs, without departing from the concept of the present invention, several simple deductions or substitutions can be made, and all should be regarded as belonging to the protection scope of the present invention.

Claims

1. An SOC calibration algorithm, characterized in that: It includes the following steps, Step S1: Obtain a Coulomb efficiency table based on the test data of the battery cell materials at different temperatures and different currents, obtain the total battery cell capacity value by using the rated total capacity value of the battery cell and the Coulomb efficiency table, and acquire and store the SOC-voltmeter of the battery cell; Step S2: Obtain the battery state; When the battery is in the charging state, go to Step S3; When the battery is in the discharging state, go to Step S4; When the battery is in the standby state, keep the SOC value unchanged; Step S3: Set a fixed limit M, compare the size of the SOC value with M, M is a fixed value, and 70% < M < 100%; If 0 < SOC value < M, still use the ampere-hour integration algorithm to calculate the SOC value; If the SOC value = M, enter the pre-calibration; The pre-calibration in Step S3 includes the following steps: Poll the current temperature and current charging current of all battery cells, select the SOC-voltmeter applicable to the current temperature and current charging current of the battery cell, and query the voltage Y when SOC = M as the calibration voltage value; Compare whether the current voltage of the battery cell reaches Y, if it reaches Y, then use the real-time voltage calibration to calculate the SOC value and quickly fit and display the SOC value as M to complete the calibration; If M < SOC value < 100%, use the real-time voltage calibration method to calculate the SOC value; Step S4: Set a fixed limit N, compare the size of the SOC value with N, N is a fixed value, and 0% < N < M < 100%; If N < SOC value < 100%, still use the ampere-hour integration algorithm to calculate the SOC value; If the SOC value = N, enter the pre-calibration; The pre-calibration in Step S4 includes the following steps: Poll the current temperature and current discharging current of all battery cells, select the SOC-voltmeter applicable to the current temperature and current discharging current of the battery cell, and query the voltage Z when SOC = N as the calibration voltage value; Compare whether the current voltage of the battery cell reaches Z, if it reaches Z, then use the real-time voltage calibration to calculate the SOC value and quickly fit and display the SOC value as N to complete the calibration; If 0 < SOC value < N, use the real-time voltage calibration method to calculate the SOC value.

2. An SOC calibration algorithm according to claim 1, characterized in that: When the voltage value of the highest single battery cell among all battery cells is less than the current calibration voltage value Y, charging still needs to be carried out until the calibration voltage value Y is reached.

3. An SOC calibration algorithm according to claim 1, characterized in that: When the voltage value of the lowest single battery cell among all battery cells is greater than the current calibration voltage value Z, discharging still needs to be carried out until the calibration voltage value Z is reached.

4. An SOC calibration algorithm according to claim 1, characterized in that: M = 90% in Step S3, and N = 10% in Step S4.

5. An SOC calibration algorithm according to claim 1, characterized in that: In Steps S3 and S4, based on the battery current and temperature within the time period t, obtain the Coulomb efficiency coefficient k from the Coulomb efficiency table, and perform the ampere-hour integration algorithm calculation based on the Coulomb efficiency coefficient k. The formula for calculating the SOC value by the ampere-hour integration algorithm is: Where SOC 0 is the initial SOC value, C N is the total capacity of the battery cell, I is the battery current; η is the discharge efficiency, k is the Coulomb efficiency coefficient, and t is the time period.

6. A SOC calibration algorithm according to claim 1, characterized in that: In the step S3, the method of real-time voltage calibration is to poll the current temperature and current charging current of the battery cell, select a SOC-voltmeter applicable to the current temperature and current charging current of the battery cell, and obtain the SOC value through the SOC corresponding to the current voltage; In the step S4, the method of real-time voltage calibration is to poll the current temperature and current discharge current of the battery cell, select a SOC-voltmeter applicable to the current temperature and current discharge current of the battery cell, and obtain the SOC value through the SOC corresponding to the current voltage.

7. A SOC calibration algorithm according to claim 1, characterized in that: In the step S3, when the SOC value calculated by the method of real-time voltage calibration is 100%, charging is stopped; In the step S4, when the SOC value calculated by the method of real-time voltage calibration is 0%, discharging is stopped.

Images