Abnormal battery cell early warning method based on battery cell internal resistance estimation and automobile

Abstract

The invention discloses an abnormal battery cell early warning method based on battery cell internal resistance estimation. The method comprises the following steps: calculating according to the internal resistance of each single battery cell in a battery to obtain the variation coefficient of the internal resistance corresponding to the single battery cell and the abnormal frequency number of the single battery cell in a set time range; and performing abnormal early warning on each single cell in the battery based on the abnormal coefficient and / or the abnormal frequency of the single cell. According to the scheme, accurate and reliable abnormal early warning can be carried out on the single battery cell, the abnormal state can be found in time, the early warning is output, and the safety of the battery is improved.

Description

Technical Field

[0001] The present invention relates to the field of new energy batteries, and in particular to an abnormal battery cell early warning method based on battery cell internal resistance estimation. Background Art

[0002] With the scarcity of oil resources and the increasing air pollution caused by other flammable gases, the development of energy-saving and environmentally friendly electric vehicles has become an increasingly hot topic in today's society. As the main power source and energy carrier of electric vehicles, the performance of batteries directly affects the performance indicators of pure electric vehicles. Battery health status (SOH) is increasingly valued as an important indicator for evaluating battery performance.

[0003] Traditional SOH estimation methods are generally based on cut-off voltage or remaining power methods. This type of battery health status monitoring is based on the evaluation of the entire battery as a whole. However, in fact, the battery is composed of multiple cells. Due to the large number of cells, if some or one of the cells fails or is abnormal, it will cause life problems for the entire battery pack.

[0004] Due to the barrel effect, a faulty or abnormal battery cell will cause abnormality or failure of the entire battery, but in fact other battery cells are still intact. Therefore, it is necessary to accurately monitor the abnormality of a certain battery cell so that the abnormality can be located in time after the telecommunications abnormality occurs, and provide basic monitoring data for repairing and replacing faulty and abnormal battery cells. However, the existing technology generally cannot monitor the abnormality of a single battery cell. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and provide an abnormal battery cell early warning method based on battery cell internal resistance estimation. The data acquisition and monitoring during the battery charging process can be used to evaluate the status of each battery cell, and abnormal batteries can be quickly and accurately identified, so as to provide basic monitoring and prediction data for subsequent battery pack replacement and battery maintenance.

[0006] In order to achieve the above-mentioned purpose, the technical solution adopted by the present invention is: an abnormal cell warning method based on cell internal resistance estimation, comprising calculating the coefficient of variation of the corresponding internal resistance of each single cell in the battery and the abnormal frequency of the single cell within a set time range according to the internal resistance of each single cell in the battery; and issuing an abnormal warning for each single cell in the battery based on the abnormal coefficient and / or abnormal frequency of the single cell.

[0007] The calculation method of the coefficient of variation of the internal resistance of a single cell includes: Where Rn is the internal resistance of the nth single cell, R 平均 To calculate the average resistance of all cells based on the resistance of each single cell, R 标准差 is the standard deviation of the resistance of all n battery cells.

[0008] When the abnormal coefficient of a single cell is greater than a set threshold value X, it is judged to be an abnormal single cell; otherwise, it is judged to be a normal single cell.

[0009] When calculating the coefficient of variation during the battery charging process, it is determined whether the coefficient of variation conditions are met. When the battery is in the charging state and the SOC of the battery is within the set range, it is determined that the coefficient of variation calculation conditions are met, and the internal resistance of each single cell is calculated and the corresponding coefficient of variation is calculated.

[0010] The method for solving the internal resistance of a single cell during battery charging includes: selecting a moment when the battery SOC is between 40% and 70% and the charging current is continuously greater than the X1C setting time and the current is stable, and recording the current I and all single cell voltages U at that moment; requesting the charging current to suddenly change to I' through the BMS, and after judging that the output current of the charger is stable and the duration is greater than the time threshold, recording the voltage U' of each single cell after the current sudden change;

[0011] Then, the difference between the voltage U' of the single cell after the current mutation and the voltage U of the corresponding cell is divided by the difference between the current I' after the mutation and the current I to obtain the internal resistance value corresponding to the single cell.

[0012] During the charging process, the internal resistance of the single battery cell in a continuous time period is obtained to determine whether the continuous time period is a valid time period. If so, the continuous time period is divided into several sub-time periods, and the internal resistance deviation and internal resistance deviation of all single battery cells in each sub-time period are calculated. The internal resistance deviation and internal resistance deviation of each single battery cell are calculated in each sub-time period. If the internal resistance deviation and internal resistance deviation are greater than the deviation threshold Y, the abnormal frequency of the corresponding single battery cell is counted once. The abnormal frequency value of each battery cell is calculated based on all sub-time periods, and the single battery cell whose abnormal frequency value is greater than the counting threshold is judged as an abnormal battery cell.

[0013] When the internal resistance at each collection moment is collected and calculated in a continuous time period, the continuous time period is determined to be a valid time period.

[0014] The internal resistance deviation of the single cell in the sub-time period and B n The calculation formula is:

[0015]

[0016] Where t1 represents the start time of the sub-time period, t2 represents the end time of the sub-time period, and R n，t R represents the internal resistance of the nth cell at time t. 0，t It represents the total internal resistance of the battery at time t, and m represents the number of single cells in the battery pack.

[0017] If a single cell is abnormal based on any one of the battery cell variation coefficient and the abnormal frequency, then the single cell is judged to be abnormal.

[0018] A new energy electric vehicle comprises the abnormal battery cell early warning method to provide early warning for the battery.

[0019] The advantages of the present invention are that the status of each battery cell can be evaluated through data collection and monitoring during the battery charging process, and abnormal cells can be quickly and accurately identified, providing basic monitoring and prediction data for subsequent battery pack cell replacement and battery repair. The health and safety of each string of cells can be evaluated by estimating the internal resistance of each string of cells. Identifying certain cells with abnormal internal resistance helps to identify cells with safety hazards and replace problematic cells in a timely manner to avoid overcharging and over-discharging caused by overcurrent in individual cells due to differences in internal resistance. This achieves the purpose of improving the overall battery life and battery safety.

[0020] Dual parameters are used for early warning. If any one parameter is abnormal, the battery cell is judged to be abnormal, so that the battery cell can be effectively estimated and reminded in time. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The following is a brief description of the contents expressed in the drawings of the present invention and the symbols in the drawings:

[0022] Figure 1 It is a schematic diagram of the principle of the internal resistance early warning of the present invention;

[0023] Figure 2 Schematic diagram of abnormal frequency of battery cells in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The specific implementation of the present invention will be further explained in detail below by describing the optimal embodiment with reference to the accompanying drawings.

[0025] The present embodiment provides an abnormal battery cell early warning method based on battery cell internal resistance estimation, which calculates the battery cell internal resistance and then respectively obtains the battery cell variation coefficient and abnormal frequency, and then performs abnormal judgment based on the battery cell variation coefficient and abnormal frequency, and sends out an early warning signal for the battery cell judged as an abnormal battery cell, so as to promptly remind the user to perform maintenance and avoid the safety impact of the battery caused by a single battery cell.

[0026] like Figure 1 As shown, an abnormal battery cell early warning method based on battery cell internal resistance estimation includes the following steps:

[0027] (1) Determine whether the battery cell internal resistance calculation conditions are met, and if so, start the single cell internal resistance calculation step;

[0028] (2) in step (1), the internal resistance of each single cell is calculated, and then the coefficient of variation of the internal resistance of each single cell in the battery and the frequency of abnormality of the single cell within a set time range are calculated based on the internal resistance of each single cell in the battery;

[0029] (3) An abnormal warning is issued to each single cell in the battery based on the abnormal coefficient and / or abnormal frequency of the single cell.

[0030] In step (1), determine whether the battery is in the fast charging process. If so, start the single cell internal resistance calculation step at this time; as for whether it is in the fast charging state, the interactive data between the charging gun and the vehicle-mounted charging port can be collected to obtain whether it is in the charging state at this time and whether the charging state at this time is fast charging or satisfied, so as to determine whether to enter the single cell internal resistance calculation step. After entering the single cell internal resistance calculation step, calculate the internal resistance of each single cell in the battery. The calculation steps include: when the battery is charging, monitor the battery SOC state in real time, select the moment when the battery SOC is between 40-70% and the charging current is continuously greater than X1 C for the set time and the current is stable, and record the current I and all single cell voltages U at this moment; request the charging current to suddenly change to I' through BMS, and after judging that the output current of the charger is stable and the duration is greater than the time threshold, record the voltage U' of each single cell after the current sudden change;

[0031] Then, the difference between the voltage U' of the single cell after the current mutation and the voltage U of the corresponding cell is divided by the difference between the current I' after the mutation and the current I to obtain the internal resistance value corresponding to the single cell. In this way, the internal resistance corresponding to each battery cell can be calculated, and the calculation formulas are: R1'=(U1'-Un1) / I'-I, R2'=(U2'-U2) / I'-I, ... Rn'=(Un'-Un) / I'-I; R1', R2'...Rn' are the internal resistances corresponding to n battery cells respectively. When calculating the internal resistance, it will be uploaded to the cloud platform for monitoring, and the internal resistance will be monitored through the cloud platform, and then the early warning information and alarm reminders will be sent through the cloud platform.

[0032] After the cloud platform obtains the internal resistance of each battery cell through monitoring, it calculates the coefficient of variation of each battery cell. After calculating the internal resistance of each battery cell, it calculates the average resistance of all cells, the standard deviation of the resistance of all n cells of the battery, and other values. Then, it calculates the variation data of the internal resistance corresponding to each battery cell. The calculation formula is:

[0033] Where An is the coefficient of variation of the nth battery cell; Rn is the internal resistance of the nth battery cell, R 平均 To calculate the average resistance of all cells based on the resistance of each single cell, R标准差 is the standard deviation of the resistance values ​​of all n battery cells. When the abnormal coefficient of a single cell is greater than the set threshold value X, it is judged as an abnormal single cell; otherwise, it is judged as a normal single cell. After judging that the battery cell is an abnormal cell, a reminder signal is sent to the user terminal through the display or instrument on the vehicle or through the cloud platform to remind the user that there is an abnormal cell in the battery and the ID of the abnormal cell, so that the user can repair or replace the abnormal cell in time.

[0034] After calculating the internal resistance of each battery cell, all the internal resistances of the cells are uploaded to the platform through TBOX for monitoring. The big data platform calculates the average resistance R of all the cells based on the resistance of each cell. 平均 、The median resistance of all cells R 中值 And the standard deviation R of the resistance of n cells 标准差 . It is used to evaluate the variation of the internal resistance of a single cell. When the coefficient of variation is higher than the variation threshold X, the corresponding cell is abnormal.

[0035] During the charging process, the internal resistance of the single battery cell in a continuous time period is obtained to determine whether the continuous time period is a valid time period. If so, the continuous time period is divided into several sub-time periods, and the internal resistance deviation and internal resistance deviation of all single battery cells in each sub-time period are calculated. The internal resistance deviation and internal resistance deviation of each single battery cell are calculated in each sub-time period. If the internal resistance deviation and internal resistance deviation are greater than the deviation threshold Y, the abnormal frequency of the corresponding single battery cell is counted once. The abnormal frequency value of each battery cell is calculated based on all sub-time periods, and the single battery cell whose abnormal frequency value is greater than the counting threshold is judged as an abnormal battery cell.

[0036] In the process of calculating the internal resistance, the calculation method of the battery internal resistance is integrated into the BMS in the form of software. When the BMS collects the voltage and current required for the internal resistance, it collects and calculates it periodically. Therefore, the internal resistance collected and calculated is the internal resistance of each battery cell of multiple battery groups that are continuously output at the same time interval. The cloud platform determines whether the uploaded internal resistance data is continuous, that is, whether all the internal resistance value data in a group of battery packs are obtained at the same time interval. If so, it means that the time period is continuous and is judged to be a valid time period. That is, when the internal resistance at each collection moment is collected and calculated within a continuous time period, the continuous time period is judged to be a valid time period, and the internal resistance of the battery cell can be calculated at each collection moment. Then, the internal resistance deviation and the internal resistance deviation of the single cell in each sub-time period can be obtained, and the internal resistance deviation and the internal resistance deviation of the single cell in the sub-time period can be obtained. n The calculation formula is:

[0037]

[0038] Where t1 represents the start time of the sub-time period, t2 represents the end time of the sub-time period, and Rn，t R represents the internal resistance of the nth cell at time t. 0，t Indicates the total internal resistance of the battery at time t, and m indicates the number of cells in the battery pack. Then, based on the internal resistance deviation and Bn, it is determined whether the nth cell in this sub-time period is abnormal. If so, an abnormality is recorded for this cell (nth cell), and then the abnormal number of times for each cell in the effective time period is counted, and the abnormal number of times is compared with the set abnormal number threshold. The cell corresponding to the abnormal number threshold is greater than the abnormal number threshold and is identified as an abnormal cell. Then, a warning signal is sent to the user to remind the user of the abnormal state of this cell.

[0039] Based on the battery cell variation coefficient and the frequency of abnormalities, if any of these two parameters is judged to be abnormal for a single cell, an alarm will be issued. This can not only achieve effective and reliable monitoring and early warning of resistance, but also double internal resistance abnormality judgment to avoid missing resistance abnormalities, thus achieving effective and reliable judgment and early warning of abnormal cells.

[0040] The embodiment also provides a new energy electric vehicle, which includes the abnormal battery cell warning method in the above embodiment to warn the battery, thereby realizing the new energy vehicle with the battery cell abnormality warning function, thereby improving the safety of the new energy vehicle.

[0041] This solution estimates the internal resistance of each string of cells and evaluates the health and safety of each string of cells. Identifying certain cells with abnormal internal resistance helps to identify cells with safety hazards and replace problematic cells in a timely manner to avoid overcharging and over-discharging caused by overcurrent in individual cells due to differences in internal resistance. This will achieve the purpose of improving the overall battery life and battery safety. Using a simple volt-ampere method to calculate the internal resistance, without considering the accuracy of the estimated value of the cell internal resistance, the coefficient of variation of the internal resistance of the single cell, as well as the internal resistance deviation and threshold of the single cell over a period of time, can be used to monitor and warn of battery safety and improve the overall battery life.

[0042] When the internal resistance abnormality warning is issued, the premise is to calculate the internal resistance. The internal resistance estimation scheme in this scheme is only triggered in the fast charging mode. After the internal resistance estimation is completed, all battery cell internal resistances, internal resistance median / normal distribution standard deviation and related numbers will be sent to the T-BOX terminal platform, and the platform will process them and issue a warning signal for warning reminders. The specific steps include:

[0043] 1. During fast charging, select a point where the battery SOC is between 40% and 70%, and determine that the charging current ≥ X1C (1C represents the current intensity when the battery is charged and discharged at its rated capacity, X1 ≥ 0.6), and the current is stable (the fluctuation within 10S < 5A), (continuously meet the requirement for t1s); Under normal circumstances, the internal resistance of a lithium battery is very small, and the slow charging current is small. When the internal resistance is very small, the voltage difference change is very small, and the error will be enlarged due to the influence of the voltage acquisition accuracy, resulting in inaccuracy. Therefore, it is necessary to estimate the internal resistance under fast charging. The current fluctuation within 10S less than 5A is a stability limit for the current value, and the continuous t1 s is a limit for the stable voltage of the battery cell.

[0044] 2. Record the current I and all the single-cell voltages U1, U2,... Un at this time;

[0045] 3. The BMS requests a sudden change in the current to I' = X2C (X2 ≤ 0.1), and determine that the current output by the charger is stable after t2min and the current is stable (the fluctuation within 10S < 5A and X2C - 5A < current < X2C + 5A); Set the maximum time (t3min) for the current in the X2C interval. If not satisfied, return to normal charging; t2min is also a limit for the stable voltage of the battery cell, and t3min is a requirement for the current time. If the current output by the charging pile cannot be stabilized at X2C, then we need to exit the process and resume normal charging. This is to solve the problem that some charging piles cannot output a very low current, but too small a current will affect the charging time and cause customer complaints.

[0046] 4. Record the current I' and all the single-cell voltages U1', U2',... Un' at this time again;

[0047] 5. Estimate the DC internal resistance R of each battery cell according to the "volt-ampere method".

[0048] R1' = (U1' - Un1) / (I' - I), R2' = (U2' - U2) / (I' - I),... Rn' = (Un' - Un) / (I' - I)

[0049] 6. Calculate the three highest internal resistances R according to the above internal resistance values max1 、R max2 、R max3 、the three lowest internal resistances R min1 、R min2 、R min3 。

[0050] 7. The three highest internal resistances R max1 、R max2 、R max3 、the three lowest internal resistances R min1 、R min2 、R min3、The resistance values ​​of all cells are uploaded to the platform through T-BOX. The big data platform calculates the average resistance value R of all cells based on the resistance value of each single cell. 平均 、The median resistance of all cells R 中值 And the standard deviation R of the resistance of n cells 标准差 . It is used to evaluate the variation of the internal resistance of a single cell. When the coefficient of variation is higher than the variation threshold X, the corresponding cell is abnormal.

[0051] 8. The calculation formula of the coefficient of variation of the internal resistance of a single cell is as follows:

[0052]

[0053] 9. The abnormal cell determination method based on the internal resistance deviation fault prediction model specifically includes the following steps: obtaining the internal resistance of each cell within a set time period; judging whether the set time period is a valid time period (a fast charging process, from plugging in the gun to pulling out the gun is a set time period, the fast charging plugging time is greater than 5 minutes and the SOC rises by 5% is a valid time period, and is judged together with the previously mentioned set time (the customer may not charge even if plugging in the gun, which is considered an invalid time period)). If the judgment result is yes, the valid time period is divided into several sub-time periods, and the internal resistance deviation and the internal resistance deviation of all the single cells in each sub-time period are calculated. If the internal resistance deviation and the internal resistance deviation are greater than the deviation threshold value Y, the corresponding single cell abnormal frequency is counted once to obtain the abnormal frequency of each single cell; the single cell whose abnormal frequency value is greater than the counting threshold is identified as an abnormal cell. Furthermore, the internal resistance deviation and B of the single cell in the sub-time period are n The calculation formula is as follows:

[0054]

[0055] Where t1 represents the start time of the sub-time period, t2 represents the end time of the sub-time period, and R n，t R represents the internal resistance of the nth cell at time t. 0，t It represents the total internal resistance of the battery at time t, and m represents the number of single cells in the battery pack.

[0056] 10. The advantage of the present invention is that we do not need to consider the absolute accuracy of the estimated value of the battery cell internal resistance, and we can study the relative data of the battery cell internal resistance:

[0057] ① The internal resistance of each cell can be compared horizontally to determine the discreteness of the resistance and the variation of the internal resistance. This can help identify the problems of some abnormal cells, help to identify cells with potential safety hazards, and replace the problematic cells in a timely manner to avoid overcharging and over-discharging caused by overcurrent in individual cells due to differences in internal resistance. This will help to improve the overall battery life and battery safety.

[0058] ② A longitudinal study can be conducted through all the internal resistance values ​​of the battery cell during its life cycle. The big data platform can record the internal resistance values ​​of the battery cell throughout its life cycle and sort out the corresponding relationship between the internal resistance and SOH during the entire life cycle. This can not only provide early warning for the battery SOH value and improve battery safety, but also collect basic information for the cascade utilization of the battery.

[0059] In the embodiment of the present invention, when calculating Bn according to the above scheme, if the time window is 1 month, such as 2023.08.15 to 2023.09.15, the effective setting time period is longer than the time window and includes the time window. The setting time period is 2023.07.15 to 2023.09.15. There must be at least 2 charging data with the SOC interval of 40%-70% in the effective setting time period. The sub-time period can be calculated in days or weeks. The effective time period is divided into several sub-time periods, according to the formula Calculate the internal resistance deviation of all single cells in each sub-time period. If the internal resistance deviation is greater than the deviation threshold, the abnormal frequency of the corresponding single cell is counted once to obtain the abnormal frequency of each single cell. The single cell whose abnormal frequency value is greater than the counting threshold is identified as an abnormal cell. Figure 2 This is a simulation statistical result diagram of abnormal cells based on the deviation fault prediction model. The horizontal axis is the cell number and the vertical axis is the abnormal frequency data. Figure 2 From the statistical data, it can be seen that the three cells with significantly higher frequency are abnormal cells and need to be warned.

[0060] Obviously, the specific implementation of the present invention is not limited to the above-mentioned methods. As long as various non-substantial improvements are made using the method concept and technical solution of the present invention, they are all within the protection scope of the present invention.

Claims

1. An abnormal battery cell early warning method based on battery cell internal resistance estimation, characterized in that: Including calculating the coefficient of variation of the corresponding internal resistance of each single cell in the battery and the frequency of abnormality of the single cell within a set time range; Based on the abnormal coefficient and / or abnormal frequency of the single cell, an abnormal warning is issued to each single cell in the battery.

2. The abnormal cell early warning method based on cell internal resistance estimation according to claim 1, characterized in that: The calculation method of the coefficient of variation of the internal resistance of a single cell includes: Where Rn is the internal resistance of the nth single cell, R 平均 To calculate the average resistance of all cells based on the resistance of each single cell, R 标准差 is the standard deviation of the resistance of all n battery cells.

3. The abnormal battery cell early warning method based on battery cell internal resistance estimation according to claim 1 or 2, characterized in that: When the abnormal coefficient of a single cell is greater than a set threshold value X, it is judged to be an abnormal single cell; otherwise, it is judged to be a normal single cell.

4. The abnormal cell early warning method based on cell internal resistance estimation according to claim 1 or 2, characterized in that: When calculating the coefficient of variation during the battery charging process, it is determined whether the coefficient of variation conditions are met. When the battery is in the charging state and the SOC of the battery is within the set range, it is determined that the coefficient of variation calculation conditions are met, and the internal resistance of each single cell is calculated and the corresponding coefficient of variation is calculated.

5. The abnormal cell early warning method based on cell internal resistance estimation according to claim 4, characterized in that: The method for solving the internal resistance of a single cell during battery charging includes: selecting a moment when the battery SOC is between 40% and 70% and the charging current is continuously greater than the X1C setting time and the current is stable, and recording the current I and all single cell voltages U at that moment; requesting the charging current to suddenly change to I' through the BMS, and after judging that the output current of the charger is stable and the duration is greater than the time threshold, recording the voltage U' of each single cell after the current sudden change; Then, the difference between the voltage U' of the single cell after the current mutation and the voltage U of the corresponding cell is divided by the difference between the current I' after the mutation and the current I to obtain the internal resistance value corresponding to the single cell.

6. The abnormal battery cell early warning method based on battery cell internal resistance estimation according to claim 1, characterized in that: During the charging process, the internal resistance of the single battery cell in a continuous time period is obtained to determine whether the continuous time period is a valid time period. If so, the continuous time period is divided into several sub-time periods, and the internal resistance deviation and internal resistance deviation of all single battery cells in each sub-time period are calculated. The internal resistance deviation and internal resistance deviation of each single battery cell are calculated in each sub-time period. If the internal resistance deviation and internal resistance deviation are greater than the deviation threshold Y, the abnormal frequency of the corresponding single battery cell is counted once. The abnormal frequency value of each battery cell is calculated based on all sub-time periods, and the single battery cell whose abnormal frequency value is greater than the counting threshold is judged as an abnormal battery cell.

7. The abnormal cell early warning method based on cell internal resistance estimation according to claim 6, characterized in that: When the internal resistance at each collection moment is collected and calculated in a continuous time period, the continuous time period is determined to be a valid time period.

8. The abnormal cell early warning method based on cell internal resistance estimation according to claim 6 or 7, characterized in that: The internal resistance deviation of the single cell in the sub-time period and B n The calculation formula is: Where t1 represents the start time of the sub-time period, t2 represents the end time of the sub-time period, and R n，t R represents the internal resistance of the nth cell at time t. 0，t It represents the total internal resistance of the battery at time t, and m represents the number of single cells in the battery pack.

9. The abnormal cell early warning method based on cell internal resistance estimation according to any one of claims 1 to 8, characterized in that: If a single cell is abnormal based on any one of the battery cell variation coefficient and the abnormal frequency, then the single cell is judged to be abnormal.

10. A new energy electric vehicle, characterized in that: The automobile includes an abnormal battery cell early warning method as described in any one of claims 1-9 for battery abnormality warning.

Images