A method for sensing the safety status of power batteries based on the accumulation of voltage center cluster deviation

Abstract

The present invention provides a method for sensing the safety status of a power battery based on the accumulation of voltage center cluster deviations, which gives full play to the advantages of high real-time performance and comprehensive and rich data samples of big data on actual operation of electric vehicles, and can achieve more precise supervision of each charging process and each cell by performing filtering, normalization and other processing on the voltage-related data of each cell in the battery pack. The method integrates parameters such as the normalized cumulative value of the risk of cell voltage deviation from the center cluster voltage, the normalized cumulative discrete entropy of the risk of cell voltage deviation from the center position voltage, and the cumulative interquartile range multiples of the unit cell voltage deviation from the center position voltage, to make risk judgments from multiple dimensions, and can provide better robustness and versatility than the existing technology.

Description

Technical Field

[0001] The present invention belongs to the technical field of power battery safety status supervision, and specifically relates to a method for calculating single cell voltage center deviation and sensing the power battery safety status by using power battery big data. Background Art

[0002] During the long-term charging and discharging process of lithium-ion power battery packs used in electric vehicles, due to the difference in charging acceptance between the individual cells in the battery pack, there is a general trend of gradually increasing and diverging charge differences between the individual cells in the pack, which can easily cause the discreteness of the batteries inside the battery pack to increase, and the performance degradation of individual batteries to increase. In severe cases, it will cause the entire battery pack to fail. For a battery pack composed of multiple cells connected in series and parallel to form a power battery, the output capacity, output power and service life of the entire battery pack often depend on the single cell with the worst performance in the battery pack, which means that the inconsistency of the single cell is the most critical factor affecting the performance of the entire power battery. In general, in the early stage of the power battery's operation, the inconsistency between the cells is not obvious, but after a certain period of repeated charging and discharging, the inconsistency between the cells will increase sharply, which will lead to the premature termination of the power battery's life.

[0003] In the prior art, the safety status of the battery is mostly sensed by thresholds of parameters such as voltage, temperature, and pressure difference. Although the safety or risk status reflected by these parameters is relatively intuitive, there is a defect that the potential risks of the battery cannot be effectively identified, and early warning cannot be provided for many faults. In some prior arts that directly obtain the cumulative risk value by accumulating the voltage and the center position, the main solution is to accumulate all voltage values, but there is a common problem of a high misjudgment rate in the identification of single safety risks. Summary of the invention

[0004] In view of this, in view of the technical problems existing in the above-mentioned field, the present invention provides a method for sensing the safety status of a power battery based on the accumulation of voltage center cluster deviation, which specifically includes the following steps:

[0005] Step 1: for the SOC interval of 10%-90% in the historical charging segment of the electric vehicle to be sensed, extract the voltage value, voltage median, voltage upper quartile and voltage lower quartile of each cell at different times, and use these parameters to calculate the cell voltage deviation of each cell at different times from the central cluster value;

[0006] Calculate the ratio of the cell voltage deviation from the central cluster value of a certain cell at different times to the sum of the cell voltage deviation from the central cluster values ​​of all cells in the battery pack, and calculate the unit normalized cumulative value of the ratio based on the charging duration and the number of data frames for a certain charging process as the unit cell voltage deviation from the central cluster voltage risk normalized cumulative value of each cell to reflect the outlier state of each cell;

[0007] Step 2: extract the median voltage values ​​of all cells in the battery pack at different times, and calculate the cell voltage deviation from the center value of each cell at different times in combination with the cell voltage value; calculate the ratio of the cell voltage deviation from the center value of a certain cell at different times to the sum of the cell voltage deviation from the center value of all cells in the battery pack, and for a certain charging process, calculate the normalized cumulative value of the ratio based on the charging duration and the number of data frames as the normalized cumulative value of the cell voltage deviation from the center voltage risk of each cell; construct a normalized cumulative matrix of the cell voltage deviation from the center voltage risk for each cell using the normalized cumulative value of the cell voltage deviation from the center voltage risk corresponding to each charge of each cell;

[0008] Divide the values ​​of the matrix elements in the normalized cumulative matrix of the risk of the cell voltage deviating from the center voltage into several intervals, calculate the probability of each cell voltage value falling into different intervals during each charging process, and calculate the normalized cumulative discrete entropy of the risk of the cell voltage deviating from the center voltage of each cell during each charging process based on the probability;

[0009] Step 3: for the normalized cumulative value of the risk of the single cell voltage deviating from the center voltage obtained in step 2, the normalized cumulative value of the risk of the single cell voltage deviating from the center voltage is calculated in combination with the number of data frames in the charging process from time t1 to time t2, the median and upper and lower quartiles of the normalized cumulative value of the risk of the single cell voltage deviating from the center voltage are calculated, and on this basis, the cumulative interquartile range multiple of the single cell voltage deviating from the center voltage of each single cell is calculated;

[0010] Step 4: In combination with the actual operating conditions, corresponding thresholds are set for the normalized cumulative value of the risk of the single cell voltage deviating from the central cluster voltage, the normalized cumulative discrete entropy of the risk of the single cell voltage deviating from the central position voltage, and the cumulative interquartile range multiples of the single cell voltage deviating from the central position voltage obtained in the above steps. When at least two results are greater than or equal to the threshold, it indicates that there is abnormal discreteness in the corresponding battery single cell, and an early warning is issued for the power battery safety risk during the charging process.

[0011] Furthermore, in step 1, the voltage value is first compared with the voltage median at different times to perform data filtering, and the following calculation of the deviation of the cell voltage from the center cluster value is performed only for the battery cells whose cell voltage deviates from the median to a predetermined degree:

[0012]

[0013] Where, ΔV ti represents the deviation of the cell voltage of cell i from the center cluster value at time t, V ti Represents the voltage value of cell i at time t, V t50％ 、V t75％ and V t25％ They represent the median and upper and lower quartiles of the voltage of cell i at time t respectively;

[0014] The ratio of the cell voltage deviation from the center cluster value of a cell at different times to the sum of all cell voltage deviations from the center cluster values ​​in the battery pack is calculated based on the following formula as the normalized cumulative value of the cell voltage deviation from the center cluster voltage risk:

[0015]

[0016] In the formula, H i It represents the normalized cumulative value of the risk of the cell voltage of cell i deviating from the central cluster voltage during a charging process from time t1 to time t2. It represents the sum of the voltage deviations of all cells in the battery pack from the center cluster at time t, and m is the total number of cells;

[0017] Combined with the number of data frames n from time t1 to time t2 during the charging process, the normalized cumulative value κ of the risk of the unit cell voltage deviating from the central cluster voltage is calculated. i :

[0018]

[0019] Furthermore, in step 2, the deviation of the cell voltage from the center position value is first calculated based on the following formula:

[0020] ΔV * ti =|V ti -V * t50％ |

[0021] Where, ΔV ti * Indicates that the cell voltage of cell i at time t deviates from the center value, V ti Represents the voltage value of cell i at time t, V * t50％ Represents the median voltage value of all cells in the battery pack at time t;

[0022] The ratio of the cell voltage deviation from the center position value of the cell i at different times to the sum of the cell voltage deviations from the center position values ​​of all cells in the battery pack is calculated based on the following formula:

[0023]

[0024] In the formula, G * ti It represents the calculated ratio result, i.e. the normalized value of the risk of the monomer voltage deviating from the center voltage, where m is the total number of monomers;

[0025] For a charging process from time t1 to time t2, the * ti The normalized cumulative value of the risk of the cell voltage of cell i deviating from the center voltage is obtained:

[0026]

[0027] All monomers H * i It is constructed as the normalized cumulative matrix of the risk of the monomer voltage deviating from the central voltage:

[0028] H * 1×m ={H * 1 ,H * 2 ,H * 3 ..........H * m}

[0029] Where m is the total number of monomers;

[0030] Extract the maximum value among the elements of the matrix H * max With the minimum value H * min Calculate the range ΔH of the normalized cumulative value of the cell voltage deviation from the center voltage risk during this charging process * :

[0031] ΔH * =H * max -H * min

[0032] With the maximum value H * max With the minimum value H * min As the endpoint, a is the determining parameter of the interval length, l is the number of intervals, and several probability intervals are divided so that the normalized cumulative value of the voltage risk of any single cell k deviating from the central voltage satisfies:

[0033]

[0034] The probability p that the normalized cumulative value of the risk of the cell voltage of cell i deviating from the center voltage falls into the jth interval i,j :

[0035]

[0036] In the formula, c i,j It indicates the number of times that the normalized cumulative value of the cell voltage deviation risk of cell i falls into the jth interval. It represents the sum of the number of times the normalized cumulative value of the cell voltage deviation risk of cell i falls into each interval, j∈l;

[0037] Using the calculated probability p i,j Calculate the normalized cumulative discrete entropy E of the risk of the cell voltage of cell i deviating from the center voltage in this charge i :

[0038]

[0039] Furthermore, for the normalized cumulative value H of the risk of the cell voltage deviating from the center voltage obtained in step 2, * i , combined with the number of data frames n in the charging process from time t1 to time t2, calculate the normalized cumulative value κ of the risk of the unit cell voltage deviating from the center voltage * i :

[0040]

[0041] Calculate the κ * i The median κ * i50％ , upper quartile κ * i75％ and the lower quartile κ * i25％ , and the cumulative interquartile range multiple η of the unit monomer voltage deviation from the center voltage is calculated based on the following formula i :

[0042]

[0043] Furthermore, based on the actual operating conditions, the normalized cumulative value κ of the risk of the cell voltage of the battery cell i deviating from the central cluster voltage is calculated. i , the normalized cumulative discrete entropy E of the risk of the single voltage deviating from the central voltage i And the cumulative interquartile range multiple of the unit cell voltage deviation from the center voltage η i Set the corresponding threshold, when κ is satisfied i ≥0.00015, Ei ≥1.5 and η i When at least two of the conditions ≥20 are met, it is determined that cell i is outlier in this charging, and an early warning of the power battery safety risk is issued.

[0044] The method for sensing the safety status of a power battery based on the cumulative deviation of the voltage center cluster provided by the present invention gives full play to the advantages of high real-time performance and comprehensive and rich data samples of the big data of actual operation of electric vehicles. By performing filtering, normalization and other processing on the voltage-related data of each cell in the battery pack, more precise supervision can be achieved for each charging process and each cell. The method integrates parameters such as the normalized cumulative value of the risk of cell voltage deviation from the center cluster voltage, the normalized cumulative discrete entropy of the risk of cell voltage deviation from the center position voltage, and the cumulative interquartile range multiples of the unit cell voltage deviation from the center position voltage, and makes risk judgments from multiple dimensions. Compared with the existing technology, it can provide better robustness and versatility. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A flow chart of the present invention for calculating the normalized cumulative value of the risk of the unit cell voltage deviating from the central cluster voltage;

[0046] Figure 2 A box diagram of the median and upper and lower quartiles of the single cell voltage data calculated by the present invention;

[0047] Figure 3 The κ calculated by the present invention * i Box plot of the median and upper and lower quartiles. DETAILED DESCRIPTION

[0048] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] Due to the different operating conditions of automotive power batteries, the vehicle driving discharge data is relatively chaotic. At the same time, the charge transfer impedance of the single cells in the battery pack is relatively large when fully charged and empty, resulting in large polarization internal resistance of the single cells and poor consistency of the single cells. The above factors will affect the accuracy of the model's calculation of single cell outliers. For this reason, in the method for sensing the safety status of a power battery based on the cumulative deviation of the voltage center cluster provided by the present invention, a unified calculation is performed for the SOC interval of 0%-90% in the battery charging segment. The method specifically includes the following steps:

[0050] Step 1: for the SOC interval of 10%-90% in the historical charging segment of the electric vehicle to be sensed, extract the voltage value, voltage median, voltage upper quartile and voltage lower quartile of each cell at different times, and use these parameters to calculate the cell voltage deviation of each cell at different times from the central cluster value;

[0051] Calculate the ratio of the cell voltage deviation from the central cluster value of a certain cell at different times to the sum of the cell voltage deviation from the central cluster values ​​of all cells in the battery pack, and calculate the unit normalized cumulative value of the ratio based on the charging duration and the number of data frames for a certain charging process as the unit cell voltage deviation from the central cluster voltage risk normalized cumulative value of each cell to reflect the outlier state of each cell;

[0052] Step 2: extract the median voltage values ​​of all cells in the battery pack at different times, and calculate the cell voltage deviation from the center value of each cell at different times in combination with the cell voltage value; calculate the ratio of the cell voltage deviation from the center value of a certain cell at different times to the sum of the cell voltage deviation from the center value of all cells in the battery pack, and for a certain charging process, calculate the normalized cumulative value of the ratio based on the charging duration and the number of data frames as the normalized cumulative value of the cell voltage deviation from the center voltage risk of each cell; construct a normalized cumulative matrix of the cell voltage deviation from the center voltage risk for each cell using the normalized cumulative value of the cell voltage deviation from the center voltage risk corresponding to each charge of each cell;

[0053] Divide the values ​​of the matrix elements in the normalized cumulative matrix of the risk of the cell voltage deviating from the center voltage into several intervals, calculate the probability of each cell voltage value falling into different intervals during each charging process, and calculate the normalized cumulative discrete entropy of the risk of the cell voltage deviating from the center voltage of each cell during each charging process based on the probability;

[0054] Step 3: for the normalized cumulative value of the risk of the monomer voltage deviating from the center voltage obtained in step 2, calculate the median and upper and lower quartiles of the normalized cumulative value of the risk of the unit monomer voltage deviating from the center voltage, and on this basis calculate the cumulative interquartile range multiple of the unit monomer voltage deviating from the center voltage of each monomer;

[0055] Step 4: In combination with the actual operating conditions, corresponding thresholds are set for the normalized cumulative value of the risk of the single cell voltage deviating from the central cluster voltage, the normalized cumulative discrete entropy of the risk of the single cell voltage deviating from the central position voltage, and the cumulative interquartile range multiples of the single cell voltage deviating from the central position voltage obtained in the above steps. When at least two results are greater than or equal to the threshold, it indicates that there is abnormal discreteness in the corresponding battery single cell, and an early warning is issued for the power battery safety risk during the charging process.

[0056] In a preferred embodiment of the present invention, in step 1, the absolute value of the difference between the voltage value and the voltage median is first calculated for different moments, and 15 mV is set as a threshold:

[0057] |V ti -V 50％ |≥15

[0058] Cells whose voltage values ​​and their median voltages have an absolute value that does not reach the threshold are considered to have a low risk and are filtered. Only cells whose voltages deviate from the median by 15 mV are filtered. Figure 1 The steps shown in the figure are as follows. The cell voltage deviation from the center cluster value is calculated based on the following formula:

[0059]

[0060] Where, ΔV ti represents the deviation of the cell voltage of cell i from the center cluster value at time t, V ti Represents the voltage value of cell i at time t, V t50％ 、V t75％ and V t25％ They represent the median and upper and lower quartiles of the cell i voltage at time t respectively; Figure 2 The box plots shown are calculated.

[0061] The ratio of the cell voltage deviation from the center cluster value of a cell at different times to the sum of all cell voltage deviations from the center cluster values ​​in the battery pack is calculated based on the following formula as the normalized cumulative value of the cell voltage deviation from the center cluster voltage risk:

[0062]

[0063] In the formula, H i It represents the normalized cumulative value of the risk of the cell voltage of cell i deviating from the central cluster voltage during a charging process from time t1 to time t2. It represents the sum of the voltage deviations of all cells in the battery pack from the center cluster at time t, and m is the total number of cells;

[0064] Combined with the number of data frames n from time t1 to time t2 during the charging process, the normalized cumulative value κ of the risk of the unit cell voltage deviating from the central cluster voltage is calculated. i :

[0065]

[0066] In a preferred embodiment of the present invention, in step 2, the deviation value of the cell voltage from the center position is first calculated based on the following formula:

[0067] ΔV * ti=|V ti -V * t50％ |

[0068] Where, ΔV ti * Indicates that the cell voltage of cell i at time t deviates from the center value, V ti Represents the voltage value of cell i at time t, V * t50％ Represents the median voltage value of all cells in the battery pack at time t;

[0069] The ratio of the cell voltage deviation from the center position value of the cell i at different times to the sum of the cell voltage deviations from the center position values ​​of all cells in the battery pack is calculated based on the following formula:

[0070]

[0071] In the formula, G * ti It represents the calculated ratio result, i.e. the normalized value of the risk of the monomer voltage deviating from the center voltage, where m is the total number of monomers;

[0072] For a charging process from time t1 to time t2, the * ti The normalized cumulative value of the risk of the cell voltage of cell i deviating from the center voltage is obtained:

[0073]

[0074] All monomers H * i It is constructed as the normalized cumulative matrix of the risk of the monomer voltage deviating from the central voltage:

[0075] H * 1×m ={H * 1 ,H * 2 ,H * 3 ..........H * m}

[0076] Where m is the total number of monomers;

[0077] Extract the maximum value among the elements of the matrix H * max With the minimum value H * min Calculate the range ΔH of the normalized cumulative value of the cell voltage deviation from the center voltage risk during this charging process * :

[0078] ΔH * =H * max -H * min

[0079] With the maximum value H * max With the minimum value H * min As the endpoint, a is the determining parameter of the interval length, l is the number of intervals, and several probability intervals are divided so that the normalized cumulative value of the voltage risk of any single cell k deviating from the central voltage satisfies:

[0080]

[0081] The probability p that the normalized cumulative value of the risk of the cell voltage of cell i deviating from the center voltage falls into the jth interval i,j :

[0082]

[0083] In the formula, c i,j It indicates the number of times that the normalized cumulative value of the cell voltage deviation risk of cell i falls into the jth interval. It represents the sum of the number of times the normalized cumulative value of the cell voltage deviation risk of cell i falls into each interval, j∈l;

[0084] Using the calculated probability p i,j Calculate the normalized cumulative discrete entropy E of the risk of the cell voltage of cell i deviating from the center voltage in this charge i :

[0085]

[0086] In a preferred embodiment of the present invention, the normalized cumulative value H of the risk of the cell voltage deviating from the center voltage obtained in step 2 is * i , combined with the number of data frames n in the charging process from time t1 to time t2, calculate the normalized cumulative value κ of the risk of the unit cell voltage deviating from the center voltage * i :

[0087]

[0088] Calculate the κ * i The median κ * i50％ , upper quartile κ * i75％ and the lower quartile κ* i25％ , which can be used Figure 3 The box plot method shown in the figure is used to calculate the cumulative interquartile range multiple η of the unit monomer voltage deviation from the central voltage of each monomer based on the following formula: i :

[0089]

[0090] In a preferred embodiment of the present invention, the normalized cumulative value κ of the risk of the cell voltage of the battery cell i deviating from the central cluster voltage is calculated based on the actual operating conditions. i , the normalized cumulative discrete entropy E of the risk of the single voltage deviating from the central voltage i And the cumulative interquartile range multiple of the unit cell voltage deviation from the center voltage η i Set the corresponding threshold, when κ is satisfied i ≥0.00015, E i ≥1.5 and η i When at least two of the conditions ≥20 are met, it is determined that cell i is outliers in this charging, indicating that there is inconsistency in the battery cells and the battery pack has safety hazards, so an early warning of the power battery safety risk is issued.

[0091] In the specific implementation of the present invention, the single cell voltage data uploaded to the electric vehicle big data platform based on the national standard 32960 can be used. Those skilled in the art should also know that other forms of power battery data can be used based on the teachings of the present invention.

[0092] It should be understood that the size of the serial numbers of the steps in the embodiment of the present invention does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0093] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for sensing the safety status of a power battery based on the accumulation of voltage center cluster deviation, characterized in that: The specific steps include: Step 1: for the SOC interval of 10%-90% in the historical charging segment of the electric vehicle to be sensed, extract the voltage value, voltage median, voltage upper quartile and voltage lower quartile of each cell at different times, and use these parameters to calculate the cell voltage deviation of each cell at different times from the central cluster value; Calculate the ratio of the cell voltage deviation from the central cluster value of a certain cell at different times to the sum of the cell voltage deviation from the central cluster values ​​of all cells in the battery pack, and calculate the unit normalized cumulative value of the ratio based on the charging duration and the number of data frames for a certain charging process as the unit cell voltage deviation from the central cluster voltage risk normalized cumulative value of each cell to reflect the outlier state of each cell; Step 2: extract the median voltage values ​​of all cells in the battery pack at different times, and calculate the cell voltage deviation from the center value of each cell at different times in combination with the cell voltage value; calculate the ratio of the cell voltage deviation from the center value of a certain cell at different times to the sum of the cell voltage deviation from the center value of all cells in the battery pack, and for a certain charging process, calculate the normalized cumulative value of the ratio based on the charging duration and the number of data frames as the normalized cumulative value of the cell voltage deviation from the center voltage risk of each cell; construct a normalized cumulative matrix of the cell voltage deviation from the center voltage risk for each cell using the normalized cumulative value of the cell voltage deviation from the center voltage risk corresponding to each charge of each cell; Divide the values ​​of the matrix elements in the normalized cumulative matrix of the risk of the cell voltage deviating from the center voltage into several intervals, calculate the probability of each cell voltage value falling into different intervals during each charging process, and calculate the normalized cumulative discrete entropy of the risk of the cell voltage deviating from the center voltage of each cell during each charging process based on the probability; Step 3: for the normalized cumulative value of the risk of the single cell voltage deviating from the center voltage obtained in step 2, the normalized cumulative value of the risk of the single cell voltage deviating from the center voltage is calculated in combination with the number of data frames in the charging process from time t1 to time t2, the median and upper and lower quartiles of the normalized cumulative value of the risk of the single cell voltage deviating from the center voltage are calculated, and on this basis, the cumulative interquartile range multiple of the single cell voltage deviating from the center voltage of each single cell is calculated; Step 4: In combination with the actual operating conditions, corresponding thresholds are set for the normalized cumulative value of the risk of the single cell voltage deviating from the central cluster voltage, the normalized cumulative discrete entropy of the risk of the single cell voltage deviating from the central position voltage, and the cumulative interquartile range multiples of the single cell voltage deviating from the central position voltage obtained in the above steps. When at least two results are greater than or equal to the threshold, it indicates that there is abnormal discreteness in the corresponding battery single cell, and an early warning is issued for the power battery safety risk during the charging process.

2. The method according to claim 1, characterized in that: In step 1, the voltage value is first compared with the voltage median at different times to perform data filtering, and the following calculation of the deviation of the cell voltage from the center cluster value is performed only for the battery cells whose cell voltage deviates from the median to a predetermined degree: Where ΔV ti represents the deviation of the cell voltage of cell i from the center cluster value at time t, V ti Represents the voltage value of cell i at time t, V t50％ 、V t75％ and V t25％ They represent the median and upper and lower quartiles of the voltage of cell i at time t respectively; The ratio of the cell voltage deviation from the center cluster value of a cell at different times to the sum of all cell voltage deviations from the center cluster values ​​in the battery pack is calculated based on the following formula as the normalized cumulative value of the cell voltage deviation from the center cluster voltage risk: In the formula, H i It represents the normalized cumulative value of the risk of the cell voltage of cell i deviating from the central cluster voltage during a charging process from time t1 to time t2. It represents the sum of the voltage deviations of all cells in the battery pack from the center cluster at time t, and m is the total number of cells; Combined with the number of data frames n from time t1 to time t2 during the charging process, the normalized cumulative value κ of the risk of the unit cell voltage deviating from the central cluster voltage is calculated. i :

3. The method according to claim 2, characterized in that: In step 2, the deviation value of the cell voltage from the center position is first calculated based on the following formula: ΔV * ti =|V ti -V * t50％ | Where, ΔV ti * Indicates that the cell voltage of cell i at time t deviates from the center value, V ti Represents the voltage value of cell i at time t, V * t50％ Represents the median voltage value of all cells in the battery pack at time t; The ratio of the cell voltage deviation from the center position value of the cell i at different times to the sum of the cell voltage deviations from the center position values ​​of all cells in the battery pack is calculated based on the following formula: In the formula, G * ti It represents the calculated ratio result, i.e. the normalized value of the risk of the monomer voltage deviating from the center voltage, where m is the total number of monomers; For a charging process from time t1 to time t2, the * ti The normalized cumulative value of the risk of the cell voltage of cell i deviating from the center voltage is obtained: All monomers H * i It is constructed as the normalized cumulative matrix of the risk of the monomer voltage deviating from the central voltage: H * 1×m ={H * 1,H * 2,H * 3..........H * m } Wherein, m is the total number of monomers; Extract the maximum value among the elements of the matrix H * max With the minimum value H * min Calculate the range ΔH of the normalized cumulative value of the cell voltage deviation from the center voltage risk during this charging process * : ΔH * =H * max -H * min With the maximum value H * max With the minimum value H * min As the endpoint, a is the determining parameter of the interval length, l is the number of intervals, and several intervals are divided so that the normalized cumulative value of the voltage risk of any single cell k deviating from the center position voltage satisfies: The probability p that the normalized cumulative value of the risk of the cell voltage of cell i deviating from the center voltage falls into the jth interval i,j : In the formula, c i,j It indicates the number of times that the normalized cumulative value of the cell voltage deviation from the center voltage risk of cell i falls into the jth interval, It represents the sum of the number of times the normalized cumulative value of the cell voltage deviation risk of cell i falls into each interval, j∈l; Using the calculated probability p i,j Calculate the normalized cumulative discrete entropy E of the risk of the cell voltage of cell i deviating from the center voltage in this charge i :

4. The method according to claim 3, characterized in that: The normalized cumulative value H of the risk of the monomer voltage deviating from the center voltage obtained in step 2 * i , combined with the number of data frames n in the charging process from time t1 to time t2, calculate the normalized cumulative value κ of the risk of the unit cell voltage deviating from the center voltage * i : Calculate the κ * i The median κ * i50％ , upper quartile κ * i75％ and the lower quartile κ * i25％ , and the cumulative interquartile range multiple η of the unit monomer voltage deviation from the center voltage of each monomer is calculated based on the following formula i :

5. The method according to claim 4, characterized in that: Combined with the actual operating conditions, the normalized cumulative value κ of the risk of the cell voltage of the battery cell i deviating from the central cluster voltage is calculated. i , the normalized cumulative discrete entropy E of the risk of the single cell voltage deviating from the central voltage i And the cumulative interquartile range multiple of the unit cell voltage deviation from the center voltage η i Set the corresponding threshold, when κ is satisfied i ≥0.00015, E i ≥1.5 and η i When at least two of the conditions ≥20 are met, it is determined that cell i is outlier in this charging, and an early warning of the power battery safety risk is issued.

Images