An active detection method and system for battery consistency monitoring

Abstract

The application is suitable for the technical field of battery detection, and particularly relates to an active detection method and system for battery consistency monitoring.The method comprises the following steps: when charging, the battery cells in the battery are numbered, the battery cells are grouped, divided into multiple sampling detection groups, and single-group power data of the sampling detection groups is acquired; single-group power coordinates are constructed, function fitting is performed on the single-group power coordinates, and a random charging control function is constructed; the battery is charged in stages, active balancing is performed in the charging process, and power transfer is performed between the battery cells; SOC values of the battery cells are recorded, a battery cell charging curve is constructed, the number of power balancing is counted, and it is determined whether the battery cells are abnormal or not.The application can control the current of charging through the random charging control function to simulate real working conditions, can improve the identification accuracy of battery consistency through charging in stages and active balancing, and can actively monitor and improve the identification speed of abnormal battery cells.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of battery detection, and particularly relates to an active detection method and system for battery consistency monitoring. BACKGROUND

[0002] Battery consistency monitoring refers to real-time monitoring and analysis of key parameters such as voltage, temperature and internal resistance of each single battery constituting a battery pack during the operation of the battery pack, to ensure that the differences between these parameters remain within a reasonable range. This process is crucial for ensuring the overall performance of the battery pack, prolonging the service life, improving safety and optimizing the charging and discharging efficiency. Through effective battery consistency monitoring, potential problems can be discovered and handled in a timely manner, avoiding the failure or damage of the entire battery pack due to the performance deviation of individual battery units.

[0003] The existing battery detection method adopts passive detection, which continuously obtains the state information of the battery cell, and then judges the battery consistency based on the obtained state information of the battery cell. This method is relatively passive, requires a long detection period, and has inaccurate detection results, cannot identify the capacity change of the battery cell in advance, and is prone to safety hazards. SUMMARY

[0004] The purpose of the present application is to provide an active detection method for battery consistency monitoring, which aims to solve the problem that the detection method provided by the prior art is relatively passive, requires a long detection period, has inaccurate detection results, cannot identify the capacity change of the battery cell in advance, and is prone to safety hazards.

[0005] The present application is implemented as follows: an active detection method for battery consistency monitoring, the method comprising:

[0006] When charging, the battery cells in the battery are numbered, the battery cells are grouped, and divided into a plurality of sampling detection groups. Single group power data of the sampling detection group is obtained, and the single group power data includes the sum of the residual SOC values of all battery cells in the sampling detection group;

[0007] Based on the single group power data, a single group power coordinate of each sampling detection group is constructed, a function fitting is performed on the single group power coordinate, and a random charging control function is constructed;

[0008] The battery is charged in stages by the random charging control function, active balancing is performed during the charging process, and power transfer is performed between the battery cells;

[0009] The SOC values of each battery cell in the battery are recorded during the charging process, a battery cell charging curve is constructed, the number of power balancing corresponding to each interval of the battery cell charging curve is counted, and it is determined whether the battery cell is abnormal.

[0010] Preferably, the step of constructing a single group power coordinate for each sampling detection group based on single group power data, and performing function fitting on the single group power coordinate to construct a random charging control function, specifically comprises:

[0011] The single group power data is arranged in descending order, and the single group power data arranged in descending order is continuously numbered;

[0012] The single group power data is arranged in descending order, and the single group power data arranged in descending order is continuously numbered;

[0013] The single group power coordinate obtained is subjected to function fitting to obtain a basic fitting function, and the basic fitting function is intercepted based on a preset charging range interval to obtain a random charging control function.

[0014] Preferably, the step of performing phase charging on the battery through the random charging control function, actively balancing during the charging process, and transferring power between the battery cells, specifically comprises:

[0015] The random charging control function is sampled to obtain a plurality of charging control parameters, and each charging control parameter is assigned a charging duration;

[0016] Before charging, the SOC values of all battery cells are adjusted to the same value through active balancing, and the charging current of the battery is controlled according to the charging duration and the charging control parameter;

[0017] During the charging process, the SOC values of the battery cells are monitored, power is transferred between the battery cells through active balancing, the number of times of active balancing of each battery cell is determined, and the SOC values of all battery cells are adjusted to the same value once every charging phase.

[0018] Preferably, the step of recording the SOC values of each battery cell in the charging process, constructing a battery cell charging curve, counting the number of times of power balancing corresponding to each interval of the battery cell charging curve, and determining whether the battery cell is abnormal, specifically comprises:

[0019] In each charging phase, the SOC values of each battery cell are recorded to construct a battery cell charging curve, the horizontal coordinate of the battery cell charging curve is time value, and the vertical coordinate is SOC value;

[0020] The number of times of active balancing in each charging phase is counted, active balancing includes positive balancing and negative balancing, and the absolute value of the difference between the number of times of positive balancing and negative balancing is the effective number of times;

[0021] The curve segment corresponding to each charging phase is compared, the offset degree between the curve segment and other curve segments is calculated, the offset degree and the corresponding effective number of times are substituted into an abnormality evaluation formula, and an abnormality determination result is output.

[0022] Preferably, the anomaly assessment formula is expressed as:

[0023] ,

[0024] in, and These are the weighting coefficients. For the first The offset corresponding to each charging stage For the first The effective number of active balancing cycles corresponding to each charging stage.

[0025] Another object of the present invention is to provide an active detection system for battery consistency monitoring, the system comprising:

[0026] The battery data acquisition module is used to number and group the battery cells in the battery during charging, dividing them into multiple sampling and detection groups, and acquiring the single-group power data of the sampling and detection group. The single-group power data includes the sum of the remaining SOC values ​​of all cells in the sampling and detection group.

[0027] The charging control function construction module is used to construct a single set of energy coordinates for each sampling detection group based on a single set of energy data, perform function fitting on the single set of energy coordinates, and construct a random charging control function.

[0028] The staged charging module is used to charge the battery in stages through a random charging control function, perform active balancing during the charging process, and transfer power between battery cells.

[0029] The anomaly detection module is used to record the SOC value of each cell in the battery during the charging process, construct the cell charging curve, count the number of power equalization times corresponding to each interval of the cell charging curve, and determine whether there is an anomaly in the cell.

[0030] Preferably, the charging control function construction module includes:

[0031] The power data numbering unit is used to arrange a single group of power data in descending order and to number the single group of power data in descending order consecutively.

[0032] The coordinate construction unit is used to retrieve each set of single-group power data in sequence, extract the recorded power data, and construct the coordinates of each set of power data.

[0033] The function fitting adjustment unit is used to perform function fitting on the obtained single set of power coordinates to obtain the basic fitting function. Based on the preset charging range, the basic fitting function is truncated to obtain the random charging control function.

[0034] Preferably, the stage charging module comprises:

[0035] a charging parameter generation unit for sampling the random charging control function to obtain a plurality of charging control parameters, and assigning each charging control parameter with a charging duration;

[0036] a pre-equalization unit for adjusting the SOC values of all battery cells to the same value by active equalization before charging, and controlling the charging current of the battery according to the charging duration and the charging control parameter;

[0037] a charging control unit for monitoring the SOC values of the battery cells during the charging process, transferring the electric quantity among the battery cells by active equalization, determining the number of times of active equalization for each battery cell, and adjusting the SOC values of all battery cells to the same value by active equalization once every charging stage.

[0038] Preferably, the anomaly identification module comprises:

[0039] a stage charging monitoring unit for recording the SOC values of the battery cells in each charging stage, and constructing a battery cell charging curve, wherein the horizontal coordinate of the battery cell charging curve is time value, and the vertical coordinate is SOC value;

[0040] an equalization statistics unit for counting the effective number of times of active equalization in each charging stage, wherein the active equalization includes positive equalization and negative equalization, and the absolute value of the difference between the number of times of positive equalization and the number of times of negative equalization is the effective number of times;

[0041] an anomaly evaluation unit for comparing the curve segment corresponding to each charging stage, calculating the offset degree between the curve segment and other curve segments, and inputting the offset degree and the corresponding effective number of times into an anomaly evaluation formula to output an anomaly determination result.

[0042] Preferably, the anomaly evaluation formula is represented as: ,

[0043] wherein, and are weight coefficients, is the offset degree corresponding to the i th charging stage, is the effective number of times of active equalization corresponding to the i th charging stage.

[0044] ​​The active detection method for battery consistency monitoring provided by this invention can control the charging current through a random charging control function to simulate real working conditions. Cell consistency is not only reflected in static parameters, but more importantly in the response differences under dynamic working conditions. By using staged charging and active balancing, the accuracy of battery consistency identification is improved, and active monitoring can be performed to improve the speed of identifying abnormal cells. Attached Figure Description

[0045] Figure 1 A flowchart of an active detection method for battery consistency monitoring provided in an embodiment of the present invention;

[0046] Figure 2 The flowchart illustrates the steps of constructing a single set of power coordinates for each sampling detection group based on a single set of power data, performing function fitting on the single set of power coordinates, and constructing a random charging control function, as provided in this embodiment of the invention.

[0047] Figure 3 The flowchart illustrates the steps of performing staged charging of the battery using a random charging control function, actively balancing the battery during the charging process, and transferring charge between battery cells, as provided in this embodiment of the invention.

[0048] Figure 4 The flowchart provided in this embodiment of the invention describes the steps of recording the SOC value of each cell in the battery during the charging process, constructing a cell charging curve, counting the number of power equalization times corresponding to each interval of the cell charging curve, and determining whether there is an abnormality in the cell.

[0049] Figure 5 This is an architecture diagram of an active detection system for battery consistency monitoring provided in an embodiment of the present invention;

[0050] Figure 6 This is an architecture diagram of the charging control function construction module provided in an embodiment of the present invention;

[0051] Figure 7 An architectural diagram of a staged charging module provided in an embodiment of the present invention;

[0052] Figure 8 This is an architecture diagram of the anomaly recognition module provided in an embodiment of the present invention. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0054] like Figure 1As shown, a flowchart of an active detection method for battery consistency monitoring provided by an embodiment of the application is shown, and the method comprises the following steps:

[0055] S100, when charging, numbering the battery cells in the battery, grouping the battery cells, dividing into a plurality of sampling detection groups, obtaining single-group power data of the sampling detection groups, and the single-group power data comprising the sum of the residual SOC values of all battery cells in the sampling detection group.

[0056] In this step, when charging, the battery cells in the battery are numbered, and each battery cell is assigned a separate number. The battery cells are grouped, and a random grouping method is used, such as dividing all battery cells into 5 sampling detection groups, each sampling detection group containing the same number of battery cells, such as a total of 100 battery cells, each sampling detection group containing 20 battery cells. The position of the battery cell is randomly determined. The current SOC value and its number of each battery cell are read, the sum of the SOC values of the battery cells in each sampling detection group is calculated, and single-group power data is obtained.

[0057] S200, based on the single-group power data, constructing single-group power coordinates of each sampling detection group, fitting a function to the single-group power coordinates, and constructing a random charging control function.

[0058] In this step, based on the single-group power data, single-group power coordinates of each sampling detection group are constructed. The residual SOC values of different battery cells are different during each charging, so the single-group power coordinates obtained are also randomly generated. The abscissa of the single-group power coordinates is the number of the single-group power coordinates, and the total SOC value recorded in the single-group power data is the ordinate of the single-group power coordinates. Since the residual SOC values of the battery cells during charging are unpredictable, the single-group power coordinates obtained are also randomly generated. According to the number of sampling detection groups, a corresponding number of single-group power coordinates are obtained. All single-group power coordinates are imported into a function fitting software, and a set of random charging control functions are fitted. The random charging control function is used to control the charging current, so that the charging current changes according to the random charging control function, and the fluctuation range of the charging current is limited.

[0059] S300, performing phase charging on the battery through the random charging control function, actively balancing during the charging process, and transferring power between the battery cells.

[0060] In this step, the battery is charged in stages by a random charging control function, and the battery SOC range for active detection is set, such as setting the battery SOC value below 50% for active detection, otherwise, active detection is not triggered. The time of active detection can be further limited, and the appropriate detection time is selected according to the user's historical power consumption time interval, such as the user's 80% charging time being 11pm and the vehicle time being after 6am, then this time period is determined as idle time, and active detection is performed when charging in this time period. The frequency of active detection is gradually increased according to the number of charges, and the greater the total number of charges, the frequency of active detection is gradually increased. When active detection is performed, the battery is controlled to be charged in stages, and the starting SOC value of each cell is adjusted by active balancing before the start of different stages to keep the same level. The active balancing method can be inductive balancing. When active balancing is performed, the amount of charge transferred by the inductor in a single time is consistent.

[0061] S400, record the SOC value of each cell in the battery during charging, construct the cell charging curve, and count the number of times of balancing the electric quantity corresponding to each interval of the cell charging curve to determine whether the cell is abnormal.

[0062] In this step, the SOC value of each cell in the battery is recorded during charging. The recorded data includes two parts. One part of the data is the change of the SOC value of the cell, such as recording the SOC value of cell A at different times. The SOC value of each cell is monitored and sampled by the BMS system, and recorded in the form of a curve. The horizontal coordinate of the curve is the charging time, and the vertical coordinate is the real-time SOC value. Each charging stage of the cell corresponds to a cell charging curve. The charging stage can be taken as an interval of 5% or 10% SOC value. The number of times of active balancing experienced by a single cell in each charging stage is counted. Active balancing is divided into positive balancing and negative balancing. Positive balancing is that the cell accepts the charge transferred from other cells, and negative balancing is that the cell outputs charge to other cells. Since the amount of charge transferred by the inductor is constant during active balancing, one positive balancing and one negative balancing cancel each other out, so that the number of times of balancing the electric quantity (the effective number of active balancing) is obtained. By comparing the cell charging curves of all cells in the same stage, the deviation degree of each cell charging curve is calculated, and then the abnormal evaluation value of the cell is calculated based on the abnormal evaluation formula combined with the number of times of balancing the electric quantity:

[0063] ,

[0064] wherein, and are weight coefficients, is the deviation degree corresponding to the th charging stage, is the number of times of balancing the electric quantity corresponding to the the effective number of active equalization corresponding to each charging stage;

[0065] When the abnormal evaluation value is greater than the preset value, it is determined that the battery cell is abnormal, and a warning is given in advance before the battery use is abnormal, avoiding the occurrence of battery pack failure and improving the safety of the electric vehicle.

[0066] As shown in Figure 2 As a preferred embodiment of the present application, the step of constructing a single-group power coordinate for each sampling detection group based on single-group power data, and performing function fitting on the single-group power coordinate to construct a random charging control function, specifically includes:

[0067] S201, arranging the single-group power data in descending order and continuously numbering the single-group power data arranged in descending order.

[0068] In this step, the single-group power data is arranged in descending order. After the arrangement, the single-group power data with a high total SOC value of the battery cell is arranged in front, and the single-group power data with a low total SOC value is arranged in back. The single-group power data is continuously numbered, and the number of the single-group power data arranged in the first position is 1. Accordingly, the numbers of the other single-group power data are determined according to the natural number sequence.

[0069] S202, sequentially calling each single-group power data, extracting the recorded power data therein, and constructing a single-group power coordinate.

[0070] S203, performing function fitting on the obtained single-group power coordinate to obtain a basic fitting function, and intercepting the basic fitting function based on a preset charging range interval to obtain a random charging control function.

[0071] In this step, each single-group power data is sequentially called, the number of the single-group power data is taken as the abscissa, and the recorded SOC value is taken as the ordinate to construct a single-group power coordinate. Through function fitting, a basic fitting function is obtained. The curve corresponding to the basic fitting function is intercepted, so that the ratio of the maximum value to the minimum value of the intercepted curve is less than a preset value, such as less than 2. The intercepted curve corresponds to a random charging control function, and the range of the independent variable is determined.

[0072] For example, the basic fitting function is , whose independent variable range is 0 to positive infinity, the part of the basic fitting function corresponding to the independent variable range [0, 10] is the random charging control function F(x), and the ratio of the maximum value to the minimum value of F(x) in the independent variable range is less than a preset value, such as 2. If the ratio of the maximum value to the minimum value of the obtained F(x) does not meet the condition, the independent variable range is shifted to the next interval, such as [10, 20], until the ratio of the maximum value to the minimum value is less than the preset value;

[0073] Briefly, the process of obtaining the random charging control function by intercepting the basic fitting function is as follows:

[0074] , [0, +∞]→F(x), [0, 10].

[0075] As shown in Figure 3 , as a preferred embodiment of the present application, the step of performing phase charging on the battery by the random charging control function, actively balancing during the charging process, and transferring power between the battery cells specifically includes:

[0076] S301, sampling the random charging control function to obtain a plurality of charging control parameters, and assigning each charging control parameter a charging duration.

[0077] In this step, the random charging control function is sampled, and the sampling length is a preset interval, such as sampling between the independent variable range 0-10, and a preset number of sampling points are selected, such as sampling every 0.1, to obtain 100 sampling points. The calculation value corresponding to each sampling point is determined, i.e. the sampling point is substituted into the random charging control function to obtain the calculation value, the maximum and minimum values of the calculation value are counted, the distribution interval [Tmin, Tmax] of the calculation value is determined, and the charging control parameter corresponding to each sampling point is determined according to the distribution interval of the calculation value. The maximum current value Imax and the minimum current value Imin are obtained, the calculation value corresponding to the sampling point is T, and the charging current (charging control parameter) I of the sampling point is Imin + (Imax-Imin)*(T-Tmin) / (Tmax-Tmin), and the charging duration is a preset value, such as one minute, i.e. charging for one minute under the charging control parameter corresponding to each sampling point.

[0078] S302, before charging, adjusting the SOC values of all battery cells to the same value by active balancing, and controlling the charging current of the battery according to the charging duration and the charging control parameter.

[0079] In this step, before charging, the SOC values of all battery cells are adjusted to the same value by active balancing. In order to facilitate the analysis of the charging condition of each battery cell, the SOC values of all battery cells are adjusted by active balancing before charging, so that all battery cells are at the same SOC value. At this time, the number of active balancing is not counted in the subsequent statistical process. When all battery cells are at the same SOC value, the charging control parameters are sequentially called, and the same charging time of each charging control parameter is executed.

[0080] S303, in the charging process, the SOC values of each battery cell are monitored, and the power is transferred between the battery cells by active balancing, and the number of active balancing of each battery cell is determined. Each time a charging stage is reached, active balancing is performed to adjust the SOC values of all battery cells to the same value.

[0081] In this step, in the charging process, the SOC values of each battery cell are monitored, and the SOC values of the battery cells are sampled by the BMS system of the vehicle to determine the real-time SOC values of the battery cells. After determining the real-time SOC values of the battery cells each time in the charging process, active balancing is performed according to the SOC value difference between the battery cells to transfer the power of the high SOC value battery cell to the low SOC value battery cell, and the number of active balancing experienced by each battery cell is recorded. Until there are a preset number of battery cells that reach the SOC upper limit value of the current charging stage, such as 10% of the battery cells reaching 35% SOC value. At this time, the charging is stopped, and the active balancing is performed to adjust the power of each battery cell to be the same. Thus, the number of active balancing of the battery cells in the current charging stage is counted, including the number of active balancing when only active balancing is performed at the end of the stage.

[0082] As shown in Figure 4 As a preferred embodiment of the present application, the step of recording the SOC values of each battery cell in the charging process, constructing a battery cell charging curve, counting the number of power balancing corresponding to each interval of the battery cell charging curve, and determining whether the battery cell is abnormal, specifically includes:

[0083] S401, in each charging stage, the SOC values of each battery cell are recorded, and a battery cell charging curve is constructed. The horizontal coordinate of the battery cell charging curve is time value, and the vertical coordinate is SOC value.

[0084] In this step, in each charging stage, the SOC values of each battery cell are recorded, and the BMS system acquires the SOC values of each battery cell at a preset time interval to construct a corresponding battery cell charging curve based on the charging time.

[0085] S402, count the number of effective active balancing operations in each charging stage. Active balancing includes positive balancing and negative balancing. The absolute value of the difference between the number of positive balancing and negative balancing operations is the number of effective operations.

[0086] In this step, the effective number of active balancing operations in each charging stage is counted. When active balancing occurs between cells, the charge between cells needs to be transferred across cells. For example, if there are 5 cells between cell A and cell G, where cell A's charge is above average, cell G's charge is below average, and the charge of the 5 cells in the middle is the average, then cell A needs to transfer charge to cell G. The transferred charge will pass through the 5 cells in the middle. For the 5 cells in the middle, the number of positive balancing operations is the same as the number of negative balancing operations. When A transfers charge to G, the effective number of active balancing operations for the 5 cells in the middle is 0. A has one negative balancing operation, and G has one positive balancing operation. The more positive balancing operations, the faster the SOC value of the cell rises under the same charging conditions, and its capacity may be smaller. Conversely, if the SOC value of the cell rises too slowly, there is a significant increase in internal resistance.

[0087] S403 compares the curve segment corresponding to each charging stage, calculates the offset between the curve segment and other curve segments, substitutes the offset and the corresponding number of valid counts into the anomaly evaluation formula, and outputs the anomaly judgment result.

[0088] In this step, the curve segments corresponding to each charging stage are compared. Specifically, each time a curve segment corresponding to a cell is selected, the Euclidean distance between this curve segment and other curve segments is calculated, the average value of this Euclidean distance is calculated, and the average distance L between this curve segment and other curve segments is obtained. Using the average distance as the offset, the abnormal evaluation value corresponding to this curve segment is retrieved. ,

[0089] in, and These are the weighting coefficients. For the first The offset corresponding to each charging stage For the first The effective number of active balancing cycles corresponding to each charging stage, among which + =1, and The value can be set empirically or based on experimental data. For example, if the deviation has a greater impact on battery safety, then... The value is larger, set to 0.7. Correspondingly, the weighting coefficient of the number of active balancing is lower, set to 0.3. The specific value is determined according to the degree of influence of the offset and the number of active balancing on the cell safety.

[0090] The step of calculating the offset degree is:

[0091] Suppose the number of battery cells is 5, and the numbers are M1, M2, M3, M4 and M5 respectively, taking the second charging stage as an example, the curve segments corresponding to the five groups of battery cells are H1, H2, H3, H4 and H5 respectively, the Euclidean distance between any two curve segments is calculated, for example, if the curve segment H1 is selected, the Euclidean distances between H1 and H2, H3, H4 and H5 are calculated respectively, and the calculated values are L1, L2, L3 and L4 respectively, and the average distance L=(L1+L2+L3+L4) / 4 is calculated, then the average distance L is the offset degree of the battery cell

[0092] The step of calculating the effective number of active balancing is:

[0093] Suppose that in the second charging stage, the positive balancing number of a certain battery cell is 10, and the negative balancing number is 5, then the effective balancing number is 10-5, that is =5;

[0094] When the abnormal evaluation value is greater than the preset value, it is determined that the battery cell is abnormal, and an alarm is issued through the vehicle.

[0095] As shown in Figure 5 , the application also provides an active detection system for battery consistency monitoring, which comprises:

[0096] A battery data acquisition module 100 is used to number the battery cells in the battery during charging, group the battery cells, divide them into a plurality of sampling detection groups, and obtain single-group power data of the sampling detection groups, wherein the single-group power data includes the sum of the residual SOC values of all battery cells in the sampling detection group.

[0097] In the system, the battery data acquisition module 100 numbers the battery cells in the battery during charging, sets a separate number for each battery cell, and groups the battery cells. When grouping, a random grouping method is used, for example, all battery cells are equally divided into 5 sampling detection groups, each sampling detection group contains the same number of battery cells, for example, a total of 100 battery cells, and each sampling detection group contains 20 battery cells. The position of the battery cell is randomly determined, the current SOC value and its number of each battery cell are read, the sum of the SOC values of the battery cells in each sampling detection group is calculated, and single-group power data is obtained.

[0098] A charging control function construction module 200 is used to construct single-group power coordinates of each sampling detection group based on the single-group power data, perform function fitting on the single-group power coordinates, and construct a random charging control function.

[0099] ​​In the system, the charge control function construction module 200 constructs a single group of power coordinates for each sampling detection group based on a single group of power data. The residual SOC values of different battery cells are different at each charging time, so the single group of power coordinates constructed are also randomly generated. The abscissa of the single group of power coordinates is the number of the single group of power coordinates, and the total SOC value recorded in the single group of power data is the ordinate of the single group of power coordinates. Since the residual SOC values of each battery cell cannot be predicted during charging, the single group of power coordinates obtained are also randomly generated. According to the number of sampling detection groups, a corresponding number of single group of power coordinates are obtained. All the single group of power coordinates are imported into a function fitting software, so as to fit a group of random charge control functions. The random charge control function is used to control the charging current, so that the charging current changes according to the random charge control function, and the fluctuation range of the charging current is limited.

[0100] The stage charging module 300 is used for performing stage charging on the battery through the random charge control function, and actively balancing during the charging process to transfer power between the battery cells.

[0101] In the system, the stage charging module 300 performs stage charging on the battery through the random charge control function. The battery SOC range for active detection is set, for example, the battery SOC value is set to be lower than 50% to perform active detection, otherwise, the active detection is not triggered. The time for active detection can be further limited. According to the historical power consumption time interval of the user, a suitable detection time is selected, for example, the user's 80% charging time is 11pm, and the vehicle use time is after 6am, so the time interval is determined as an idle time. When charging in this time interval, active detection is performed. The frequency of active detection is gradually increased according to the number of charging times. The greater the total number of charging times, the more the frequency of active detection is gradually increased. When active detection is performed, the battery is controlled to be charged in stages. Before the start of different stages, the starting SOC values of the battery cells are adjusted by active balancing to keep the same level. The active balancing method can be inductive balancing. When active balancing is performed, the amount of charge transferred by the inductor in a single transfer is kept consistent.

[0102] The abnormality identification module 400 is used for recording the SOC values of each battery cell in the battery during the charging process, constructing a battery cell charging curve, and counting the number of power balancing corresponding to each interval of the battery cell charging curve to determine whether the battery cell is abnormal.

[0103] In the system, the abnormality identification module 400 records the SOC values of each cell in the battery during the charging process. The recorded data includes two parts. One part is the SOC value change of the cell, such as recording the SOC value of cell A at different times. The SOC values of each cell are monitored and sampled through the BMS system, and recorded in the form of a curve. The horizontal coordinate of the curve is the charging time, and the vertical coordinate is the real-time SOC value. Each charging stage of the cell corresponds to a cell charging curve. The charging stage can be divided into intervals of 5% or 10% SOC value. The number of active balancing experienced by a single cell in each charging stage is counted. Active balancing includes positive balancing and negative balancing. Positive balancing refers to the cell accepting charge transferred from other cells, and negative balancing refers to the cell outputting charge to other cells. Since the amount of charge transferred by the inductor remains unchanged during active balancing, one positive balancing and one negative balancing cancel each other out, resulting in the number of power balancing (effective number of active balancing). By comparing the cell charging curves of all cells in the same stage, the deviation of each cell charging curve is calculated, and the abnormality evaluation value of the cell is calculated based on the abnormality evaluation formula combined with the number of power balancing:

[0104] ,

[0105] wherein, and are weight coefficients, is the deviation corresponding to the th charging stage, is the effective number of active balancing corresponding to the th charging stage;

[0106] When the abnormality evaluation value is greater than the preset value, it is determined that the cell has an abnormality. Early warning is performed before the battery usage appears abnormal, avoiding the occurrence of battery pack failure and improving the safety of electric vehicles.

[0107] As shown in Figure 6 , as a preferred embodiment of the present application, the charging control function construction module 200 includes:

[0108] The power data numbering unit 201 is used to arrange a single set of power data in descending order and continuously number the single set of power data arranged in descending order.

[0109] In this module, the power data numbering unit 201 arranges a single set of power data in descending order. After arranging, the single set of power data with a high total SOC value of the cell is arranged in front, and the single set of power data with a low total SOC value is arranged in back. The single set of power data is continuously numbered. The number of the single set of power data arranged in the first position is 1. Accordingly, the numbers of the subsequent other single sets of power data are determined according to the natural number sequence.

[0110] The coordinate construction unit 202 is used to retrieve each set of single-group power data in sequence, extract the recorded power data, and construct the coordinates of each set of power data.

[0111] The function fitting adjustment unit 203 is used to perform function fitting on the obtained single set of power coordinates to obtain a basic fitting function. Based on the preset charging range interval, the basic fitting function is truncated to obtain a random charging control function.

[0112] In this module, individual sets of power data are retrieved sequentially. The power data number of each set is used as the x-axis, and the recorded SOC value is used as the y-axis to construct a power coordinate system for each set. By performing function fitting, a basic fitting function is obtained. The curve corresponding to the basic fitting function is then truncated so that the ratio of the maximum value to the minimum value of the truncated curve is less than a preset value, such as a ratio of the maximum value to the minimum value being less than 2. The truncated curve corresponds to the random charging control function, and its independent variable range is determined.

[0113] like Figure 7 As shown, in a preferred embodiment of the present invention, the staged charging module 300 includes:

[0114] The charging parameter generation unit 301 is used to sample the random charging control function to obtain multiple charging control parameters and assign a charging duration to each charging control parameter.

[0115] In this module, the charging parameter generation unit 301 samples the random charging control function. The sampling length is a preset interval, such as sampling within the range of 0-10 for the independent variable. A preset number of sampling points are selected, such as sampling once every 0.1, resulting in 100 sampling points. The calculated value corresponding to each sampling point is determined by substituting the sampling point into the random charging control function to obtain the calculated value. The maximum and minimum values ​​of the calculated values ​​are statistically analyzed to determine the distribution interval of the calculated values ​​[Tmin, Tmax]. Based on the distribution interval of the calculated values, the charging control parameters corresponding to each sampling point are determined. The maximum charging current value Imax and the minimum current value Imin are obtained. The calculated value corresponding to the sampling point is T. Then, the charging current (charging control parameter) at the sampling point is I = Imin + (Imax - Imin) * (T - Tmin) / (Tmax - Tmin). The charging time is a preset value, such as one minute, that is, charging for one minute under the charging control parameters corresponding to each sampling point.

[0116] The pre-balancing unit 302 is used to adjust the SOC value of all cells to the same value through active balancing before charging, and to control the charging current of the battery according to the charging time and charging control parameters.

[0117] In the module, the pre-equalization unit 302 adjusts the SOC values of all the battery cells to the same value by active equalization before charging. In order to facilitate the analysis of the charging conditions of each battery cell, the SOC values of all the battery cells are adjusted by active equalization before charging, so that all the battery cells are at the same SOC value. At this time, the number of active equalization is not counted in the subsequent statistical process. When all the battery cells are at the same SOC value, the charging control parameters are sequentially called, and the charging time of the same charging duration is performed under each charging control parameter.

[0118] The charging control unit 303 is used for monitoring the SOC values of each battery cell during charging, transferring the electric quantity between the battery cells by active equalization, and determining the number of active equalization of each battery cell. Active equalization is performed once every charging stage to adjust the SOC values of all the battery cells to the same value.

[0119] In the module, the charging control unit 303 monitors the SOC values of each battery cell during charging, samples the SOC values of the battery cells by the BMS system of the vehicle, determines the real-time SOC values of the battery cells, and performs active equalization according to the differences in the SOC values between the battery cells after determining the real-time SOC values of the battery cells each time during charging. The electric quantity of the battery cell with a high SOC value is transferred to the battery cell with a low SOC value, and the number of active equalization experienced by each battery cell is recorded. Until there are a preset number of battery cells that reach the SOC upper limit value of the current charging stage, such as 10% of the battery cells reaching 35% SOC value. At this time, charging is stopped, active equalization is performed to adjust the electric quantity of each battery cell to be the same. Thus, the number of active equalization of the battery cells in the current charging stage is counted, including the number of active equalization when active equalization is performed at the end of the stage.

[0120] As shown in FIG. 4, Figure 8 As a preferred embodiment of the present application, the abnormality identification module 400 includes:

[0121] The stage charging monitoring unit 401 is used for recording the SOC values of each battery cell in each charging stage, and constructing a battery cell charging curve. The horizontal coordinate of the battery cell charging curve is the time value, and the vertical coordinate is the SOC value.

[0122] In the module, the stage charging monitoring unit 401 records the SOC values of each battery cell in each charging stage. The BMS system obtains the SOC values of each battery cell at a preset time interval, and constructs a corresponding battery cell charging curve based on the charging time.

[0123] The equalization statistical unit 402 is used for counting the effective number of active equalization in each charging stage. Active equalization includes positive equalization and negative equalization. The absolute value of the difference between the number of positive equalization and the number of negative equalization is the effective number.

[0124] In this module, the equalization statistics unit 402 counts the effective number of active equalizations performed in each charging stage. When active equalization is performed between cells, the charge between cells needs to be transferred across cells. For example, if there are 5 cells between cell A and cell G, where cell A's charge is higher than the average, cell G's charge is lower than the average, and the charge of the 5 cells in the middle is the average, then cell A needs to transfer charge to cell G. The transferred charge will pass through the 5 cells in the middle. For the 5 cells in the middle, the number of positive equalizations and the number of negative equalizations are the same. When A transfers charge to G, the effective number of active equalizations for the 5 cells in the middle is 0. A has one negative equalization, and G has one positive equalization. The more positive equalizations, the faster the SOC value of the cell rises under the same charging conditions, and its capacity may be smaller. Conversely, if the SOC value of the cell rises too slowly, there is a significant increase in internal resistance.

[0125] The anomaly assessment unit 403 is used to compare the curve segment corresponding to each charging stage, calculate the offset between the curve segment and other curve segments, substitute the offset and the corresponding number of valid times into the anomaly assessment formula, and output the anomaly judgment result.

[0126] In this module, the anomaly assessment unit 403 compares the curve segments corresponding to each charging stage. Specifically, it selects a curve segment corresponding to a battery cell each time, calculates the Euclidean distance between this curve segment and other curve segments, calculates the average value of this Euclidean distance, obtains the average distance L between this curve segment and other curve segments, uses the average distance as the offset, and retrieves the anomaly assessment value corresponding to this curve segment.

[0127] ,

[0128] in, and These are the weighting coefficients. For the first The offset corresponding to each charging stage For the first The effective number of active balancing cycles corresponding to each charging stage;

[0129] When the abnormal assessment value is greater than the preset value, it is determined that the battery cell is abnormal, and an alarm is issued through the vehicle.

[0130] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An active detection method for battery consistency monitoring, characterized in that, The method includes: During charging, the cells in the battery are numbered, grouped, and divided into multiple sampling and detection groups. The power data of a single sampling and detection group is obtained. The power data of a single sampling and detection group includes the sum of the remaining SOC values ​​of all cells in the sampling and detection group. Based on a single set of power data, a single set of power coordinates for each sampling and detection group is constructed. A function is fitted to the single set of power coordinates to construct a random charging control function. The battery is charged in stages through a random charging control function, and active balancing is performed during the charging process to transfer power between the cells. During the charging process, the SOC value of each cell in the battery is recorded, a cell charging curve is constructed, the number of times the power equalization is performed for each interval of the cell charging curve, and it is determined whether there is any abnormality in the cell. The steps of constructing a single set of power coordinates for each sampling detection group based on a single set of power data, performing function fitting on the single set of power coordinates, and constructing a random charging control function specifically include: Arrange the individual sets of electricity data in descending order, and then number the individual sets of electricity data consecutively after arranging them in descending order. Retrieve each set of individual power data in sequence, extract the recorded power data, and construct the power coordinates for each set. The obtained single set of power coordinates is fitted with a function to obtain a basic fitting function. The basic fitting function is then truncated based on a preset charging range to obtain a random charging control function.

2. The active detection method for battery consistency monitoring according to claim 1, characterized in that, The steps of performing staged charging of the battery through a random charging control function, actively balancing the battery during the charging process, and transferring power between battery cells specifically include: The random charging control function is sampled to obtain multiple charging control parameters, and each charging control parameter is assigned a charging duration. Before charging, the SOC value of all cells is adjusted to the same value through active balancing, and the charging current of the battery is controlled according to the charging time and charging control parameters. During the charging process, the SOC value of each cell is monitored, and power is transferred between the cells through active balancing. The number of active balancing operations for each cell is determined. Active balancing is performed once at each charging stage to adjust the SOC value of all cells to the same value.

3. The active detection method for battery consistency monitoring according to claim 1, characterized in that, The steps of recording the SOC value of each cell in the battery during charging, constructing a cell charging curve, counting the number of power equalization cycles corresponding to each interval of the cell charging curve, and determining whether there is an abnormality in the cell specifically include: In each charging stage, the SOC value of each cell is recorded, and a cell charging curve is constructed. The horizontal axis of the cell charging curve is the time value, and the vertical axis is the SOC value. The effective number of active balancing operations is counted in each charging stage. Active balancing includes positive balancing and negative balancing. The absolute value of the difference between the number of positive balancing and negative balancing operations is the effective number of operations. The curve segment corresponding to each charging stage is compared, the offset between the curve segment and other curve segments is calculated, and the offset and the corresponding number of valid counts are substituted into the anomaly evaluation formula to output the anomaly judgment result.

4. The active detection method for battery consistency monitoring according to claim 3, characterized in that, The anomaly assessment formula is expressed as follows: ; in, and These are the weighting coefficients. For the first The offset corresponding to each charging stage For the first The effective number of active balancing cycles corresponding to each charging stage.

5. An active detection system for battery consistency monitoring, characterized in that, The system includes: The battery data acquisition module is used to number and group the battery cells in the battery during charging, dividing them into multiple sampling and detection groups, and acquiring the single-group power data of the sampling and detection group. The single-group power data includes the sum of the remaining SOC values ​​of all cells in the sampling and detection group. The charging control function construction module is used to construct a single set of energy coordinates for each sampling detection group based on a single set of energy data, perform function fitting on the single set of energy coordinates, and construct a random charging control function. The staged charging module is used to charge the battery in stages through a random charging control function, perform active balancing during the charging process, and transfer power between battery cells. The anomaly detection module is used to record the SOC value of each cell in the battery during the charging process, construct the cell charging curve, count the number of power equalization times corresponding to each interval of the cell charging curve, and determine whether there is an anomaly in the cell. The charging control function construction module includes: The power data numbering unit is used to arrange a single group of power data in descending order and to number the single group of power data in descending order consecutively. The coordinate construction unit is used to retrieve each set of single-group power data in sequence, extract the recorded power data, and construct the coordinates of each set of power data. The function fitting adjustment unit is used to perform function fitting on the obtained single set of power coordinates to obtain the basic fitting function. Based on the preset charging range, the basic fitting function is truncated to obtain the random charging control function.

6. The active detection system for battery consistency monitoring according to claim 5, characterized in that, The stage charging module includes: The charging parameter generation unit is used to sample the random charging control function to obtain multiple charging control parameters and assign a charging duration to each charging control parameter; The pre-balancing unit is used to actively balance all the cells to the same SOC value before charging, and to control the charging current of the battery according to the charging time and charging control parameters. The charging control unit monitors the SOC value of each cell during the charging process, transfers power between cells through active balancing, determines the number of times active balancing is performed for each cell, and performs active balancing once at each charging stage to adjust the SOC value of all cells to the same value.

7. The active detection system for battery consistency monitoring according to claim 5, characterized in that, The anomaly detection module includes: The stage charging monitoring unit is used to record the SOC value of each cell in each charging stage and construct the cell charging curve. The horizontal axis of the cell charging curve is the time value and the vertical axis is the SOC value. The balancing statistics unit is used to count the effective number of active balancing operations in each charging stage. Active balancing includes positive balancing and negative balancing. The absolute value of the difference between the number of positive balancing and negative balancing operations is the effective number of operations. The anomaly assessment unit is used to compare the curve segment corresponding to each charging stage, calculate the offset between the curve segment and other curve segments, substitute the offset and the corresponding number of valid counts into the anomaly assessment formula, and output the anomaly judgment result.

8. The active detection system for battery consistency monitoring according to claim 7, characterized in that, The anomaly assessment formula is expressed as follows: ; in, and These are the weighting coefficients. For the first The offset corresponding to each charging stage For the first The effective number of active balancing cycles corresponding to each charging stage.

Images