A method for predicting the battery capacity of a vehicle battery system

Abstract

This invention discloses a method for predicting the battery capacity of a vehicle battery system, comprising: acquiring multiple sets of state data for the battery system; the battery system includes multiple individual cells, and the state data sets include data reporting time, total current, voltage of each cell, state of charge (SOC), cumulative mileage, vehicle speed, and battery system temperature; determining the operating condition factor and historical battery capacity of the battery system based on each set of state data; wherein, the operating condition factor includes the historical average SOC range, the historical average temperature of the battery system, the historical average rate capability, the historical cumulative mileage of the vehicle, characteristic capacity consistency, historical average vehicle speed, the proportion of vehicle storage time, the proportion of high SOC time, and the proportion of fast charging times; determining a battery capacity prediction model based on the operating condition factor and historical battery capacity; and predicting the battery capacity of the battery system based on the battery capacity prediction model. This invention can accurately estimate and predict the battery capacity, thereby solving the problems of difficulty and low accuracy in battery capacity estimation.

Description

Technical Field

[0001] This invention relates to the field of battery technology, and more particularly to a method for predicting the battery capacity of a vehicle battery system. Background Technology

[0002] Lithium-ion batteries, as the most promising energy storage component, are now widely used in the electric vehicle field. Accurate capacity estimation of lithium-ion batteries plays a crucial role in improving the safety and driving range of electric vehicles.

[0003] Currently, the main method for estimating the capacity of lithium batteries is to conduct cyclic charge-discharge experiments on lithium batteries, control the differences in test conditions (such as temperature and rate), carry out group experiments, collect and analyze data, and fit curves to establish a capacity estimation model.

[0004] Empirical methods rely on fitting empirical data to estimate battery capacity. However, this approach is overly dependent on the accuracy of the fitted curve model and requires specialized equipment to conduct the experiments. Summary of the Invention

[0005] This invention provides a method for predicting the battery capacity of a vehicle battery system, which can accurately estimate and predict the battery capacity, thereby solving the problems of difficulty and low accuracy in battery capacity estimation.

[0006] This invention provides a method for predicting the battery capacity of a vehicle battery system, comprising: acquiring multiple sets of state data for the battery system; wherein each set of state data corresponds to a certain operating condition of the battery system, the battery system includes multiple individual cells, and the state data sets include data reporting time, total current, voltage of each individual cell, SOC, vehicle cumulative mileage, vehicle speed, and battery system temperature; determining the operating condition factors and historical battery capacity of the battery system based on each set of state data; wherein the operating condition factors include historical average SOC range, historical average temperature of the battery system, historical average rate capability, historical cumulative mileage of the vehicle, characteristic capacity consistency, historical average vehicle speed, vehicle storage time percentage, high SOC time percentage, and fast charging frequency percentage; determining a battery capacity prediction model based on the operating condition factors and historical battery capacity; and predicting the battery capacity of the battery system based on the battery capacity prediction model.

[0007] Optionally, the operating condition factor of the battery system is determined based on each set of state data, including: determining the SOC range under the corresponding operating condition of each set of state data based on the maximum and minimum SOC values ​​in each set of state data; and determining the historical average SOC range under the corresponding operating condition based on the SOC range and the total time of the corresponding operating condition of each set of state data.

[0008] Optionally, each set of state data includes multiple state data entries. When the operating condition corresponding to each set of state data is a discharge condition or a charging condition, the operating condition factor and historical battery capacity of the battery system are determined based on each set of state data, including: calculating the charge value of each state data entry based on the total current recorded in each state data entry and the time difference between two adjacent state data entries; determining the historical battery capacity under the corresponding operating condition based on the charge value of each state data entry; determining the average rate under the corresponding operating condition based on the maximum value of the historical battery capacity; and determining the historical average rate under the corresponding operating condition based on the average rate and the total time of the corresponding operating condition.

[0009] Optionally, when the operating condition corresponding to each set of state data is a charging condition, the operating condition factor of the battery system is determined based on each set of state data, including: determining whether the average rate under the operating condition corresponding to each set of state data is greater than a preset value; when the average rate under the operating condition corresponding to each set of state data is less than the preset value, calculating the change in the charge value and the change in the voltage of each individual cell according to a preset offset condition; and determining the characteristic capacity consistency under the charging condition based on the historical battery capacity of the battery system corresponding to the maximum value of the ratio of the change in the voltage of each individual cell to the change in the charge value.

[0010] Optionally, when the operating condition corresponding to each set of state data is a charging condition, the operating condition factor of the battery system is determined according to each set of state data, including: when the average rate under the operating condition corresponding to each set of state data is greater than a preset value, the proportion of fast charging times is determined according to the ratio of the number of charging conditions to the total number of historical charging times.

[0011] Optionally, when the operating condition corresponding to each set of state data is a discharge condition, the operating condition factor of the battery system is determined according to each set of state data, including: determining the historical average vehicle speed under the discharge condition based on the average vehicle speed corresponding to the discharge condition and the total time of the discharge condition; preferably, when the operating condition corresponding to each set of state data is a stationary condition or a charging condition, the operating condition factor of the battery system is determined according to each set of state data, including: determining the historical average vehicle speed under the stationary condition and the charging condition based on the historical average vehicle speed under the previous discharge condition.

[0012] Optionally, when the operating condition corresponding to each set of state data is a static operating condition, the operating condition factor of the battery system is determined according to each set of state data, including: determining the proportion of vehicle storage time based on the ratio of the total time of static operating condition to the total time of vehicle access to the big data platform.

[0013] Optionally, the operating condition factors of the battery system are determined based on each set of state data, including: determining the proportion of high SOC duration based on the ratio of the duration when SOC is greater than or equal to the preset SOC to the total time the vehicle is connected to the big data platform.

[0014] Optionally, the operating condition factor of the battery system is determined based on each set of state data, including: determining the historical average temperature of the battery system under each operating condition based on the average temperature of the battery system corresponding to each set of state data and the total time of the operating condition corresponding to each set of state data.

[0015] Optionally, the operating condition factor of the battery system is determined based on each set of state data, including: taking the cumulative mileage of any vehicle recorded in each set of state data as the historical cumulative mileage of the vehicle under the corresponding operating condition of each set of state data.

[0016] The method for predicting the battery capacity of a vehicle battery system provided in this invention first acquires multiple sets of state data for the battery system. Then, based on each set of state data, it determines the operating condition factors and historical battery capacity of the battery system. Next, it determines a battery capacity prediction model based on the operating condition factors and historical battery capacity. Finally, it predicts the battery capacity of the battery system based on the battery capacity prediction model. This method can determine the operating condition factors and historical battery capacity under different operating conditions. The operating condition factors include the historical average SOC range, the historical average temperature of the battery system, the historical average rate, the vehicle's historical cumulative mileage, characteristic capacity consistency, the historical average vehicle speed, the proportion of vehicle storage time, the proportion of high SOC time, and the proportion of fast charging times. Based on the operating condition factors and historical battery capacity, an accurate battery capacity prediction model can be obtained, thereby accurately estimating and predicting the battery capacity and solving the problems of difficulty and low accuracy in battery capacity estimation.

[0017] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart of a method for predicting the battery capacity of a vehicle battery system provided in an embodiment of the present invention;

[0020] Figure 2 This is a flowchart of another method for predicting the battery capacity of a vehicle battery system provided in an embodiment of the present invention;

[0021] Figure 3 This is a flowchart of another method for predicting the battery capacity of a vehicle battery system provided in an embodiment of the present invention;

[0022] Figure 4 This is a flowchart of another method for predicting the battery capacity of a vehicle battery system provided in an embodiment of the present invention;

[0023] Figure 5 This is a flowchart of another method for predicting the battery capacity of a vehicle battery system provided in an embodiment of the present invention;

[0024] Figure 6 This is a flowchart of another method for predicting the battery capacity of a vehicle battery system provided in an embodiment of the present invention;

[0025] Figure 7 This is a flowchart of another method for predicting the battery capacity of a vehicle battery system provided in an embodiment of the present invention;

[0026] Figure 8 This is a feature importance evaluation diagram of a battery capacity prediction model provided in an embodiment of the present invention;

[0027] Figure 9 This is a comparison chart of the predicted value and the actual value of a battery capacity prediction model provided in an embodiment of the present invention;

[0028] Figure 10 This is a probability density distribution diagram of the prediction error value of a battery capacity prediction model provided in an embodiment of the present invention. Detailed Implementation

[0029] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0030] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0031] Figure 1This is a flowchart of a method for predicting the battery capacity of a vehicle battery system according to an embodiment of the present invention. The method can be executed by a battery system battery capacity prediction device, which can be implemented in software and / or hardware and can be integrated into the vehicle's battery management system.

[0032] like Figure 1 As shown, the method for predicting the battery capacity of a vehicle battery system provided in this embodiment includes the following steps:

[0033] S101. Obtain multiple sets of status data for the battery system.

[0034] Each set of status data corresponds to a battery system operating condition. The battery system includes multiple individual cells, and the status data set includes data reporting time, total current, voltage of each individual cell, SOC, vehicle cumulative mileage, vehicle speed, and battery system temperature.

[0035] A vehicle refers to a wheeled vehicle that is driven or towed by a power unit, travels on roads, and is used for carrying people, transporting goods, or performing specialized engineering operations. For example, a vehicle may be an electric vehicle that includes a battery pack.

[0036] Operating conditions include charging, discharging, and resting.

[0037] Each set of status data corresponds to a table, and each set of status data includes multiple status data entries. Each status data entry is a single row of status data, and each status data entry includes the data reporting time, total current, voltage of each cell, SOC, vehicle cumulative mileage, vehicle speed, and battery system temperature.

[0038] The state of charge (SOC) of a battery system refers to the ratio of the current remaining capacity of the vehicle battery to the capacity of the vehicle battery when fully charged.

[0039] Optionally, the vehicle is connected to a big data platform. The big data platform can collect real-time operating data from the vehicle through onboard sensors and BMS, preprocess the real-time operating data, remove points with large errors in the real-time operating data, and thus obtain effective real-time operating data (i.e., status data set).

[0040] S102. Determine the operating condition factor and historical battery capacity of the battery system based on each set of state data.

[0041] Among them, the operating condition factors include the historical average SOC range, the historical average temperature of the battery system, the historical average rate, the historical cumulative mileage of the vehicle, the consistency of characteristic capacity, the historical average vehicle speed, the proportion of vehicle storage time, the proportion of high SOC time, and the proportion of fast charging times.

[0042] The historical average SOC range refers to the average SOC range of all previous state data groups under the corresponding operating conditions. The historical average SOC range of each state data group can be determined based on the SOC in each state data group.

[0043] The historical average temperature of the battery system refers to the average temperature of the operating conditions corresponding to all previous state data groups. The historical average temperature of the battery system under the corresponding operating conditions of each state data group can be determined based on the temperature of the battery system in each state data group.

[0044] Rate refers to the current required for a battery system to discharge its rated capacity within a specified time. Historical average rate refers to the average rate under the same operating conditions for all previous state data groups. The historical average rate can be determined based on the data reporting time and total current in each state data group.

[0045] Vehicle historical cumulative mileage refers to the cumulative mileage of any vehicle recorded in each set of status data.

[0046] Characteristic capacity consistency refers to the fact that after individual cells of the same specification and model are assembled into a battery system, the charging capacity of each individual cell may have certain differences. The characteristic capacity consistency of each state data group under the corresponding operating condition can be determined based on the voltage of each individual cell in each state data group.

[0047] Historical average vehicle speed refers to the average vehicle speed of the state data group corresponding to all discharge conditions of the battery system. The historical average vehicle speed of the state data group corresponding to the discharge condition can be determined based on the vehicle speed in the state data group corresponding to the discharge condition.

[0048] The vehicle storage time percentage refers to the proportion of the vehicle's storage time to the total time the vehicle is connected to the big data platform. This percentage can be determined based on the data reporting time in the status data group corresponding to the static working condition.

[0049] The high SOC duration percentage refers to the proportion of the total time when the SOC reaches or exceeds the preset SOC to the total time when the vehicle is connected to the big data platform. The high SOC duration percentage for each working condition can be determined based on the SOC and data reporting time in each group of status data.

[0050] The percentage of fast charging times refers to the proportion of historical fast charging times to historical charging times. Fast charging means that the average rate corresponding to each set of status data is greater than a preset value. The percentage of fast charging times can be determined based on the total current of each set of status data.

[0051] Historical battery capacity refers to the total charge of the battery system. The historical battery capacity of each state data group can be determined based on the total current and data reporting time in each state data group.

[0052] S103. Determine the battery capacity prediction model based on operating condition factors and historical battery capacity.

[0053] Operating condition factors and historical battery capacity are integrated into training data. This training data is then used to train the battery capacity estimation model, resulting in a battery capacity prediction model. Optionally, the battery capacity estimation model can be built based on the XGBoost regression algorithm.

[0054] S104. Predict the battery capacity of the battery system based on the battery capacity prediction model.

[0055] By inputting operating condition factors under different operating conditions into the battery capacity prediction model, the battery capacity prediction model can obtain the predicted battery capacity of the battery system.

[0056] The method for predicting the battery capacity of a vehicle battery system provided in this invention first acquires multiple sets of state data for the battery system. Then, based on each set of state data, it determines the operating condition factors and historical battery capacity of the battery system. Next, it determines a battery capacity prediction model based on the operating condition factors and historical battery capacity. Finally, it predicts the battery capacity of the battery system based on the battery capacity prediction model. This method can determine the operating condition factors and historical battery capacity under different operating conditions. The operating condition factors include the historical average SOC range, the historical average temperature of the battery system, the historical average rate, the vehicle's historical cumulative mileage, characteristic capacity consistency, the historical average vehicle speed, the proportion of vehicle storage time, the proportion of high SOC time, and the proportion of fast charging times. Based on the operating condition factors and historical battery capacity, an accurate battery capacity prediction model can be obtained, thereby accurately estimating and predicting the battery capacity and solving the problems of difficulty and low accuracy in battery capacity estimation.

[0057] Figure 2 This is a flowchart of another method for predicting the battery capacity of a vehicle battery system provided in an embodiment of the present invention, such as... Figure 2 As shown, the method for predicting the battery capacity of a vehicle battery system provided in this embodiment includes the following steps:

[0058] S201. Obtain multiple sets of status data for the battery system.

[0059] S202. Determine the SOC range for each working condition based on the maximum and minimum SOC values ​​in each set of state data.

[0060] Calculate the SOC range for each group of state data using the following formula:

[0061] SOCrange =SOC max -SOC min ;

[0062] Among them, SOC range This represents the SOC range for each group of state data. max SOC represents the maximum SOC value in each group of state data. min This represents the minimum SOC value in each group of state data.

[0063] S203. Determine the historical average SOC range for each operating condition based on the SOC range and the total time of the corresponding operating condition for each group of state data.

[0064] The SOC range of each set of state data is converted into the historical average SOC range using the following formula:

[0065]

[0066] Among them, y i x represents the historical average SOC range under the operating condition corresponding to the i-th state data group. j t represents the SOC range of the j-th state data group. j Represents the total time for the working condition corresponding to the j-th state data group, where i is a positive integer greater than or equal to 2.

[0067] If the j-th status data group includes N status data, then the total time corresponding to the working condition in the j-th status data group is equal to the data reporting time of the Nth status data in the j-th status data group minus the data reporting time of the first status data.

[0068] For example, the historical average SOC range for the operating condition corresponding to the second state data group is:

[0069]

[0070] Where y2 represents the historical average SOC range under the operating condition corresponding to the second state data group, x1 represents the SOC range of the first state data group, t1 represents the total time of the operating condition corresponding to the first state data group, x2 represents the SOC range of the second state data group, and t2 represents the total time of the operating condition corresponding to the second state data group. In other words, determining the historical average SOC range under the operating condition corresponding to the i-th state data group requires the SOC ranges from the first to the i-th state data group. When calculating the historical SOC range under the operating condition corresponding to each state data group, the SOC ranges of all previous state data groups are included in the calculation process to eliminate the inaccuracy of simple averaging; a weighted average is closer to the actual situation.

[0071] S204. Determine the battery capacity prediction model based on operating condition factors and historical battery capacity.

[0072] S205. Predict the battery capacity of the battery system based on the battery capacity prediction model.

[0073] Figure 3 This is a flowchart of another method for predicting the battery capacity of a vehicle battery system provided by an embodiment of the present invention. This embodiment is based on the above embodiments, such as... Figure 3 As shown, when the operating condition corresponding to each set of state data is a discharge condition or a charging condition, the operating condition factor and historical battery capacity of the battery system are determined based on each set of state data, including:

[0074] S301. Calculate the charge value of each status data based on the total current recorded in each status data and the time difference between two adjacent status data.

[0075] Calculate the battery value for each status data point using the following formula:

[0076] q j =I j ×Δtime j ;

[0077] Where, q j I represents the battery level value of the j-th status data. j Δtime represents the total current of the j-th state data record. j This represents the time difference between two adjacent state data points, where j is a positive integer greater than or equal to 1.

[0078] The time difference between two adjacent status data is equal to the data reporting time of the j-th status data minus the data reporting time of the (j-1)-th status data. Specifically, the energy value of the first status data is equal to the total current recorded in the first status data multiplied by a fixed time. For example, the fixed time can be 10 seconds.

[0079] S302. Determine the historical battery capacity under the corresponding operating condition for each set of status data based on the battery capacity value of each set of status data corresponding to the operating condition.

[0080] The historical battery capacity under the corresponding operating condition for each set of state data is calculated using the following formula:

[0081]

[0082] Among them, Q i q represents the historical battery capacity under the corresponding operating condition for each set of state data. j This represents the battery value corresponding to each row of status data.

[0083] S303. Determine the average rate of each state data group under the corresponding operating condition based on the maximum historical battery capacity.

[0084] The average multiplier for each set of state data under the corresponding operating condition is determined using the following formula:

[0085]

[0086] Among them, Q i c_rate represents the historical battery capacity under the corresponding operating condition for each set of state data, c_rate represents the average rate under the corresponding operating condition for each set of state data, and Δtime represents the total time under the corresponding operating condition for each set of state data.

[0087] Specifically, Q i It only represents the numerical value of charge / discharge capacity and is not needed for identifying charging / discharging, therefore Q is not retained. i The symbol is used. Because the data reporting time is constantly changing, the historical battery capacity corresponding to each group of status data is constantly changing, meaning that there is a maximum value for the historical battery capacity corresponding to each group of status data.

[0088] S304. Determine the historical average multiplier for each working condition based on the average multiplier and the total time of the corresponding working condition for each group of state data.

[0089] Convert the average multiplier of each set of state data into the historical average multiplier using the following formula:

[0090]

[0091] Among them, y i c represents the historical average multiplier under the operating condition corresponding to the i-th state data group. j t represents the average multiple of the j-th group of state data. j Represents the total time for the working condition corresponding to the j-th state data group, where i is a positive integer greater than or equal to 2.

[0092] For example, the historical average multiple for the operating condition corresponding to the second state data group is:

[0093]

[0094] Where y2 represents the historical average multiplier under the working condition corresponding to the second state data group, c1 represents the SOC range of the first state data group, t1 represents the total time of the working condition corresponding to the first state data group, c2 represents the historical average multiplier of the second state data group, and t2 represents the total time of the working condition corresponding to the second state data group.

[0095] Optionally, when the operating condition corresponding to each set of state data is a static operating condition, the operating condition factor of the battery system is determined according to each set of state data, including: determining the historical average rate under the static operating condition based on the historical average rate of the previous non-static operating condition.

[0096] The historical average multiplier under stationary conditions is equal to the historical average multiplier under non-stationary conditions. Specifically, a vehicle being parked for more than half an hour is considered a stationary condition. In other words, a time difference of more than half an hour between two adjacent data points is considered a stationary condition.

[0097] Figure 4 This is a flowchart of another method for predicting the battery capacity of a vehicle battery system provided by an embodiment of the present invention. This embodiment is based on the above embodiments, such as... Figure 4 As shown, when the operating condition corresponding to each set of state data is the charging condition, the operating condition factor of the battery system is determined based on each set of state data, including:

[0098] S401. Determine whether the average multiplier under the corresponding working condition for each group of state data is greater than the preset value.

[0099] Specifically, when the average rate of charge under the corresponding operating condition of each set of status data is less than a preset value, the operating condition corresponding to each set of status data is a slow charging condition; when the average rate of charge under the corresponding operating condition of each set of status data is greater than the preset value, the operating condition corresponding to each set of status data is a fast charging condition. Therefore, when the average rate of charge under the corresponding operating condition of each set of status data is less than the preset value, step S4021 is executed, and when the average rate of charge under the corresponding operating condition of each set of status data is greater than the preset value, step S4022 is executed.

[0100] S4021. When the average multiplier under the corresponding operating condition of each group of status data is less than the preset value, calculate the change in the power value and the change in the voltage of each individual cell according to the preset offset condition.

[0101] The changes in electrical charge and voltage of each individual cell are calculated using the following formulas:

[0102] dq = q - q_shift;

[0103] dV i =V i -V i _shift;

[0104] Where dq represents the change in battery level for each group of state data, q represents the battery level for each group of state data, q_shift represents the battery level for each group of state data after offsetting according to a preset offset condition, and dV i This represents the change in the voltage of the i-th cell in each set of state data, V. iThis represents the voltage of the i-th cell in each state data group, in V. i _shift represents the voltage of the i-th column in each state data group after offsetting according to the preset offset conditions.

[0105] S4022. When the average rate under the corresponding working condition of each state data group is greater than the preset value, the proportion of fast charging times is determined according to the ratio of the number of charging working conditions to the total number of historical charging times.

[0106] S403. Determine the characteristic capacity consistency under charging conditions based on the historical battery capacity of the battery system corresponding to the maximum value of the ratio of the change in voltage of each individual cell to the change in charge value.

[0107] The ratio of the change in voltage to the change in charge of each individual cell is denoted as dV. i / dq, where i represents the sequence number of each individual unit voltage. Based on dV i A curve is plotted showing the historical battery capacity of the battery system and its / dq value. The vertical axis of the curve is dV. i / dq, the horizontal axis represents the historical battery capacity of the battery system (i.e., the total charge of the battery system).

[0108] When each state data group includes multiple state data entries, each state data group includes multiple columns of individual cell voltages. That is, each column of individual cell voltages includes multiple state data entries, and each state data entry corresponds to a charge value. Since the historical battery capacity of the battery system in each state data group is the accumulation of the charge values ​​of each state data entry, each column of individual cell voltages corresponds to the historical battery capacity of the battery system, i.e., dV. i / dq corresponds to the historical battery capacity of a battery system, therefore the dV of each cell voltage column i / dq can plot a curve. If each state data set has 98 individual voltage columns, then 98 curves can be plotted. Find the dV for each individual voltage column. i The total charge of the battery system corresponding to the maximum value of / dq is used to calculate the voltage column dV of each individual cell. i The range of the total charge of the battery system corresponding to the maximum value of / dq, that is, the characteristic capacity consistency corresponding to each group of state data.

[0109] Figure 5 This is a flowchart of another method for predicting the battery capacity of a vehicle battery system provided by an embodiment of the present invention. This embodiment is based on the above embodiments, such as... Figure 5 As shown, the method for predicting the battery capacity of a vehicle battery system provided in this embodiment includes the following steps:

[0110] S501, Obtain multiple sets of status data from the battery system.

[0111] S502. When the operating condition corresponding to each group of state data is the discharge operating condition, determine the historical average vehicle speed under the discharge operating condition based on the average vehicle speed corresponding to the discharge operating condition and the total time of the discharge operating condition.

[0112] The historical average vehicle speed under discharge conditions is determined using the following formula:

[0113]

[0114] Where speed represents the historical average vehicle speed corresponding to the current state data group under the corresponding operating condition, v i t represents the average speed corresponding to the working condition in the i-th group of state data. i I(soc_status) represents the total time for the operating condition corresponding to the i-th group of status data. i =1) indicates that the i-th group of state data is in the discharge condition.

[0115] S503. When the operating condition corresponding to each group of state data is a stationary operating condition or a charging operating condition, the historical average vehicle speed for the stationary operating condition and the charging operating condition is determined based on the historical average vehicle speed under the previous discharge operating condition.

[0116] Among them, the historical average vehicle speed under stationary and charging conditions is equal to the historical average vehicle speed under the previous discharge condition.

[0117] S504. Determine the battery capacity prediction model based on operating condition factors and historical battery capacity.

[0118] S505. Predict the battery capacity of the battery system based on the battery capacity prediction model.

[0119] Figure 6 This is a flowchart of another method for predicting the battery capacity of a vehicle battery system provided by an embodiment of the present invention. This embodiment is based on the above embodiments, such as... Figure 6 As shown, the method for predicting the battery capacity of a vehicle battery system provided in this embodiment includes the following steps:

[0120] S601, Obtain multiple sets of status data from the battery system.

[0121] S602. When the working condition corresponding to each group of status data is a static working condition, the proportion of vehicle storage time is determined according to the ratio of the total time of static working condition to the total time of vehicle access to the big data platform.

[0122] The percentage of storage time under static conditions is determined using the following formula:

[0123]

[0124] Wherein, storeTimes represents the percentage of storage time in the current inactive state, t iI(soc_status) represents the total time for the operating condition corresponding to the i-th group of status data. i =0) indicates that the working condition corresponding to the i-th group of state data is the static working condition, T i This represents the total time the vehicle spends connected to the big data platform, which is the total time for all status data groups corresponding to the operating conditions.

[0125] S603. Determine the battery capacity prediction model based on operating condition factors and historical battery capacity.

[0126] S604. Predict the battery capacity of the battery system based on the battery capacity prediction model.

[0127] Figure 7 This is a flowchart of another method for predicting the battery capacity of a vehicle battery system provided by an embodiment of the present invention. This embodiment is based on the above embodiments, such as... Figure 7 As shown, the method for predicting the battery capacity of a vehicle battery system provided in this embodiment includes the following steps:

[0128] S701, Obtain multiple sets of status data from the battery system.

[0129] S702. Determine the proportion of high SOC duration based on the ratio of the duration when SOC is greater than or equal to the preset SOC to the total time when the vehicle accesses the big data platform.

[0130] For example, if the preset SOC is 80% of the vehicle battery's capacity in a fully charged state, the percentage of high SOC duration under the corresponding operating condition for each set of state data can be calculated using the following formula:

[0131]

[0132] Where highSOC represents the percentage of high SOC duration under the corresponding operating condition in the current state data group, t i .I(SOC≥80) represents the total time when the SOC is greater than or equal to 80 under the corresponding operating condition of the i-th state data group, T i This represents the total time the vehicle spends connected to the big data platform, which is the total time for all state data groups under the corresponding operating conditions. In other words, to calculate the percentage of high SOC time for the operating conditions corresponding to the Nth state data group, we need to use the total time for the operating conditions corresponding to the Nth state data group with an SOC greater than or equal to 80.

[0133] S703. Determine the historical average temperature of the battery system under each operating condition based on the average temperature of the battery system corresponding to each set of state data and the total time of the operating condition corresponding to each set of state data.

[0134]

[0135] Where T_mean represents the historical average temperature of the battery system under the corresponding operating condition of the current state data group, T ei t represents the average temperature of the battery system under the corresponding operating condition of the i-th set of state data. i This represents the total time for the operating condition corresponding to the i-th group of state data.

[0136] S704. The cumulative mileage of any vehicle recorded in each set of status data is taken as the historical cumulative mileage of the vehicle under the corresponding working condition of each set of status data.

[0137] Optionally, the vehicle's historical cumulative mileage under the corresponding operating condition for each state data group is the vehicle's cumulative mileage in each state data group, and the value can be the value of the first line of state data record in each state data group.

[0138] S705. Determine the battery capacity prediction model based on operating condition factors and historical battery capacity.

[0139] S706. Predict the battery capacity of the battery system based on the battery capacity prediction model.

[0140] Battery capacity prediction models can assess the characteristic importance of each characteristic factor among operating condition factors. Figure 8 This is a feature importance evaluation diagram for a battery capacity prediction model provided in an embodiment of the present invention. (Reference) Figure 8 The feature importance of vehicle historical cumulative mileage (miles) is approximately 0.35, the feature importance of high SOC duration (highSOC) percentage ...

[0141] Figure 9 This is a comparison chart of the predicted value and the actual value of a battery capacity prediction model provided in an embodiment of the present invention. Figure 10 This is a probability density distribution diagram of the prediction error value of a battery capacity prediction model provided in an embodiment of the present invention. Combined with... Figure 9 and Figure 10 It can be seen that the battery capacity prediction model obtained by the vehicle battery system battery capacity prediction method provided in any embodiment of the present invention can accurately estimate and predict the battery capacity.

[0142] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0143] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for predicting the battery capacity of a vehicle battery system, characterized in that, include: Multiple sets of status data for the battery system are acquired; wherein each set of status data corresponds to an operating condition of the battery system, the battery system includes multiple individual cells, and the status data set includes data reporting time, total current, voltage of each individual cell, SOC, vehicle cumulative mileage, vehicle speed, and temperature of the battery system; The operating condition factors and historical battery capacity of the battery system are determined based on each set of state data. The operating condition factors include the historical average SOC range, the historical average temperature of the battery system, the historical average rate, the historical cumulative mileage of the vehicle, the characteristic capacity consistency, the historical average vehicle speed, the proportion of vehicle storage time, the proportion of high SOC time, and the proportion of fast charging times. A battery capacity prediction model is determined based on the operating condition factors and the historical battery capacity. Predict the battery capacity of the battery system based on the battery capacity prediction model; When the operating condition corresponding to each group of the aforementioned state data is a discharge condition, the operating condition factor of the battery system is determined based on each group of the aforementioned state data, including: The historical average vehicle speed under the discharge condition is determined based on the average vehicle speed corresponding to the discharge condition and the total time of the discharge condition. When the operating condition corresponding to each group of state data is a resting condition or a charging condition, the operating condition factor of the battery system is determined according to each group of state data, including: The historical average vehicle speeds for the stationary and charging conditions are determined based on the historical average vehicle speeds under the discharge conditions described above.

2. The method for predicting the battery capacity of a vehicle battery system according to claim 1, characterized in that, The operating condition factors of the battery system are determined based on each set of state data, including: The SOC range for each working condition is determined based on the maximum and minimum SOC values ​​in each group of state data. The historical average SOC range for each operating condition is determined based on the SOC range and the total time corresponding to each group of state data.

3. The method for predicting the battery capacity of a vehicle battery system according to claim 1, characterized in that, Each set of state data includes multiple state data entries. When the operating condition corresponding to each set of state data is a discharge condition or a charging condition, determining the operating condition factor and historical battery capacity of the battery system based on each set of state data includes: The energy value of each state data is calculated based on the total current of each state data record and the time difference between two adjacent state data records; The historical battery capacity under the corresponding operating condition for each group of state data is determined based on the power value of each state data item. The average rate of each state data group under the corresponding operating condition is determined based on the maximum value of the historical battery capacity. The historical average multiplier for a given operating condition is determined based on the average multiplier and the total time corresponding to each group of state data.

4. The method for predicting the battery capacity of a vehicle battery system according to claim 3, characterized in that, When the operating condition corresponding to each group of the aforementioned state data is a charging condition, the operating condition factor of the battery system is determined based on each group of the aforementioned state data, including: Determine whether the average multiplier under the corresponding working condition for each group of state data is greater than a preset value; When the average multiplier under the corresponding operating condition of each group of state data is less than the preset value, the change in the power value and the change in the voltage of each individual cell are calculated according to the preset offset condition. The characteristic capacity consistency under the charging condition is determined based on the historical battery capacity of the battery system corresponding to the maximum value of the ratio of the change in voltage of each individual cell to the change in charge value.

5. The method for predicting the battery capacity of a vehicle battery system according to claim 3, characterized in that, When the operating condition corresponding to each group of the aforementioned state data is a charging condition, the operating condition factor of the battery system is determined based on each group of the aforementioned state data, including: When the average multiplier under the corresponding working condition of each group of state data is greater than a preset value, the proportion of fast charging times is determined according to the ratio of the number of times the charging working condition is used to the total number of historical charging times.

6. The method for predicting the battery capacity of a vehicle battery system according to claim 1, characterized in that, When the operating condition corresponding to each group of the aforementioned state data is a static operating condition, the operating condition factor of the battery system is determined based on each group of the aforementioned state data, including: The vehicle storage time percentage is determined by the ratio of the total time spent in the static operating condition to the total time the vehicle spends accessing the big data platform.

7. The method for predicting the battery capacity of a vehicle battery system according to claim 1, characterized in that, The operating condition factors of the battery system are determined based on each set of state data, including: The percentage of high SOC duration is determined by the ratio of the duration during which the SOC is greater than or equal to a preset SOC to the total time the vehicle spends accessing the big data platform.

8. The method for predicting the battery capacity of a vehicle battery system according to claim 1, characterized in that, The operating condition factors of the battery system are determined based on each set of state data, including: The historical average temperature of the battery system under each operating condition is determined based on the average temperature of the battery system corresponding to each group of state data and the total time of the operating condition corresponding to each group of state data.

9. The method for predicting the battery capacity of a vehicle battery system according to claim 1, characterized in that, The operating condition factors of the battery system are determined based on each set of state data, including: The cumulative mileage of any vehicle recorded in each of the state data groups is taken as the historical cumulative mileage of the vehicle under the corresponding operating condition of each state data group.

Images