Intelligent prediction and compensation method for life attenuation of new energy automobile power battery

Abstract

The application discloses a new energy automobile power battery life attenuation intelligent prediction and compensation method and relates to the technical field of new energy automobiles. The application effectively solves the problems of jump and loss existing in the vehicle-mounted original data by multi-dimensional data cleaning, abnormality elimination and timestamp alignment; ensures standardization based on a bench test mapping table in the low mileage stage, fuses the Arrhenius correction algorithm of the discharge depth and the average temperature in the high mileage stage, and accurately captures the aging characteristics under the variable working conditions of the actual vehicle; cooperates with the monthly self-updating mechanism of the model parameters, significantly improves the accuracy and long-term adaptability of the residual life prediction, adjusts the charging and discharging power limit dynamically according to the RUL, starts the accurate balanced charging based on the voltage standard deviation, and calibrates the state of charge in combination with the OCV-SOC deviation. The method realizes the transformation from 'passive management' to 'active compensation', effectively delays the deterioration of the battery consistency, and maximizes the utilization rate of the whole life cycle of the battery.

Description

Technical Field

[0001] This invention relates to the field of new energy vehicle technology, specifically to a method for intelligent prediction and compensation of the lifespan degradation of power batteries in new energy vehicles. Background Technology

[0002] With the rapid development of the new energy vehicle industry, the performance of power batteries, as the core energy source, directly affects the vehicle's range, safety, and operating costs. During use, power batteries gradually experience capacity decay and increased internal resistance, a process known as battery life degradation. When the battery capacity decays to below 80% of its rated capacity, it is generally considered to have reached the end of life (EOL). Therefore, accurately predicting the remaining useful life (RUL) of power batteries and effectively managing and compensating based on the prediction results are key technologies for battery management systems (BMS) and onboard big data platforms.

[0003] In the existing technological system, the prediction of power battery life mainly relies on laboratory bench test data and simple empirical models. Traditional prediction methods are usually based on the correspondence between a constant number of charge-discharge cycles and capacity decay (i.e., the SN curve), estimating life by converting vehicle mileage into equivalent cycle counts. However, real-world vehicle operating conditions are extremely complex, and battery degradation is not only related to the number of cycles, but also closely related to the depth of discharge (DOD), charge-discharge rate, ambient temperature, and the internal temperature field distribution of the battery pack.

[0004] The existing technology has the following main drawbacks: 1. Limited data dimensions and coarse cleaning: Raw data collected from vehicles often contains a large number of abnormal jumps caused by electromagnetic interference or sensor jitter, as well as missing segments due to communication packet loss. Existing methods mostly use simple filtering or direct removal, lacking a mechanism for aligning and reconstructing timestamps of multidimensional physical quantities (such as voltage, current, temperature, and SOC), resulting in low quality of basic data for feature extraction.

[0005] 2. The predictive model is out of sync with real vehicle operating conditions: Traditional bench test models are based on an ideal constant temperature and constant current environment, which cannot reflect the real stress of varying operating conditions and temperatures during actual vehicle operation. For older vehicles with high mileage, the degradation mechanism is significantly different from that of new vehicles. If the linear mapping relationship based on bench tests is still used, the prediction error will increase sharply with the increase of mileage.

[0006] 3. Lack of dynamic correction and adaptive capabilities: Most existing life prediction models are static parameter models with fixed coefficients. As battery materials age and the vehicle's operating environment changes, the model parameters will drift. If not updated in real time, the prediction results will gradually deviate from the true values.

[0007] 4. Disconnect between prediction and control strategies: Most BMS can only perform passive equalization based on the current voltage difference or simple power limiting based on the current SOC, lacking a forward-looking compensation strategy based on future lifetime prediction (RUL). For example, it cannot dynamically adjust charging and discharging power limits according to the predicted degradation trend, nor can it combine RUL and consistency status to generate an accurate SOC calibration scheme, leading to problems such as consistency deterioration and a sharp drop in usable power in the later stages of battery life. Summary of the Invention

[0008] To address the aforementioned technical problems, this technical solution provides an intelligent prediction and compensation method for the lifespan degradation of power batteries in new energy vehicles, resolving the issues raised in the background section.

[0009] To achieve the above objectives, the technical solution adopted by the present invention is as follows: In a first aspect of the present invention, a method for intelligent prediction and compensation of the lifespan degradation of power batteries for new energy vehicles is also provided, comprising: The vehicle's power battery historical operating data is collected in real time through the vehicle big data platform. The data includes time series data such as single cell voltage, total voltage, total current, bus current, battery pack temperature, ambient temperature, SOC state of charge, and driving range. The collected raw data is cleaned and reconstructed in multiple dimensions, abnormal jump data is removed, missing data segments are filled in, and timestamps are aligned. Based on the cleaned data, the fixed depth of discharge (DOD), average temperature during discharge, current integral capacity of charge and discharge segments, and voltage drop characteristics of the battery were extracted within equal interval mileage ranges. A lifespan degradation prediction model based on real vehicle data is constructed. Depending on whether the historical mileage of the target vehicle is greater than a preset threshold, the corresponding prediction algorithm is selected to calculate the degradation percentage at the target mileage. If the historical mileage is greater than the threshold, a correction algorithm that integrates fixed depth of discharge and average temperature is used for prediction. If the historical mileage is less than or equal to the threshold, the prediction is made based on the relationship between the number of cycles and the degradation percentage established by bench test data. The remaining life (RUL) of the power battery is calculated based on the predicted percentage of degradation, and a life degradation trend chart is generated. Based on the remaining life prediction results and the current battery consistency status, compensation control instructions for the battery management system are generated. These instructions include equalization charging strategies, charge and discharge power limits, and SOC calibration schemes.

[0010] Preferably, the collection of historical operating data of the target vehicle's power battery specifically includes: When the vehicle is in motion, the sampling frequency of individual unit voltage, total voltage, total current, and bus current is set to 10Hz, that is, once every 100 milliseconds. The sampling frequency for battery pack temperature and ambient temperature is set to 1Hz, meaning it is sampled once every 1 second. The SOC state of charge update frequency is synchronized with the total voltage sampling frequency, which is 10Hz. Mileage data is obtained by integrating the vehicle speed signal, which is sampled at a frequency of 50Hz. The cumulative mileage is calculated in real time using a formula. : ; in, Indicates the total mileage traveled. This indicates the vehicle's real-time speed. Indicates the sampling time interval; The cumulative mileage is uploaded to the big data platform every 100 meters. When the vehicle accelerates or decelerates rapidly, and the absolute value of the acceleration is greater than 0.5g, the high-frequency acquisition mode is triggered, temporarily increasing the sampling frequency of voltage and current to 100Hz for 10 seconds to capture transient voltage drop characteristics.

[0011] Preferably, the multidimensional cleaning and reconstruction of the collected raw data specifically includes: Set a sliding time window with a width of 5 seconds and a step size of 1 second, and calculate the standard deviation of the individual voltage data within the window; if the difference between the individual voltage value at a certain moment and the average value within the window is greater than 3 times the standard deviation, then the voltage at that moment is determined to be abnormal jump data and is removed. For time breakpoints generated after removing outlier data, if the breakpoint duration is less than or equal to 10 seconds, linear interpolation is used to complete the breakpoints. Let the time of the missing point be... The previous moment was The next moment is The corresponding values ​​are respectively and Then missing values The calculation formula is: ; in, This represents the voltage padding value at the missing moment. This represents the voltage sample value at the moment immediately preceding the missing point. This represents the voltage sample value at the moment following the missing point. The timestamp representing the missing point; If the breakpoint duration is greater than 10 seconds, the data segment is marked as invalid and will not be completed. Based on the vehicle GPS timestamp, the timestamps of total voltage, total current, bus current, battery pack temperature, ambient temperature, SOC and mileage are uniformly aligned to the same millisecond-level time axis to ensure that all physical quantity data are included under the same timestamp.

[0012] Preferably, the extraction of the fixed depth of discharge (DOD) and average temperature during the discharge process of the battery within equal mileage intervals specifically includes: Divide the driving distance into segments of 10 kilometers each to obtain the first segment. Mileage segment ; Extract the first The SOC value at the start time of each mileage segment is denoted as... The SOC value at the end time is extracted and denoted as . ; Calculate the fixed depth of discharge for this mileage section. The calculation formula is: ; in, Indicates the first Depth of discharge for each mileage segment, This indicates the percentage of the state of charge at the start of that mileage segment. This indicates the percentage of the state of charge at the end of the mileage segment; like This is then marked as a discharge process, and only the discharge process is retained. ; Extract the temperature data from all battery pack sampling points during the discharge process period, and calculate the arithmetic mean as the average temperature during the discharge process for that mileage range. The calculation formula is: ; in, Indicates the first Average temperature of each discharge mileage segment This indicates the total number of temperature sampling points within this mileage segment. Indicates the first The battery pack temperature value at each sampling time.

[0013] Preferably, the extraction of the current integral capacity and voltage drop characteristics of the charge / discharge segment specifically includes: Select the discharge start time Until the end of discharge The time period; Total current during this time period By performing time integration, the current integral capacity is obtained. The formula is calculated using the trapezoidal integral method: ; in, This represents the current integral capacity of the discharge segment. This represents the total number of current sampling points within the discharge segment. Indicates the first The total current value at each sampling time. Indicates the first The total current value at each sampling time. Indicates the first The timestamp of each sampling moment Indicates the first The timestamp of each sampling moment; Extract the start time of discharge Average unit voltage and the end of discharge Average unit voltage ; Calculate voltage drop characteristics The formula is: ; in, This represents the voltage drop during the discharge process. This represents the average value of the individual cell voltage at the moment of discharge initiation. This represents the average voltage of a single cell at the moment the discharge ends. Simultaneously calculate the voltage drop rate. The formula is: ; in, Indicates the rate of voltage drop. Indicates the duration of the discharge.

[0014] Preferably, the step of selecting a corresponding prediction algorithm to calculate the attenuation percentage at the target mileage based on whether the historical mileage of the target vehicle is greater than a preset threshold specifically includes: The preset threshold for historical mileage is set to 100,000 kilometers; When the target vehicle's historical mileage At 10,000 kilometers, perform predictions based on bench test data: Retrieve the pre-stored bench test mapping table, which records different number of loops. Corresponding capacity decay percentage ; Historical driving mileage Converted to equivalent loop count The conversion formula is: ; in, This indicates the equivalent number of cycles corresponding to the historical mileage. This indicates the historical mileage of the target vehicle. This refers to the average driving range of a vehicle on a full charge. Search in the mapping table For the corresponding interval, linear interpolation is used to calculate the current attenuation percentage. ; Based on target mileage Calculate the total equivalent number of iterations. The formula is: ; in, This represents the total equivalent number of loops corresponding to the target mileage. This indicates the target predicted mileage set by the user. Search in the mapping table The target attenuation percentage is obtained from the corresponding interval. ; When the target vehicle's historical mileage At 10,000 kilometers, execute the fusion correction algorithm: The modified model of the Arrhenius equation is established, and the calculation formula is as follows: ; in, This indicates the predicted percentage of degradation at the target mileage. These are coefficients obtained in advance by fitting data from actual vehicles. Boltzmann's constant, The average temperature of the historical discharge process. The average depth of discharge within the target mileage range. Historical mileage It is a natural constant; Under the target mileage and Substituting into the above equation, the percentage of attenuation at the target mileage can be directly calculated.

[0015] Preferably, the step of calculating the remaining life (RUL) of the power battery based on the predicted degradation percentage and generating a life degradation trend chart specifically includes: Obtain the rated capacity of the power battery and the current measured maximum capacity ; Calculate the current actual attenuation percentage The formula is: ; in, This indicates the current actual percentage of decay. This indicates the battery's rated capacity at the time of manufacture. This indicates the currently measured maximum available capacity; The end-of-life threshold is set at 80% of the rated capacity, that is, when When RUL is 0; Using the predicted percentage of attenuation at the target mileage Combined with the current percentage of decay Calculate the remaining tolerable attenuation. The formula is: ; in, It represents the percentage increase in degradation that can be tolerated from the current state until the end of the life (80% degradation); Calculate the total predicted attenuation increment The formula is: ; in, This indicates the total percentage decrease expected from the current mileage to the target mileage; Calculate remaining life mileage The formula is: ; in, Indicates remaining lifespan mileage. Indicates the target mileage. Indicates historical mileage; Using historical mileage as the horizontal axis and cumulative attenuation percentage as the vertical axis, plot the attenuation data points that have occurred, and extend the calculated future attenuation trend by fitting a linear or exponential curve to generate a continuous lifespan attenuation trend chart.

[0016] Preferably, the generated equalization charging strategy specifically includes: Real-time acquisition of voltage values ​​for all individual cells within the battery pack, and calculation of voltage standard deviation. The formula is: ; in, The standard deviation of the individual cell voltage is represented by the standard deviation of the individual cell voltage. This indicates the total number of individual cells in the battery pack. Indicates the first The voltage value of each individual cell. This represents the arithmetic mean of the voltages of all individual cells. If standard deviation If the battery consistency is poor, then equalization charging will be initiated. The cell with the highest voltage is selected as the benchmark, and the equalization activation threshold is set to the benchmark voltage minus... ; For all individual cells with voltages higher than the equalization threshold, passive discharge is performed through a parallel resistor until their voltage drops to the reference voltage minus [a certain value]. Within the range; During the equalization process, if any single cell voltage is detected to be lower than the preset safety lower limit voltage, the equalization circuit will be immediately disconnected.

[0017] Preferably, the generation of charge / discharge power limits and SOC calibration scheme specifically includes: Based on the Remaining Life Limit (RUL) and the current battery consistency status, query the pre-stored power limit mapping table; If the rated power limit (RUL) is greater than 30%, the rated power remains unchanged; if the RUL is between 10% and 30%, the charging power is limited to 70% of the rated power and the discharging power is limited to 80% of the rated power; if the RUL is less than 10%, the charging power is limited to 40% of the rated power and the discharging power is limited to 50% of the rated power. For the SOC calibration procedure, it should be performed when the vehicle is stationary and the battery pack temperature is within 25℃±5℃: Read the ampere-hour integral SOC value calculated by the current BMS, and denot it as... ; Measure the current open-circuit voltage (OCV) of the battery pack, and obtain the corresponding OCV-SOC curve value from the table, denoted as . ; Calculate calibration coefficients The formula is: ; in, Indicates the SOC calibration coefficient. This represents the state of charge value obtained by looking up the open-circuit voltage in a table. This represents the state of charge value calculated by the BMS using the ampere-hour integration method; like Then the SOC value inside the BMS will be forcibly corrected to... The corrected formula is: ; in, This represents the revised new SOC value, with 0.5 being the preset correction factor. like If not, no correction will be made.

[0018] In a second aspect of the invention, a smart prediction and compensation system for the life degradation of power batteries in new energy vehicles is also provided, comprising: The data acquisition module is used to collect historical operating data of the target vehicle's power battery in real time through the vehicle big data platform. The data includes time series data such as single cell voltage, total voltage, total current, bus current, battery pack temperature, ambient temperature, SOC state of charge, and driving range. The data processing module is used to perform multi-dimensional cleaning and reconstruction on the collected raw data, remove abnormal jump data, fill in missing data segments, and perform timestamp alignment. The extraction module is used to extract, based on the cleaned data, the fixed depth of discharge (DOD), average temperature during discharge, current integral capacity of charge and discharge segments, and voltage drop characteristics of the battery within equal interval mileage segments. The prediction module is used to construct a life decay prediction model based on real vehicle data. It selects the corresponding prediction algorithm to calculate the decay percentage at the target mileage based on whether the historical mileage of the target vehicle is greater than a preset threshold. If the historical mileage is greater than the threshold, a correction algorithm that integrates fixed discharge depth and average temperature is used for prediction. If the historical mileage is less than or equal to the threshold, prediction is made based on the correspondence between the number of cycles and the decay percentage established by bench test data. The first generation module is used to calculate the remaining life (RUL) of the power battery based on the predicted degradation percentage and generate a life degradation trend chart. The second generation module is used to generate compensation control instructions for the battery management system based on the remaining life prediction results and the current battery consistency status. The instructions include equalization charging strategy, charge and discharge power limit and SOC calibration scheme.

[0019] Compared with existing technologies, this invention provides a method for intelligent prediction and compensation of the lifespan degradation of power batteries for new energy vehicles, which has the following beneficial effects: This invention first effectively solves the problems of jumps and missing data in original vehicle data through multi-dimensional data cleaning, anomaly removal, and timestamp alignment, ensuring high synchronization of multiple physical quantities such as voltage, current, temperature, and SOC on the time axis, providing a high-quality data foundation for feature extraction. In the low-mileage stage, standardization is ensured based on a bench test mapping table; in the high-mileage stage, an Arrhenius correction algorithm integrating depth of discharge and average temperature is used to accurately capture aging characteristics under varying vehicle operating conditions. Combined with a monthly self-updating mechanism for model parameters, the accuracy and long-term adaptability of remaining battery life prediction are significantly improved. Furthermore, charging and discharging power limits are dynamically adjusted based on RUL, precise equalization charging is initiated based on voltage standard deviation, and state of charge calibration is performed using OCV-SOC deviation. This method achieves a shift from "passive management" to "active compensation," effectively delaying battery consistency deterioration and maximizing battery utilization throughout its entire lifecycle. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the method flow for S101-S106 in this invention.

[0021] Figure 2 This is a schematic diagram of the method flow for S201-S206 in this invention.

[0022] Figure 3 This is a schematic diagram of the method flow for S301-S305 in this invention.

[0023] Figure 4 This is a schematic diagram of the method flow for S401-S405 in this invention.

[0024] Figure 5 This is a schematic diagram of the method flow for S501-S503 in this invention. Detailed Implementation

[0025] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0026] Example 1 Please refer to Figure 1 As shown, in a first aspect of the present invention, a method for intelligent prediction and compensation of life degradation of power batteries for new energy vehicles is also provided, comprising: S101. Real-time collection of historical operating data of the target vehicle's power battery through the vehicle big data platform. The data includes time series data such as single cell voltage, total voltage, total current, bus current, battery pack temperature, ambient temperature, SOC state of charge, and driving range. S102. Perform multidimensional cleaning and reconstruction on the collected raw data, remove abnormal jump data, fill in missing data segments, and align timestamps. S103. Based on the cleaned data, extract the fixed depth of discharge (DOD), average temperature during discharge, current integral capacity of charge and discharge segments, and voltage drop characteristics of the battery within equal interval mileage segments. S104. Construct a lifespan degradation prediction model based on real vehicle data. Select the corresponding prediction algorithm to calculate the degradation percentage under the target mileage based on whether the historical mileage of the target vehicle is greater than a preset threshold. If the historical mileage is greater than the threshold, a correction algorithm that integrates fixed discharge depth and average temperature is used for prediction. If the historical mileage is less than or equal to the threshold, prediction is made based on the relationship between the number of cycles and the degradation percentage established by bench test data. S105. Calculate the remaining life (RUL) of the power battery based on the predicted degradation percentage and generate a life degradation trend chart. S106. Based on the remaining life prediction results and the current battery consistency status, generate compensation control instructions for the battery management system. The instructions include equalization charging strategy, charging and discharging power limit and SOC calibration scheme.

[0027] As will be understood by those skilled in the art, this invention firstly addresses the issues of jumps and missing data in original vehicle data through multi-dimensional data cleaning, anomaly removal, and timestamp alignment. This ensures high synchronization of multiple physical quantities such as voltage, current, temperature, and SOC on the time axis, providing a high-quality data foundation for feature extraction. In the low-mileage stage, standardization is ensured based on a bench test mapping table. In the high-mileage stage, an Arrhenius correction algorithm integrating depth of discharge (DOD) and average temperature is used to accurately capture aging characteristics under varying vehicle operating conditions. Combined with a monthly self-updating mechanism for model parameters, this significantly improves the accuracy and long-term adaptability of remaining battery life (RUL) prediction. Furthermore, it dynamically adjusts charging and discharging power limits based on RUL, initiates precise equalization charging based on voltage standard deviation, and performs state of charge calibration using OCV-SOC deviation. This method achieves a shift from "passive management" to "active compensation," effectively delaying battery consistency deterioration and maximizing battery utilization throughout its entire lifecycle.

[0028] Please refer to Figure 2 As shown, historical operating data of the target vehicle's power battery is collected, specifically including: S201. When the vehicle is in motion, the sampling frequency of individual unit voltage, total voltage, total current and bus current is set to 10Hz, that is, once every 100 milliseconds. S202. The sampling frequency for battery pack temperature and ambient temperature is set to 1Hz, that is, once every 1 second. The update frequency of S203 and SOC state of charge is synchronized with the total voltage sampling frequency, which is 10Hz; S204. Mileage data is obtained based on the integration of the vehicle speed signal, which is sampled at a frequency of 50Hz. The cumulative mileage is calculated in real time using a formula. : ; in, Indicates the total mileage traveled. This indicates the vehicle's real-time speed. Indicates the sampling time interval; S205: The cumulative mileage value is uploaded to the big data platform every 100 meters. S206. When the vehicle accelerates or decelerates rapidly, and the absolute value of the acceleration is greater than 0.5g, the high-frequency acquisition mode is triggered, and the sampling frequency of voltage and current is temporarily increased to 100Hz for 10 seconds to capture transient voltage drop characteristics.

[0029] Please refer to Figure 3 As shown, the collected raw data undergoes multidimensional cleaning and reconstruction, specifically including: S301. Set a sliding time window with a width of 5 seconds and a step size of 1 second. Calculate the standard deviation of the individual voltage data within the window. If the difference between the individual voltage value at a certain moment and the average value within the window is greater than 3 times the standard deviation, then the voltage at that moment is determined to be abnormal jump data and is removed. S302. For time breakpoints generated after removing outlier data, if the breakpoint duration is less than or equal to 10 seconds, linear interpolation should be used to complete the breakpoints. S303, Let the time of the missing point be... The previous moment was The next moment is The corresponding values ​​are respectively and Then missing values The calculation formula is: ; in, This represents the voltage padding value at the missing moment. This represents the voltage sample value at the moment immediately preceding the missing point. This represents the voltage sample value at the moment following the missing point. The timestamp representing the missing point; S304. If the breakpoint duration is greater than 10 seconds, mark the data segment as invalid and do not perform completion. S305 uses the vehicle GPS timestamp as a reference to align the timestamps of total voltage, total current, bus current, battery pack temperature, ambient temperature, SOC, and mileage to the same millisecond-level time axis, ensuring that all physical quantity data are included under the same timestamp.

[0030] Please refer to Figure 4 As shown, the fixed depth of discharge (DOD) and average temperature during discharge are extracted from the battery over equally spaced mileage intervals, specifically including: S401. Divide the driving mileage into segments of 10 kilometers each to obtain the first segment. Mileage segment ; S402, Extract the first The SOC value at the start time of each mileage segment is denoted as... The SOC value at the end time is extracted and denoted as . ; S403. Calculate the fixed depth of discharge for this mileage section. The calculation formula is: ; in, Indicates the first Depth of discharge for each mileage segment, This indicates the percentage of the state of charge at the start of that mileage segment. This indicates the percentage of the state of charge at the end of the mileage segment; S404, if This is then marked as a discharge process, and only the discharge process is retained. ; S405. Extract the temperature data of all battery pack sampling points during the discharge process period, and calculate the arithmetic mean as the average temperature of the discharge process for that mileage segment. The calculation formula is: ; in, Indicates the first Average temperature of each discharge mileage segment This indicates the total number of temperature sampling points within this mileage segment. Indicates the first The battery pack temperature value at each sampling time.

[0031] Please refer to Figure 5 As shown, the current integral capacity and voltage drop characteristics of the charge / discharge segment are extracted, specifically including: S501, Select the discharge start time Until the end of discharge The time period; S502, Total current during this time period By performing time integration, the current integral capacity is obtained. The formula is calculated using the trapezoidal integral method: ; in, This represents the current integral capacity of the discharge segment. This represents the total number of current sampling points within the discharge segment. Indicates the first The total current value at each sampling time. Indicates the first The total current value at each sampling time. Indicates the first The timestamp of each sampling moment Indicates the first The timestamp of each sampling moment; S503, Extract the discharge start time Average unit voltage and the end of discharge Average unit voltage ; Calculate voltage drop characteristics The formula is: ; in, This represents the voltage drop during the discharge process. This represents the average value of the individual cell voltage at the moment of discharge initiation. This represents the average voltage of a single cell at the moment the discharge ends. Simultaneously calculate the voltage drop rate. The formula is: ; in, Indicates the rate of voltage drop. Indicates the duration of the discharge.

[0032] Based on whether the target vehicle's historical mileage exceeds a preset threshold, the corresponding prediction algorithm is selected to calculate the attenuation percentage at the target mileage, specifically including: The preset threshold for historical mileage is set to 100,000 kilometers; When the target vehicle's historical mileage At 10,000 kilometers, perform predictions based on bench test data: Retrieve the pre-stored bench test mapping table, which records different number of loops. Corresponding capacity decay percentage ; Historical driving mileage Converted to equivalent loop count The conversion formula is: ; in, This indicates the equivalent number of cycles corresponding to the historical mileage. This indicates the historical mileage of the target vehicle. This refers to the average driving range of a vehicle on a full charge. Search in the mapping table For the corresponding interval, linear interpolation is used to calculate the current attenuation percentage. ; Based on target mileage Calculate the total equivalent number of iterations. The formula is: ; in, This represents the total equivalent number of loops corresponding to the target mileage. This indicates the target predicted mileage set by the user. Search in the mapping table The target attenuation percentage is obtained from the corresponding interval. ; When the target vehicle's historical mileage At 10,000 kilometers, execute the fusion correction algorithm: The modified model of the Arrhenius equation is established, and the calculation formula is as follows: ; in, This indicates the predicted percentage of degradation at the target mileage. These are coefficients obtained in advance by fitting data from actual vehicles. Boltzmann's constant, The average temperature of the historical discharge process. The average depth of discharge within the target mileage range. Historical mileage It is a natural constant; Under the target mileage and Substituting into the above equation, the percentage of attenuation at the target mileage can be directly calculated.

[0033] The remaining life (RUL) of the power battery is calculated based on the predicted percentage of degradation, and a life degradation trend chart is generated, including: Obtain the rated capacity of the power battery and the current measured maximum capacity ; Calculate the current actual attenuation percentage The formula is: ; in, This indicates the current actual percentage of decay. This indicates the battery's rated capacity at the time of manufacture. This indicates the currently measured maximum available capacity; The end-of-life threshold is set at 80% of the rated capacity, that is, when When RUL is 0; Using the predicted percentage of attenuation at the target mileage Combined with the current percentage of decay Calculate the remaining tolerable attenuation. The formula is: ; in, It represents the percentage increase in degradation that can be tolerated from the current state until the end of the life (80% degradation); Calculate the total predicted attenuation increment The formula is: ; in, This indicates the total percentage decrease expected from the current mileage to the target mileage; Calculate remaining life mileage The formula is: ; in, Indicates remaining lifespan mileage. Indicates the target mileage. Indicates historical mileage; Using historical mileage as the horizontal axis and cumulative attenuation percentage as the vertical axis, plot the attenuation data points that have occurred, and extend the calculated future attenuation trend by fitting a linear or exponential curve to generate a continuous lifespan attenuation trend chart.

[0034] Generate a balanced charging strategy, specifically including: Real-time acquisition of voltage values ​​for all individual cells within the battery pack, and calculation of voltage standard deviation. The formula is: ; in, The standard deviation of the individual cell voltage is represented by the standard deviation of the individual cell voltage. This indicates the total number of individual cells in the battery pack. Indicates the first The voltage value of each individual cell. This represents the arithmetic mean of the voltages of all individual cells. If standard deviation If the battery consistency is poor, then equalization charging will be initiated. The cell with the highest voltage is selected as the benchmark, and the equalization activation threshold is set to the benchmark voltage minus... ; For all individual cells with voltages higher than the equalization threshold, passive discharge is performed through a parallel resistor until their voltage drops to the reference voltage minus [a certain value]. Within the range; During the equalization process, if any single cell voltage is detected to be lower than the preset safety lower limit voltage, the equalization circuit will be immediately disconnected.

[0035] Generate a charge / discharge power limitation and SOC calibration scheme, specifically including: Based on the Remaining Life Limit (RUL) and the current battery consistency status, query the pre-stored power limit mapping table; If the rated power limit (RUL) is greater than 30%, the rated power remains unchanged; if the RUL is between 10% and 30%, the charging power is limited to 70% of the rated power and the discharging power is limited to 80% of the rated power; if the RUL is less than 10%, the charging power is limited to 40% of the rated power and the discharging power is limited to 50% of the rated power. For the SOC calibration procedure, it should be performed when the vehicle is stationary and the battery pack temperature is within 25℃±5℃: Read the ampere-hour integral SOC value calculated by the current BMS, and denot it as... ; Measure the current open-circuit voltage (OCV) of the battery pack, and obtain the corresponding OCV-SOC curve value from the table, denoted as . ; Calculate calibration coefficients The formula is: ; in, Indicates the SOC calibration coefficient. This represents the state of charge value obtained by looking up the open-circuit voltage in a table. This represents the state of charge value calculated by the BMS using the ampere-hour integration method; like Then the SOC value inside the BMS will be forcibly corrected to... The corrected formula is: ; in, This represents the revised new SOC value, with 0.5 being the preset correction factor. like If not, no correction will be made.

[0036] In a second aspect of the invention, a smart prediction and compensation system for the life degradation of power batteries in new energy vehicles is also provided, comprising: The data acquisition module is used to collect historical operating data of the target vehicle's power battery in real time through the vehicle big data platform. The data includes time series data such as single cell voltage, total voltage, total current, bus current, battery pack temperature, ambient temperature, SOC state of charge, and driving range. The data processing module is used to perform multi-dimensional cleaning and reconstruction of the collected raw data, remove abnormal jump data, fill in missing data segments, and perform timestamp alignment. The extraction module is used to extract the fixed depth of discharge (DOD), average temperature during discharge, current integral capacity and voltage drop characteristics of the battery within equal interval mileage ranges based on the cleaned data. The prediction module is used to build a life decay prediction model based on real vehicle data. It selects the corresponding prediction algorithm to calculate the decay percentage at the target mileage based on whether the historical mileage of the target vehicle is greater than a preset threshold. If the historical mileage is greater than the threshold, a correction algorithm that integrates fixed discharge depth and average temperature is used for prediction. If the historical mileage is less than or equal to the threshold, prediction is made based on the correspondence between the number of cycles and the decay percentage established by bench test data. The first generation module is used to calculate the remaining life (RUL) of the power battery based on the predicted degradation percentage and generate a life degradation trend chart. The second generation module is used to generate compensation control instructions for the battery management system based on the remaining life prediction results and the current battery consistency status. The instructions include equalization charging strategy, charging and discharging power limit and SOC calibration scheme.

[0037] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent prediction and compensation of life degradation of power batteries for new energy vehicles, characterized in that, include: The vehicle's power battery historical operating data is collected in real time through the vehicle big data platform. The data includes time series data such as single cell voltage, total voltage, total current, bus current, battery pack temperature, ambient temperature, SOC state of charge, and driving range. The collected raw data is cleaned and reconstructed in multiple dimensions, abnormal jump data is removed, missing data segments are filled in, and timestamps are aligned. Based on the cleaned data, the fixed depth of discharge (DOD), average temperature during discharge, current integral capacity of charge and discharge segments, and voltage drop characteristics of the battery were extracted within equal interval mileage ranges. A lifespan degradation prediction model based on real vehicle data is constructed. Depending on whether the historical mileage of the target vehicle is greater than a preset threshold, the corresponding prediction algorithm is selected to calculate the degradation percentage at the target mileage. If the historical mileage is greater than the threshold, a correction algorithm that integrates fixed depth of discharge and average temperature is used for prediction. If the historical mileage is less than or equal to the threshold, the prediction is made based on the relationship between the number of cycles and the degradation percentage established by bench test data. The remaining life (RUL) of the power battery is calculated based on the predicted percentage of degradation, and a life degradation trend chart is generated. Based on the remaining life prediction results and the current battery consistency status, compensation control instructions for the battery management system are generated. These instructions include equalization charging strategies, charge and discharge power limits, and SOC calibration schemes.

2. The intelligent prediction and compensation method for the life degradation of power batteries in new energy vehicles according to claim 1, characterized in that, The historical operating data of the target vehicle's power battery collected specifically includes: When the vehicle is in motion, the sampling frequency of individual unit voltage, total voltage, total current, and bus current is set to 10Hz, that is, once every 100 milliseconds. The sampling frequency for battery pack temperature and ambient temperature is set to 1Hz, meaning it is sampled once every 1 second. The SOC state of charge update frequency is synchronized with the total voltage sampling frequency, which is 10Hz. Mileage data is obtained by integrating the vehicle speed signal, which is sampled at a frequency of 50Hz. The cumulative mileage is calculated in real time using a formula. : ； in, Indicates the total mileage traveled. This indicates the vehicle's real-time speed. Indicates the sampling time interval; The cumulative mileage is uploaded to the big data platform every 100 meters. When the vehicle accelerates or decelerates rapidly, and the absolute value of the acceleration is greater than 0.5g, the high-frequency acquisition mode is triggered, temporarily increasing the sampling frequency of voltage and current to 100Hz for 10 seconds to capture transient voltage drop characteristics.

3. The intelligent prediction and compensation method for the life degradation of power batteries in new energy vehicles according to claim 1, characterized in that, The multidimensional cleaning and reconstruction of the collected raw data specifically includes: Set a sliding time window with a width of 5 seconds and a step size of 1 second, and calculate the standard deviation of the individual voltage data within the window; if the difference between the individual voltage value at a certain moment and the average value within the window is greater than 3 times the standard deviation, then the voltage at that moment is determined to be abnormal jump data and is removed. For time breakpoints generated after removing outlier data, if the breakpoint duration is less than or equal to 10 seconds, linear interpolation is used to complete the breakpoints. Let the time of the missing point be... The previous moment was The next moment is The corresponding values ​​are respectively and Then missing values The calculation formula is: ； in, This represents the voltage padding value at the missing moment. This represents the voltage sample value at the moment immediately preceding the missing point. This represents the voltage sample value at the moment following the missing point. The timestamp representing the missing point; If the breakpoint duration is greater than 10 seconds, the data segment is marked as invalid and will not be completed. Based on the vehicle GPS timestamp, the timestamps of total voltage, total current, bus current, battery pack temperature, ambient temperature, SOC and mileage are uniformly aligned to the same millisecond-level time axis to ensure that all physical quantity data are included under the same timestamp.

4. The intelligent prediction and compensation method for the life degradation of power batteries in new energy vehicles according to claim 1, characterized in that, The extraction of the fixed depth of discharge (DOD) and average temperature during the discharge process of the battery within equally spaced mileage intervals specifically includes: Divide the driving distance into segments of 10 kilometers each to obtain the first segment. Mileage segment ; Extract the first The SOC value at the start time of each mileage segment is denoted as... The SOC value at the end time is extracted and denoted as . ; Calculate the fixed depth of discharge for this mileage section. The calculation formula is: ； in, Indicates the first Depth of discharge for each mileage segment, This indicates the percentage of the state of charge at the start of that mileage segment. This indicates the percentage of the state of charge at the end of the mileage segment; like This is then marked as a discharge process, and only the discharge process is retained. ; Extract the temperature data from all battery pack sampling points during the discharge process period, and calculate the arithmetic mean as the average temperature during the discharge process for that mileage range. The calculation formula is: ； in, Indicates the first Average temperature of each discharge mileage segment This indicates the total number of temperature sampling points within this mileage segment. Indicates the first The battery pack temperature value at each sampling time.

5. The intelligent prediction and compensation method for the life degradation of power batteries in new energy vehicles according to claim 1, characterized in that, The extraction of the current integral capacity and voltage drop characteristics of the charge / discharge segment specifically includes: Select the discharge start time Until the end of discharge The time period; Total current during this time period By performing time integration, the current integral capacity is obtained. The formula is calculated using the trapezoidal integral method: ； in, This represents the current integral capacity of the discharge segment. This represents the total number of current sampling points within the discharge segment. Indicates the first The total current value at each sampling time. Indicates the first The total current value at each sampling time. Indicates the first The timestamp of each sampling moment Indicates the first The timestamp of each sampling moment; Extract the start time of discharge Average unit voltage and the end of discharge Average unit voltage ; Calculate voltage drop characteristics The formula is: ； in, This represents the voltage drop during the discharge process. This represents the average value of the individual cell voltage at the moment of discharge initiation. This represents the average voltage of a single cell at the moment the discharge ends. Simultaneously calculate the voltage drop rate. The formula is: ； in, Indicates the rate of voltage drop. Indicates the duration of the discharge.

6. The intelligent prediction and compensation method for the life degradation of power batteries in new energy vehicles according to claim 1, characterized in that, The step of selecting a corresponding prediction algorithm to calculate the attenuation percentage under the target mileage based on whether the historical mileage of the target vehicle is greater than a preset threshold specifically includes: The preset threshold for historical mileage is set to 100,000 kilometers; When the target vehicle's historical mileage At 10,000 kilometers, perform predictions based on bench test data: Retrieve the pre-stored bench test mapping table, which records different number of loops. Corresponding capacity decay percentage ; Historical driving mileage Converted to equivalent loop count The conversion formula is: ； in, This indicates the equivalent number of cycles corresponding to the historical mileage. This indicates the historical mileage of the target vehicle. This refers to the average driving range of a vehicle on a full charge. Search in the mapping table For the corresponding interval, linear interpolation is used to calculate the current attenuation percentage. ; Based on target mileage Calculate the total equivalent number of iterations. The formula is: ； in, This represents the total equivalent number of loops corresponding to the target mileage. This indicates the target predicted mileage set by the user. Search in the mapping table The target attenuation percentage is obtained from the corresponding interval. ; When the target vehicle's historical mileage At 10,000 kilometers, execute the fusion correction algorithm: The modified model of the Arrhenius equation is established, and the calculation formula is as follows: ； in, This indicates the predicted percentage of degradation at the target mileage. These are coefficients obtained in advance by fitting data from actual vehicles. Boltzmann's constant, The average temperature of the historical discharge process. The average depth of discharge within the target mileage range. Historical mileage It is a natural constant; Under the target mileage and Substituting into the above equation, the percentage of attenuation at the target mileage can be directly calculated.

7. The intelligent prediction and compensation method for the life degradation of power batteries in new energy vehicles according to claim 1, characterized in that, The process of calculating the remaining life (RUL) of the power battery based on the predicted percentage of degradation and generating a life degradation trend chart specifically includes: Obtain the rated capacity of the power battery and the current measured maximum capacity ; Calculate the current actual attenuation percentage The formula is: ； in, This indicates the current actual percentage of decay. This indicates the battery's rated capacity at the time of manufacture. This indicates the currently measured maximum available capacity; The end-of-life threshold is set at 80% of the rated capacity, that is, when When RUL is 0; Using the predicted percentage of attenuation at the target mileage Combined with the current percentage of decay Calculate the remaining tolerable attenuation. The formula is: ； in, It represents the percentage increase in degradation that can be tolerated from the current state until the end of the life (80% degradation); Calculate the total predicted attenuation increment The formula is: ； in, This indicates the total percentage decrease expected from the current mileage to the target mileage; Calculate remaining life mileage The formula is: ； in, Indicates remaining lifespan mileage. Indicates the target mileage. Indicates historical mileage; Using historical mileage as the horizontal axis and cumulative attenuation percentage as the vertical axis, plot the attenuation data points that have occurred, and extend the calculated future attenuation trend by fitting a linear or exponential curve to generate a continuous lifespan attenuation trend chart.

8. The intelligent prediction and compensation method for the life degradation of power batteries in new energy vehicles according to claim 1, characterized in that, The generated equalization charging strategy specifically includes: Real-time acquisition of voltage values ​​for all individual cells within the battery pack, and calculation of voltage standard deviation. The formula is: ； in, The standard deviation of the individual cell voltage is represented by the standard deviation of the individual cell voltage. This indicates the total number of individual cells in the battery pack. Indicates the first The voltage value of each individual cell. This represents the arithmetic mean of the voltages of all individual cells. If standard deviation If the battery consistency is poor, then equalization charging will be initiated. The cell with the highest voltage is selected as the benchmark, and the equalization activation threshold is set to the benchmark voltage minus... ; For all individual cells with voltages higher than the equalization threshold, passive discharge is performed through a parallel resistor until their voltage drops to the reference voltage minus [a certain value]. Within the range; During the equalization process, if any single cell voltage is detected to be lower than the preset safety lower limit voltage, the equalization circuit will be immediately disconnected.

9. The intelligent prediction and compensation method for the life degradation of power batteries in new energy vehicles according to claim 1, characterized in that, The aforementioned charge / discharge power limitation and SOC calibration scheme specifically includes: Based on the Remaining Life Limit (RUL) and the current battery consistency status, query the pre-stored power limit mapping table; If the rated power limit (RUL) is greater than 30%, the rated power remains unchanged; if the RUL is between 10% and 30%, the charging power is limited to 70% of the rated power and the discharging power is limited to 80% of the rated power; if the RUL is less than 10%, the charging power is limited to 40% of the rated power and the discharging power is limited to 50% of the rated power. For the SOC calibration procedure, it should be performed when the vehicle is stationary and the battery pack temperature is within 25℃±5℃: Read the ampere-hour integral SOC value calculated by the current BMS, and denot it as... ; Measure the current open-circuit voltage (OCV) of the battery pack, and obtain the corresponding OCV-SOC curve value from the table, denoted as . ; Calculate calibration coefficients The formula is: ； in, Indicates the SOC calibration coefficient. This represents the state of charge value obtained by looking up the open-circuit voltage in a table. This represents the state of charge value calculated by the BMS using the ampere-hour integration method; like Then the SOC value inside the BMS will be forcibly corrected to... The corrected formula is: ； in, This represents the revised new SOC value, with 0.5 being the preset correction factor. like If not, no correction will be made.

10. A smart prediction and compensation system for the life degradation of power batteries in new energy vehicles, used to implement the smart prediction and compensation method for the life degradation of power batteries in new energy vehicles as described in any one of claims 1-9, characterized in that, include: The data acquisition module is used to collect historical operating data of the target vehicle's power battery in real time through the vehicle big data platform. The data includes time series data such as single cell voltage, total voltage, total current, bus current, battery pack temperature, ambient temperature, SOC state of charge, and driving range. The data processing module is used to perform multi-dimensional cleaning and reconstruction on the collected raw data, remove abnormal jump data, fill in missing data segments, and perform timestamp alignment. The extraction module is used to extract, based on the cleaned data, the fixed depth of discharge (DOD), average temperature during discharge, current integral capacity of charge and discharge segments, and voltage drop characteristics of the battery within equal interval mileage segments. The prediction module is used to construct a life decay prediction model based on real vehicle data. It selects the corresponding prediction algorithm to calculate the decay percentage at the target mileage based on whether the historical mileage of the target vehicle is greater than a preset threshold. If the historical mileage is greater than the threshold, a correction algorithm that integrates fixed discharge depth and average temperature is used for prediction. If the historical mileage is less than or equal to the threshold, prediction is made based on the correspondence between the number of cycles and the decay percentage established by bench test data. The first generation module is used to calculate the remaining life (RUL) of the power battery based on the predicted degradation percentage and generate a life degradation trend chart. The second generation module is used to generate compensation control instructions for the battery management system based on the remaining life prediction results and the current battery consistency status. The instructions include equalization charging strategy, charge and discharge power limit and SOC calibration scheme.

Images