Method and device for evaluating use performance of vehicle battery

Abstract

The invention discloses a method and a device for evaluating the use performance of a vehicle battery, and relates to the technical field of evaluation of the use performance of a battery. According to the technical scheme of the invention, the target data of the multiple vehicles are obtained on the big data platform, and the target data are the data of the use performance of the power batteries of the vehicles along with the change in the process that the vehicles execute the historical driving tasks; according to the method, survival analysis and danger analysis can be carried out on target data of a plurality of vehicles to obtain a survival curve and a danger curve of a power battery; performing index item analysis on the survival curve and the danger curve according to the performance evaluation time period of the power battery to obtain a survival index and a danger index of the power battery; and determining the use performance of the power battery in the performance evaluation period according to the survival index and the danger index. According to the technical scheme, scientific and comprehensive evaluation of the use performance of the power battery is realized, and the accuracy of evaluation of the use performance of the power battery is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of battery performance evaluation, and in particular to a vehicle battery performance evaluation method and device. Background Art

[0002] With the rapid development of new energy vehicle technology, the performance advantages and application prospects of power batteries have attracted much attention. As the core component of new energy vehicles, the accuracy of power battery performance evaluation is crucial to the safety, reliability and user experience of the vehicle. However, existing power battery evaluation methods often rely on limited experimental data and empirical models, which are difficult to fully reflect the performance changes of power batteries in actual use.

[0003] Therefore, how to improve the accuracy of power battery performance evaluation is a technical problem that needs to be solved urgently. Summary of the invention

[0004] In view of the above technical problems, the present invention provides a vehicle battery performance evaluation method and device, which can implement accurate evaluation of power battery performance based on big data, thereby improving the accuracy of power battery performance evaluation.

[0005] The embodiment of the present invention provides the following solution:

[0006] In a first aspect, an embodiment of the present invention provides a method for evaluating the performance of a vehicle battery, the method comprising:

[0007] Obtain target data of multiple vehicles on the big data platform, wherein the target data is data on changes in the performance of the vehicle's power battery during the vehicle's historical driving mission;

[0008] Perform survival analysis and hazard analysis on target data of multiple vehicles to obtain the survival curve and hazard curve of the power battery;

[0009] According to the performance evaluation period of the power battery, the survival curve and the danger curve are analyzed for indicators respectively to obtain the survival index and danger index of the power battery;

[0010] Based on the survival index and danger index, determine the performance of the power battery during the performance evaluation period.

[0011] In an optional embodiment, obtaining target data of multiple vehicles on a big data platform includes:

[0012] Extract historical driving task data on the big data platform according to the preset statistical table template to obtain a statistical table for each vehicle;

[0013] According to the data represented by the data statistics table of each vehicle, first data of the driving time of each vehicle changing with the use performance of the power battery and second data of the driving mileage of each vehicle changing with the use performance of the power battery are obtained;

[0014] Target data for each vehicle is obtained based on the first data and the second data.

[0015] In an optional embodiment, obtaining target data of each vehicle according to the first data and the second data includes:

[0016] Cleaning the first data and the second data of each vehicle to remove duplicate data and invalid data;

[0017] Perform missing value filling on the first data and the second data after data cleaning to obtain the first data and the second data after data filling;

[0018] The first data and the second data after the data filling are normalized to obtain the target data of the corresponding vehicle.

[0019] In an optional embodiment, survival analysis and hazard analysis are performed on target data of multiple vehicles respectively to obtain a survival curve and a hazard curve of the power battery, including:

[0020] Obtaining target faults of the multiple vehicles according to data features represented by target data of the multiple vehicles, wherein the target faults are faults related to the performance of the power battery;

[0021] Determine the survival probability of the power battery under different usage cycles based on the target failures of multiple vehicles and the preset survival function;

[0022] According to the corresponding change relationship between the service life of the power battery and the survival probability, the survival curve of the power battery is obtained;

[0023] According to the survival curve and the preset hazard function, the hazard curve of the power battery is obtained.

[0024] In an optional embodiment, obtaining target faults of multiple vehicles according to data features represented by target data of multiple vehicles includes:

[0025] According to the fault items of the performance of the power battery associated with different component systems on each vehicle, the corresponding characteristic data is determined in the target data, wherein the component systems include the transmission system, the braking system, the electrical system, the cooling system and the battery system;

[0026] The characteristic data of all fault items of each vehicle are counted to obtain the target faults of multiple vehicles.

[0027] In an optional embodiment, the survival curve and the danger curve are respectively analyzed for index items according to the performance evaluation period of the power battery to obtain the survival index and the danger index of the power battery, including:

[0028] Determine the median survival time, average survival time and risk ratio of the power battery under preset comparison items during the performance evaluation period according to the survival curve, wherein the preset comparison items include mileage comparison items and / or fault type comparison items;

[0029] Determine the cumulative risk value and / or curve slope of the power battery during the performance evaluation period according to the hazard curve;

[0030] Characterize the survival indicators of power batteries based on median survival time, mean survival time, and hazard ratio under preset comparison items;

[0031] Characterize the danger index of the power battery based on the cumulative risk value and / or curve slope.

[0032] In an optional embodiment, after determining the use performance of the power battery in the performance evaluation period according to the survival index and the danger index, the method further includes:

[0033] When the performance evaluation period is the operation management period, output the management strategy of the operation vehicle in the subsequent management period;

[0034] Determine the maintenance and replacement time of power batteries on operating vehicles based on management strategies.

[0035] In a second aspect, an embodiment of the present invention further provides a vehicle battery performance evaluation device, the device comprising:

[0036] An acquisition module is used to acquire target data of multiple vehicles on a big data platform, wherein the target data is data on changes in the performance of the vehicle's power battery during the vehicle's execution of historical driving tasks;

[0037] The first acquisition module is used to perform survival analysis and hazard analysis on target data of multiple vehicles to obtain a survival curve and a hazard curve of the power battery;

[0038] A second obtaining module is used to analyze the index items of the survival curve and the danger curve respectively according to the performance evaluation period of the power battery to obtain the survival index and the danger index of the power battery;

[0039] The first determination module is used to determine the use performance of the power battery during the performance evaluation period according to the survival index and the danger index.

[0040] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising a processor and a memory, wherein the memory is coupled to the processor and stores instructions, and when the instructions are executed by the processor, the electronic device executes the steps of any one of the methods in the first aspect.

[0041] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any one of the methods in the first aspect.

[0042] Compared with the prior art, the vehicle battery performance evaluation method and device of the present invention have the following advantages:

[0043] The technical solution of the present invention obtains target data of multiple vehicles on a big data platform. Since the target data is data on the performance changes of the power battery of the vehicle during the vehicle's execution of historical driving tasks, survival analysis and hazard analysis can be performed on the target data of multiple vehicles to obtain the survival curve and hazard curve of the power battery; the survival curve and hazard curve are analyzed for index items according to the performance evaluation period of the power battery to obtain the survival index and hazard index of the power battery; the performance of the power battery in the performance evaluation period is determined according to the survival index and hazard index. The technical solution implements survival analysis and hazard analysis based on big data related to the performance of the power battery, accurately counts the performance change characteristics of the power battery during use, realizes a scientific and comprehensive evaluation of the performance of the power battery, and improves the accuracy of the evaluation of the performance of the power battery. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 A flowchart of a method for evaluating the performance of a vehicle battery provided by an embodiment of the present invention;

[0046] Figure 2-1 A schematic diagram of a basic information table provided in an embodiment of the present invention;

[0047] Figure 2-2 Schematic diagram of a travel result table provided in an embodiment of the present invention Figure 1 ;

[0048] Figure 2-3 Schematic diagram 2 of a travel result table provided in an embodiment of the present invention;

[0049] Figure 2-4 A schematic diagram of a fault information statistics table provided in an embodiment of the present invention;

[0050] Figure 3 A flowchart of data preprocessing provided by an embodiment of the present invention;

[0051] Figure 4 A schematic diagram of a survival curve provided by an embodiment of the present invention;

[0052] Figure 5 A schematic diagram of a curve of a danger function provided by an embodiment of the present invention;

[0053] Figure 6 A schematic diagram of a danger curve provided by an embodiment of the present invention;

[0054] Figure 7 A schematic diagram of the output result of the risk ratio provided by an embodiment of the present invention;

[0055] Figure 8 A schematic diagram of the structure of a vehicle battery performance evaluation device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0056] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention belong to the scope of protection of the embodiments of the present invention.

[0057] See also Figure 1 , Figure 1 This is a flow chart of a vehicle battery performance evaluation method provided by an embodiment of the present invention. The evaluation method can be applied to computer equipment or server equipment to implement power battery performance evaluation. The operating equipment can run the evaluation method, and no specific limitation is made here. The evaluation method includes:

[0058] S11. Acquire target data of multiple vehicles on a big data platform, wherein the target data is data on changes in the performance of the power batteries of the vehicles during the execution of historical driving tasks.

[0059] Specifically, the big data platform can be a demonstration operation platform for vehicles, or a monitoring platform for data management of power batteries, which can collect data on the performance of power batteries when the number of driving tasks of a large number of vehicles changes. The target data can be the number of times the power battery is charged, or the mileage and travel time represented by the historical driving tasks. For example, as the number of historical driving tasks increases, the charging time and discharge time of the power battery change accordingly, and the data represented by the charging time and discharge time are determined as the target data. As the use time of the power battery increases, the charging time and discharge time will change accordingly, and the performance of the power battery can be determined accordingly. The target data can include multiple types of data related to the change of the use performance of the power battery, and corresponding labels are established for each type of data, which are stored in correspondence with the data related to the change of the use performance of the power battery. It can be understood that the power battery can be a solid-state battery or other types of batteries, and no specific restrictions are made here. Big data analysis plays a vital role. It can not only reveal the laws of driving behavior and power battery usage habits, but also provide strong support for fault diagnosis and early warning.

[0060] Exemplarily, step S11 includes sub-steps S11-1 to S11-3, which are specifically described as follows:

[0061] S11-1. Extract the data of historical driving tasks on the big data platform according to the preset statistical table template to obtain the data statistical table of each vehicle. The big data platform uses the vehicle-mounted sensors, wireless communication equipment and Internet of Things technology to collect vehicle operation and power battery usage data, and conducts a series of calculations and analyses to finally obtain vehicle operation and battery usage analysis results, providing users with driving behavior analysis, battery usage habit analysis, endurance analysis, fault diagnosis and early warning and other functions. The template for collecting data on the big data platform can be found at Figure 2-1 to Figure 2-4 , including the basic information table vin_base_info, the trip result table trip_result_info, the trip result table trip_speed_statistics and the fault information statistics table alarm_base. Through various tables, data from historical driving tasks can be comprehensively collected. The statistical table template can be set based on the extraction requirements of the target data, such as various types of vehicle faults, fault time, mileage, driving time, etc. The statistical table template can accurately extract the data required for subsequent analysis.

[0062] S11-2. According to the data represented by the data statistics table of each vehicle, obtain the first data of the driving time of each vehicle as the performance of the power battery changes, and the second data of the driving mileage of each vehicle as the performance of the power battery changes. The first data can be represented as the data of the driving time changing with various vehicle faults, such as the battery failure fault and short circuit fault that occur in the vehicle as the driving time increases. Similarly, the second data is the data of the power battery performance changing with the increase of the driving mileage, and the corresponding electrical system faults can be collected as the driving time increases.

[0063] S11-3, according to the first data and the second data, obtain the target data of each vehicle. The target data can be represented as a data set. The first data and the second data reflect the performance change characteristics of the power battery from different dimensions. Using them as target data can make subsequent analysis more comprehensive and accurate.

[0064] In practical applications, since the first data and the second data are obtained through various sensors or estimation algorithms, if the sensor fails or the vehicle system fails, the data may be inaccurate. Based on this, sub-step S11-3 is implemented based on the following steps. Specifically, it includes:

[0065] The first step is to clean the first and second data of each vehicle to remove duplicate and invalid data. Figure 3 , Figure 3 The flowchart of data preprocessing provided by the embodiment of the present invention. The collected raw data needs to be cleaned to remove duplicate and invalid data records, including abnormal data, to ensure the quality and integrity of the data. The embodiment of the present invention uses the IQR (Interquartile Range) method to process outliers.

[0066] #Example: Detect and remove outliers using the IQR method

[0067] Q1 = data.quantile(0.25)

[0068] Q3 = data.quantile(0.75)

[0069] IQR=Q3-Q1

[0070] data=data[~((data<(Q1-1.5*IQR))|(data>(Q3+1.5*IQR))).any(axis=1)]

[0071] The second step is to fill missing values ​​in the first data and the second data after data cleaning to obtain the first data and the second data after data filling. For the missing values ​​in the data, use an applicable method to fill the data, such as using an interpolation method or filling based on statistical information of historical data to reduce the impact of missing data on the analysis results; check whether there are missing values ​​in the data set and perform processing. The embodiment of the present invention uses the mean filling method for processing. The core code is as follows:

[0072] data = df[col]

[0073] print(data.isnull().sum()) / / Check the missing values ​​in the data. Assume that the original data is in DataFrame format, the variable name is df, and the field to be analyzed is col

[0074] data.fillna(data.mean(), inplace=True) / / fill missing values ​​with mean

[0075] The third step is to normalize the first and second data after data filling to obtain the target data of the corresponding vehicle. Data standardization / normalization is a key step in data preprocessing, which aims to convert the original data into a unified scale, thereby eliminating the dimensional differences between different data features and improving the stability and accuracy of the model. The core code is as follows:

[0076] from sklearn.preprocessing import StandardScaler

[0077] scaler = StandardScaler()

[0078] data_scaled=scaler.fit_transform(data)

[0079] After this series of data preprocessing steps, high-quality, standardized target data can be obtained, laying a solid foundation for subsequent analysis. After obtaining the target data of multiple vehicles, step S12 is entered.

[0080] S12. Perform survival analysis and hazard analysis on target data of multiple vehicles respectively to obtain a survival curve and a hazard curve of the power battery.

[0081] Specifically, survival analysis and hazard analysis are both statistical techniques. Survival analysis is used to analyze life expectancy or the time to an event. A survival curve shows the probability of survival at different time points. See Figure 4 , Figure 4This is a schematic diagram of the survival curve. The survival curve usually starts from 1 and gradually decreases over time, reflecting the decrease in the survival probability over time. The survival curve can be used to characterize the survival probability of the power battery at different usage times to assess the risk of subsequent use.

[0082] Hazard analysis is closely related to survival analysis, which focuses on the risk or rate of an event occurring. In survival analysis, the hazard function describes the conditional probability of an event occurring at a given point in time or time period, given that the individual has survived to that point in time. The hazard curve is a graphical representation of hazard analysis that shows the hazard rate over time. Unlike the survival curve, the hazard curve is not necessarily monotonically decreasing over time; it may increase, decrease, or remain constant over time, depending on the data characteristics represented by the target data. The hazard curve can help identify risk factors and how they affect the likelihood of an event occurring over time.

[0083] Survival analysis is a statistical method that makes statistical inferences on one or more non-negative random variables, studies survival phenomena and response time data and their statistical laws. It is a statistical method that considers both the results and the survival time, and can make full use of the incomplete information provided by the data to describe the distribution characteristics of the survival time and analyze the main factors affecting the survival time. The main difference between survival analysis and other multivariate analyses is that survival analysis takes into account the length of time it takes for each observation to have a certain outcome.

[0084] In survival analysis and hazard analysis, the analysis strategy is closely related to the survival function S(t), hazard function h(t), cumulative hazard function H(t), etc. The survival function S(t) describes the probability that an event occurs at a time greater than time point t; the hazard function h(t) gives the instantaneous potential of the event per unit time, assuming that the individual has survived until time t. This function can be a constant risk or change over time, such as rising, falling, falling first and then rising, etc.; and the cumulative hazard function H(t) represents the cumulative number of events experienced before time t. These concepts together constitute the basic framework of survival analysis and hazard analysis, which are used to study the relationship between influencing factors and survival time and outcomes, and predict individual survival predictions at different factor levels.

[0085] Exemplarily, step S12 includes sub-steps S12-1 to S12-4, which are specifically described as follows:

[0086] S12-1. According to the data features represented by the target data of multiple vehicles, the target faults of multiple vehicles are obtained, wherein the target faults are faults related to the performance of the power battery. The target data is monitoring data associated with the driving time and / or mileage. If the data is abnormal, it means that the vehicle may have a fault. Based on the above analysis method, various types of vehicle faults can be obtained, and the faults associated with battery performance are target faults. For example, battery heat dissipation faults, charging faults, etc.

[0087] Specifically, the target faults of multiple vehicles can be obtained based on the following method, including the following steps:

[0088] The first step is to determine the corresponding feature data in the target data according to the fault items of the performance of the power battery associated with different component systems on each vehicle, where the component systems include the transmission system, brake system, electrical system, cooling system and battery system. In the survival analysis of driving time and mileage of big data, the faults involved refer to the faults of the main component systems of the vehicle, which will affect the normal operation, safety or performance of the vehicle. These component systems involve the engine, transmission system, brake system, electrical system, cooling system, suspension system, etc. of the vehicle. Engine faults include engine stall, ignition system fault, fuel system problem, etc.; transmission system faults include transmission fault, clutch fault, drive shaft problem, etc.; brake system faults include brake failure, brake fluid leakage, brake disc / pad wear, etc.; electrical system faults include battery failure, generator fault, line short circuit or open circuit, etc.; cooling system faults include radiator fault, water pump fault, coolant leakage, etc.; battery pack faults include battery temperature, voltage abnormality, pressure difference abnormality, etc.

[0089] It should be noted that scrapes, minor scratches or damage to the paint of the vehicle are not considered "faults" that affect the performance of the power battery, but fall into the category of appearance or body damage. In survival analysis, this type of damage is usually not taken into account unless it is directly related to the failure of the main system or component of the vehicle. Target faults can be set for different vehicles. For example, for hybrid vehicles, the engine will drive the generator to charge the power battery. At this time, if the engine fails, it will affect the performance of the power battery, and the engine failure will be included in the category of target failure; if the vehicle is a pure electric vehicle and the power source is only the power battery, there is no relevant data on engine failure.

[0090] In the second step, the characteristic data of all fault items of each vehicle are counted to obtain the target faults of multiple vehicles. Each vehicle has a frame number (vin), total mileage (total_mile), total running time (minutes) (total_run_time), total charging time (minutes) (total_charge_time), vehicle use (vehicleuse), mileage of the trip (kilometers) (run_mile), driving time of the trip (minutes) (run_time), maximum speed (km / h) (max_speed), maximum speed (r / min) (max_rpm), average torque / torque (N*m) (average_rpm), average battery temperature (average_battery_temperature), average motor temperature (average_driver_motor_temperature), fault level (alarm_level), fault name (alarm_name), fault occurrence province or municipality (alarm_pro), fault occurrence city (alarm_city), fault occurrence time (alarm_time), fault occurrence date (date_time), fault occurrence hour (alarm_hour), SOC (State of Charge), etc.

[0091] The key derived feature quantities involved in survival analysis include: the number of vehicles (n_vehicles), the time from driving to failure (days) (failure_times), the observation cut-off time (days) (censor_times), the actual observed time (observed_times), whether the event occurred (1 for failure, 0 for no failure but the observation ended) (event_observed), the mileage at the time of failure (mileage_at_event), etc.

[0092] The number of vehicles (n_vehicles) can be obtained from the basic information table (vin_base_info). The total number of vehicles used for survival analysis can be obtained from the vehicle frame number (vin) records with a total mileage (total_mile) greater than 50km. The time (days) until failure (failure_times), the final mileage of the vehicle, and the total mileage before the first failure can be obtained from the cumulative mileage (mileage_at_event) at the end of the failure in the fault information statistics table (alarm_base).

[0093] The observation cut-off time (days) (censor_times) can be obtained from the data date (data_time) field in the vehicle basic information table (vin_base_info) according to the vehicle frame number (vin). The actual observed time (observed_times) is calculated based on the three feature quantities of the vehicle frame number, the time from driving to failure (days) (failure_times), and the observation cut-off time (days) (censor_times).

[0094] Whether the event occurs (event_observed) is based on whether the vehicle fault event occurs: 1 is a fault; 0 is no fault but the observation is over. The fault classification (alarm_class) can be obtained from the fault information statistics table (alarm_base) according to the vehicle's frame number (vin) and the earliest fault occurrence date (date_time). Specifically including 1: engine failure; 2: transmission system failure; 3: brake system failure; 4: electrical system failure; 5: cooling system failure; 6: battery pack failure; 7: other failures. The cumulative mileage at the time of failure (mileage_at_event), that is, the total mileage accumulated by the vehicle when the first fault occurs. It can be obtained from the fault information statistics table (alarm_base) according to the vehicle's frame number (vin) and the earliest fault occurrence date (date_time).

[0095] So far, the characteristic data of all fault items of each vehicle have been counted, and the target faults of multiple vehicles have been obtained.

[0096] S12-2. According to the target failures of multiple vehicles and the preset survival function, determine the survival probability of the power battery under different use cycles. The survival probability represents the probability that a patient or event survives longer than time t, and is represented by S(t): S(t) = P(T≥t), such as the 5-year survival rate: S(5) = P(T>5). The curve drawn with time t as the horizontal axis and S(t) as the vertical axis is called the survival rate curve. It is a descending curve. The steeper the slope, the lower the survival rate or the shorter the survival time. Its slope represents the death rate.

[0097] Specifically, the survival analysis algorithm is used to perform survival analysis. The relevant data of the target fault can be imported into the Python third-party software package lifelines library. In the lifelines library, the fit() method of the KaplanMeierFitter class is used to calculate the Kaplan-Meier survival function estimate. The Kaplan-Meier method is a non-parametric statistical method used to estimate the probability of survival (or non-occurrence of an event) from the time of the event until a specific time point. The event here in survival analysis usually refers to the occurrence of the result of interest, such as a vehicle failure.

[0098] Initialize the Kaplan-Meier estimator. KaplanMeierFitter is a class in the lifelines library that is used to perform Kaplan-Meier survival analysis. This class can help estimate the survival curve and compare the survival differences between different groups. The Kaplan-Meier survival function is a non-parametric method used to estimate the proportion of individuals who survive at different time points. Specifically include:

[0099] Step 1: Define and initialize variables, including time and event_observed. The variable time is an array or list of the survival time of each observed object. The variable event_observed is a Boolean array or list, indicating whether the event is observed (usually True) or not observed (usually False, possibly due to right censoring, etc.) at each time point.

[0100] Step 2: Sorting and grouping. The fit() method will sort the observations according to the time array. The observations are divided into different risk sets (also called risk groups) according to the sorted order. Each risk set contains individuals that are still under observation at the current time point.

[0101] Step 3: Calculate the survival probability. For each time point t (starting from the smallest time point), calculate the proportion of individuals who did not experience the event at all time points before that time point, that is, the survival probability.

[0102] If an event is observed at time point t (i.e. event_observed[i] == True), the number of individuals in the current risk set is used to calculate the decrease in the survival probability. Specifically, the calculation formula for the survival probability S(t) is:

[0103] S(t)=S(t-Δt)×(1-dt / nt)

[0104] Among them, dt is the number of individuals who have experienced the event at time point t, nt is the number of individuals who have not experienced the event before (including) time point t (that is, the size of the current risk set), and S(t-Δt) is the survival probability at the previous time point (for the first time point, it is usually assumed to be 1).

[0105] Step 4: Handle right censoring. If no event is observed at time point t (i.e. event_observed[i] == False), the survival probability remains unchanged because there is no information indicating that the survival probability will decrease after point t. This usually occurs when the individual is still alive at the end of data collection, or when the individual loses follow-up (lost to follow-up).

[0106] Step 5: Output. After the above steps, the fit() method will calculate the survival probability of a series of time points, which will be used for subsequent survival curve drawing.

[0107] S12-3. Obtain the survival curve of the power battery based on the corresponding change relationship between the service life of the power battery and the survival probability. The fit() method of KaplanMeierFitter uses the Kaplan-Meier method to calculate the survival probability at each time point by combining time data and event observation status, thereby obtaining a non-parametric estimate of the survival function. These calculations are crucial to understanding the time and event patterns in survival data.

[0108] Survival curve: A graph that connects the survival rates corresponding to each time point, with the observation (follow-up) time as the horizontal axis and the survival rate as the vertical axis. The survival curve is a descending curve, and the height and the slope of the decline should be noted during analysis. A flat survival curve indicates a high survival rate or a longer survival period, while a steep survival curve indicates a low survival rate or a shorter survival period.

[0109] Survival function S(t): P(T≥t), which represents the probability that an individual’s survival time is greater than t.

[0110]

[0111] Please continue reading Figure 4 The curve in the figure shows the change in the survival probability of the power battery as the vehicle's driving time goes by. In the early stages of vehicle operation (the front part of the time axis), the survival probability is high, close to 1.0, indicating that the vehicle is running well at this stage and is unlikely to fail. However, as time goes by (the horizontal axis moves to the right), from 0 days to 6000 days, the survival probability gradually decreases, indicating that the possibility of power battery failure or failure increases; at the same time, the slope of the curve reflects the rate of vehicle failure. The larger the slope, the faster the failure rate.

[0112] The upper boundary (the highest point on the vertical axis, i.e., the position where the survival probability is 1.0) represents the ideal state where the vehicle has not experienced any failures or any situations that render the vehicle inoperable at any given time point. In short, if the survival probability is 1.0, it means the power battery is completely alive or not yet phased out. The lower boundary (the lowest point on the vertical axis, i.e., the position where the survival probability is 0.0) represents the state where the vehicle has experienced a failure or is no longer available before a given time point. A survival probability of 0.0 means the power battery has failed or been phased out.

[0113] The survival curve is of great significance for understanding the reliability and service life of the power battery. It can help us predict the likelihood of the power battery failing at a certain future time point and make corresponding decisions based on this.

[0114] S12-4. Obtain the hazard curve of the power battery based on the survival curve and a preset hazard function. The hazard function h(t) is an important concept in hazard analysis, used to describe the risk of a person or object dying or being damaged at a specific moment t. The value of this function represents the probability of an event occurring per unit time at time point t.

[0115] The hazard function h(t) characterizes the instantaneous probability of an object dying immediately after surviving for a certain number of years.

[0116]

[0117] Please refer to Figure 5 , h1(t) is an upward curve, indicating that the hazard rate increases over time, such as in the case of a severe failure that cannot be repaired, where the hazard rate shows an increasing trend over time; the curve of h2(t) shows a downward trend, indicating that the hazard rate gradually decreases, such as in the case of an injury caused by an accident, where the risk of death gradually decreases after repair; h3(t) is a stable hazard rate function, such as for some vehicles after repair during the stable period, where the hazard rate remains basically unchanged.

[0118] The hazard function h(t) describes the instantaneous probability or rate of an event (such as death, failure, etc.) occurring in the next small time period from t to t+Δt given that the object has survived until time t. Specifically, h(t) is the conditional probability of an individual experiencing an event in the next unit time at time t.

[0119] The hazard function h(t) is mathematically defined as: h(t) = limΔt→0P(t≤T<t+Δt∣T≥t) / Δt, where T is the time of the event occurrence (such as the time of death), and P(t≤T<t+Δt∣T≥t) represents the probability of an event occurring in the time period from t to t+Δt given that the individual has survived until time t.

[0120] The survival function S(t) represents the probability that an individual survives to time t or longer, S(t) = P(T ≥ t). The calculation method has been described above and will not be further described here. The cumulative hazard function is the integral of the hazard function from time 0 to t, which represents the cumulative risk of events occurring before t: H(t) = ∫0th(s)ds. Its relationship with the survival function S(t) is: S(t) = eH(t).

[0121] Therefore, the hazard function can be derived from the survival function by taking the natural logarithm of both sides of S(t) = eH(t) to obtain lnS(t) = -H(t), taking the derivative of both sides with respect to t, and using the chain rule and the definition of H(t) (i.e., H(t) is the integral of h(t)) to obtain dlnS(t) / dt = S′(t) / S(t) = -h(t)

[0122] Since S′(t) is the derivative of S(t), and S(t) is a decreasing function of t (i.e., the probability of survival decreases as time increases), S′(t) < 0. Therefore, h(t) can be solved as: h(t) = -S′(t) / S(t), which means that in order to calculate h(t), it is necessary to know the specific form of S(t) and find its derivative S′(t).

[0123] Therefore, the ordinate value of the hazard function h(t) is calculated by the survival function S(t) and its derivative S′(t). In practical applications, these data usually come from experimental observations or historical data and may be estimated by statistical models (such as parametric models, nonparametric models, or semiparametric models).

[0124] Similarly, the data can be imported into the Python third-party package lifelines library to obtain the hazard curve. NelsonAalenFitter is a class in the lifelines library that is used to perform Nelson-Aalen estimation. The Nelson-Aalen estimator is a nonparametric method for estimating the cumulative hazard function. Unlike the Kaplan-Meier estimator (used to estimate the survival function), the Nelson-Aalen estimator estimates the cumulative hazard function, which describes the cumulative amount of risk of an event occurring before a given time point.

[0125] Core code: naf.fit(data['observed_time'], event_observed = data['event'], label = 'Predicted Vehicles(Nelson-Aalen)'), where data is a DataFrame containing 'observed_time' and 'event' columns of all vehicles based on the big data of solid-state battery passenger car demonstration operation obtained through data processing; observed_time is the time when the individual vehicle is observed or tracked; event_observed is a Boolean or binary value indicating whether an event of interest (e.g., death, failure, etc.) occurred at that time, 1 or True indicates that the event occurred, 0 or False indicates that the individual is still alive at the end of the observation or the event has not occurred, and the binary value (0, 1) is used here.

[0126] See also Figure 6 , Figure 6 This is a schematic diagram of the risk curve. The curve in the middle of the figure represents the cumulative risk predicted by the Nelson-Aalen method, which increases over time (X axis, number of days). The shadows above and below the curve represent the actual cumulative risk, which may show different trends due to fluctuations in actual data.

[0127] The shaded upper boundary represents the maximum value that the cumulative hazard may reach at a given point in time. This value is an upper limit based on actual observations or statistical model predictions, taking into account variability, uncertainty, and possible extreme cases in the data. In practical applications, the upper boundary provides an estimate of the worst case scenario, indicating that the cumulative hazard is unlikely to exceed this level before a certain point in time, although this does not mean that it is completely impossible. The shaded lower boundary usually represents the starting value of the cumulative hazard, which is 0. In survival analysis, this means that at the time point when the observation begins (such as the date of purchase of a new car, when the equipment is put into use, etc.), there is no risk accumulation of the event. From a technical point of view, the lower boundary is not always strictly 0, especially in some complex statistical processes, some prior information or basic risk levels may be taken into account. However, in most cases, in order to simplify understanding and analysis, the cumulative hazard can be considered to start from 0. The existence of the lower boundary is mainly to contrast with the upper boundary and jointly define the possible range of changes in the cumulative hazard.

[0128] The uncertainty of the risk is expressed by the shaded upper and lower boundaries, which together form a confidence interval that reflects the uncertainty of the cumulative risk estimate. The wider the interval, the greater the uncertainty of the estimate. By understanding the upper and lower boundaries of the cumulative risk, decision makers can better assess the risk level and formulate corresponding risk mitigation measures or resource allocation strategies.

[0129] So far, survival analysis and hazard analysis have been performed on the target data to obtain the survival curve and hazard curve of the power battery respectively.

[0130] S13. Performing index item analysis on the survival curve and the danger curve respectively according to the performance evaluation period of the power battery to obtain the survival index and danger index of the power battery.

[0131] Specifically, the survival index and the danger index can be set separately based on the analysis requirements. For example, the index can be set by the slopes represented by the survival curve and the danger curve respectively. The slope of the curve represents the changing characteristics of the life of the power battery as the use time increases. Taking the slope of the danger curve as an example, the larger the slope, the higher the cumulative risk; if the slope is small, the lower the cumulative risk. It can be understood that both the survival curve and the danger curve are characterized as a whole based on the changing characteristics of the data, but the performance evaluation results of the power battery in the performance evaluation period can be concretized through the survival index and the danger index. The performance evaluation period can be set freely, and it can be a future period that needs to be predicted or a historical period that needs to be analyzed.

[0132] Exemplarily, step S13 includes sub-steps S13-1 to S13-4, which are specifically described as follows:

[0133] S13-1. Determine the median survival time, average survival time and risk ratio of the power battery under the preset comparison items during the performance evaluation period based on the survival curve. The preset comparison items include mileage comparison items and / or fault type comparison items. The median survival time is the time point when the survival probability (or cumulative risk) drops to the threshold. The threshold can be set to 0.5. The median survival time can be obtained based on the survival threshold to determine the time point corresponding to the survival probability. The average survival time is the average of the survival times of the power batteries on all vehicles. The average survival time can be obtained directly by calculating the average.

[0134] The hazard ratio is used to compare the hazard ratio between two or more groups in the Cox proportional hazard model. It can be obtained by comparing the mileage range as a comparison item, or by grouping and comparing based on the fault type. Taking the mileage range (km) as an example, the grouping results are shown in Table 1 below.

[0135] Table 1

[0136] Starting mileage Termination Mileage Group Number 0 10000 1 10001 20000 2 20001 30000 3 30001 40000 4 40001 50000 5 50001 60000 6 … … …

[0137] Table 1 includes multiple mileage intervals, which are estimated by the Cox model. The Cox proportional hazard model is a statistical method in survival analysis that is used to study how one or more predictor variables (also called covariates or risk factors) affect the time of an event. Import the Python third-party package lifelines library. CoxPHFitter() is a class in the lifelines Python library, which is used to fit the Cox proportional hazard model (CoxProportional Hazards Model). Core code: from lifelines import CoxPHFitter

[0138] cph=CoxPHFitter() / / Initialize CoxPHFitter()

[0139] cph.fit(data,duration_col='observed_time',event_col='event',formula="group_num") / / Use CoxPHFitter to fit the model

[0140] The group number group_num is a grouping based on the mileage interval. In addition, more groups can be made to calculate the risk ratio according to actual needs, so as to further locate the specific fault category based on the influencing factors of the time of occurrence of a certain event.

[0141] In the embodiment of the present invention, the risk ratio of the grouping is calculated based on the driving mileage interval of 10000 km. Figure 7 , Figure 7 The calculated hazard ratio is shown in the middle box. cph.summary is a DataFrame returned after fitting the Cox Proportional Hazards Model in the lifelines library, which contains the results of model parameter estimation and statistical test information.

[0142] The following is an analysis of this summary (or output page):

[0143] coef is the coefficient estimate of group_num, which indicates the average change in the log hazard ratio when group_num increases by one unit (from group 1 to group 2, or from group 2 to group 3, etc.). In this example, coef is 0.065302, which means that when group_num increases from 0 (group 1) to 10000km (group 2), the log hazard ratio increases by 0.065302 on average.

[0144] exp(coef) is an estimate of the hazard ratio (HR). It indicates the relative risk of one group to another. In this example, exp(coef) is 1.065302, which means that the risk of group 2 relative to group 1 is 1.065302 times that of group 1.

[0145] se(coef) is the standard error of the coefficient estimate, which indicates the uncertainty of the coefficient estimate. In this example, the standard error of the coefficient estimate is 0.772495.

[0146] coef lower 95% and coef upper 95% are 95% confidence intervals for the coefficient estimate. These two values ​​are the lower and upper limits of the 95% confidence interval (CI) for coef. In this example, the 95% CI for coef is (-1.450805, 1.577321). Since this interval includes 0, it is not certain that group_num has a significant effect on risk.

[0147] exp(coef)lower 95% and exp(coef)upper 95% are the 95% confidence intervals for the hazard ratio, the lower and upper limits of the 95% CI for the hazard ratio (0.234381, 4.841969).

[0148] cmp to indicates which base level this covariate is compared to. Here, it is compared to 1 (i.e., group 1).

[0149] z is a statistic associated with the normal distribution (or standard normal distribution) that describes the number of standard deviations between an observation or sample mean and a theoretical value (such as the population mean), in this case 0.081888.

[0150] p is whether the observed effect is likely to be due to random error alone. If the p-value is less than a certain significance level (such as 0.05), the null hypothesis (that the covariate has a significant relationship with risk) is rejected. In this example, the p-value is 0.934736, which is greater than 0.05, so the null hypothesis cannot be rejected, that is, there is no significant relationship between group_num and risk (but note that this may be due to small data or the effect does not really exist).

[0151] log2(p) is the negative logarithm of the p-value and is sometimes used to quantify the strength of evidence.

[0152] S13-2. Determine the cumulative risk value and / or curve slope of the power battery during the performance evaluation period based on the hazard curve. The cumulative risk value is used to describe the cumulative risk of a person or object being damaged or killed before a certain moment. The cumulative risk is the integral of the risk over time, reflecting the cumulative risk from the starting point to the current time point t. The slope of the curve characterizes the changing characteristics of the risk accumulation.

[0153] S13-3. Characterize the survival index of the power battery according to the median survival time, average survival time and the risk ratio under the preset comparison item. A corresponding label can be set for each survival index, and the corresponding median survival time, average survival time and risk ratio can be filled in under the corresponding label to facilitate staff to read the data.

[0154] S13-4. Characterize the danger index of the power battery according to the cumulative risk value and / or the curve slope. Similarly, a label can be created through a template for characterizing the danger index, and the corresponding cumulative risk value and / or curve slope can be filled in under the label.

[0155] So far, the survival indicators and danger indicators of power batteries have been obtained.

[0156] S14. Determine the performance of the power battery during the performance evaluation period based on the survival index and the danger index.

[0157] Specifically, the performance of the power battery during the performance evaluation period can be directly reflected through the survival index and the danger index. For example, after the power battery is used for 1000 days, its survival probability is less than 0.4, which means that it is very easy to fail, there is a great risk in its operation, and the performance is poor. The power battery can be scored in the implementation of the performance. For example, the score range is 1-10. When the score is 8-10, it means that the performance of the power battery during the performance evaluation period is good; conversely, when the score is less than 5, it means that the performance of the power battery during the performance evaluation period is poor.

[0158] By evaluating the performance of power batteries, vehicle operations can be optimized. For example, when the performance evaluation period is the operation management period, the management strategy for the operating vehicles in the subsequent management period is output. The maintenance time and replacement time of the power batteries on the operating vehicles are determined according to the management strategy. The management strategy can guide the maintenance and replacement time of the power batteries, reduce operational risks and costs, and provide a scientific basis for decision makers. Of course, based on the evaluated performance, the advantages and disadvantages of power battery technology can also be quantified to provide a decision-making basis for product development and market promotion.

[0159] Based on the same technical concept as the evaluation method, the embodiment of the present invention also provides a vehicle battery performance evaluation device, see Figure 8 , Figure 8The diagram is a schematic diagram of the structure of the evaluation device. The evaluation device includes:

[0160] The acquisition module 801 is used to acquire target data of multiple vehicles on the big data platform, wherein the target data is data on the usage performance of the power battery of the vehicle during the vehicle's execution of historical driving tasks;

[0161] A first obtaining module 802 is used to perform survival analysis and hazard analysis on target data of multiple vehicles to obtain a survival curve and a hazard curve of the power battery;

[0162] A second obtaining module 803 is used to analyze the index items of the survival curve and the danger curve respectively according to the performance evaluation period of the power battery to obtain the survival index and the danger index of the power battery;

[0163] The first determination module 804 is used to determine the use performance of the power battery during the performance evaluation period according to the survival index and the danger index.

[0164] In an optional embodiment, the acquisition module includes:

[0165] The first acquisition submodule is used to extract the data of historical driving tasks on the big data platform according to a preset statistical table template to obtain a data statistical table for each vehicle;

[0166] A second obtaining submodule is used to obtain first data of the driving time of each vehicle changing with the use performance of the power battery and second data of the driving mileage of each vehicle changing with the use performance of the power battery according to the data represented by the data statistics table of each vehicle;

[0167] The third obtaining submodule is used to obtain target data of each vehicle according to the first data and the second data.

[0168] In an optional embodiment, the third obtaining submodule includes:

[0169] A first preprocessing unit, used for performing data cleaning on the first data and the second data of each vehicle to remove duplicate data and invalid data;

[0170] A second preprocessing unit is used to fill missing values ​​in the first data and the second data after data cleaning to obtain the first data and the second data after data filling;

[0171] The third preprocessing unit is used to normalize the first data and the second data after the data filling to obtain the target data of the corresponding vehicle.

[0172] In an optional embodiment, the first obtaining module includes:

[0173] A fourth obtaining submodule is used to obtain target faults of the plurality of vehicles according to data features represented by target data of the plurality of vehicles, wherein the target fault is a fault related to the performance of the power battery;

[0174] A first determination submodule is used to determine the survival probability of the power battery under different use cycles according to the target faults of multiple vehicles and a preset survival function;

[0175] A fifth obtaining submodule is used to obtain a survival curve of the power battery according to the corresponding change relationship between the service life of the power battery and the survival probability;

[0176] The sixth obtaining submodule is used to obtain the danger curve of the power battery according to the survival curve and a preset danger function.

[0177] In an optional embodiment, the fourth obtaining submodule includes:

[0178] A determination unit, configured to determine corresponding characteristic data in the target data according to the fault items of the performance of the power battery associated with different component systems on each vehicle, wherein the component systems include a transmission system, a braking system, an electrical system, a cooling system, and a battery system;

[0179] The obtaining unit is used to collect statistics on the characteristic data of all fault items of each vehicle to obtain target faults of multiple vehicles.

[0180] In an optional embodiment, the second obtaining module includes:

[0181] A second determination submodule is used to determine the median survival time, the average survival time and the risk ratio of the power battery under the preset comparison items during the performance evaluation period according to the survival curve, wherein the preset comparison items include the mileage comparison item and\or the fault type comparison item;

[0182] A third determination submodule is used to determine the cumulative risk value and / or the curve slope of the power battery during the performance evaluation period according to the risk curve;

[0183] The first characterization submodule is used to characterize the survival index of the power battery according to the median survival time, the average survival time and the risk ratio under the preset comparison item;

[0184] The second characterization submodule is used to characterize the danger index of the power battery according to the cumulative risk value and / or the curve slope.

[0185] In an optional embodiment, the evaluation device further includes:

[0186] An output module, used for outputting the management strategy of the operating vehicle in the subsequent management period when the performance evaluation period is the operation management period;

[0187] The second determination module is used to determine the maintenance time and replacement time of the power battery on the operating vehicle according to the management strategy.

[0188] Based on the same technical concept as the evaluation method, an embodiment of the present invention also provides an electronic device, including a processor and a memory, wherein the memory is coupled to the processor and stores instructions, and when the instructions are executed by the processor, the electronic device executes the steps of any one of the evaluation methods.

[0189] Based on the same technical concept as the evaluation method, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of any one of the evaluation methods are implemented.

[0190] The technical solution provided in the embodiments of the present invention has at least the following technical effects or advantages:

[0191] By acquiring the target data of multiple vehicles on the big data platform, since the target data is the data of the performance changes of the vehicle's power battery during the vehicle's historical driving mission, survival analysis and hazard analysis can be performed on the target data of multiple vehicles to obtain the survival curve and hazard curve of the power battery; according to the performance evaluation period of the power battery, the survival curve and hazard curve are analyzed for indicators to obtain the survival index and hazard index of the power battery; according to the survival index and hazard index, the performance of the power battery in the performance evaluation period is determined. This technical solution implements survival analysis and hazard analysis based on big data related to the performance of the power battery, accurately counts the performance change characteristics of the power battery during use, realizes a scientific and comprehensive evaluation of the performance of the power battery, and improves the accuracy of the evaluation of the performance of the power battery.

[0192] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0193] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (modules, systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0194] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0195] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0196] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0197] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for evaluating the performance of a vehicle battery, characterized in that: The method comprises: Acquire target data of multiple vehicles on a big data platform, wherein the target data is data on changes in the performance of the power batteries of the vehicles during the execution of historical driving tasks; Performing survival analysis and hazard analysis on the target data of the plurality of vehicles respectively to obtain a survival curve and a hazard curve of the power battery; performing index item analysis on the survival curve and the danger curve respectively according to the performance evaluation period of the power battery to obtain a survival index and a danger index of the power battery; The use performance of the power battery in the performance evaluation period is determined according to the survival index and the danger index.

2. The method for evaluating the performance of a vehicle battery according to claim 1, characterized in that: The step of obtaining target data of multiple vehicles on the big data platform includes: Extracting the data of the historical driving tasks on the big data platform according to a preset statistical table template to obtain a data statistical table for each vehicle; According to the data represented by the data statistics table of each vehicle, first data indicating that the driving time of each vehicle varies with the use performance of the power battery, and second data indicating that the driving mileage of each vehicle varies with the use performance of the power battery are obtained; The target data of each vehicle is obtained according to the first data and the second data.

3. The method for evaluating the performance of a vehicle battery according to claim 2, characterized in that: The step of obtaining the target data of each vehicle according to the first data and the second data includes: Cleaning the first data and the second data of each vehicle to remove duplicate data and invalid data; Perform missing value filling on the first data and the second data after data cleaning to obtain the first data and the second data after data filling; The first data and the second data after the data filling are normalized to obtain the target data of the corresponding vehicle.

4. The method for evaluating the performance of a vehicle battery according to claim 1, characterized in that: The performing survival analysis and hazard analysis on the target data of the plurality of vehicles respectively to obtain a survival curve and a hazard curve of the power battery includes: Obtaining target faults of the multiple vehicles according to data features represented by the target data of the multiple vehicles, wherein the target faults are faults associated with the performance of the power battery; Determining the survival probability of the power battery under different usage cycles according to the target failures of the multiple vehicles and a preset survival function; Obtaining a survival curve of the power battery according to a corresponding change relationship between the service life of the power battery and the survival probability; A danger curve of the power battery is obtained according to the survival curve and a preset danger function.

5. The method for evaluating the performance of a vehicle battery according to claim 4, characterized in that: The step of obtaining target faults of the plurality of vehicles according to data features represented by target data of the plurality of vehicles comprises: Determining corresponding characteristic data in the target data according to the fault items of the performance of the power battery associated with different component systems on each vehicle, wherein the component systems include a transmission system, a braking system, an electrical system, a cooling system, and a battery system; The characteristic data of all fault items of each vehicle are counted to obtain target faults of the multiple vehicles.

6. The method for evaluating the performance of a vehicle battery according to claim 1, characterized in that: The performing index item analysis on the survival curve and the danger curve respectively according to the performance evaluation period of the power battery to obtain the survival index and the danger index of the power battery includes: Determine, according to the survival curve, the median survival time, the average survival time, and the risk ratio of the power battery under the preset comparison items during the performance evaluation period, wherein the preset comparison items include a mileage comparison item and / or a fault type comparison item; Determining, according to the risk curve, a cumulative risk value and / or a curve slope of the power battery during the performance evaluation period; Characterizing the survival index of the power battery according to the median survival time, the average survival time and the risk ratio under the preset comparison item; The danger index of the power battery is characterized according to the cumulative risk value and / or the slope of the curve.

7. The method for evaluating the performance of a vehicle battery according to claim 1, characterized in that: After determining the use performance of the power battery in the performance evaluation period according to the survival index and the danger index, the method further includes: When the performance evaluation period is an operation management period, outputting a management strategy for the operation vehicle in a subsequent management period; The maintenance time and replacement time of the power battery on the operating vehicle are determined according to the management strategy.

8. A vehicle battery performance evaluation device, characterized in that: The device comprises: An acquisition module, used to acquire target data of multiple vehicles on a big data platform, wherein the target data is data on changes in the performance of the power battery of the vehicle during the vehicle's execution of historical driving tasks; A first acquisition module is used to perform survival analysis and hazard analysis on the target data of the plurality of vehicles respectively, so as to obtain a survival curve and a hazard curve of the power battery; a second obtaining module, configured to perform index item analysis on the survival curve and the danger curve respectively according to a performance evaluation period of the power battery, so as to obtain a survival index and a danger index of the power battery; The first determination module is used to determine the use performance of the power battery in the performance evaluation period according to the survival index and the danger index.

9. An electronic device, characterized in that: The electronic device comprises a processor and a memory, wherein the memory is coupled to the processor and stores instructions, and when the instructions are executed by the processor, the electronic device executes the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 7 are implemented.

Images