A battery status evaluation method and device

Abstract

The present application provides a battery status assessment method and device. The method provided by the present application includes: obtaining the driving data of the vehicle and the battery data of the battery; using the driving data as the judgment standard, filtering the first associated data caused by the driving behavior from the battery data, using the battery maintenance data as the judgment standard, filtering the second associated data caused by the battery usage behavior from the battery data, and generating target associated data based on the first associated data and the second associated data; establishing a static analysis model, processing the target associated data of the battery based on the static analysis model, and obtaining the target battery risk factor; establishing a dynamic early warning model, and evaluating the real-time health status of the battery based on the dynamic early warning model according to the target battery risk factor and the real-time operation data of the battery. The battery status assessment method and device provided by the present application are used to perform accurate battery risk assessment in real time.

Description

Technical Field

[0001] The present application relates to the field of battery management technology, and in particular to a battery status evaluation method and device. Background Art

[0002] Lithium batteries have been widely used in electric vehicles, consumer electronics, energy storage systems and other fields. Their performance and safety directly affect the reliability of these applications and user experience. In order to ensure the safety and stability of lithium batteries during use, the battery management system is responsible for monitoring the battery's operating parameters such as voltage, current, and temperature. However, in actual applications, vehicles cannot use this data for fault warning and health management, making it difficult to trace the cause of vehicle battery failure, resulting in a high scrap / repair ratio, and vehicle drivers are unable to know the impact of their driving style on the vehicle battery.

[0003] Existing lithium battery failure warning algorithms mainly rely on models based on laboratory environments. These models can usually only warn based on obvious battery characteristics a short time before the failure occurs. For example, when the battery temperature or voltage exceeds the set safety range, the warning system will sound an alarm. These models usually ignore the complex factors in the actual use environment and cannot solve the capture and prevention of potential risks in the actual use process, resulting in limited application of the warning system. Summary of the invention

[0004] In view of this, the present application provides a battery status assessment method and device for performing accurate battery risk assessment in real time.

[0005] Specifically, the present application is implemented through the following technical solutions:

[0006] A first aspect of the present application provides a battery status assessment method, the method comprising:

[0007] Acquire driving data of the vehicle and battery data of the battery; use the driving data as a judgment standard to filter out first associated data caused by driving behavior from the battery data; use the battery maintenance data as a judgment standard to filter out second associated data caused by battery use behavior from the battery data; and generate target associated data based on the first associated data and the second associated data;

[0008] Establishing a static analysis model, and processing the target associated data of the battery based on the static analysis model to obtain a target battery risk factor; wherein the target battery risk factor is used to characterize the contribution of different battery characteristics to different abnormal events;

[0009] A dynamic early warning model is established, and based on the dynamic early warning model, the real-time health status of the battery is evaluated according to the target battery risk factor and the real-time operation data of the battery.

[0010] A second aspect of the present application provides a battery status evaluation device, the device comprising an acquisition module, a generation module and a processing module; wherein:

[0011] The acquisition module is used to acquire the driving data of the vehicle and the battery data of the battery;

[0012] The generating module is configured to use the driving data as a judgment standard to filter out first associated data caused by driving behavior from the battery data, use the battery maintenance data as a judgment standard to filter out second associated data caused by battery use behavior from the battery data, and generate target associated data based on the first associated data and the second associated data;

[0013] The processing module is used to establish a static analysis model, and process the target associated data of the battery based on the static analysis model to obtain a target battery risk factor; wherein the target battery risk factor is used to characterize the contribution of different battery characteristics to different abnormal events;

[0014] The processing module is used to establish a dynamic early warning model, and evaluate the real-time health status of the battery based on the dynamic early warning model according to the target battery risk factor and the real-time operation data of the battery.

[0015] The third aspect of the present application provides a battery status assessment device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any one of the methods provided in the first aspect of the present application are implemented.

[0016] The fourth aspect of the present application 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 provided in the first aspect of the present application.

[0017] The battery status evaluation method and device provided by the present application first obtain the driving data of the vehicle and the battery data of the battery; take the driving data as the judgment standard, filter the first associated data caused by the driving behavior from the battery data, take the battery maintenance data as the judgment standard, filter the second associated data caused by the battery use behavior from the battery data, and generate the target associated data based on the first associated data and the second associated data. In this way, by comparing the abnormalities in the driving process with the abnormalities in the battery use process, when the driving process and the battery use process are abnormal, the target associated data can be determined, which can ensure the accurate identification of the target associated data. Compared with the single judgment condition, which may cause misjudgment, the target associated data generated based on the first associated data and the second associated data ensures the accuracy of the target associated data. Then establish a static analysis model, process the target associated data of the battery based on the static analysis model, and obtain the target battery risk factor. Finally, establish a dynamic early warning model, and evaluate the real-time health status of the battery based on the dynamic early warning model according to the target battery risk factor and the real-time operation data of the battery. In this way, in addition to capturing the relevant data of the battery, the target associated data also captures the data related to the driving behavior during the driving process, so that the battery problem can be analyzed without being limited to the single data of the battery, ensuring the accuracy of the battery status evaluation.

[0018] Furthermore, by using a static analysis model to conduct an in-depth analysis of the target-related data and calculating the contribution corresponding to each battery feature, potential risk factors can be discovered. Furthermore, by using a dynamic early warning model to monitor vehicle and battery data in real time, it is possible to quickly respond to changes in the battery, and combine the target battery risk factor given by the static analysis model to quickly assess the battery's health status and failure risk. Furthermore, the target battery risk factor obtained by the static analysis model can characterize the weight values ​​of different abnormal events (characterized by contribution). In this way, the degree of impact of abnormal events on the vehicle battery is statically marked, and then the dynamic early warning model is used to detect the vehicle's actual operating data in real time, and the dynamic real-time analysis is completed in combination with the weight value. In this way, the health status of the vehicle battery can be detected in real time through static contribution and dynamic detection calculations. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flowchart of a first embodiment of a battery status evaluation method provided in the present application;

[0020] Figure 2 A flowchart of a second embodiment of a battery status assessment method provided in this application;

[0021] Figure 3 A schematic diagram of an implementation example 1 of the application process of the dynamic early warning model provided in this application;

[0022] Figure 4A flowchart of a second embodiment of a battery status assessment method provided in this application;

[0023] Figure 5 A flowchart of a third embodiment of the battery status assessment method provided in this application;

[0024] Figure 6 A flowchart of a fourth embodiment of the battery status assessment method provided in this application;

[0025] Figure 7 A hardware structure diagram of a battery status assessment device in which the battery status assessment device of the present application is located;

[0026] Figure 8 This is a schematic diagram of the structure of the first embodiment of the battery status assessment device provided in this application. DETAILED DESCRIPTION

[0027] Here, exemplary embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application.

[0028] The terms used in this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in this article refers to and includes any or all possible combinations of one or more associated listed items.

[0029] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0030] The present application provides a battery status assessment method and device for performing accurate battery risk assessment in real time.

[0031] Specific embodiments are given below to introduce the technical solution of the present application in detail.

[0032] Figure 1 This is a flowchart of the first embodiment of the battery status evaluation method provided by this application. Please refer to Figure 1 , the method provided in this embodiment may include:

[0033] S101. Acquire driving data of a vehicle and battery data of a battery; use the driving data as a judgment standard to filter out first associated data caused by driving behavior from the battery data; use the battery maintenance data as a judgment standard to filter out second associated data caused by battery usage behavior from the battery data; and generate target associated data based on the first associated data and the second associated data.

[0034] Specifically, the specific data included in the vehicle's driving data is determined according to actual needs, and this embodiment is not limited to this. For example, the vehicle's driving data may include information such as the vehicle's own speed, acceleration, and mileage. For another example, the vehicle's driving data may also include road condition information on which the vehicle is traveling, vehicle driving slope information, driver's operation instruction information, etc.

[0035] It should be noted that the irregular data in the driving data can be determined as the first associated data. In specific implementation, when a specific driving behavior that affects the battery health occurs, the corresponding associated data is determined as the first associated data. For example, in one embodiment, the first associated data includes high power consumption and high power feed caused by long uphill and downhill driving and sudden braking / acceleration, as well as long high and low temperatures under extreme working conditions.

[0036] It is understandable that since the use of batteries will lead to battery aging, when there is no abnormality in the use of the batteries, the driving data can also be determined as the first associated data. In specific implementation, the data of all vehicles during normal driving can be saved to obtain the first associated data. For example, in one embodiment, the driving acceleration, vehicle speed, driving habits, and battery daily average accumulated mileage of the vehicle are saved to obtain the first associated data.

[0037] Furthermore, the driving data of the vehicle can be obtained through an on-board diagnostic system, a vehicle sensor network, and a vehicle data reading platform. For example, in one embodiment, the driving data of the vehicle is obtained through a vehicle network carried by the vehicle.

[0038] Furthermore, the specific data included in the battery data of the vehicle is determined according to actual needs, and in this embodiment, there is no limitation on this. For example, the battery data of the vehicle may include the battery's charging current, charging voltage, charging temperature, charging SOC, battery rated capacity, and battery capacity decay rate. For another example, in another embodiment, the battery data of the vehicle may include timestamp information, battery fault status information, battery scrap status information, battery attribute information, battery direct observation data, battery indirect evaluation index, battery business attribute information, and battery maintenance status information.

[0039] It should be noted that the non-standard data in the battery data can be determined as the judgment standard, and the data that meets the judgment standard can be used as the second associated data. In other words, the non-standard data in the battery data is determined as the second associated data. For example, in one embodiment, the second associated data is used to characterize the anomalies caused during the use of the battery. When a specific battery use behavior that affects the health of the battery occurs, the corresponding associated data is determined as the second associated data. For example, in one embodiment, the second associated data includes extremely rapid battery temperature rise, excessive current, unstable voltage, excessive power feed, frequent charge and discharge cycles, etc.

[0040] It should be noted that the vehicle's battery data can be obtained through the battery management system and the battery box data at the same time. In specific implementation, the battery management system can obtain battery information, timestamp, battery temperature, battery voltage, battery current, preset fault code, and battery state of charge (State of Charge, referred to as SOC). The battery box data can obtain battery charging status, battery discharge status, battery charging mode, water cooling working status, and battery SOC, etc.

[0041] It should be noted that both battery data and driving data can be obtained from the data reported by the battery box. In some special vehicles, the battery data and driving data can also be determined through active reporting by the vehicle.

[0042] In specific implementation, in this embodiment, the associated data caused by the driving behavior and the battery use behavior are combined to obtain the target associated data.

[0043] S102, establishing a static analysis model, and processing the target associated data of the battery based on the static analysis model to obtain a target battery risk factor; wherein the target battery risk factor is used to characterize the contribution of different battery characteristics to different abnormal events.

[0044] Specifically, the static analysis model can process the target-related data, and determine the contribution of the battery feature to the battery anomaly based on the probability of the battery feature in the target-related data causing the abnormal event, and characterize it with the target battery risk factor. In other words, the static analysis model "scores" different battery features to determine the contribution of different battery features to the abnormal event. It can be understood that different battery features correspond to different target battery risk factors.

[0045] Further, the target associated data at least includes data related to the battery during use, the attribute data of the battery itself, the driving data generated by the vehicle during driving, the attribute data of the vehicle itself, and the environmental data corresponding to the vehicle and the battery during use (the environmental data may include data such as ambient temperature, ambient humidity, air sand content, ambient wind resistance, weather conditions, etc.), and determines the impact of the target associated data on the battery. It can be understood that since the battery characteristics correspond to the target associated data, the data dimensions contained in the battery characteristics are not less than the data dimensions contained in the battery characteristics. In specific implementation, all data that can be collected by the vehicle and the battery during use can be analyzed, all data can be determined as target associated data, and the impact of the target associated data on the battery can be analyzed. It should be noted that the static analysis model can quickly determine the contribution of each battery feature, and the contribution can be directly used when the dynamic early warning model performs subsequent calculations. The target battery risk factor calculated by the static analysis model conforms to the impact of different battery characteristics, which is more reasonable and accurate than the existing method of presetting the same weight for all battery features. At the same time, when processing the same battery feature, there is no need to calculate again, which can improve the efficiency of the operation. Furthermore, by analyzing the results of the dynamic early warning model combined with the static analysis model output, we can fully utilize the contribution of the battery characteristics, accurately obtain the impact of different battery characteristics on the vehicle battery, and obtain the accurate battery status.

[0046] Figure 2 This is a schematic diagram of the contribution of battery characteristics provided in this application. Figure 2 In this embodiment, the static analysis model forms weights for battery characteristics such as battery brand, average daily battery usage time, total battery consistency score, and average daily cumulative battery mileage.

[0047] Furthermore, the contribution of all battery characteristics calculated by the static analysis model can be output in different forms, which is not limited in this embodiment. Figure 2 ,The static analysis model presents the contribution of the battery characteristics in the form of a ,nomogram, and corresponds different data to the same scoring criterion based on segments of different ,lengths. In this way, the total score can be obtained by calculating the scores of the data of battery ,characteristics in different dimensions.

[0048] In the specific implementation, for example, in one embodiment, please continue to refer to Figure 2, for the static analysis model related to scrapping, the average daily battery usage time of 4 hours and the average daily cumulative battery mileage of 100 kilometers contribute the same amount to the risk of battery scrapping. For another example, in another embodiment, the brand also has a significant impact on the risk of battery scrapping, and brand I batteries are purchased instead of brand III. For another example, in another embodiment, when the battery consistency has always been good, and the average daily usage time is less than 14 hours, and the average daily cumulative battery mileage is less than 350 kilometers, the risk of battery scrapping within 5 years of warranty is almost zero.

[0049] Furthermore, according to different usage requirements, the battery condition can be estimated based on the total score.

[0050] It should be noted that static analysis models can have different outcome variables (the outcome variable is the dependent variable), and static analysis models can be divided into different action types according to different outcome variables. In specific implementation, when studying the scrapping of batteries, a static analysis model related to scrapping is formed based on the impact of target-related data on battery scrapping; when studying battery failure, a static analysis model related to failure is formed based on the impact of target-related data on battery failure.

[0051] A specific embodiment is given below to introduce in detail the process of obtaining the target battery risk factor using a static analysis model:

[0052] The static analysis model provided in this embodiment includes a data processing module, a contribution analysis module and a risk assessment module.

[0053] (1) The data processing module analyzes and evaluates the battery characteristics according to the battery outcome data, and determines the target battery risk factor of the battery characteristics for each of the abnormal events based on the contribution analysis module.

[0054] Specifically, the battery outcome data is a record of all events in the battery from manufacturing to retirement. When an event occurs in the battery, the corresponding event is recorded in the battery. For example, in one embodiment, the battery outcome data includes retirement records, maintenance records, return records, and failure records.

[0055] Furthermore, the data processing module is used to preprocess the target associated data of the input model and extract features. For example, in one embodiment, the data processing module performs data alignment on the target associated data, matches the first associated data with the second associated data, and then filters out abnormal features in the target associated data.

[0056] Furthermore, the contribution analysis module is responsible for analyzing the contribution of battery characteristics to different abnormal events. It should be noted that an abnormal event is an event corresponding to an abnormality of the vehicle. It is understandable that abnormal events may include abnormal battery temperature events, battery operation failure events, battery capacity failure events, battery charging failure events, and failure events of specified battery parts.

[0057] In specific implementation, for example, in one embodiment, abnormal events include battery failure and battery retirement. The contribution analysis module determines that the target battery risk factor of battery feature a for battery failure is 3%, and the target battery risk factor of battery retirement is 5%, the target battery risk factor of battery feature b for battery failure is 1%, and the target battery risk factor of battery retirement is 2%, and the target battery risk factor of battery feature c for battery failure is 6%, and the target battery risk factor of battery retirement is 9%.

[0058] It can be understood that the battery characteristics are input into the static analysis model, and after being analyzed and processed by the static analysis model, the target battery risk factor corresponding to the battery characteristics is output.

[0059] The battery status assessment method provided in this embodiment first analyzes and evaluates the battery characteristics based on the data processing module, and determines the contribution of the battery characteristics to each abnormal event based on the contribution analysis module, and then obtains the target battery risk factor. In this way, the battery characteristics are processed through the static analysis model, and the contribution calculation of different battery characteristics is completed. Since the static analysis model accurately analyzes the characteristics of different battery characteristics and their impact on abnormal events, an accurate and reliable target battery risk factor can be obtained through contribution calculation, thus ensuring the reliability of the risk estimate obtained through the target battery risk factor.

[0060] Specifically, the battery characteristics and abnormal events are determined, and the contribution of each battery characteristic under each abnormal event is calculated, wherein one battery characteristic corresponds to one or more abnormal events, and the comprehensive weight is determined based on the abnormal events. The comprehensive contribution of each abnormal characteristic under each abnormal event is calculated based on the comprehensive weight to obtain the target battery risk factor corresponding to the battery characteristic.

[0061] S103: Establish a dynamic early warning model, and evaluate the real-time health status of the battery based on the dynamic early warning model according to the target battery risk factor and the real-time operation data of the battery.

[0062] Specifically, the dynamic early warning model can combine the target battery risk factor based on the static analysis model to monitor and analyze the battery in real time and obtain the real-time health status.

[0063] It should be noted that for different dependent variables, corresponding to the static analysis model, different types of dynamic early warning models can be established. In specific implementation, combined with the above example, a dynamic early warning model related to scrapping is established for the static analysis model related to scrapping; a dynamic early warning model related to fault is established for the static analysis model related to fault.

[0064] Figure 3 This is a schematic diagram of the first embodiment of the application process of the dynamic early warning model provided by this application. Figure 3 ,For any battery at any time, the weight output by the static analysis model can be used for calculation.

[0065] Please continue to refer to Figure 3 , combined with the above example, the weights of each target-related data output by the static analysis model are represented by a nomogram, and the target-related data corresponding to the actual battery in use is input into the dynamic warning model. By comparing the nomogram, the current state of the battery is obtained. For example, in one embodiment, battery A is a brand II, the average daily usage time is 8 hours, the total consistency score is stable above 500, and the average daily mileage is 100 kilometers. The values ​​of these target-related data are substituted into the nomogram for calculation to obtain their respective scores. The total score of battery A shows that its survival rate in 5 years and 8 years is almost 100%, that is, the scrap rate ≈ 0%, which means that battery A is in good condition at present. Under the premise of maintaining the current working conditions, it can be used until the rated equivalent cycle number of its power battery is exhausted, and even after the end of the power battery life cycle, it can be transformed into a storage battery for continued use. For another example, in another embodiment, battery B is a brand III, the average daily usage time is 12 hours, the total consistency score is about 140, and the average daily cumulative mileage of the battery is 350 kilometers. Substituting the values ​​of these target-related data into the nomogram, we get the scores for each. The total score of battery B shows that its 5-year survival rate is about 88% and its 8-year survival rate is about 82%. The corresponding scrap rates are 12% and 18%. This means that the current state of battery B is poor. If it continues to maintain the current operating conditions, there is a 12% probability that problems will occur in the next 5 years that will lead to scrapping or the state of the components will be reduced to the scrap threshold. If the next 8 years are used as the observation window, the above probability is 18%.

[0066] The battery status assessment method provided in this embodiment combines the dynamic warning model with the static analysis model to obtain the ability to actuate risks, which can realize diversified applications of batteries. Furthermore, by monitoring the fluctuation of risks in the short term, we can issue an alarm when the risk suddenly rises to a threshold, and provide solution suggestions based on the risk factor score of the battery.

[0067] A specific embodiment is given below to introduce in detail the process of the dynamic early warning model evaluating the real-time health status of the battery according to the target battery risk factor and the real-time operation data of the battery:

[0068] (1) Predicting a target battery risk factor based on the static analysis model.

[0069] The specific implementation process of this step can refer to the explanation in step S102, which will not be repeated here.

[0070] (2) Determining an abnormal event threshold corresponding to each of the abnormal events based on the battery data of the battery.

[0071] Specifically, the abnormal event threshold corresponding to the abnormal event is used to determine the condition for the abnormal event to occur. For example, in one embodiment, when an abnormal event reaches the abnormal event threshold, it is determined that the abnormal event occurs.

[0072] It should be noted that the specific value of the abnormal event threshold is set according to the actual situation, and is not limited in this embodiment. For example, in one embodiment, the abnormal event threshold for the occurrence of an abnormal event can be predicted based on regression analysis. For another example, in another embodiment, the abnormal event threshold can be obtained by statistically analyzing the battery usage habits in real scenarios and the historical data of the battery obtained from the battery data.

[0073] (3) Acquire real-time associated data, use the target battery risk factor as a weight value, identify the real-time associated data based on the dynamic early warning model, and obtain a real-time risk value corresponding to each of the abnormal events.

[0074] Specifically, the real-time associated data may be obtained through a battery management system or a vehicle management system, or may be obtained through a dedicated data detection device, which is not limited in this embodiment.

[0075] It should be noted that the real-time associated data may include short-term excessive temperature rise, short-term excessive current, long-term voltage instability, battery management system output fault code, frequent charge and discharge cycles, excessive power supply, normal water cooling operation and high temperature, etc.

[0076] Furthermore, the real-time risk value indicates the risk level of the battery status at the current moment. The real-time risk value can be used to timely detect and warn of potential battery problems, so that appropriate measures can be taken.

[0077] When this step is implemented, the method for calculating the real-time risk value based on the target battery risk factor and the real-time associated data can be selected according to actual needs, and this is not limited in this embodiment. For example, in one embodiment, the target battery risk factor is used as a weight value, and the real-time associated data is weighted by the weight value to obtain the real-time risk value.

[0078] (4) Accumulating the real-time risk value corresponding to each abnormal event to obtain a cumulative risk value.

[0079] It should be noted that the real-time risk values ​​corresponding to the same abnormal event need to be superimposed. For example, in one embodiment, the abnormal event is battery retirement, and the factual risk values ​​corresponding to all battery retirements obtained by the dynamic early warning model are superimposed to obtain the cumulative risk value corresponding to battery retirement.

[0080] (5) When any of the accumulated risk values ​​exceeds the abnormal event threshold of the corresponding abnormal event, determining that the real-time health status is unhealthy.

[0081] In this step, the cumulative risk value corresponding to the same abnormal event is compared with the abnormal event threshold in real time. When the cumulative risk value is greater than the abnormal event threshold, the abnormal event is judged to have occurred, and the actual health status of the battery is determined to be unhealthy.

[0082] Optionally, when the cumulative risk value in any item exceeds the abnormal event threshold of the corresponding abnormal event, the step of determining that the real-time health status is unhealthy further includes:

[0083] Step 1: Evaluate the failure risk and scrap risk corresponding to the battery based on the real-time health status.

[0084] Specifically, the failure risk is the possibility of battery failure, and the scrap risk is the possibility of battery scrapping. After determining that the real-time health status of the battery is unhealthy, the failure risk of the battery can be calculated using a pre-trained failure risk model; the scrap risk of the battery can be calculated using a pre-trained scrap risk model.

[0085] Step 2: Based on the failure risk and the scrap risk, dynamic early warning reminders and static insight attribution are performed on the battery.

[0086] Specifically, the battery status can be analyzed by the failure risk and the scrap risk, and a dynamic early warning reminder of the battery can be immediately issued after the failure risk and the scrap risk are obtained. It should be noted that static insight attribution is used to determine the cause of the unhealthy real-time health status. Static insight attribution can be performed immediately after the failure risk and the scrap risk are obtained, or at a specified time after the failure risk and the scrap risk are obtained. This is not limited in this embodiment.

[0087] The method provided in this embodiment ensures dynamic early warning reminders and static insight attribution through a failure risk model and a scrap risk model, which can ensure the rational application of batteries and improve the user experience.

[0088] The battery status assessment method provided in this embodiment first predicts the target battery risk factor according to the static analysis model, then determines the abnormal event threshold corresponding to each abnormal event based on the battery data of the battery, and then obtains real-time associated data, takes the target battery risk factor as the weight value, identifies the real-time associated data based on the dynamic early warning model, obtains the real-time risk value corresponding to each abnormal event, and then accumulates the real-time risk value corresponding to each abnormal event to obtain the cumulative risk value, and finally determines that the real-time health state is unhealthy when any cumulative risk value exceeds the abnormal event threshold of the corresponding abnormal event. In this way, the target battery risk factor and the real-time associated data obtained by real-time detection are combined through the dynamic early warning model to jointly calculate the cumulative risk value of the battery. In this way, the target battery risk factor output by the static analysis model can accurately reflect the contribution value of different associated data, so the health state of the battery can be continuously measured through the real-time associated data. Further, since the dynamic early warning model is combined with the accurate target battery risk factor output by the static analysis model, the accuracy and reliability of the battery status judgment can be guaranteed.

[0089] The battery status assessment method provided in this embodiment not only captures battery-related data through target-related data, but also captures data related to driving behavior during driving. It is not limited to analyzing battery problems with single battery data, thus ensuring the accuracy of battery status assessment. Furthermore, by conducting an in-depth analysis of the target-related data through a static analysis model and calculating the contribution corresponding to each battery feature, potential risk factors can be discovered. Furthermore, by real-time monitoring of vehicle and battery data through a dynamic early warning model, it is possible to quickly respond to changes in the battery, and quickly assess the health status and failure risk of the battery in combination with the target battery risk factor given by the static analysis model.

[0090] Optionally, based on the above embodiment, the step of establishing a static analysis model, processing the target associated data of the battery based on the static analysis model, and obtaining the target battery risk factor further includes:

[0091] (1) Compare the target battery risk factors obtained by the static analysis model in adjacent cycles.

[0092] Specifically, the specific duration of each cycle is determined according to actual needs, and is not limited in this embodiment. For example, in one embodiment, each cycle is set to one hour, and at this time, the target battery risk factor output by the static analysis model is compared every hour.

[0093] (2) Identify the mutation time of the target battery risk factor based on the comparison results.

[0094] In specific implementation, when the comparison result value is large, the time at this time is determined as the target battery risk factor mutation time. In other words, when the target battery risk factor values ​​obtained in the two adjacent cycles differ greatly, the moment of the latter cycle is determined as the target battery risk factor mutation time.

[0095] It should be noted that when a target battery risk factor mutates, it indicates that the battery's health status may change significantly.

[0096] (3) Determine whether to enter an iteration step based on the driving data of the vehicle and the battery data of the battery at the target battery risk factor mutation time.

[0097] When this step is implemented, after obtaining the driving data and battery data of the target battery risk factor mutation time, the performance of the current static analysis model can be evaluated, and whether to enter the iteration step can be determined by the accuracy, stability, and predictive ability of the static analysis model. For example, in one embodiment, when the accuracy, stability, and predictive ability of the static analysis model are all lower than the corresponding performance setting values, it is determined that the static analysis model enters the iteration step.

[0098] (4) If so, evaluate the analysis performance corresponding to the static analysis model.

[0099] Specifically, by determining the analysis performance of the static analysis model, its performance in processing target-related data can be understood. In specific implementation, a traditional analysis performance acquisition method can be used to determine the accuracy of the static analysis model in processing data to determine the performance corresponding to the static analysis model. For example, in one embodiment, the analysis performance of the static analysis model can be determined based on the Brill score and the log-likelihood number.

[0100] (5) Determining sample data based on the analysis performance, and training the static analysis model based on the sample data.

[0101] Specifically, the sample data is obtained based on the abnormal vehicle events that occurred from the previous iterative training to the current one. In specific implementation, the data and contribution corresponding to all abnormal vehicle events in the specified time period before the iterative training can be determined as sample data; representative data can also be determined as sample data by analyzing performance. For example, in one embodiment, data containing significant features in the data before and after the mutation time of the target battery risk factor is determined as sample data, and the static analysis model is retrained based on the new sample data to complete the iteration process of the static analysis model.

[0102] The process provided in this embodiment first compares the target battery risk factors obtained by the static analysis model of adjacent cycles, then identifies the mutation time of the target battery risk factor based on the comparison results, and then determines whether to enter the iteration step based on the driving data of the vehicle and the battery data of the battery at the mutation time of the target battery risk factor. After determining to enter the iteration step, the analysis performance corresponding to the static analysis model is evaluated, and finally the sample data is determined based on the analysis performance, and the static analysis model is trained based on the sample data. In this way, it is determined whether to iterate the static analysis model based on the comparison results of adjacent cycles of the static analysis model, ensuring that the static model is iterated at the appropriate time. Furthermore, by iterating the static analysis model, the model can always provide the most accurate evaluation, ensuring the reliability and accuracy of the model processing data.

[0103] Figure 4 This is a flowchart of the second embodiment of the battery status evaluation method provided by this application. Please refer to Figure 4 Based on the above embodiment, the step of evaluating the real-time health status of the battery includes:

[0104] S401. If the real-time health status is unhealthy, generate warning information according to the real-time health status, and the warning information at least includes a warning part, a warning reason and a warning behavior.

[0105] Specifically, the abnormal situation of the battery can be reported in time through the early warning information. It should be noted that the early warning information may include the early warning location, the early warning reason and the early warning behavior, so as to help the user take appropriate response measures.

[0106] Furthermore, the warning location is a specific location in the battery where the problem occurs. For example, in one embodiment, the warning location may be the entire battery, or a battery module.

[0107] Furthermore, the warning reason is the main reason causing the unhealthy state of the battery. For example, in one embodiment, the warning reason may be abnormal voltage, excessive temperature, battery overcharge, battery overdischarge, etc.

[0108] Furthermore, the warning action is a remedial measure to deal with the unhealthy state of the battery. For example, in one embodiment, the warning action may be to stop charging, limit the discharge rate, cool the battery, or return the battery to the factory for repair.

[0109] In specific implementation, warning information can be generated based on the conclusions obtained by the dynamic warning model. In specific implementation, corresponding thresholds can be preset for each target-related data. When the dynamic warning model determines that the battery is in a poor state (i.e., the possibility of result variables such as scrapping and failure is high), it is detected whether the values ​​of each target-related data of the battery exceed the threshold, and corresponding warning information is generated for the target-related data that exceeds the threshold. For example, in one embodiment, when the 5-year survival rate of battery B is too low, a warning message is generated: repair battery B to solve the problem of low consistency, reduce the use load of battery B, increase the battery replacement / charging ratio of battery B, expand the number of spare batteries in the battery B battery replacement station, or invest in more battery replacement stations to balance the battery load, etc.

[0110] It is understandable that the warning information can be used to remind multiple aspects. In specific implementation, corresponding maintenance warning information can be generated for battery maintenance stations, battery use warning information can be generated for battery holders, and operation warning information can be provided to battery swap station operators.

[0111] S402: Generate a battery management plan based on the warning information, generate a maintenance plan based on the battery management plan, and send the maintenance plan to the user end.

[0112] Specifically, the battery management plan may include the future use plan and maintenance plan of the battery, and the battery is maintained through the maintenance plan. In specific implementation, in this step, the maintenance requirements of the battery can be determined based on the early warning information, and the maintenance requirements include maintenance steps, required maintenance tools, estimated maintenance time and other related content.

[0113] It should be noted that the maintenance plan can be customized and expanded according to actual needs to meet the application requirements of batteries in different scenarios.

[0114] The battery status assessment method provided in this embodiment generates warning information according to the real-time health status when the real-time health status is unhealthy, and then generates a battery management plan based on the warning information, generates a maintenance plan based on the battery management plan, and sends the maintenance plan to the user end. In this way, generating a maintenance plan and sending it to the user end can effectively identify and solve problems in the battery system and ensure the safe and efficient operation of the battery.

[0115] Figure 5 This is a flowchart of the third embodiment of the battery status evaluation method provided by this application. Please refer to Figure 5 Based on the above embodiment, the real-time health status is published on the battery management platform; the step of the battery management platform obtaining battery usage suggestions based on the real-time health status includes:

[0116] S501: In response to a battery status checking request, obtain the driving data.

[0117] Specifically, the battery management platform is used to manage the health status of the battery. Through the battery management platform, users can monitor the status of the battery in real time and evaluate the battery.

[0118] Furthermore, the battery status check request is a request actively initiated by the user when the user needs to manage the battery. After receiving the battery status check request, the battery management extracts the driving data before the current time.

[0119] S502: Determine driving behavior habits based on the driving data.

[0120] Specifically, the driving behavior habit is the user's habit of using the vehicle battery. The driving behavior habit can be obtained through driving data. It should be noted that the specific data type included in the driving behavior habit is determined according to actual needs, and this is not limited in this embodiment. For example, in one embodiment, the driving behavior habit may include the frequency of sudden acceleration, the frequency of sudden braking, and the frequency of high-speed driving.

[0121] S503: Determine target usage suggestions and battery maintenance suggestions according to the battery usage suggestions and the driving behavior habits, and display the target usage suggestions and battery maintenance suggestions based on the battery management platform.

[0122] Specifically, the target usage suggestion is based on the driver's battery usage habits, providing targeted follow-up usage suggestions. It should be noted that since different batteries correspond to different driving behavior habits, the generated target usage suggestions are also different. The target usage suggestions can improve the health and life of the battery.

[0123] In specific implementation, the target usage recommendations may include real-time display of battery failure risks, real-time display of battery life loss, real-time battery failure risk warning, battery failure attribution and life loss recommendations, and battery health consultation information.

[0124] Furthermore, battery maintenance recommendations are suggestions for battery repair and subsequent maintenance.

[0125] Furthermore, when the target usage suggestions and battery maintenance suggestions are displayed based on the battery management platform, the specific display form can be determined according to actual needs, and this is not limited in this embodiment. For example, the target usage suggestions and battery maintenance suggestions can be displayed in the form of text, audio, and animation.

[0126] A specific embodiment is given below to describe in detail the process of determining target usage suggestions:

[0127] (1) Determining historical usage habits of the battery based on the real-time health status.

[0128] Specifically, based on the real-time health status of the battery and past usage data, the historical usage habits of the battery are determined. For example, in one embodiment, the historical usage habits of the battery may include the charging range of the battery, the frequency of sudden acceleration of the vehicle, the frequency of sudden braking of the vehicle, etc.

[0129] (2) Determine the usage difference between the historical usage habits and the standard usage habits of the battery.

[0130] Specifically, the standard usage habits of the battery are determined according to the model and specifications of the battery, etc. It should be noted that the corresponding standard usage habits can be added to the battery management platform when the battery leaves the factory.

[0131] In specific implementation, the historical usage habits can be compared with the preset standard usage habits to obtain the usage difference between the two.

[0132] (3) Generating the target usage recommendation based on the usage difference.

[0133] In specific implementation, for example, in one embodiment, personalized target usage suggestions are generated based on usage differences: "The historical charging range of this battery is not within the standard range, and it is recommended to adjust the charging strategy."

[0134] The battery status assessment method provided in this embodiment first responds to a battery status viewing request to obtain driving data, then determines driving behavior habits based on the driving data, and then determines target usage suggestions based on the battery usage suggestions and driving behavior habits, and displays the target usage suggestions based on the battery management platform. In this way, through the battery management platform, users can understand the historical usage habits of the battery and compare them with the standard usage habits to generate personalized usage suggestions. Furthermore, when the user needs to understand the status of the battery, the battery management platform can be used to actively understand the battery status. The battery management platform not only helps users optimize the battery usage and maintenance strategies, but also improves the safety and performance of the battery.

[0135] Figure 6 This is a flowchart of the fourth embodiment of the battery status evaluation method provided by this application. Please refer to Figure 6 Based on the above embodiment, the step of obtaining the driving data of the vehicle and the battery data of the battery includes:

[0136] S601. Extract target features based on data generated during driving of the vehicle and data generated by the battery; wherein the target features include timestamp information, and the timestamp information is used to determine a time point for extracting the target features.

[0137] Specifically, the target features are data generated by the vehicle, data generated by the battery, and data related to the battery.

[0138] Furthermore, the timestamp information is the time point when the target feature is extracted, and the time synchronization of the data can be ensured by the timestamp information.

[0139] S602: Determine the time point of battery abnormality according to the battery maintenance information in the battery data.

[0140] Specifically, when the battery is inspected and maintained, the inspection and maintenance information of the battery is obtained, and the inspection and maintenance information of the battery includes all the information whether the battery remains normal or abnormal.

[0141] Furthermore, the battery maintenance information also includes the occurrence time of each event. For example, in one embodiment, when generating battery maintenance information, the time point when the battery abnormality occurs is determined as the battery abnormality time point. It is understandable that when the target associated data has a greater impact on the battery, the time point when it is obtained can be determined as the battery abnormality time point. For example, in one embodiment, when the battery is working in an extremely harsh environment, the start time of this period is determined as the battery abnormality time point.

[0142] S603. Determine the target battery feature in which the abnormality occurs in the target feature based on the battery abnormality time point and the timestamp information, determine the portion of the target battery feature related to vehicle driving as the driving data of the vehicle, and determine the portion of the target battery feature corresponding to the battery as the battery data of the battery.

[0143] In the specific implementation, in this embodiment, the time point in the timestamp information that is the same as the battery abnormality time point is extracted, the corresponding target feature is determined as the target battery feature, and the target battery feature is compared with the judgment standard to determine that the first associated data is the target battery feature that meets the judgment standard of the driving data, and the second associated data is determined to meet the judgment standard of the battery maintenance data.

[0144] Furthermore, after obtaining the first associated data and the second associated data, the two may be directly superimposed to obtain the target associated data.

[0145] The battery status assessment method provided in this embodiment first extracts the target features based on the data generated during vehicle driving and the data generated by the battery, then determines the battery abnormality time point based on the battery maintenance information in the battery data, and finally determines the target battery features where the abnormality occurs in the target features based on the battery abnormality time point and timestamp information, and determines the part of the target battery features related to vehicle driving as the vehicle driving data, and determines the part of the target battery features corresponding to the battery as the battery data of the battery. In this way, the battery features can be extracted from the target features, and these features can be divided into driving data and battery data. Through the timestamp information, it can be ensured that the driving data and the battery data correspond to and verify each other, reducing the possibility of misjudgment and ensuring the accuracy of the extracted target-related data. In this way, it is helpful to accurately identify the abnormal behavior of the battery and provide a basis for fault warning and maintenance decisions.

[0146] Corresponding to the aforementioned embodiment of a battery status evaluation method, the present application also provides an embodiment of a battery status evaluation device.

[0147] The present application provides an embodiment of a battery status evaluation device that can be applied to a battery status evaluation device. The device embodiment can be implemented by software, or by hardware, or by a combination of software and hardware. Taking software implementation as an example, as a logical device, the processor of the battery status evaluation device in which it is located reads the corresponding computer program instructions in the non-volatile memory into the memory and runs them. From a hardware perspective, if Figure 7 As shown, it is a hardware structure diagram of the battery status evaluation device in which the battery status evaluation device of the present application is located, except Figure 7 In addition to the processor, memory, network interface, and non-volatile memory shown, the battery status evaluation device in the embodiment where the device is located may also include other hardware according to the actual function of the battery status evaluation device, which will not be described in detail.

[0148] Figure 8 This is a schematic diagram of the structure of the first embodiment of the battery status evaluation device provided by this application. Please refer to Figure 8 The device provided in this embodiment includes an acquisition module 810, a generation module 820 and a processing module 830; wherein:

[0149] The acquisition module 810 is used to acquire the driving data of the vehicle and the battery data of the battery;

[0150] The generating module 820 is configured to use the driving data as a judgment standard to filter out first associated data caused by driving behavior from the battery data, use the battery maintenance data as a judgment standard to filter out second associated data caused by battery use behavior from the battery data, and generate target associated data based on the first associated data and the second associated data;

[0151] The processing module 830 is used to establish a static analysis model, and process the target associated data of the battery based on the static analysis model to obtain a target battery risk factor; wherein the target battery risk factor is used to characterize the contribution of different battery characteristics to different abnormal events;

[0152] The processing module 830 is used to establish a dynamic early warning model, and evaluate the real-time health status of the battery based on the dynamic early warning model according to the target battery risk factor and the real-time operation data of the battery.

[0153] The device of this embodiment can be used to perform Figure 1 The steps, specific implementation principles and implementation processes of the method embodiment shown are similar and will not be repeated here.

[0154] Optionally, the processing module 830 is specifically used for establishing the static analysis model, and processing the target associated data of the battery based on the static analysis model to obtain a target battery risk factor, including:

[0155] The processing module 830 is also specifically used for the data processing module to analyze and evaluate the battery characteristics according to the battery outcome data, and to determine the target battery risk factor of the battery characteristics for each of the abnormal events based on the contribution analysis module.

[0156] Optionally, the processing module 830 is further specifically used to compare the target battery risk factors obtained by the static analysis model in adjacent cycles;

[0157] The processing module 830 is further specifically used to identify the mutation time of the target battery risk factor according to the comparison result;

[0158] The processing module 830 is further specifically used to determine whether to enter an iteration step based on the driving data of the vehicle and the battery data of the battery at the target battery risk factor mutation time;

[0159] The processing module 830 is further specifically configured to evaluate the analysis performance corresponding to the static analysis model after determining to enter the iteration step;

[0160] The processing module 830 is further specifically configured to determine sample data based on the analysis performance, and train the static analysis model based on the sample data.

[0161] Optionally, the processing module 830 is further specifically configured to predict a target battery risk factor according to the static analysis model;

[0162] The processing module 830 is further specifically configured to determine an abnormal event threshold corresponding to each abnormal event based on the battery data of the battery;

[0163] The processing module 830 is further specifically used to obtain real-time associated data, use the target battery risk factor as a weight value, identify the real-time associated data based on the dynamic early warning model, and obtain a real-time risk value corresponding to each abnormal event;

[0164] The processing module 830 is further specifically configured to accumulate the real-time risk value corresponding to each abnormal event to obtain a cumulative risk value;

[0165] The processing module 830 is further specifically configured to determine that the real-time health status is unhealthy when any of the accumulated risk values ​​exceeds the abnormal event threshold of the corresponding abnormal event.

[0166] Optionally, the processing module 830 is further specifically configured to generate warning information according to the real-time health status if the real-time health status is unhealthy, wherein the warning information at least includes a warning part, a warning reason and a warning behavior;

[0167] The processing module 830 is further specifically configured to generate a battery management plan based on the warning information, generate a maintenance plan based on the battery management plan, and send the maintenance plan to a user terminal.

[0168] Optionally, the processing module 830 is further specifically configured to obtain the driving data in response to a battery status viewing request;

[0169] The processing module 830 is further specifically configured to determine a driving behavior habit based on the driving data;

[0170] The processing module 830 is further specifically used to determine target usage suggestions and battery maintenance suggestions based on the battery usage suggestions and the driving behavior habits, and to display the target usage suggestions and battery maintenance suggestions based on the battery management platform.

[0171] Optionally, the processing module 830 is further specifically configured to determine a historical usage habit of the battery based on the real-time health status;

[0172] The processing module 830 is further specifically used to determine the usage difference between the historical usage habit and the standard usage habit of the battery;

[0173] The processing module 830 is further specifically configured to generate the target usage suggestion based on the usage difference.

[0174] Optionally, the acquisition module 810 is specifically used to extract target features based on the data generated during the driving of the vehicle and the data generated by the battery; wherein the target features include timestamp information, and the timestamp information is used to determine the time point of extraction of the target features;

[0175] The acquisition module 810 is further specifically used to determine the battery abnormality time point according to the battery maintenance information in the battery data;

[0176] The acquisition module 810 is also specifically used to determine the target battery feature where the abnormality occurs in the target feature based on the battery abnormality time point and the timestamp information, determine the part of the target battery feature related to vehicle driving as the driving data of the vehicle, and determine the part of the target battery feature corresponding to the battery as the battery data of the battery.

[0177] Please continue to refer to Figure 7 The present application also provides a battery status evaluation device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any one of the methods provided in the first aspect of the present application are implemented.

[0178] The present application also provides a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the steps of any one of the methods provided in the present application are implemented.

[0179] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0180] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present application scheme. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0181] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A battery status assessment method, characterized in that: The method comprises: Acquire driving data of the vehicle and battery data of the battery; use the driving data as a judgment standard to filter out first associated data caused by driving behavior from the battery data; use the battery maintenance data as a judgment standard to filter out second associated data caused by battery use behavior from the battery data; and generate target associated data based on the first associated data and the second associated data; Establishing a static analysis model, and processing the target associated data of the battery based on the static analysis model to obtain a target battery risk factor; wherein the target battery risk factor is used to characterize the contribution of different battery characteristics to different abnormal events; A dynamic early warning model is established, and based on the dynamic early warning model, the real-time health status of the battery is evaluated according to the target battery risk factor and the real-time operation data of the battery.

2. The method according to claim 1, characterized in that The static analysis model includes a data processing module, a contribution analysis module and a risk assessment module; the establishment of the static analysis model, processing the target associated data of the battery based on the static analysis model, and obtaining the target battery risk factor includes: The data processing module analyzes and evaluates the battery characteristics according to the battery outcome data, and determines the target battery risk factor of the battery characteristics for each of the abnormal events based on the contribution analysis module.

3. The method according to claim 1, characterized in that: After establishing the static analysis model and processing the target associated data of the battery based on the static analysis model to obtain the target battery risk factor, the method further includes: Comparing the target battery risk factors obtained by the static analysis model in adjacent cycles; Identify the mutation time of the target battery risk factor based on the comparison results; Determining whether to enter an iteration step based on the driving data of the vehicle and the battery data of the battery at the target battery risk factor mutation time; If yes, then evaluating the analysis performance corresponding to the static analysis model; Sample data is determined based on the analysis performance, and the static analysis model is trained based on the sample data.

4. The method according to claim 1, characterized in that: The establishing of a dynamic early warning model, and evaluating the real-time health status of the battery based on the dynamic early warning model according to the target battery risk factor and the real-time operation data of the battery, includes: Predicting a target battery risk factor according to the static analysis model; Determine an abnormal event threshold corresponding to each of the abnormal events based on the battery data of the battery; Acquire real-time associated data, use the target battery risk factor as a weight value, identify the real-time associated data based on the dynamic early warning model, and obtain a real-time risk value corresponding to each abnormal event; Accumulating the real-time risk value corresponding to each abnormal event to obtain a cumulative risk value; When any of the accumulated risk values ​​exceeds the abnormal event threshold of the corresponding abnormal event, the real-time health status is determined to be unhealthy.

5. The method according to claim 1, characterized in that The real-time health status of the battery is evaluated, including: If the real-time health status is unhealthy, generate warning information according to the real-time health status, and the warning information at least includes the warning part, the warning reason and the warning behavior; A battery management plan is generated based on the early warning information, a maintenance plan is generated based on the battery management plan, and the maintenance plan is sent to a user end.

6. The method according to claim 1, characterized in that The real-time health status is published on the battery management platform; The battery management platform obtains battery usage recommendations based on the real-time health status, and the method further includes: In response to a battery status checking request, acquiring the driving data; determining a driving behavior habit based on the driving data; Target usage suggestions and battery maintenance suggestions are determined according to the battery usage suggestions and the driving behavior habits, and the target usage suggestions and battery maintenance suggestions are displayed based on the battery management platform.

7. The method according to claim 6, characterized in that The battery management platform obtains battery usage suggestions based on the real-time health status, including: determining historical usage habits of the battery based on the real-time health status; determining a usage difference between the historical usage habits and a standard usage habit of the battery; The target usage suggestion is generated based on the usage difference.

8. The method according to claim 1, characterized in that The obtaining of the driving data of the vehicle and the battery data of the battery includes: Extracting target features based on the data generated during the driving of the vehicle and the data generated by the battery; wherein the target features include timestamp information, and the timestamp information is used to determine the time point of extraction of the target features; Determine the time point of battery abnormality based on battery maintenance information in the battery data; The target battery feature in which the abnormality occurs in the target feature is determined based on the battery abnormality time point and the timestamp information, the portion of the target battery feature related to vehicle driving is determined as driving data of the vehicle, and the portion of the target battery feature corresponding to the battery is determined as battery data of the battery.

9. The method according to claim 2, characterized in that: The determining, based on the contribution analysis module, a target battery risk factor of the battery feature for each of the abnormal events includes: Identify battery characteristics and abnormal events; Calculate the contribution of each battery feature to each abnormal event, where one battery feature corresponds to multiple abnormal events; Determine the comprehensive weight based on abnormal events; The comprehensive contribution of each abnormal feature under each abnormal event is calculated based on the comprehensive weight to obtain the target battery risk factor corresponding to the battery feature.

10. A battery status evaluation device, characterized in that: The device comprises an acquisition module, a generation module and a processing module; wherein: The acquisition module is used to acquire the driving data of the vehicle and the battery data of the battery; The generating module is configured to use the driving data as a judgment standard to filter out first associated data caused by driving behavior from the battery data, use the battery maintenance data as a judgment standard to filter out second associated data caused by battery use behavior from the battery data, and generate target associated data based on the first associated data and the second associated data; The processing module is used to establish a static analysis model, and process the target associated data of the battery based on the static analysis model to obtain a target battery risk factor; wherein the target battery risk factor is used to characterize the contribution of different battery characteristics to different abnormal events; The processing module is used to establish a dynamic early warning model, and evaluate the real-time health status of the battery based on the dynamic early warning model according to the target battery risk factor and the real-time operation data of the battery.

Images