A vehicle battery life management method and device, electronic equipment and storage medium

Abstract

This application provides a vehicle battery life management method, apparatus, electronic device, and storage medium. The method includes: collecting vehicle usage data; quantifying the usage behavior into an impact coefficient on the battery health degradation rate based on a pre-built risk quantification model; determining a warning threshold for each usage behavior in each cycle based on the impact coefficient; generating personalized battery management suggestions and pushing them to an interactive terminal when a usage behavior reaches the warning threshold in any cycle; determining the user's adoption of the management suggestions and the battery improvement based on the usage data, and optimizing the suggestion strategy and calibrating the parameters of the risk quantification model based on the adoption and improvement. This application's embodiments, through the above method, help to achieve precise correlation and dynamic optimization between user behavior and battery life management.

Description

Technical Field

[0001] This application relates to the field of vehicle battery life management technology, and more specifically, to a vehicle battery life management method, apparatus, electronic device, and storage medium. Background Technology

[0002] Battery health is a key indicator affecting the driving range, safety performance, and resale value of new energy vehicles. Precise management of battery life helps slow down battery degradation and improve the user experience.

[0003] Research has revealed that existing battery management methods primarily rely on the battery's physical parameters (such as voltage and temperature) for monitoring and protection, resulting in limited accuracy in battery life management.

[0004] Therefore, there is a need to provide an improved method for managing vehicle battery life. Summary of the Invention

[0005] In view of this, embodiments of this application provide a vehicle battery life management method, apparatus, electronic device, and storage medium, which helps to achieve accurate correlation and dynamic optimization between user behavior and battery life management.

[0006] In a first aspect, embodiments of this application provide a vehicle battery life management method, the method comprising: Collect vehicle usage data; the usage data is used to characterize the user's vehicle usage behavior; Based on a pre-built risk quantification model, the vehicle usage behavior is quantified as an impact coefficient on the rate of battery health degradation; Based on the aforementioned impact coefficient, the warning threshold for each of the aforementioned vehicle usage behaviors in each period is determined; During any given period, when a vehicle usage behavior is detected to reach the warning threshold, a personalized battery management suggestion is generated and pushed to the interactive terminal. Based on the vehicle usage data, the user's adoption of the management recommendations and the improvement of the battery are determined. Based on the adoption and improvement, the recommendation strategy is optimized and the parameters of the risk quantification model are calibrated.

[0007] In one feasible implementation, the risk quantification model is constructed in the following ways: Acquire historical usage data and corresponding battery health degradation data for multiple sample vehicles; Based on the historical vehicle usage data, the vehicle usage behavior characteristics of the sample vehicles in each period are extracted. The battery health degradation rate of each of the sample vehicles is calculated based on the battery health degradation data. Regression analysis was performed using the vehicle usage behavior characteristics as independent variables and the battery health degradation rate as dependent variable to obtain the influence coefficient of each vehicle usage behavior on the battery health degradation rate.

[0008] In one feasible implementation, the vehicle usage behavior includes at least one of the following: Risky charging behavior, which includes at least one of fast charging behavior, full charging behavior, or deep charging behavior; Risky discharge behavior, wherein the risky discharge behavior includes at least one of rapid acceleration behavior, rapid deceleration behavior, or undervoltage discharge behavior; Risky parking behavior, which includes at least one of the following: parking with a full charge, parking with a low charge, or parking in a high-temperature environment; Risky environmental interaction behaviors include at least one of the following: charging behavior in a high-temperature environment, charging behavior in a low-temperature environment, or using substandard charging piles.

[0009] In one feasible implementation, determining the warning threshold for each of the vehicle usage behaviors in each period based on the influence coefficient includes: Obtain a preset upper limit for loss for a reference period, wherein the duration of the reference period is greater than the period. Based on the influence coefficient corresponding to each of the aforementioned vehicle usage behaviors, the upper limit of the reference period loss is allocated to each vehicle usage behavior to obtain the allowable loss amount of each vehicle usage behavior within the reference period. Based on the allowable wear and tear of each vehicle use behavior within the reference period and the corresponding influence coefficient, calculate the maximum allowable number of occurrences of each vehicle use behavior within the reference period. Based on the maximum permissible number of occurrences of each vehicle usage behavior within the reference period, the warning threshold for each period within the reference period is determined.

[0010] In one feasible implementation, the method further includes: Get the current state parameters of the battery; When the status parameters meet the preset maintenance trigger conditions, the idle time window when the user does not use the vehicle is predicted based on the vehicle usage data. During the idle time window, a maintenance negotiation request is pushed to the interactive terminal; Upon receiving a confirmation instruction in response to the maintenance negotiation request, a battery maintenance operation is performed within the idle time window. The maintenance operation includes at least one of battery resting, temperature adjustment, charging, or parameter calibration.

[0011] In one feasible implementation, determining the user's adoption of the management recommendations based on the vehicle usage data includes: Set an observation window corresponding to the management suggestion, the length of the observation window being one or more of the aforementioned periods; Within the observation window, the actual behavior value of the target vehicle usage behavior corresponding to the management suggestion is extracted from the vehicle usage data; Obtain the suggested target value for the target vehicle usage behavior from the management recommendations; The adoption status of the management recommendations is determined based on the difference between the actual behavior value and the suggested target value.

[0012] In one feasible implementation, optimizing the recommended strategy and calibrating the parameters of the risk quantification model based on the adoption and improvement outcomes includes: Based on the adoption status and the improvement status, an effectiveness score is calculated for each suggested strategy; the effectiveness score is positively correlated with the improvement status of the battery. The push priority of different suggestion strategies is adjusted based on the aforementioned performance score; The impact coefficient in the risk quantification model is updated using the accumulated adoption and improvement data.

[0013] Secondly, embodiments of this application also provide a vehicle battery life management device, the device comprising: The data collection module is used to collect vehicle usage data; the usage data is used to characterize the user's vehicle usage behavior. The quantification module is used to quantify the vehicle usage behavior into an impact coefficient on the rate of battery health degradation based on a pre-built risk quantification model. The determination module is used to determine the warning threshold for each of the vehicle usage behaviors in each period based on the influence coefficient; The push module is used to generate personalized battery management suggestions and push them to the interactive terminal when a certain vehicle use behavior reaches the warning threshold within any period. The optimization module is used to determine the user's adoption of the management suggestions and the improvement of the battery based on the vehicle usage data, and to optimize the suggestion strategy and calibrate the parameters of the risk quantification model based on the adoption and improvement.

[0014] In one feasible implementation, the risk quantification model is constructed in the following ways: Acquire historical usage data and corresponding battery health degradation data for multiple sample vehicles; Based on the historical vehicle usage data, the vehicle usage behavior characteristics of the sample vehicles in each period are extracted. The battery health degradation rate of each of the sample vehicles is calculated based on the battery health degradation data. Regression analysis was performed using the vehicle usage behavior characteristics as independent variables and the battery health degradation rate as dependent variable to obtain the influence coefficient of each vehicle usage behavior on the battery health degradation rate.

[0015] In one feasible implementation, the vehicle usage behavior includes at least one of the following: Risky charging behavior, which includes at least one of fast charging behavior, full charging behavior, or deep charging behavior; Risky discharge behavior, wherein the risky discharge behavior includes at least one of rapid acceleration behavior, rapid deceleration behavior, or undervoltage discharge behavior; Risky parking behavior, which includes at least one of the following: parking with a full charge, parking with a low charge, or parking in a high-temperature environment; Risky environmental interaction behaviors include at least one of the following: charging behavior in a high-temperature environment, charging behavior in a low-temperature environment, or using substandard charging piles.

[0016] In one feasible implementation, the push module is used to determine the warning threshold for each of the vehicle usage behaviors in each period based on the influence coefficient, for the following purposes: Obtain a preset upper limit for loss for a reference period, wherein the duration of the reference period is greater than the period. Based on the influence coefficient corresponding to each of the aforementioned vehicle usage behaviors, the upper limit of the reference period loss is allocated to each vehicle usage behavior to obtain the allowable loss amount of each vehicle usage behavior within the reference period. Based on the allowable wear and tear of each vehicle use behavior within the reference period and the corresponding influence coefficient, calculate the maximum allowable number of occurrences of each vehicle use behavior within the reference period. Based on the maximum permissible number of occurrences of each vehicle usage behavior within the reference period, the warning threshold for each period within the reference period is determined.

[0017] In one feasible implementation, the device further includes: The parameter acquisition module is used to acquire the current status parameters of the battery; The prediction module is used to predict the idle time window when the status parameters meet the preset maintenance trigger conditions based on the vehicle usage data. The maintenance push module is used to push maintenance negotiation requests to the interactive terminal during the idle time window. The maintenance module is used to perform battery maintenance operations within the idle time window after receiving a confirmation instruction in response to the maintenance negotiation request. The maintenance operations include at least one of battery resting, temperature adjustment, charging, or parameter calibration.

[0018] In one feasible implementation, the optimization module is configured to determine the user's adoption rate of the management recommendations based on the vehicle usage data, and is configured to: Set an observation window corresponding to the management suggestion, the length of the observation window being one or more of the aforementioned periods; Within the observation window, the actual behavior value of the target vehicle usage behavior corresponding to the management suggestion is extracted from the vehicle usage data; Obtain the suggested target value for the target vehicle usage behavior from the management recommendations; The adoption status of the management recommendations is determined based on the difference between the actual behavior value and the suggested target value.

[0019] In one feasible implementation, the optimization module is configured to optimize the recommended strategy and calibrate the parameters of the risk quantification model based on the adoption status and the improvement status, for the following purposes: Based on the adoption status and the improvement status, an effectiveness score is calculated for each suggested strategy; the effectiveness score is positively correlated with the improvement status of the battery. The push priority of different suggestion strategies is adjusted based on the aforementioned performance score; The impact coefficient in the risk quantification model is updated using the accumulated adoption and improvement data.

[0020] Thirdly, embodiments of this application also provide an electronic device, including: a processor, a storage medium, and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the vehicle battery life management method as described in any one of the first aspects.

[0021] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the vehicle battery life management method as described in any one of the first aspects.

[0022] This application provides a vehicle battery life management method, device, electronic device, and storage medium. By collecting vehicle usage data to characterize user vehicle usage behavior, and quantifying the usage behavior into an impact coefficient on the battery health degradation rate based on a risk quantification model, a quantitative correlation between user behavior and battery life is achieved. By determining the warning threshold for each usage behavior in each cycle based on the impact coefficient, personalized battery management suggestions can be generated and pushed in a timely manner when user behavior reaches the preset threshold, achieving accurate identification and proactive intervention of user risky behavior. By determining the user's adoption of management suggestions and battery improvement based on vehicle usage data, and optimizing the suggestion strategy and calibrating the parameters of the risk quantification model accordingly, a closed-loop management mechanism of "identification-intervention-feedback-optimization" is formed.

[0023] Compared with the prior art, the embodiments of this application can incorporate user vehicle usage behavior into the battery life management system, realize accurate quantitative correlation between user behavior and battery life, and continuously optimize management strategies based on user feedback, thereby effectively improving the accuracy of battery life management and the effectiveness of user behavior guidance.

[0024] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0025] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 A flowchart of a vehicle battery life management method provided in an embodiment of this application is shown.

[0027] Figure 2 A flowchart of another vehicle battery life management method provided in an embodiment of this application is shown.

[0028] Figure 3 A schematic diagram of the structure of a vehicle battery life management device provided in an embodiment of this application is shown.

[0029] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of this application is shown. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0031] Battery health is a key indicator affecting the driving range, safety performance, and resale value of new energy vehicles. Precise management of battery life helps slow down battery degradation and improve the user experience. Besides the inherent characteristics of the cell materials, battery life degradation is also significantly influenced by the user's actual driving behavior—such as charging habits, driving style, and parking environment. Theoretically, scientific management and guidance of user electricity consumption behavior can effectively extend battery life.

[0032] However, most users lack systematic knowledge about battery health and are unaware of the cumulative impact of daily driving habits on battery life. If vehicle systems do not provide proactive and timely intervention prompts, users are highly likely to unconsciously continue engaging in risky behaviors that damage battery health, such as prolonged periods of full-charge parking, frequent deep fast charging, and charging in high-temperature environments. These behaviors can lead to irreversible degradation of battery life. Therefore, effectively identifying and monitoring user risky behaviors and generating user-friendly and actionable driving suggestions are crucial for improving battery life management.

[0033] Research has revealed that traditional battery management methods primarily rely on monitoring and protecting the battery's physical parameters. One approach involves the battery management system (BMS) monitoring parameters such as voltage and temperature in real time, triggering alarms or protection mechanisms when these parameters exceed safety thresholds. However, this approach fails to incorporate user-controllable behavioral variables as independent assessment dimensions into the risk quantification system. This results in the system's inability to identify and assess the long-term cumulative damage to battery life caused by specific user behaviors, such as the hidden degradation caused by prolonged full-charge storage. Another approach utilizes some advanced vehicle infotainment systems with simple user behavior recording functions, capable of sending alerts to users when specific scenarios (such as full charging or aggressive driving) are detected. However, this approach stops at simply providing symptom alerts, failing to offer clear and actionable suggestions or personalize the system to suit individual user habits. This leads to poor usability and difficulty in user compliance. More importantly, existing solutions lack tracking of user adoption of suggestions and evaluation feedback on the battery improvement effects after adoption, failing to establish a closed-loop management process of "suggestion—adoption—optimization."

[0034] In summary, since the cumulative impact of user behavior on battery life is difficult to quantify and identify, existing prompting methods lack personalization and enforceability, and the lack of a closed-loop feedback mechanism prevents continuous system optimization, an improved vehicle battery life management method is needed to address these issues.

[0035] Based on this, embodiments of this application provide a vehicle battery life management method, apparatus, electronic device, and storage medium, which are described below through embodiments.

[0036] To facilitate understanding of this embodiment, a vehicle battery life management method disclosed in this application will first be described in detail. For example... Figure 1 As shown, it includes the following steps: Step 101: Collect vehicle usage data; the usage data is used to characterize the user's vehicle usage behavior.

[0037] During the daily operation and use of a vehicle, a large amount of vehicle usage data is generated, reflecting user operating habits and usage patterns. This data is not an isolated physical signal, but a direct manifestation of user interaction with the vehicle. For example, when a user charges the vehicle, what charging method is used, the intensity of acceleration and deceleration during driving, and the environment in which the vehicle is parked all fall under the category of vehicle usage behavior. While these behaviors themselves are not directly equivalent to changes in the physical state of the battery, their long-term cumulative effect significantly impacts the battery's health and degradation rate.

[0038] Therefore, it is necessary to collect this type of vehicle usage data. The collected data can include various signals transmitted in the vehicle's internal network (such as the CAN bus), as well as records generated during interactions between the vehicle and external devices (such as charging stations and cloud platforms). The types of data collected can be flexibly determined according to actual needs; for example, it can include battery charging and discharging data, vehicle driving status data, and data related to the vehicle's environment. These data collectively depict the user's vehicle usage behavior patterns from different perspectives, providing a foundation for subsequent analysis and processing.

[0039] It should be noted that the collected vehicle usage data should be raw data or data that has undergone preliminary cleaning to preserve the true details of the behavior. The frequency and timing of data collection can also be adjusted according to the specific scenario, for example, data can be collected every time the vehicle is started, every time it is charged, or during periodic reporting. Different vehicle models and configurations may have different types of data that can be collected. Those skilled in the art can choose appropriate data sources according to the actual situation, as long as the data can effectively characterize the user's vehicle usage behavior.

[0040] Step 102: Based on the pre-built risk quantification model, the vehicle usage behavior is quantified into an impact coefficient on the rate of battery health degradation.

[0041] After collecting vehicle usage data that characterizes user behavior, it's necessary to establish a quantifiable correlation between these behaviors and battery life. Different usage behaviors have varying degrees of impact on battery health. For example, frequent fast charging versus occasional fast charging, and deep discharge versus shallow discharge, have significantly different effects on accelerating battery degradation. Simply relying on qualitative judgments of "whether a behavior is good or bad" is insufficient for precise battery life management. Therefore, a risk quantification model is needed to transform user behavior into specific, calculable quantitative indicators.

[0042] The risk quantification model is pre-built, and its construction idea can be understood as: analyzing the statistical regularity between different driving behaviors and the actual battery degradation rate through historical data of a large number of sample vehicles. There are various specific implementation methods. For example, regression analysis can be used, with characteristics of various driving behaviors (such as monthly fast charging frequency, average depth of discharge, etc.) as independent variables and the battery health degradation rate as the dependent variable, to fit the contribution of each behavior to the degradation rate and obtain the corresponding influence coefficient. This influence coefficient reflects the magnitude of the impact of each occurrence of a certain driving behavior or each increase in intensity on the battery health degradation rate. Other machine learning algorithms or statistical models can also be used, as long as they can achieve a quantitative mapping from behavior to impact.

[0043] It's important to note that the "impact coefficient" in the risk quantification model is a broad concept; it can be positive, negative, or zero. A positive number indicates that the behavior accelerates battery degradation, a negative number indicates that the behavior slows down battery degradation (although in reality, most adverse behaviors accelerate degradation), and zero indicates no significant impact. The specific numerical meaning of the impact coefficient can be determined based on the model's output format. For example, it can be expressed as the monthly increase in the rate of battery degradation caused by a unit number of occurrences of the behavior, or as a percentage change relative to the baseline rate of degradation. After the risk quantification model is established, inputting the current vehicle usage behavior data into the model will yield various impact coefficients corresponding to the user's current behavior pattern, thus achieving a quantitative assessment of user behavior. This quantitative result provides a scientific basis for subsequently determining warning thresholds and generating personalized recommendations.

[0044] Step 103: Based on the influence coefficient, determine the warning threshold for each of the vehicle usage behaviors in each cycle.

[0045] After obtaining the impact coefficients of various vehicle usage behaviors on the rate of battery health degradation, a further question needs to be answered: to what extent should a certain behavior trigger a user alert? The impact coefficient itself reflects the "magnitude of the impact of a unit behavior," but it does not directly provide a standard for judging "how much is too much." Therefore, a warning threshold needs to be introduced as a benchmark to measure whether user behavior has exceeded the limit.

[0046] The determination of the warning threshold is closely related to the impact coefficient. The core idea can be understood as follows: based on the expected goals of battery life management and the impact coefficient of various behaviors, deduce the reasonable upper limit for each behavior within a given time period. The time period can be flexibly set according to management needs, for example, it can be based on a month, week, or quarter. Different vehicle usage behaviors have different impact coefficients, and the corresponding warning thresholds will also differ. Behaviors with larger impact coefficients mean that their effect on accelerating battery degradation is more significant, and the corresponding warning thresholds may be set more strictly; conversely, behaviors with smaller impact coefficients may have relatively lenient warning thresholds.

[0047] There are several ways to determine the specific impact of a battery degradation target. One feasible approach is to first set an overall management goal, such as controlling the additional battery degradation caused by user behavior within a certain range over a relatively long reference period (e.g., one year). This range can be understood as a "degradation budget." Then, based on the impact coefficients of various behaviors, this total budget is allocated to each behavior to obtain the allowable total amount of each behavior within the reference period. By combining the distribution of each behavior within the reference period, or by adjusting it according to a certain ratio, the warning threshold for each behavior within a shorter period (e.g., one month) can be obtained. When allocating the budget, the actual degree of harm of different behaviors to the battery can be considered, as well as the adjustability of user behavior habits. Those skilled in the art can set allocation weights or rules according to actual needs.

[0048] Another approach is to determine the baseline threshold directly based on historical statistical data. For example, the average level or reasonable upper limit of various behaviors in a large group of users with healthy batteries can be statistically analyzed and used as a reference to set the threshold. This method does not rely on a specific "damage budget" but is based on statistical patterns within the group. Regardless of the method used, the common point is that the determination of the warning threshold is based on the influence coefficient, reflecting the differences in the impact of different behaviors on battery degradation, thereby achieving differentiated and targeted threshold setting.

[0049] Warning thresholds can be a single value or multiple levels. For example, a relatively lenient mild warning threshold and a relatively strict severe warning threshold can be set, corresponding to different warning intensities or frequencies. When user behavior approaches or exceeds different levels of thresholds, the system can adopt different response strategies, thereby achieving gradual behavior guidance without unduly disturbing the user.

[0050] Step 104: In any given period, when a vehicle usage behavior is detected to reach the warning threshold, a personalized battery management suggestion is generated and pushed to the interactive terminal.

[0051] The purpose of setting warning thresholds is to alert users at appropriate times. During actual vehicle use, various driving behaviors are continuously monitored, and the actual frequency or intensity of each behavior within each cycle is recorded. At the end of the current cycle, or during real-time monitoring in the middle of the cycle, the user's actual behavior values ​​can be compared with the preset warning thresholds. When the occurrence of a certain driving behavior reaches or exceeds the corresponding warning threshold, it means that the user's behavior in the current cycle has reached a point requiring attention, and it is necessary to issue a reminder to the user.

[0052] The approach isn't simply to inform users that "a certain behavior has exceeded the limit," but rather to generate personalized battery management suggestions with guidance. Personalization means that the suggestions should match the user's current behavior, usage scenario, and the vehicle's actual condition. For example, regarding the same issue of exceeding the fast charging limit, for users with long commutes and limited charging convenience, the suggestion might be "Try to use slow charging overnight for the rest of the week"; for users with short commutes and home charging stations, the suggestion might be "Set the charging limit to 90% in the future." Suggestions can also incorporate historical user behavior patterns. For instance, if a user typically travels long distances on weekends, and fast charging is detected as exceeding the limit on Friday, the suggestion might be "Before your weekend long trip, we recommend using slow charging to fully charge today, rather than frequent fast charging." This personalized approach is more easily understood and accepted by users.

[0053] Suggestions should be specific and actionable, avoiding vague statements. Besides telling users "what to do," you can also explain "why to do it" and "what the benefits are." For example, you could suggest, "Frequent fast charging accelerates battery degradation. We've already fast-charged quite a few times this week; we recommend using slow charging tonight to help slow down battery aging." This positive guidance is more motivating than simple warnings.

[0054] The interactive interface for push notifications can be the vehicle's infotainment screen, a user's mobile application, or other channels that can reach the user, such as SMS or voice assistants. The timing of push notifications also needs consideration, avoiding forced notifications while the user is driving or when it's inconvenient for them to view information. Push notifications can be sent appropriately after the vehicle is turned off, during charging, or when the user checks the vehicle's status. The form of the push notifications can be text prompts, suggestion cards combining text and images, or voice announcements; those skilled in the art can choose the appropriate presentation method based on the actual interaction scenario.

[0055] It's important to note that the same content doesn't need to be pushed every time the warning threshold is reached. For instances where the same behavior repeatedly triggers the warning threshold, the tone and content of the suggestions can be gradually adjusted, transitioning from gentle reminders to stronger ones. Alternatively, based on user feedback history, suggestions with higher adoption rates can be prioritized to further improve their relevance and effectiveness.

[0056] Step 105: Based on the vehicle usage data, determine the user's adoption of the management suggestions and the battery improvement status, and optimize the suggestion strategy and calibrate the parameters of the risk quantification model based on the adoption and improvement status.

[0057] Pushing management suggestions to users is not the end of the process. If suggestions are simply pushed without monitoring their effectiveness, it's impossible to know whether the suggestions were adopted by users, whether adoption truly benefits the battery, and thus, the quality of the suggestions cannot be continuously improved. Therefore, after pushing suggestions, it's necessary to continuously track changes in user behavior and battery status.

[0058] User adoption of management suggestions can be reflected through continued collection of vehicle usage data. Once a suggestion is pushed out, an observation window can be set up to focus on changes in vehicle usage behavior related to that suggestion. For example, if the suggestion is "reduce the number of fast charging sessions," then during the subsequent observation period, the actual number of fast charging sessions performed by the user can be counted and compared with the target value in the suggestion to determine whether the user has adopted the suggestion.

[0059] The improvement in battery performance needs to be assessed in conjunction with long-term changes in battery health. Battery degradation is a slow, cumulative process, and significant changes are difficult to observe in a short period. Therefore, the assessment of improvement can be based on battery health data over a longer time span, such as comparing the rate of battery degradation several months before and after adopting the recommendations. Alternatively, a predictive approach can be used, based on the degree of improvement in user behavior and incorporating the impact coefficients in a risk quantification model, to estimate the expected benefit to battery life over a future period if the user continues to maintain the improved behavior pattern. This predicted benefit can serve as a quantitative indicator of improvement, providing feedback to users to create positive incentives.

[0060] After obtaining information on adoption and improvement, subsequent management strategies and the model itself can be optimized and adjusted.

[0061] One aspect is optimizing suggestion strategies. Different suggestion types, push timings, and wording can lead to varying degrees of user adoption. By accumulating a large amount of user adoption feedback data, an effectiveness score can be calculated for each suggestion strategy. This score considers both adoption rate and improvement effect; for example, strategies with high adoption rates and significant improvement effects receive higher scores, while strategies with low adoption rates or poor improvement effects receive lower scores. Based on the effectiveness score, the system can dynamically adjust the priority of subsequent suggestion pushes, prioritizing those proven and more effective strategies. For ineffective strategies, adjustments to wording, push timing, or replacement with other suggestion types can be attempted.

[0062] On the other hand, it involves calibrating the parameters of the risk quantification model. The impact coefficients in the risk quantification model are constructed based on historical sample data, reflecting the statistical regularities of the group. However, the actual impact of user behavior on the battery may differ depending on the vehicle model, battery type, and even the user's usage environment. As the system accumulates more and more user adoption feedback data, it can obtain real data pairs on "battery improvement after user behavior improvement." Using this real feedback data, the original impact coefficients can be corrected and calibrated. For example, if a large amount of feedback data shows that the actual impact of a certain behavior is less than the model's predicted impact, the impact coefficient of that behavior can be appropriately lowered; conversely, if the actual impact is greater than the prediction, the impact coefficient can be increased. This continuous calibration based on real feedback allows the risk quantification model to continuously approach the reality, and the prediction accuracy will become increasingly higher.

[0063] Through the aforementioned closed-loop mechanism, the embodiments of this application are no longer a one-way "monitoring-reminder" tool, but a management system with self-learning and continuous evolution capabilities. Every piece of user feedback and every change in behavior helps the system become smarter and more accurate, thereby achieving continuous optimization of battery life management.

[0064] This application provides a vehicle battery life management method, device, electronic device, and storage medium. By collecting vehicle usage data to characterize user vehicle usage behavior, and quantifying the usage behavior into an impact coefficient on the battery health degradation rate based on a risk quantification model, a quantitative correlation between user behavior and battery life is achieved. By determining the warning threshold for each usage behavior in each cycle based on the impact coefficient, personalized battery management suggestions can be generated and pushed in a timely manner when user behavior reaches the preset threshold, achieving accurate identification and proactive intervention of user risky behavior. By determining the user's adoption of management suggestions and battery improvement based on vehicle usage data, and optimizing the suggestion strategy and calibrating the parameters of the risk quantification model accordingly, a closed-loop management mechanism of "identification-intervention-feedback-optimization" is formed.

[0065] Compared with the prior art, the embodiments of this application can incorporate user vehicle usage behavior into the battery life management system, realize accurate quantitative correlation between user behavior and battery life, and continuously optimize management strategies based on user feedback, thereby effectively improving the accuracy of battery life management and the effectiveness of user behavior guidance.

[0066] In a feasible implementation plan, such as Figure 2 As shown, the construction method of the risk quantification model includes the following steps: Step 201: Obtain historical usage data and corresponding battery health degradation data for multiple sample vehicles.

[0067] The construction of a risk quantification model needs to be based on real data. Therefore, it is necessary to obtain historical data from a large number of sample vehicles as the basis for modeling. The number of sample vehicles should be sufficient to cover different vehicle types, usage scenarios, and user behavior patterns to ensure that the risk quantification model has statistical significance and generalization ability.

[0068] Historical vehicle usage data includes various usage records of sample vehicles over a relatively long period (e.g., more than two years), covering multiple dimensions such as charging behavior, discharging behavior, and parking behavior. Battery health degradation data refers to the record of changes in the battery health status of sample vehicles within the same time period, usually presented in the form of SOH values. These two types of data need to be matched one-to-one, that is, the usage data of the same vehicle within the same time range must be matched with its battery health degradation data to establish a correlation between behavior and degradation.

[0069] Step 202: Extract the vehicle usage behavior characteristics of the sample vehicles in each period based on the historical vehicle usage data.

[0070] Raw vehicle usage data is typically fine-grained time-series data, which cannot be directly used for modeling and analysis. This raw data needs to be processed to extract feature indicators that can characterize user behavior patterns. The extraction time period can be set as needed, for example, on a monthly basis, summarizing various behaviors that occur within a month.

[0071] The method for extracting vehicle usage behavior features depends on the specific behavior type. For charging behavior, features such as the number of fast charges per month, the number of full charges per month, and the average depth of charge can be extracted; for discharging behavior, features such as the number of rapid accelerations per month, the number of rapid decelerations per month, and the average depth of discharge can be extracted; for parking behavior, features such as the average duration of parking after each full charge and the number of times the vehicle is parked in high temperatures per month can be extracted. These features quantify user behavior habits from different perspectives and constitute the set of independent variables for subsequent modeling.

[0072] Step 203: Calculate the battery health degradation rate of each of the sample vehicles based on the battery health degradation data.

[0073] Battery health degradation data is typically represented as a time-varying SOH (State of Health) sequence. To quantify the rate of battery degradation for each sample vehicle, a degradation rate indicator needs to be extracted from this sequence. Since the decrease in SOH is a slow, cumulative process, short-term fluctuations do not represent the true degradation trend; therefore, calculations based on data over a longer time span are necessary.

[0074] A common approach is to use linear regression to fit the SOH (State of Health) trend of each vehicle over time, and then use the slope obtained from the regression as the average monthly degradation rate of that vehicle. This slope value reflects the rate at which the battery health of the vehicle degrades within the statistical period. For vehicles with non-linear degradation characteristics, piecewise fitting or other methods can also be used, but the core objective remains the same: to use a quantifiable value to represent the battery degradation rate of each vehicle as the dependent variable for subsequent modeling.

[0075] Step 204: Perform regression analysis with the vehicle usage behavior characteristics as independent variables and the battery health degradation rate as dependent variable to obtain the influence coefficient of each vehicle usage behavior on the battery health degradation rate.

[0076] After obtaining the usage behavior characteristics and corresponding battery degradation rates of a large number of sample vehicles, a mathematical relationship between the two can be established. Regression analysis is a commonly used modeling method. Its basic idea is to find a set of coefficients that allow the relationship between various usage behavior characteristics and battery degradation rates to be approximated by a linear or nonlinear mathematical expression.

[0077] Taking multiple linear regression as an example, the battery degradation rate of each vehicle can be represented as a weighted sum of various driving behavior characteristics, plus a constant term and an error term. Through optimization algorithms such as least squares, the weight coefficients that minimize the prediction error can be calculated. These coefficients are the influence coefficients of each driving behavior on the battery degradation rate, and their values ​​reflect the degree to which the behavior contributes to battery degradation. A positive influence coefficient indicates that the behavior accelerates battery degradation, and the larger the coefficient, the more significant the accelerating effect; a negative coefficient indicates that the behavior slows down degradation, although this is less common in practice. In this way, the originally abstract relationship between behavior and degradation is transformed into concrete, calculable coefficients, providing a scientific basis for subsequent threshold setting and personalized recommendations.

[0078] In constructing the risk quantification model, this embodiment acquires historical vehicle usage data and corresponding battery health degradation data from a large number of sample vehicles, extracts usage behavior characteristics and calculates battery degradation rates, and then uses regression analysis to establish a quantitative correlation between the two, obtaining the influence coefficients of each usage behavior on the battery health degradation rate. This process transforms the previously difficult-to-quantify relationship between user behavior and battery life into specific, calculable influence coefficients, providing a scientific basis and foundation for subsequent precise warning threshold setting, personalized suggestion generation, and continuous model calibration, thereby significantly improving the accuracy and reliability of battery life management.

[0079] In one feasible implementation, the vehicle usage behavior includes at least one of the following: Risky charging behavior, which includes at least one of fast charging behavior, full charging behavior, or deep charging behavior.

[0080] Risky discharge behavior, which includes at least one of rapid acceleration behavior, rapid deceleration behavior, or undervoltage discharge behavior.

[0081] Risky parking behavior, which includes at least one of the following: parking with a full charge, parking with a low charge, or parking in a high-temperature environment.

[0082] Risky environmental interaction behaviors include at least one of the following: charging behavior in a high-temperature environment, charging behavior in a low-temperature environment, or using substandard charging piles.

[0083] In actual vehicle use, various risky behaviors can affect battery health. To achieve comprehensive monitoring and accurate identification of user behavior, this application's implementation plan categorizes the vehicle usage behaviors requiring attention into four main types. The first type is risky charging behavior, mainly referring to actions that may damage the battery during vehicle charging, such as frequent use of high-power fast charging, always charging the battery to full capacity, or frequently performing deep charging. The second type is risky discharging behavior, mainly referring to actions that may damage the battery during driving, such as frequent rapid acceleration or deceleration leading to high-current discharge, or frequently depleting the battery to an undervoltage state. The third type is risky parking behavior, mainly referring to states that may damage the battery while the vehicle is parked, such as prolonged parking in a fully charged state, prolonged parking in a depleted state, or prolonged parking in high-temperature environments. The fourth type is risky environmental interaction behavior, mainly referring to situations where user use of the vehicle in specific external environments may damage the battery, such as charging in high-temperature environments, charging in low-temperature environments, or frequently using substandard charging stations that do not meet standards.

[0084] By systematically identifying and classifying the aforementioned risk behaviors, we can more comprehensively cover various scenarios that may damage battery health during daily vehicle use. This definition provides a clear scope for subsequent behavior quantification, threshold setting, and personalized recommendations, enabling battery life management to move beyond a few typical scenarios and effectively monitor complex and diverse risk behaviors, thereby further improving the comprehensiveness and precision of management.

[0085] In one feasible implementation, determining the warning threshold for each of the vehicle usage behaviors in each period based on the influence coefficient includes: A preset upper limit for loss is obtained for a reference period, the length of which is longer than the period. Based on the influence coefficient corresponding to each vehicle usage behavior, the upper limit for loss in the reference period is allocated to each vehicle usage behavior to obtain the allowable loss amount for each vehicle usage behavior within the reference period. Based on the allowable loss amount and corresponding influence coefficient for each vehicle usage behavior within the reference period, the maximum allowable occurrence number of each vehicle usage behavior within the reference period is calculated. Based on the maximum allowable occurrence number of each vehicle usage behavior within the reference period, a warning threshold for each period within the reference period is determined for each vehicle usage behavior.

[0086] When determining warning thresholds for various vehicle usage behaviors, a core issue needs to be considered: what standard should be used to judge "how much is too much"? Relying solely on impact coefficients only tells us "how harmful" a behavior is, but not "how many times it can occur." Therefore, a feasible implementation plan introduces the concept of a "wear and tear limit" as a benchmark for judgment.

[0087] The upper limit for battery degradation is a preset allowable total degradation over a relatively long reference period (e.g., one year), which can be understood as an "annual budget" for battery life management. This budget represents the expectation of controlling the additional battery degradation caused by user behavior within an acceptable range. Since different driving behaviors cause varying degrees of battery damage, this total budget needs to be allocated based on the "impact coefficient" corresponding to each behavior. Behaviors with a high impact coefficient mean that the "budget" consumed per unit of behavior is higher, and therefore the allocated allowable degradation amount will be correspondingly constrained; behaviors with a low impact coefficient can have a relatively more lenient allowable degradation amount. In this way, the total budget within the reference period is reasonably broken down into various specific behaviors, resulting in the allowable degradation amount for each behavior within the reference period.

[0088] After obtaining the permissible losses for each action within the reference period, and combining this with the corresponding "impact coefficient," the maximum permissible number of times each action can occur within the reference period can be deduced. This number reflects the maximum number of times a particular action can occur within the entire reference period, provided it does not exceed the total budget. For example, if an action incurs a loss of 0.002% of SOH each time it occurs, and the permissible loss for that action within the reference period is 0.012% of SOH, then its maximum permissible number of occurrences within the reference period is 6 times.

[0089] Finally, since actual monitoring and alerts are conducted on shorter periods (e.g., one month), the maximum allowable number of occurrences within the reference period needs to be broken down into smaller periods. This breakdown can be done evenly or non-evenly based on seasonality or user behavior habits. The resulting warning thresholds for each short period are the actual standards used to determine whether user behavior has exceeded the limits. Through this hierarchical method from "total budget" to "single behavior threshold," the setting of warning thresholds is no longer arbitrary but has clear, quantifiable basis, achieving a scientific mapping from lifespan management goals to specific behavioral interventions.

[0090] In one feasible implementation, the method further includes: Obtain the current state parameters of the battery; when the state parameters meet the preset maintenance trigger conditions, predict the idle time window when the user does not use the vehicle based on the vehicle usage data; within the idle time window, push a maintenance negotiation request to the interactive terminal; after receiving the confirmation instruction for the maintenance negotiation request, perform battery maintenance operations within the idle time window, the maintenance operations including at least one of battery rest, temperature adjustment, charging, or parameter calibration.

[0091] Building upon the aforementioned embodiments, the vehicle can also possess an active maintenance function. During daily use, certain key battery parameters can deviate over time and with accumulated usage. For example, the battery charge display may become inaccurate due to a lack of long-term calibration, and the battery health status estimate may deviate from the true value due to a lack of calibration conditions. If such parameter deviations accumulate, they not only affect user experience but may also pose safety hazards. However, traditional solutions often rely on specific passive usage scenarios for parameter calibration, such as requiring the vehicle to remain idle for a sufficient period at a specific charge level—conditions that are difficult to meet naturally in actual user use.

[0092] To address this issue, a feasible implementation plan involves acquiring the battery's current status parameters in real time and determining whether preset maintenance trigger conditions are met. These trigger conditions can be set based on the characteristics of different parameter types. For example, a trigger might occur when the accumulated error in battery estimation exceeds a certain threshold, or when the time since the last health status calibration or the equivalent number of cycles exceeds a baseline value. When the conditions are met, maintenance is not performed immediately. Instead, the system first learns the user's usage patterns based on historical vehicle usage data to predict idle time windows during which the user will not use the vehicle and will be connected to a charging station. For example, if it is discovered that the user charges their vehicle at home every Wednesday evening after 10 PM until 7 AM the following morning, this period can be identified as a usable idle window.

[0093] After predicting a suitable idle time window, the system will push a maintenance consultation request to the user via the vehicle's infotainment screen or mobile app, informing the user that the system plans to perform battery maintenance during the upcoming idle period, and explaining the approximate time required for maintenance and that it will not affect the user's expected driving time. The user can choose to allow this time, always allow, or decline. The maintenance operation will only be automatically executed within the agreed idle time window after receiving confirmation from the user. The specific content of the maintenance operation can be configured as needed, such as allowing the battery to rest for a period of time to stabilize its state, adjusting the battery temperature to a suitable range if necessary, and then charging it in a specific manner, while calibrating the battery's charge or health parameters during the process.

[0094] This proactive scheduling, user consultation, and automatic execution approach not only solves the problem of calibration conditions not being met naturally in traditional solutions, but also avoids forcibly occupying vehicle resources without the user's knowledge. It enables regular maintenance of battery parameters while ensuring the user's vehicle usage needs, which helps maintain the accuracy of battery management in the long term.

[0095] In one feasible implementation, determining the user's adoption of the management recommendations based on the vehicle usage data includes: An observation window is set up to correspond to the management suggestion, and the length of the observation window is one or more of the periods. Within the observation window, the actual behavior value of the target vehicle use behavior corresponding to the management suggestion is extracted from the vehicle use data. The suggested target value for the target vehicle use behavior in the management suggestion is obtained. Based on the difference between the actual behavior value and the suggested target value, the adoption status of the management suggestion is determined.

[0096] After pushing management suggestions to users, it's necessary to know whether they have actually adopted them in order to evaluate their effectiveness and provide a basis for subsequent optimization. To this end, a feasible implementation plan sets a corresponding observation window for each management suggestion. The length of the observation window can be flexibly set according to the type of behavior targeted by the suggestion. For example, for behaviors like charging habits that can be changed in the short term, a week can be set as the observation window; for behaviors like fully charged parking that require a longer time to observe changes, a month or longer can be set. The length of the observation window can be one or more of the aforementioned monitoring periods to maintain consistency with the overall management system's cycle settings.

[0097] Within the observation window, user vehicle usage data is continuously collected, and actual behavioral values ​​related to the target vehicle usage behavior in relation to the current management recommendations are extracted from this data. For example, if the previously pushed recommendation was "limit fast charging to no more than 3 times per week," then the observation window will count the actual number of fast charging incidents and obtain the actual behavioral value. If the recommendation was "avoid parking with a fully charged battery," then the total or average duration of parking with a fully charged battery within the observation window will be counted. At the same time, it is also necessary to obtain the target value set for this target vehicle usage behavior in the original management recommendation. This target value can be a specific number of times, duration, or other quantifiable indicators.

[0098] After obtaining the actual behavior value and the suggested target value, the two are compared, and the difference between them determines the user's adoption of the management suggestion. This difference can be measured in several ways, such as calculating the ratio of the actual value to the target value, or calculating the absolute difference between the two. If the actual value is close to or better than the suggested target value, it indicates that the user has adopted the suggestion well; if the actual value is not much different from the suggested target value, it indicates that the user has partially adopted the suggestion; if the actual value is significantly worse than the suggested target value or even worse than before the suggestion, it indicates that the user has not adopted the suggestion. Through this quantitative evaluation of adoption, the actual effect of each suggestion can be clearly understood, providing a reliable data foundation for subsequent strategy optimization.

[0099] In one feasible implementation, optimizing the recommended strategy and calibrating the parameters of the risk quantification model based on the adoption and improvement outcomes includes: Based on the adoption and improvement status, the effectiveness score of each suggested strategy is calculated; the effectiveness score is positively correlated with the improvement status of the battery; the push priority of different suggested strategies is adjusted according to the effectiveness score; and the impact coefficient in the risk quantification model is updated using the accumulated adoption and improvement status.

[0100] After obtaining user feedback on management suggestions and battery improvements, the system can optimize its operational logic. In a feasible implementation, an effectiveness score needs to be calculated for each suggested strategy. This score quantifies the actual effectiveness of the strategy. The calculation of the effectiveness score needs to consider two aspects: user adoption and the actual battery improvement after adoption. Generally, strategies with high adoption rates and significant improvement results receive higher scores; strategies with low adoption rates or insignificant improvement results receive lower scores. Specifically, the effectiveness score should be positively correlated with battery improvement; that is, the better the improvement, the higher the score. This guides the system to adopt strategies that are truly beneficial to battery health.

[0101] After calculating the effectiveness scores of each suggested strategy, the push priority of different strategies can be adjusted based on the scores. Strategies with high scores have proven to be more effective in practice and can be prioritized in subsequent notifications. For strategies with low scores, their push frequency can be reduced, or the specific content, wording, and timing of the push can be adjusted to improve their effectiveness. Through this dynamic adjustment, suggested strategies can be continuously optimized and gradually approach the most suitable model for the current user group.

[0102] Furthermore, the parameters of the risk quantification model itself can be calibrated using accumulated adoption and improvement data. The impact coefficients in the risk quantification model are initially constructed based on historical sample data, reflecting group statistical patterns. As the running time increases, a large amount of real user feedback data accumulates, forming real samples of "battery changes before and after behavior improvement." Using these new samples, regression analysis can be performed again to update the impact coefficients of various driving behaviors on the rate of battery health degradation. For example, if a large amount of feedback data indicates that the actual impact of a certain behavior is less than the model's original prediction, the impact coefficient of that behavior can be appropriately lowered; conversely, it can be increased. Through this continuous calibration, the risk quantification model can continuously approach reality, and the prediction accuracy will improve accordingly, thus enabling the entire battery life management system to have the ability to learn and continuously evolve.

[0103] This application provides a vehicle battery life management method, device, electronic device, and storage medium. By collecting vehicle usage data to characterize user vehicle usage behavior, and quantifying the usage behavior into an impact coefficient on the battery health degradation rate based on a risk quantification model, a quantitative correlation between user behavior and battery life is achieved. By determining the warning threshold for each usage behavior in each cycle based on the impact coefficient, personalized battery management suggestions can be generated and pushed in a timely manner when user behavior reaches the preset threshold, achieving accurate identification and proactive intervention of user risky behavior. By determining the user's adoption of management suggestions and battery improvement based on vehicle usage data, and optimizing the suggestion strategy and calibrating the parameters of the risk quantification model accordingly, a closed-loop management mechanism of "identification-intervention-feedback-optimization" is formed.

[0104] Compared with the prior art, the embodiments of this application can incorporate user vehicle usage behavior into the battery life management system, realize accurate quantitative correlation between user behavior and battery life, and continuously optimize management strategies based on user feedback, thereby effectively improving the accuracy of battery life management and the effectiveness of user behavior guidance.

[0105] Based on the same technical concept, embodiments of this application also provide a vehicle battery life management device, such as... Figure 3 As shown, the device includes: The data acquisition module 301 is used to collect vehicle usage data; the usage data is used to characterize the user's vehicle usage behavior.

[0106] The quantification module 302 is used to quantify the vehicle usage behavior into an impact coefficient on the rate of battery health degradation based on a pre-built risk quantification model.

[0107] The determination module 303 is used to determine the warning threshold for each of the vehicle usage behaviors in each period based on the influence coefficient.

[0108] The push module 304 is used to generate personalized battery management suggestions and push them to the interactive terminal when a certain vehicle use behavior is detected to reach the warning threshold in any cycle.

[0109] The optimization module 305 is used to determine the user's adoption of the management suggestions and the improvement of the battery based on the vehicle usage data, and to optimize the suggestion strategy and calibrate the parameters of the risk quantification model based on the adoption and improvement.

[0110] In one feasible implementation, the risk quantification model is constructed in the following ways: Obtain historical usage data and corresponding battery health degradation data for multiple sample vehicles.

[0111] Based on the historical vehicle usage data, the usage behavior characteristics of sample vehicles in each period are extracted.

[0112] The battery health degradation rate of each sample vehicle is calculated based on the battery health degradation data.

[0113] Regression analysis was performed using the vehicle usage behavior characteristics as independent variables and the battery health degradation rate as dependent variable to obtain the influence coefficient of each vehicle usage behavior on the battery health degradation rate.

[0114] In one feasible implementation, the vehicle usage behavior includes at least one of the following: Risky charging behavior, which includes at least one of fast charging behavior, full charging behavior, or deep charging behavior.

[0115] Risky discharge behavior, which includes at least one of rapid acceleration behavior, rapid deceleration behavior, or undervoltage discharge behavior.

[0116] Risky parking behavior, which includes at least one of the following: parking with a full charge, parking with a low charge, or parking in a high-temperature environment.

[0117] Risky environmental interaction behaviors include at least one of the following: charging behavior in a high-temperature environment, charging behavior in a low-temperature environment, or using substandard charging piles.

[0118] In one feasible implementation, the push module is used to determine the warning threshold for each of the vehicle usage behaviors in each period based on the influence coefficient, for the following purposes: Obtain a preset upper limit for loss for a reference period, wherein the duration of the reference period is greater than the period.

[0119] Based on the influence coefficient corresponding to each of the aforementioned vehicle usage behaviors, the upper limit of the reference period loss is allocated to each vehicle usage behavior to obtain the allowable loss amount of each vehicle usage behavior within the reference period.

[0120] Based on the allowable wear and tear of each vehicle usage behavior within the reference period and the corresponding influence coefficient, the maximum allowable number of occurrences of each vehicle usage behavior within the reference period is calculated.

[0121] Based on the maximum permissible number of occurrences of each vehicle usage behavior within the reference period, the warning threshold for each period within the reference period is determined.

[0122] In one feasible implementation, the device further includes: The parameter acquisition module is used to acquire the current status parameters of the battery.

[0123] The prediction module is used to predict the idle time window when the status parameters meet the preset maintenance trigger conditions, based on the vehicle usage data.

[0124] The maintenance push module is used to push maintenance negotiation requests to the interactive terminal during the idle time window.

[0125] The maintenance module is used to perform battery maintenance operations within the idle time window after receiving a confirmation instruction in response to the maintenance negotiation request. The maintenance operations include at least one of battery resting, temperature adjustment, charging, or parameter calibration.

[0126] In one feasible implementation, the optimization module is configured to determine the user's adoption rate of the management recommendations based on the vehicle usage data, and is configured to: Set an observation window corresponding to the management suggestion, the length of the observation window being one or more of the aforementioned periods.

[0127] Within the observation window, the actual behavior value of the target vehicle usage behavior corresponding to the management suggestion is extracted from the vehicle usage data.

[0128] Obtain the suggested target value for the target vehicle usage behavior from the management recommendations.

[0129] The adoption status of the management recommendations is determined based on the difference between the actual behavior value and the suggested target value.

[0130] In one feasible implementation, the optimization module is configured to optimize the recommended strategy and calibrate the parameters of the risk quantification model based on the adoption status and the improvement status, for the following purposes: Based on the adoption status and the improvement status, an effectiveness score is calculated for each suggested strategy; the effectiveness score is positively correlated with the improvement status of the battery.

[0131] The push priority of different suggestion strategies is adjusted based on the performance score.

[0132] The impact coefficient in the risk quantification model is updated using the accumulated adoption and improvement data.

[0133] This application provides a vehicle battery life management method, device, electronic device, and storage medium. By collecting vehicle usage data to characterize user vehicle usage behavior, and quantifying the usage behavior into an impact coefficient on the battery health degradation rate based on a risk quantification model, a quantitative correlation between user behavior and battery life is achieved. By determining the warning threshold for each usage behavior in each cycle based on the impact coefficient, personalized battery management suggestions can be generated and pushed in a timely manner when user behavior reaches the preset threshold, achieving accurate identification and proactive intervention of user risky behavior. By determining the user's adoption of management suggestions and battery improvement based on vehicle usage data, and optimizing the suggestion strategy and calibrating the parameters of the risk quantification model accordingly, a closed-loop management mechanism of "identification-intervention-feedback-optimization" is formed.

[0134] Compared with the prior art, the embodiments of this application can incorporate user vehicle usage behavior into the battery life management system, realize accurate quantitative correlation between user behavior and battery life, and continuously optimize management strategies based on user feedback, thereby effectively improving the accuracy of battery life management and the effectiveness of user behavior guidance.

[0135] Figure 4 A schematic diagram of an electronic device provided in this application embodiment includes: a processor 401, a storage medium 402, and a bus 403. The storage medium 402 stores machine-readable instructions executable by the processor 401. When the electronic device runs the vehicle battery life management method as described in the embodiment, the processor 401 communicates with the storage medium 402 via the bus 403, and the processor 401 executes the machine-readable instructions to perform the steps as described in the embodiment.

[0136] In this embodiment, the storage medium 402 may also execute other machine-readable instructions to perform other methods as described in the embodiment. For details on the specific execution steps and principles, please refer to the description of the embodiment, which will not be repeated here.

[0137] This application also provides a computer-readable storage medium storing a computer program that is executed by a processor to perform the steps as described in the embodiments.

[0138] In this embodiment, the computer program, when run by the processor, can also execute other machine-readable instructions to perform other methods as described in the embodiments. For details on the specific execution steps and principles, please refer to the description of the embodiments, which will not be repeated here.

[0139] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interface; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0140] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0141] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0142] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0143] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for managing the lifespan of a vehicle battery, characterized in that, The method includes: Collect vehicle usage data; the usage data is used to characterize the user's vehicle usage behavior; Based on a pre-built risk quantification model, the vehicle usage behavior is quantified as an impact coefficient on the rate of battery health degradation; Based on the aforementioned impact coefficient, the warning threshold for each of the aforementioned vehicle usage behaviors in each period is determined; During any given period, when a vehicle usage behavior is detected to reach the warning threshold, a personalized battery management suggestion is generated and pushed to the interactive terminal. Based on the vehicle usage data, the user's adoption of the management recommendations and the improvement of the battery are determined. Based on the adoption and improvement, the recommendation strategy is optimized and the parameters of the risk quantification model are calibrated.

2. The method according to claim 1, characterized in that, The risk quantification model is constructed in the following ways: Acquire historical usage data and corresponding battery health degradation data for multiple sample vehicles; Based on the historical vehicle usage data, the vehicle usage behavior characteristics of the sample vehicles in each period are extracted. The battery health degradation rate of each of the sample vehicles is calculated based on the battery health degradation data. Regression analysis was performed using the vehicle usage behavior characteristics as independent variables and the battery health degradation rate as dependent variable to obtain the influence coefficient of each vehicle usage behavior on the battery health degradation rate.

3. The method according to claim 1, characterized in that, The vehicle use behavior includes at least one of the following: Risky charging behavior, which includes at least one of fast charging behavior, full charging behavior, or deep charging behavior; Risky discharge behavior, wherein the risky discharge behavior includes at least one of rapid acceleration behavior, rapid deceleration behavior, or undervoltage discharge behavior; Risky parking behavior, which includes at least one of the following: parking with a full charge, parking with a low charge, or parking in a high-temperature environment; Risky environmental interaction behaviors include at least one of the following: charging behavior in a high-temperature environment, charging behavior in a low-temperature environment, or using substandard charging piles.

4. The method according to claim 1, characterized in that, The determination of the early warning threshold for each of the vehicle usage behaviors in each period based on the influence coefficient includes: Obtain a preset upper limit for loss for a reference period, wherein the duration of the reference period is greater than the period. Based on the influence coefficient corresponding to each of the aforementioned vehicle usage behaviors, the upper limit of the reference period loss is allocated to each vehicle usage behavior to obtain the allowable loss amount of each vehicle usage behavior within the reference period. Based on the allowable wear and tear of each vehicle use behavior within the reference period and the corresponding influence coefficient, calculate the maximum allowable number of occurrences of each vehicle use behavior within the reference period. Based on the maximum permissible number of occurrences of each vehicle usage behavior within the reference period, the warning threshold for each period within the reference period is determined.

5. The method according to claim 1, characterized in that, The method further includes: Get the current state parameters of the battery; When the status parameters meet the preset maintenance trigger conditions, the idle time window when the user does not use the vehicle is predicted based on the vehicle usage data. During the idle time window, a maintenance negotiation request is pushed to the interactive terminal; Upon receiving a confirmation instruction in response to the maintenance negotiation request, a battery maintenance operation is performed within the idle time window. The maintenance operation includes at least one of battery resting, temperature adjustment, charging, or parameter calibration.

6. The method according to claim 1, characterized in that, The process of determining the user's adoption of the management suggestions based on the vehicle usage data includes: Set an observation window corresponding to the management suggestion, the length of the observation window being one or more of the aforementioned periods; Within the observation window, the actual behavior value of the target vehicle usage behavior corresponding to the management suggestion is extracted from the vehicle usage data; Obtain the suggested target value for the target vehicle usage behavior from the management recommendations; The adoption status of the management recommendations is determined based on the difference between the actual behavior value and the suggested target value.

7. The method according to claim 1, characterized in that, The process of optimizing the recommended strategy and calibrating the parameters of the risk quantification model based on the adoption and improvement outcomes includes: Based on the adoption status and the improvement status, an effectiveness score is calculated for each suggested strategy; the effectiveness score is positively correlated with the improvement status of the battery. The push priority of different suggestion strategies is adjusted based on the aforementioned performance score; The impact coefficients in the risk quantification model are updated using the accumulated adoption and improvement data.

8. A vehicle battery life management device, characterized in that, The device includes: The data collection module is used to collect vehicle usage data; the usage data is used to characterize the user's vehicle usage behavior. The quantification module is used to quantify the vehicle usage behavior into an impact coefficient on the rate of battery health degradation based on a pre-built risk quantification model. The determination module is used to determine the warning threshold for each of the vehicle usage behaviors in each period based on the influence coefficient; The push module is used to generate personalized battery management suggestions and push them to the interactive terminal when a certain vehicle use behavior reaches the warning threshold within any cycle. The optimization module is used to determine the user's adoption of the management suggestions and the improvement of the battery based on the vehicle usage data, and to optimize the suggestion strategy and calibrate the parameters of the risk quantification model based on the adoption and improvement.

9. An electronic device, characterized in that, include: The device includes a processor, a storage medium, and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is in operation, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the vehicle battery life management method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the vehicle battery life management method as described in any one of claims 1 to 7.

Images