New energy vehicle battery use health degree evaluation method, device and equipment

Abstract

The invention discloses a new energy vehicle battery use health degree evaluation method, device and equipment, and relates to the technical field of new energy vehicle big data analysis. According to the method, the new energy vehicles are grouped according to the vehicle basic information, and then the user behavior characteristics representing the user charging and discharging habits and the driving behaviors in the user behavior data of each new energy vehicle in each group of new energy vehicles are extracted; the difference value between the ideal battery health degree attenuation and the actual battery health degree attenuation under the user behavior data is determined, and the relative aging rate difference value is obtained to serve as the label of the user behavior characteristics, so that the neural network model is trained and applied; therefore, for the target new energy vehicle, the relative aging rate difference value can be predicted based on the user behavior characteristics so as to evaluate the battery use health degree, the specific influence and quantitative influence of the user behavior mode on the battery health can be visually reflected, the bad use habit of the user can be improved, and the service life of the new energy vehicle can be prolonged.

Description

Technical Field

[0001] This invention relates to the field of big data analysis technology for new energy vehicles, and in particular to a method, apparatus, and equipment for assessing the health of new energy vehicle batteries. Background Technology

[0002] Currently, the State of Health (SOH) of power batteries for new energy vehicles is a key indicator for evaluating battery performance, which usually represents the degree of degradation in capacity or internal resistance of the battery relative to its initial state.

[0003] In existing technologies, current mainstream SOH estimation techniques, whether based on equivalent circuit models, electrochemical models, or data-driven algorithms, mostly focus on the physical and electrochemical states inside the battery. These methods analyze operating data such as voltage, current, and temperature to accurately track microscopic degradation mechanisms such as active lithium ion loss, electrode material phase transitions, and solid electrolyte interface film thickening.

[0004] However, during the use of new energy vehicles, users' usage habits often have a significant impact on battery aging. Existing methods for assessing the health of new energy vehicle batteries are difficult to intuitively reflect the specific and quantitative impact of user behavior patterns on battery health. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, device, and equipment for assessing the health of new energy vehicle batteries to address the aforementioned technical issues.

[0006] The present invention adopts the following technical solution: This invention provides a method for assessing the health of new energy vehicle batteries, comprising: Acquire basic vehicle information of several new energy vehicles, user behavior data within a historical time window, and battery internal status data within a historical time window; the user behavior data includes at least one of charging and discharging event data, driving behavior data, and operating condition data; New energy vehicles are grouped based on their basic vehicle information. For each new energy vehicle in each group, user behavior features are extracted from the user behavior data of that new energy vehicle. Based on the battery internal state data within the historical time window, determine the actual battery health degradation per unit mileage of the new energy vehicle within the historical time window, and determine the ideal battery health degradation per unit mileage under ideal usage conditions. Based on the difference between the ideal battery health degradation and the actual battery health degradation, determine the relative aging rate difference corresponding to the user behavior data. Using user behavior features as input to a neural network model, the relative aging rate difference corresponding to user behavior data is predicted to train the neural network model. Acquire basic vehicle information and user behavior data of the target new energy vehicle, extract user behavior features of the target new energy vehicle, and predict the relative aging rate difference through a trained neural network model. Determine the battery health score of the target new energy vehicle based on the relative aging rate difference.

[0007] Optionally, the basic vehicle information includes: vehicle identification number, vehicle model, vehicle age, cumulative mileage, and operating area; The battery internal status data includes: the battery health assessed internally by the battery management system; The charge / discharge event data includes at least one of the following: charging start / end timestamp, charging start / end state of charge, charging amount, charging power curve, charging type, average ambient temperature during charging, and average battery temperature during charging. The driving behavior data includes at least one of the following: trip start / end timestamp, mileage, instantaneous vehicle speed curve, accelerator pedal opening curve, brake pedal opening curve, and average ambient temperature during driving. The operating condition data includes at least one of the following: hill climbing conditions, load conditions, and road conditions.

[0008] Optionally, the extraction of user behavior features representing user charging and discharging habits and driving behavior from the user behavior data of the new energy vehicle specifically includes: Based on charge and discharge event data, extract at least one of the following: frequency of charging the new energy vehicle to a value greater than a first state of charge threshold, frequency of starting charging from a value lower than a second state of charge threshold, percentage of charging times exceeding a charging rate threshold, average charging rate, percentage of charging duration exceeding a charging rate threshold, percentage of standby time exceeding a third state of charge threshold, percentage of charging above a first temperature threshold, and percentage of charging below a second temperature threshold; and / or, Based on driving behavior data, extract at least one of the following: the percentage of accelerations exceeding a first threshold, the percentage of decelerations exceeding a second threshold, the average vehicle speed, the percentage of driving below a fourth charge threshold, and the percentage of driving time at an ambient temperature exceeding a third temperature threshold; and / or, Based on the operating condition data, extract at least one of the following for the new energy vehicle: heavy-load mileage percentage, hill-climbing mileage percentage, and highway mileage percentage.

[0009] Optionally, the battery health score of the target new energy vehicle can be determined based on the difference in relative aging rates using the following formula: ; in, Rate the battery's health status. For weighted parameters, This represents the difference in relative aging rates. This is the segmentation threshold.

[0010] Optionally, the method further includes: The contribution of different user behavior characteristics to the relative aging rate difference of the target new energy vehicle was analyzed by the SHAP method, and several user behavior characteristics with the greatest impact on the battery health of the target new energy vehicle were identified. When it is determined that some values ​​of several user behavior features fall within an optimizable range, several optimization suggestions are determined based on the corresponding user behavior. Based on the impact of user behavior characteristics corresponding to each optimization suggestion on the battery health of the target new energy vehicle and the optimization space of the user behavior characteristics, the optimization suggestions are weighted, ranked, and fed back to the user.

[0011] This invention provides a device for assessing the health of new energy vehicle batteries, comprising: The acquisition module is used to acquire basic vehicle information of several new energy vehicles, user behavior data within a historical time window, and battery internal status data within a historical time window; the user behavior data includes at least one of charging and discharging event data, driving behavior data, and operating condition data. The extraction module is used to group new energy vehicles according to their basic vehicle information, and extract user behavior features from the user behavior data of each new energy vehicle in each group. The annotation module is used to determine the actual battery health degradation per unit mileage of the new energy vehicle within the historical time window based on the battery internal state data within the historical time window, and to determine the ideal battery health degradation per unit mileage under ideal use conditions. Based on the difference between the ideal battery health degradation and the actual battery health degradation, the relative aging rate difference corresponding to the user behavior data is determined. The training module is used to predict the relative aging rate difference corresponding to user behavior data by using user behavior features as input to the neural network model, so as to train the neural network model. The evaluation module is used to acquire basic vehicle information and user behavior data of the target new energy vehicle, extract user behavior features of the target new energy vehicle, predict the relative aging rate difference through a trained neural network model, and determine the battery health score of the target new energy vehicle based on the relative aging rate difference.

[0012] The present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for assessing the health of new energy vehicle batteries.

[0013] The present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the above-mentioned method for assessing the health of new energy vehicle batteries.

[0014] The above-mentioned at least one technical solution adopted in this invention can achieve the following beneficial effects: This invention first groups new energy vehicles based on their basic vehicle information. Then, for each new energy vehicle in each group, it extracts user behavior features from the user behavior data that represent the user's charging and discharging habits and driving behavior. Next, it determines the difference between the ideal battery health degradation and the actual battery health degradation under the user behavior data, obtaining the relative aging rate difference as a label for the user behavior feature. This label is then used to train and apply a neural network model. Thus, for a target new energy vehicle, the relative aging rate difference can be predicted based on the user behavior feature to assess the battery health. This can intuitively reflect the specific and quantitative impact of user behavior patterns on battery health, which is beneficial for users to improve bad usage habits and extend the service life of new energy vehicles. Attached Figure Description

[0015] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0016] Figure 1 This invention provides a schematic flowchart of a method for assessing the health of new energy vehicle batteries. Figure 2 This invention provides a flowchart illustrating the process from battery health analysis to providing personalized optimization suggestions. Figure 3 A schematic diagram of a new energy vehicle battery health assessment device provided by the present invention; Figure 4 This invention provides a schematic diagram of a computer device for implementing a method for assessing the health of new energy vehicle batteries. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0018] Generally, battery aging is affected by a variety of factors, including the number of charge-discharge cycles, temperature, current rate, and calendar time. User habits, such as frequent high-rate charge-discharge, prolonged exposure to high or low temperature environments, and frequent stays at high or low SOC states, have a significant impact on battery aging.

[0019] However, existing SOH assessment methods often focus on the internal physical or electrochemical state of the battery, making it difficult to intuitively reflect the specific and quantitative impact of user behavior patterns on battery health. On the other hand, changes in battery SOH are a slow and irreversible process, which makes traditional SOH display or prediction methods unable to reflect the quality of recent user behavior in a timely manner, and unable to provide users with short-term, perceptible behavioral feedback.

[0020] Furthermore, existing battery maintenance recommendations based on battery health assessments are mostly general guidelines, such as "avoid overcharging and over-discharging" and "use slow charging whenever possible." These recommendations lack specificity. Users cannot know which of their specific behaviors have the greatest impact on battery health, nor can they obtain tailored, easier-to-implement optimization suggestions based on their own usage patterns. This significantly limits the initiative and effectiveness of users in improving their behavior.

[0021] The technical solutions provided by the various embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] Figure 1 This is a schematic diagram of a method for assessing the health of new energy vehicle batteries according to the present invention, which specifically includes the following steps: S101: Obtain basic vehicle information of several new energy vehicles, user behavior data within a historical time window, and battery internal status data within a historical time window; the user behavior data includes at least one of charging and discharging event data, driving behavior data, and operating condition data.

[0023] S102: Group new energy vehicles according to their basic vehicle information, and extract user behavior features from the user behavior data of each new energy vehicle in each group.

[0024] S103: Based on the battery internal state data within the historical time window, determine the actual battery health degradation per unit mileage of the new energy vehicle within the historical time window, and determine the ideal battery health degradation per unit mileage under ideal usage conditions. Based on the difference between the ideal battery health degradation and the actual battery health degradation, determine the relative aging rate difference corresponding to the user behavior data.

[0025] S104: Using user behavior features as input to the neural network model, predict the relative aging rate difference corresponding to user behavior data in order to train the neural network model.

[0026] S105: Obtain the basic vehicle information and user behavior data of the target new energy vehicle, extract the user behavior characteristics of the target new energy vehicle, and predict the relative aging rate difference through the trained neural network model. Determine the battery health score of the target new energy vehicle based on the relative aging rate difference.

[0027] For ease of explanation, the following description focuses solely on the server as the executing entity. The server mentioned in this invention can be a server set up on a business platform, or a device such as a desktop computer or laptop computer capable of executing the solution of this invention.

[0028] This invention provides a method for assessing the health of new energy vehicle batteries, overcoming the shortcomings of existing battery health indicators in terms of user behavior feedback. By collecting and analyzing multi-dimensional user behavior data, the method quantifies the impact of user behavior on battery aging rate within a preset period (such as a week or month), generating a periodic indicator that sensitively reflects the quality of user behavior.

[0029] Based on this, the periodic indicator can be transformed into a user-friendly battery health user behavior score with a positive incentive mechanism. Simultaneously, a deep analysis and quantification of the specific contribution of each user behavior characteristic to the score can be conducted, and key behaviors leading to accelerated battery degradation can be accurately identified. This allows for the provision of users with clearly targeted and highly actionable personalized battery usage optimization suggestions, ultimately helping users improve bad habits, effectively slowing down battery degradation, and extending battery life.

[0030] This invention primarily uses a neural network model to assess the health of new energy vehicle batteries; therefore, the neural network model is trained first. This allows for the acquisition of training data and annotations.

[0031] In one or more embodiments of the present invention, the server can obtain basic vehicle information of several new energy vehicles, user behavior data within a historical time window, and battery internal state data within a historical time window.

[0032] The vehicle's basic information includes: Vehicle Identification Number (VIN), vehicle model, vehicle age, cumulative mileage, and operating area. The operating area refers to the main locations where the vehicle operates, such as different cities at the city level, or even larger regions. Internal battery status data includes: battery health assessed internally by the Battery Management System (BMS). Charge / discharge event data includes at least one of the following: charging start / end timestamps, charging start / end state of charge, charge amount, charging power curve, charging type, average ambient temperature during charging, and average battery temperature during charging. Driving behavior data includes at least one of the following: trip start / end timestamps, mileage, instantaneous speed curve, accelerator pedal opening curve, brake pedal opening curve, and average ambient temperature during driving. Operating condition data includes at least one of the following: hill climbing conditions, load conditions, and road conditions. All this data can be obtained through the vehicle terminal T-BOX and the Battery Management System (BMS). All data can include timestamps, and all curve-type data can be aligned by timestamps for calculating percentages, averages, and other statistical characteristics.

[0033] Furthermore, the server can group vehicles based on cumulative mileage, vehicle age, operating area, and vehicle type; the data from different vehicle groups can then be used to train different neural network models and make predictions through these models. It can also extract and quantify user behavior data, preprocessing the collected raw data and extracting various predefined behavioral features within a preset time window that quantify user charging and discharging habits, driving behavior, and operating conditions. These features are controllable by the user or influenced by their driving or charging habits.

[0034] In one or more embodiments of the present invention, the server may extract at least one of the following based on the charging and discharging event data of the new energy vehicle: the frequency of charging to a value greater than a first state of charge threshold, the frequency of charging from a value lower than a second state of charge threshold, the percentage of charging times greater than a charging rate threshold, the average charging rate, the percentage of charging duration greater than a charging rate threshold, the percentage of idle time greater than a third state of charge threshold, the percentage of charging above a first temperature threshold, and the percentage of charging below a second temperature threshold; and / or, based on driving behavior data, extract at least one of the following: the percentage of acceleration greater than a first threshold, the percentage of deceleration greater than a second threshold, the average vehicle speed, the percentage of driving below a fourth state of charge threshold, and the percentage of running time at an ambient temperature greater than a third temperature threshold; and / or, based on operating condition data, extract at least one of the following: the percentage of heavy-load mileage, the percentage of uphill mileage, and the percentage of high-speed mileage.

[0035] For example, the raw user behavior data collected in step S101 can be preprocessed, including data cleaning, synchronization, interpolation, and filtering. Then, for a preset time window (one week or one month), features that can quantitatively characterize the user's charging and discharging habits and driving behavior can be extracted. Charging habit characteristics: frequency of charging to high SOC (e.g., >95%), frequency of charging from low SOC (e.g., <15%), percentage of fast charging usage, average charging rate, percentage of high charging rate duration, percentage of time left uncharged at high SOC (e.g., >90%), percentage of charging at high temperatures, percentage of charging at low temperatures, etc.

[0036] Driving habit characteristics: percentage of rapid acceleration from 100 km / h, percentage of rapid deceleration from 100 km / h, average vehicle speed, percentage of driving at low SOC (e.g., <20%), percentage of driving time at high temperatures, etc.

[0037] Operating conditions: percentage of heavy-load mileage, percentage of mileage climbing, percentage of mileage on highways, etc.

[0038] The extracted features can be used as sample data for training the neural network model. Furthermore, the difference in relative aging rate corresponding to user behavior data can be determined based on the difference between the ideal battery health degradation and the actual battery health degradation, and this can be used as a label for the sample data.

[0039] Specifically, the unit SOH decay of vehicles within the group can be calculated within the time window. And the ideal SOH decay determined based on the minimum decay within the group. ; Calculate the Aging Acceleration Index (AAI), where .

[0040] Based on this, a regression model is constructed that takes user behavior characteristics as input and the relative aging rate difference (AAI) as output, such as the neural network model Multilayer Perceptron (MLP). This invention does not limit the specific type of regression model.

[0041] For example, for vehicles within a group, calculate the required data: This includes the behavioral feature vector within the window, the estimated SOH values ​​at the beginning and end of the window, and the total mileage within the window.

[0042] Calculation window for unit SOH attenuation .

[0043] The ideal SOH change is calculated. The ideal SOH change refers to the expected amount of battery degradation after a certain period of use under ideal conditions. It can be obtained based on battery aging mechanisms (calendar aging, cycle aging) or a large amount of laboratory / historical data. Here, the ideal SOH degradation is determined based on the minimum degradation within the group. .

[0044] Calculate the difference in relative aging rates .

[0045] Training the regression model: Using the historical data above, select a suitable regression model for training. Independent variable (X): User behavior feature vector within each window. Dependent variable (y): Difference in relative aging rate within that window. This enables it to accurately predict the difference in relative aging rates under a given behavioral feature vector. .

[0046] Furthermore, in one or more embodiments of the present invention, the predicted AAI can also be converted into a rating of 0-100. The rating cycle is short (weekly / monthly), and the rating is based on the impact of user behavior on the aging rate, which can promptly reflect changes in recent user behavior, resulting in significant rating changes and providing immediate feedback and incentives.

[0047] For example, a monotonically decreasing function can be used, Score = max(A, 100 - C×AAI), or a piecewise function can be used, where, Rate the battery's health status. For weighted parameters, This represents the minimum score threshold (e.g., 80). A score approaches 100 when AAI is low, and drops rapidly when AAI is high. Parameter C and the segmentation threshold of the piecewise function need to be adjusted based on the actual data and the desired score distribution. The goal is to achieve high scores when AAI is 0 (or close to 0, i.e., user behavior is consistent with or better than the baseline behavior), and low scores when AAI is very high.

[0048] Furthermore, to directly quantify the impact of user behavior on aging rate and to quantitatively analyze and inform users of the specific negative impacts of their behaviors (such as excessive fast charging, prolonged storage at high SOC, and aggressive driving) on ​​battery health, allowing users to clearly understand the consequences of their habits, in one or more embodiments of this invention, the server can also analyze the contribution of different user behavior characteristics to the relative aging rate difference of the target new energy vehicle using the SHAP method, identifying several user behavior characteristics that have the greatest impact on the battery health of the target new energy vehicle; when it is determined that the values ​​of several user behavior characteristics fall within an optimizable range, several optimization suggestions are determined based on the corresponding user behavior; based on the degree of impact of the user behavior characteristics corresponding to each optimization suggestion on the battery health of the target new energy vehicle and the optimizable space of the user behavior characteristics, the optimization suggestions are weighted, ranked, and fed back to the user.

[0049] For the target user's score in the current period, the SHAP (Shapley Additive Explanations) value analysis method is used to explain which behavioral features led to this score. Since the user behavior score is derived from AAI (Average AI), and the score is negatively correlated with AAI, the SHAP value is directly calculated using a trained neural network model. SHAP value analysis calculates the contribution of each feature to the deviation of its AAI prediction from the baseline (average AAI value) for each user's prediction for each period. Positive values ​​indicate that the feature's performance leads to a higher AAI, while negative values ​​indicate that the feature's performance helps maintain a lower AAI. The specific direction (positive / negative) and magnitude of the contribution of each user's behavioral feature to its current AAI prediction value are obtained and ranked.

[0050] Features with the highest positive SHAP values ​​can be identified, such as "high SOC charging frequency," "high percentage of fast charging usage," and "frequent rapid acceleration." These are the main culprits behind low user ratings. Features with high negative SHAP values ​​can also be identified to praise the user's positive aspects.

[0051] For example, for a user J, the model predicts an AAI of 0.25. SHAP analysis shows that the baseline average predicted AAI for users in this category is 0.10. The analysis of user J's SHAP_values, ranked by their positive values, identifies the features with the greatest negative impact on AAI. For instance, SHAP_value(F_fast_charge) = 0.015 and SHAP_value(F_high_accel) = 0.008, indicating that the user's higher frequency of fast charging increased their predicted AAI by 1.5%, and more aggressive acceleration increased it by 0.8%. Both factors combined resulted in an AAI above average, with fast charging having the greatest impact.

[0052] Furthermore, personalized and actionable suggestions can be provided to users based on the impact. Based on quantitative impact analysis, the generated suggestions are no longer generic guidelines, but personalized suggestions that are proposed for each user’s most prominent bad habits, with clear directions for improvement and expected results, which significantly improves the relevance, acceptance and implementation effectiveness of the suggestions.

[0053] For example, based on the aforementioned quantitative results of behavioral impact, especially the characteristics that have a significant negative impact on the current user, and combined with a pre-set battery maintenance knowledge base and suggestion rule engine, one or more specific and personalized usage optimization suggestions can be generated for the user.

[0054] The suggested generation logic is as follows: Identify the list of behavioral characteristics that have the most significant negative impact on the current situation, identifying the Top-K characteristics with the highest SHAP (Shape, Scale, and Optimization). Examine the specific values ​​of users on these negative impact characteristics to determine whether they fall into the "undesirable" or "optimizable" range, such as a fast charging rate far exceeding the average or excessively long high SOC (State of Charge) dwell time. From the suggestion template library, match and select the most relevant optimization suggestions based on the identified undesirable behavioral characteristics. Prioritize the suggestions according to their impact and the user's potential for improvement.

[0055] Based on this, by guiding users to adopt personalized suggestions and improve bad usage habits, the actual aging rate of the battery can be effectively slowed down, the effective lifespan of the battery can be extended, the replacement cost for users can be reduced, and user satisfaction can be improved. Figure 2 This is a schematic diagram illustrating a process from battery health analysis to providing personalized optimization suggestions in this invention.

[0056] based on Figure 1 The proposed method for assessing the battery health of new energy vehicles involves first grouping new energy vehicles based on their basic information. Then, for each new energy vehicle in each group, user behavior features representing user charging and discharging habits and driving behavior are extracted from the user behavior data. The difference between the ideal battery health degradation and the actual battery health degradation under the user behavior data is then determined, resulting in a relative aging rate difference. This difference is used as a label for the user behavior features to train and apply a neural network model. Thus, for a target new energy vehicle, the relative aging rate difference can be predicted based on user behavior features to assess the battery health. This method can intuitively reflect the specific and quantitative impact of user behavior patterns on battery health, which is beneficial for users to improve poor usage habits and extend the service life of new energy vehicles.

[0057] When applying the new energy vehicle battery health assessment method provided by this invention, it is not necessary to consider... Figure 1 The steps shown are executed in sequence. The specific execution order of each step can be determined as needed, and this invention does not impose any restrictions on it.

[0058] The above describes a method for assessing the health of new energy vehicle batteries according to one or more embodiments of the present invention. Based on the same idea, the present invention also provides a corresponding device for assessing the health of new energy vehicle batteries, such as... Figure 3 As shown.

[0059] Figure 3 A schematic diagram of a new energy vehicle battery health assessment device provided by the present invention includes: The acquisition module 201 is used to acquire basic vehicle information of several new energy vehicles, user behavior data within a historical time window, and battery internal status data within a historical time window; the user behavior data includes at least one of charging and discharging event data, driving behavior data, and operating condition data. Extraction module 202 is used to group new energy vehicles according to vehicle basic information, and extract user behavior features from the user behavior data of each new energy vehicle in each group. The annotation module 203 is used to determine the actual battery health degradation per unit mileage of the new energy vehicle within the historical time window based on the battery internal state data within the historical time window, and to determine the ideal battery health degradation per unit mileage under ideal use conditions. Based on the difference between the ideal battery health degradation and the actual battery health degradation, the relative aging rate difference corresponding to the user behavior data is determined. The training module 204 is used to predict the relative aging rate difference corresponding to user behavior data by using user behavior features as input to the neural network model, so as to train the neural network model. The evaluation module 205 is used to acquire the basic vehicle information and user behavior data of the target new energy vehicle, extract the user behavior characteristics of the target new energy vehicle, predict the relative aging rate difference through the trained neural network model, and determine the battery health score of the target new energy vehicle based on the relative aging rate difference.

[0060] Specific limitations regarding the health assessment device for new energy vehicle batteries can be found in the limitations on the health assessment method for new energy vehicle batteries mentioned above, and will not be repeated here. Each module in the aforementioned health assessment device for new energy vehicle batteries can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0061] The present invention also provides a computer-readable storage medium storing a computer program that can be used to execute the above-described... Figure 1 The provided method for assessing the health of new energy vehicle batteries.

[0062] The present invention also provides Figure 4 The schematic diagram of the computer device shown is as follows: Figure 4As shown, at the hardware level, this computer device includes a processor, internal bus, network interface, memory, and non-volatile memory, and may also include other hardware required for business operations. The processor reads the corresponding computer program from the non-volatile memory into memory and then executes it to achieve the above. Figure 1 The provided method for assessing the health of new energy vehicle batteries.

[0063] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0064] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this invention.

Claims

1. A method for assessing the health of batteries used in new energy vehicles, characterized in that, include: Acquire basic vehicle information of several new energy vehicles, user behavior data within a historical time window, and battery internal status data within a historical time window; The user behavior data includes at least one of the following: charging and discharging event data, driving behavior data, and operating condition data; New energy vehicles are grouped based on their basic vehicle information. For each new energy vehicle in each group, user behavior features are extracted from the user behavior data of that new energy vehicle. Based on the battery internal state data within the historical time window, determine the actual battery health degradation per unit mileage of the new energy vehicle within the historical time window, and determine the ideal battery health degradation per unit mileage under ideal usage conditions. Based on the difference between the ideal battery health degradation and the actual battery health degradation, determine the relative aging rate difference corresponding to the user behavior data. Using user behavior features as input to a neural network model, the relative aging rate difference corresponding to user behavior data is predicted to train the neural network model. Acquire basic vehicle information and user behavior data of the target new energy vehicle, extract user behavior features of the target new energy vehicle, and predict the relative aging rate difference through a trained neural network model. Determine the battery health score of the target new energy vehicle based on the relative aging rate difference.

2. The method for assessing the health of new energy vehicle batteries as described in claim 1, characterized in that, The basic vehicle information includes: vehicle identification number, vehicle model, vehicle age, cumulative mileage, and operating area; The battery internal status data includes: the battery health assessed internally by the battery management system; The charge / discharge event data includes at least one of the following: charging start / end timestamp, charging start / end state of charge, charging amount, charging power curve, charging type, average ambient temperature during charging, and average battery temperature during charging. The driving behavior data includes at least one of the following: trip start / end timestamp, mileage, instantaneous vehicle speed curve, accelerator pedal opening curve, brake pedal opening curve, and average ambient temperature during driving. The operating condition data includes at least one of the following: hill climbing conditions, load conditions, and road conditions.

3. The method for assessing the health of new energy vehicle batteries as described in claim 1, characterized in that, The extraction of user behavior features representing user charging and discharging habits and driving behavior from the user behavior data of the new energy vehicle specifically includes: Based on charge and discharge event data, extract at least one of the following: frequency of charging the new energy vehicle to a value greater than a first state of charge threshold, frequency of starting charging from a value lower than a second state of charge threshold, percentage of charging times exceeding a charging rate threshold, average charging rate, percentage of charging duration exceeding a charging rate threshold, percentage of standby time exceeding a third state of charge threshold, percentage of charging above a first temperature threshold, and percentage of charging below a second temperature threshold; and / or, Based on driving behavior data, extract at least one of the following: the percentage of accelerations exceeding a first threshold, the percentage of decelerations exceeding a second threshold, the average vehicle speed, the percentage of driving below a fourth charge threshold, and the percentage of driving time at an ambient temperature exceeding a third temperature threshold; and / or, Based on the operating condition data, extract at least one of the following for the new energy vehicle: heavy-load mileage percentage, hill-climbing mileage percentage, and highway mileage percentage.

4. The method for assessing the health of new energy vehicle batteries as described in claim 1, characterized in that, The determination of the battery health score of the target new energy vehicle based on the difference in relative aging rates specifically includes: The battery health score of the target new energy vehicle is determined based on the difference in relative aging rates using the following formula: ； in, Rate the battery's health status. For weighted parameters, This represents the difference in relative aging rates. This is the minimum score threshold.

5. The method for assessing the health of new energy vehicle batteries as described in claim 1, characterized in that, The method further includes: The contribution of different user behavior characteristics to the relative aging rate difference of the target new energy vehicle was analyzed by the SHAP method, and several user behavior characteristics with the greatest impact on the battery health of the target new energy vehicle were identified. When it is determined that some values ​​of several user behavior features fall within an optimizable range, several optimization suggestions are determined based on the corresponding user behavior. Based on the impact of user behavior characteristics corresponding to each optimization suggestion on the battery health of the target new energy vehicle and the optimization space of the user behavior characteristics, the optimization suggestions are weighted, ranked, and fed back to the user.

6. A device for assessing the health of batteries used in new energy vehicles, characterized in that, include: The acquisition module is used to acquire basic vehicle information of several new energy vehicles, user behavior data within a historical time window, and battery internal status data within a historical time window. The user behavior data includes at least one of the following: charging and discharging event data, driving behavior data, and operating condition data; The extraction module is used to group new energy vehicles according to their basic vehicle information, and extract user behavior features from the user behavior data of each new energy vehicle in each group. The annotation module is used to determine the actual battery health degradation per unit mileage of the new energy vehicle within the historical time window based on the battery internal state data within the historical time window, and to determine the ideal battery health degradation per unit mileage under ideal use conditions. Based on the difference between the ideal battery health degradation and the actual battery health degradation, the relative aging rate difference corresponding to the user behavior data is determined. The training module is used to predict the relative aging rate difference corresponding to user behavior data by using user behavior features as input to the neural network model, so as to train the neural network model. The evaluation module is used to acquire basic vehicle information and user behavior data of the target new energy vehicle, extract user behavior features of the target new energy vehicle, predict the relative aging rate difference through a trained neural network model, and determine the battery health score of the target new energy vehicle based on the relative aging rate difference.

7. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the method as described in any one of claims 1 to 5.

8. A computer device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method as described in any one of claims 1 to 5.

Images