Method and device for determining charging duration of energy storage equipment in vehicle and vehicle

Abstract

The invention discloses a method and device for determining the charging duration of energy storage equipment in a vehicle and the vehicle. The method comprises the steps that in response to the situation that energy storage equipment and energy supply equipment are in a connection state, vehicle type information of a vehicle is obtained; a target prediction model corresponding to the vehicle type information is determined from the multiple candidate prediction models, and the accuracy of the charging duration of the energy storage equipment determined by using the target prediction model is greater than the accuracy of the charging duration of the energy storage equipment determined by using candidate prediction models except the target prediction model in the multiple candidate prediction models; the target prediction model is used for predicting the charging duration based on the state information of the energy storage equipment and the charging demand information of the energy storage equipment, the state information is used for representing the charging state of the energy storage equipment in the environment where the vehicle is located, and the charging demand information is used for representing the demand for charging the energy storage equipment through the energy supply equipment. The technical problem that the charging duration determination accuracy of the energy storage equipment is low is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle charging, in particular to a method and device for determining charging duration of an energy storage device in a vehicle and a vehicle. BACKGROUND

[0002] In the related art, the charging duration prediction method mainly depends on the current power grid electricity load and the user required charging amount to estimate the charging completion time. Although the above method is simple, it still has the technical problem of low accuracy of determining the charging duration of the energy storage device in the vehicle, because the differences in battery characteristics between different electric vehicles are ignored.

[0003] At present, there is no effective solution to the above technical problems. SUMMARY

[0004] The embodiments of the present application provide a method and device for determining charging duration of an energy storage device in a vehicle and a vehicle to at least solve the technical problem of low accuracy of determining the charging duration of the energy storage device.

[0005] According to an aspect of an embodiment of the present application, a method for determining charging duration of an energy storage device in a vehicle is provided. The vehicle includes an energy storage device, and the method includes: in response to the energy storage device being in a connected state with a power supply device, obtaining vehicle model information; from a plurality of candidate prediction models, determining a target prediction model corresponding to the vehicle model information, wherein the accuracy of determining the charging duration of the energy storage device using the target prediction model is greater than the accuracy of determining the charging duration of the energy storage device using candidate prediction models other than the target prediction model in the plurality of candidate prediction models; and using the target prediction model, predicting the charging duration based on state information of the energy storage device and charging demand information of the energy storage device, wherein the state information is used to indicate the charging state of the energy storage device in the environment of the vehicle, and the charging demand information is used to indicate the demand for charging the energy storage device using the power supply device.

[0006] Optionally, using the target prediction model, predicting the charging duration based on the state information of the energy storage device and the charging demand information of the energy storage device includes: using the target prediction model, determining power change information of the energy storage device based on the state information, wherein the power change information is used to indicate the degree of change in charging power of the energy storage device in the charging state; and using the target prediction model, predicting the charging duration based on the power change information and the charging demand information.

[0007] Optionally, the method further includes: obtaining vehicle model samples, state information samples, and environmental state samples for training an initial prediction model; and training the initial prediction model using the vehicle model samples, the state information samples, and the environmental state samples to obtain a candidate prediction model corresponding to the vehicle model samples.

[0008] Optionally, the initial prediction model is trained by using the vehicle type sample, the state information sample and the environment state sample to obtain a candidate prediction model corresponding to the vehicle type sample, including: sending the vehicle type sample, the state information sample and the environment state sample to a cloud of a vehicle manufacturer corresponding to the vehicle; performing grouping processing on the state information sample and the environment state sample according to the vehicle type sample by using the cloud of the vehicle manufacturer to obtain grouped state information sample and grouped environment state sample; performing denoising processing on the grouped state information sample and the grouped environment state sample respectively to obtain denoised state information sample and denoised environment state sample; and training the initial prediction model corresponding to the vehicle type sample by using the denoised state information sample and the denoised environment state sample to obtain the candidate prediction model.

[0009] Optionally, the vehicle type sample includes a model sample and a configuration sample, the grouping processing on the state information sample and the environment state sample according to the vehicle type sample by using the cloud of the vehicle manufacturer includes: performing grouping on the state information sample and the environment state sample according to the model sample by using the cloud of the vehicle manufacturer to obtain grouped first-layer state information sample and grouped first-layer environment state sample; and performing grouping on the first-layer state information sample and the first-layer environment state sample according to the configuration sample by using the cloud of the vehicle manufacturer to obtain grouped second-layer state information sample and grouped second-layer environment state sample; and the method further includes: performing grouping on the second-layer state information sample and the second-layer environment state sample according to a maximum current output by a matched energy supply device sample of a vehicle sample corresponding to the vehicle type sample to obtain grouped third-layer state information sample and grouped third-layer environment state sample.

[0010] Optionally, the method further includes: updating the vehicle type sample, the state information sample and the environment state sample according to a target period to obtain updated vehicle type sample, updated state information sample and updated environment state sample.

[0011] Optionally, the initial prediction model is trained by using the vehicle type sample, the state information sample and the environment state sample to obtain a candidate prediction model corresponding to the vehicle type sample, including: training the initial prediction model by using the updated vehicle type sample, the updated state information sample and the updated environment state sample to obtain the candidate prediction model.

[0012] According to another aspect of the embodiments of the present application, a device for determining charging duration of an energy storage device in a vehicle is provided. The device can include: an obtaining unit configured to obtain vehicle model information in response to the energy storage device of the vehicle being in a connected state with a power supply device; a determining unit configured to determine, from a plurality of candidate prediction models, a target prediction model corresponding to the vehicle model information, wherein an accuracy of determining the charging duration of the energy storage device using the target prediction model is greater than an accuracy of determining the charging duration of the energy storage device using a candidate prediction model other than the target prediction model from the plurality of candidate prediction models; and a predicting unit configured to predict the charging duration using the target prediction model based on state information of the energy storage device and charging demand information of the energy storage device, wherein the state information is used to indicate a charging state of the energy storage device in an environment of the vehicle, and the charging demand information is used to indicate a demand for charging the energy storage device using the power supply device.

[0013] According to another aspect of the embodiments of the present application, a computer readable storage medium is provided. The computer readable storage medium includes a stored program, wherein the program, when executed, controls a device in which the computer readable storage medium is located to perform the above method according to the embodiments of the present application.

[0014] According to another aspect of the embodiments of the present application, a processor is provided. The processor is configured to execute a program, wherein the program, when executed, performs the above method according to the embodiments of the present application.

[0015] According to another aspect of the embodiments of the present application, an electronic device is provided. The electronic device includes a memory and a processor. The memory stores a computer program, and the processor is configured to execute the computer program to perform the above method according to the embodiments of the present application.

[0016] According to another aspect of the embodiments of the present application, a computer program product is provided. The computer program product includes a computer program, and the computer program, when executed by a processor, implements the above method according to the embodiments of the present application.

[0017] According to another aspect of the embodiments of the present application, a vehicle is provided. The vehicle includes a memory and a processor. The memory stores an executable program, and the processor is configured to execute the program, and the program, when executed, implements the above method according to the embodiments of the present application.

[0018] In this embodiment of the invention, in response to the connection between the energy storage device and the energy supply device, the vehicle model information is acquired. From multiple candidate prediction models, a target prediction model corresponding to the vehicle model information is determined. The accuracy of determining the charging time of the energy storage device using the target prediction model is greater than the accuracy of determining the charging time of the energy storage device using candidate prediction models other than the target prediction model. Using the target prediction model, the charging time is predicted based on the state information and charging demand information of the energy storage device. The state information represents the charging state of the energy storage device in the vehicle's environment, and the charging demand information represents the need to charge the energy storage device using the energy supply device. In other words, in this embodiment, the accuracy of determining the vehicle charging time is significantly improved by implementing a personalized prediction model. Addressing the problem of inaccurate predictions caused by neglecting individual vehicle differences and operating condition variables in related technologies, an improvement measure is proposed. When the vehicle establishes a connection with the charging pile, the system immediately collects the vehicle model information and selects the most suitable prediction model for that vehicle model from a pre-set model library, ensuring that the accuracy of this model in estimating charging time surpasses all other models in the library. By employing the aforementioned prediction model, which comprehensively considers the battery's real-time state and the user's charging requirements (such as the amount of charge needed), the charging completion time is dynamically predicted. This overcomes the limitations of general prediction methods, enabling customized and highly accurate charging time prediction based on different vehicle characteristics and environmental conditions. This significantly optimizes the charging experience and efficiency, solves the technical problem of low accuracy in determining the charging time of energy storage devices in vehicles, and achieves the technical effect of improving the accuracy of determining the charging time of energy storage devices in vehicles. Attached Figure Description

[0019] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0020] Figure 1 This is a flowchart of a method for determining the charging time of an energy storage device in a vehicle according to an embodiment of the present invention;

[0021] Figure 2 This is a schematic diagram of a charging time estimation process based on big data according to an embodiment of the present invention;

[0022] Figure 3 This is a schematic diagram of a charging time determination device for an energy storage device in a vehicle according to an embodiment of the present invention. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0024] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] According to an embodiment of the present invention, an embodiment of a method for determining the charging time of an energy storage device in a vehicle is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0026] Figure 1 This is a flowchart of a method for determining the charging time of an energy storage device in a vehicle according to an embodiment of the present invention, as shown below. Figure 1 As shown, the method may include the following steps:

[0027] In step S102, in response to the connection between the energy storage device and the energy supply device, the vehicle model information is obtained.

[0028] In the technical solution provided by step S102 of the present invention, the vehicle includes an energy storage device.

[0029] Optionally, energy storage devices can refer to on-board batteries or power batteries, serving as energy storage units for vehicles to store electrical energy for use by the vehicle's drive system. The performance of these energy storage devices (such as capacity, charging efficiency, and health status) directly affects the vehicle's range and charging characteristics. Energy supply devices can be charging piles, including but not limited to public charging piles and home charging piles. Charging piles are infrastructure that provides power to electric vehicle batteries, establishing an electrical connection with the electric vehicle's energy storage devices to transmit electrical energy and charge the vehicle. The connection state refers to the electrical and communication contact state established between the energy storage device and the energy supply device. In a charging scenario, when the electric vehicle's charging interface is connected to the charging gun of the charging pile, and the two can communicate data, they enter a connection state. In this connection state, the charging pile can read the current status of the electric vehicle's battery, such as remaining charge and battery temperature, to optimize the charging strategy.

[0030] Optionally, vehicle model information may include specific vehicle configurations and models, such as vehicle brand, model, battery type, battery capacity, and vehicle electrical system configuration. This vehicle model information is used to select a suitable predictive model, as different vehicle models may have significantly different battery and charging characteristics, affecting charging time. By accurately obtaining vehicle model information, the predictive system can call the most suitable model, improving prediction accuracy.

[0031] In this embodiment, if the energy storage device and the functional device are in a connected state, the vehicle model information can be obtained.

[0032] Optionally, when an electric vehicle approaches a charging station to begin charging, the charging station's control system begins listening for signals at the charging interface to detect any connection request signals. The charging station's charging gun is inserted into the electric vehicle's charging socket, forming a physical connection. At this point, the charging station's mechanical locking mechanism may activate to ensure a secure connection. The charging station and the electric vehicle establish a communication link through the connected charging interface. The charging station can authenticate the electric vehicle to confirm whether it is on the authorized charging list or to verify permissions through a vehicle-to-everything (V2X) service platform. These steps ensure the security of the charging service.

[0033] Optionally, once the charging pile's control system confirms the connection's legitimacy, it can send a request to the electric vehicle's Battery Management System (BMS) or onboard information system to inquire about the vehicle's model information. Upon receiving the request, the electric vehicle extracts model information, including vehicle brand, model, and battery specifications, from its onboard database or BMS and sends it back to the charging pile via a communication interface. After receiving the model information, the charging pile can parse it and extract key parameters.

[0034] Optionally, the charging station stores the parsed vehicle information in a local database or on a cloud server for use in subsequent steps. Based on the received vehicle information, the charging station's control system or backend server searches a pre-built predictive model library for the charging time prediction model that best matches the vehicle information. This is to improve the accuracy of the charging time prediction. The predictive model selected by the charging station will be used in the next round of charging time estimation processing, ensuring that the charging service is personalized based on the specific attributes of the vehicle.

[0035] Step S104: Determine the target prediction model corresponding to the vehicle model information from multiple candidate prediction models.

[0036] In the technical solution provided by step S104 of the present invention, the accuracy of determining the charging time of the energy storage device using the target prediction model is greater than the accuracy of determining the charging time of the energy storage device using candidate prediction models other than the target prediction model among multiple candidate prediction models.

[0037] Optionally, a candidate prediction model can refer to a set of pre-developed and trained mathematical models or algorithms designed to predict the charging time of electric vehicles. Each candidate prediction model is designed to solve the problem of estimating charging time under specific conditions, such as different vehicle models, battery types, and charging conditions (temperature, charging rate, etc.). The candidate prediction models constitute a model library containing candidate prediction models for various scenarios.

[0038] Optionally, the target prediction model can be the optimal model selected from candidate prediction models, tailored to specific vehicle model information. This target prediction model is chosen because it performs best among numerous candidate prediction models in terms of accuracy in predicting vehicle charging time for specific vehicle model information. The selection criterion for the target prediction model is to better reflect the battery characteristics and charging behavior related to that vehicle model, thereby providing the most accurate charging time prediction result. This target prediction model can be a charging time estimation model.

[0039] In this embodiment, after obtaining the vehicle model information, the target prediction model corresponding to the current vehicle model information can be determined from multiple candidate prediction models.

[0040] Optionally, when an electric vehicle connects to a charging station and begins communication, the charging station's control system or backend server initiates a model library retrieval process to select the most suitable prediction model based on vehicle information. The charging station receives vehicle model information from the electric vehicle's BMS or onboard information system, including but not limited to key parameters such as vehicle model, battery type, and battery capacity. Subsequently, the system analyzes this vehicle model information to determine the vehicle's basic configuration and battery characteristics. Based on the vehicle model information analysis results, the system performs a preliminary screening from a pre-set candidate prediction model library to identify a subset of models that match or are closest to the vehicle model information. These models are typically already labeled or categorized for rapid identification.

[0041] Optionally, for the initially selected matching models, the predictive performance of each candidate prediction model is further evaluated. This includes historical prediction accuracy, model adaptability, and performance under specific operating conditions. The evaluation process may use machine learning algorithms, statistical analysis, or other quantitative methods to ensure the objectivity and accuracy of the evaluation. Based on the model performance evaluation results, a target prediction model is determined, which significantly outperforms other candidate models in the accuracy of predicting charging time. Selection criteria may include the model's generalization ability, prediction accuracy under specific conditions, and its matching degree with the current vehicle condition.

[0042] In this embodiment of the application, the above method can dynamically select the most suitable target prediction model from multiple candidate prediction models based on the vehicle model information and current operating conditions, thereby achieving high-precision prediction of the charging time of energy storage devices and solving the problem of inaccurate prediction caused by insufficient model universality in traditional methods.

[0043] Step S106: Using the target prediction model, based on the state information of the energy storage device and the charging demand information of the energy storage device, predict the charging time.

[0044] In the technical solution of step S106 of the present invention, the status information is used to indicate the charging status of the energy storage device in the environment where the vehicle is located. The charging demand information is used to indicate the demand for charging the energy storage device using the energy supply device.

[0045] Optionally, the status information may include, but is not limited to: charging map information, battery capacity, battery cycle count, battery state of charge (SOC), maximum power after vehicle-charging station matching, SOC changes during charging, and cell temperature changes. The charging map information can be a detailed record of charging characteristics, including the battery's charging response and characteristics under different charging conditions (such as current and voltage), used to understand the battery's charging behavior patterns. Battery capacity refers to the current electrical energy stored in the battery, usually expressed as a percentage of remaining capacity or a specific capacity value. Battery cycle count indicates the number of charge-discharge cycles the battery has undergone since manufacturing, affecting the battery's health and performance. SOC represents the ratio of the current battery capacity to a fully charged capacity. Maximum power after vehicle-charging station matching refers to the maximum power the charging station can charge the vehicle at under the current vehicle-charging station matching condition, affecting charging speed and charging time. SOC changes during charging record the rate or trend of SOC change during charging, helping to predict remaining charging time. The above cell temperature changes can be used to represent the temperature changes of the cells inside the battery.

[0046] Optionally, charging demand information can be used to represent a user's specific charging requirements, which may include, but are not limited to, charging restrictions. These charging restrictions can refer to a user-set charging deadline, a desired percentage of charge, or a limit on maximum charging power, and are one of the important factors the system needs to consider when predicting charging duration.

[0047] In this embodiment, after determining the target prediction model corresponding to the vehicle model information from multiple candidate prediction models, the target prediction model can be used to predict the charging time based on the state information of the energy storage device and the charging demand of the energy storage device.

[0048] Optionally, comprehensive status information about the energy storage device (battery) and user charging demand information can be collected. Status information includes charging characteristic mapping (map information), current battery capacity, battery cycle count, battery state of charge (SOC), maximum charging power after vehicle-charging station matching, and SOC and cell temperature changes during charging. Charging demand information mainly involves user-set charging cutoff conditions, such as desired SOC level, charging time limits, or power demand limits. The collected information may require preprocessing, including data cleaning, format standardization, and validity verification, to ensure the accuracy and consistency of all input information and avoid the impact of abnormal data on prediction results.

[0049] Optionally, a target prediction model is loaded, and its parameters are adjusted based on the current environment and vehicle conditions. This may include fine-tuning the model's internal structure or weights based on the latest temperature, battery health status, or charging demand to achieve more accurate predictions. The processed state information and charging demand information are input as features into the target prediction model. These steps ensure that the model fully considers all factors affecting charging time, including but not limited to the current battery state, external environmental conditions, and the user's specific charging needs. Based on the input features, the target prediction model calculates the predicted charging time using its internal algorithms or mathematical relationships. This prediction is based on all collected and processed information, as well as the model's predictive capabilities.

[0050] In steps S102 to S106 of this application, in response to the connection between the energy storage device and the energy supply device, vehicle model information is obtained; from multiple candidate prediction models, a target prediction model corresponding to the vehicle model information is determined, wherein the accuracy of determining the charging time of the energy storage device using the target prediction model is greater than the accuracy of determining the charging time of the energy storage device using candidate prediction models other than the target prediction model; using the target prediction model, the charging time is predicted based on the state information and charging demand information of the energy storage device, wherein the state information represents the charging state of the energy storage device in the vehicle's environment, and the charging demand information represents the need for the energy supply device to charge the energy storage device. In other words, in this embodiment, the accuracy of determining the vehicle charging time is significantly improved by implementing a personalized prediction model. To address the issue of inaccurate predictions caused by neglecting individual vehicle differences and operating condition variables in related technologies, an improvement measure is proposed. Upon establishing a connection between the vehicle and the charging station, the system immediately collects the vehicle's model information and selects the most suitable prediction model from a pre-built model library. This ensures that the model's accuracy in estimating charging time surpasses all other models in the library. Using this prediction model, the system dynamically predicts charging completion time by comprehensively considering the battery's real-time state and the user's charging requirements (such as the required charging amount). This overcomes the limitations of general prediction methods, achieving customized and highly accurate charging time prediction based on different vehicle characteristics and environmental conditions. This significantly optimizes the charging experience and efficiency, solves the technical problem of low accuracy in determining the charging time of energy storage devices in vehicles, and achieves the technical effect of improving the accuracy of charging time determination for energy storage devices in vehicles.

[0051] The method described in this embodiment will be further described below.

[0052] As an optional embodiment, step S106, using a target prediction model, predicts the charging time based on the state information and charging demand information of the energy storage device, including: using the target prediction model, based on the state information, determining the power change information of the energy storage device, wherein the power change information is used to represent the degree of change in the charging power of the energy storage device during the charging state; and using the target prediction model, based on the power change information and the charging demand information, predicting the charging time.

[0053] In this embodiment, the power change information can be a power curve, which describes the change in charging power of an energy storage device (such as the battery of an electric vehicle) over time during the charging process. This power change information can be represented as a graph, with the horizontal axis representing time and the vertical axis representing charging power, thus visually demonstrating the dynamic changes in power during charging. The power change information reflects the nonlinear characteristics of the charging process and is closely related to the battery's physical characteristics, charging strategy, and external factors (such as power limitations of the power supply equipment and ambient temperature). In predicting charging time, power change information is one of the core input parameters because it directly determines the charging rate and the time required to complete charging. By comprehensively analyzing battery state information and power change information, the target prediction model can more accurately predict the time required for the battery to reach the user-set charging demand from its current state throughout the entire charging cycle, thereby providing more accurate charging time predictions and optimizing the charging experience.

[0054] Optionally, in the process of predicting charging duration using a target prediction model based on the state information and charging demand information of the energy storage device, the target prediction model can be used to determine the power change information of the energy storage device based on the state information. The target prediction model can then be used to predict charging duration based on power change information and charging demand information.

[0055] Optionally, the current state information of the energy storage device is collected, including but not limited to battery charge, SOC status, battery temperature, number of charging cycles, and the maximum charging power matched between the vehicle and the charging station. This state information is preprocessed to ensure data accuracy and consistency, preparing it for model prediction. The previously determined optimal target prediction model is selected. This model, trained with large datasets, can handle various state information and predict power changes under different conditions. The collected and processed state information is used as input, and the target prediction model is applied to generate power change information, i.e., a power curve. In the above steps, the target prediction model can consider battery characteristics, environmental factors, and real-time status, outputting a predicted charging power value that varies over time.

[0056] Optionally, the predicted power change information is analyzed, and the power curve may be further optimized by combining battery safety thresholds and charging station power limitations to ensure that the prediction conforms to both physical laws and practical operational constraints. Simultaneously, the user's set charging requirements are confirmed, such as the desired state of charge (SOC), power limitations during charging, and time limitations. Based on the optimized power change information and charging requirements, the system uses a target prediction model to perform comprehensive calculations to predict the time required to complete charging. This typically involves integral calculations, calculating the area under the power curve to obtain the accumulated charging energy, and then, based on user needs and battery characteristics, estimating the total time required to reach the charging target. The predicted charging time is post-processed, potentially converting the results into a more intuitive format, such as converting time to hours and minutes, and then presenting the results to the user through the vehicle's display system or the charging station's user interface. During charging, battery status and environmental changes are continuously monitored, new data is collected, and the power change information and predicted charging time are adjusted in real time. This dynamic update mechanism ensures the real-time nature and accuracy of the prediction results, helps to promptly respond to changes that may occur during charging, and provides users with updated charging time estimates.

[0057] In the embodiments of this application, the above method can generate a dynamic change prediction of charging power based on the real-time status information of the energy storage device and the user's specific charging needs, and accurately predict the time required to complete charging, providing strong support for the optimization of the charging process and the user's time arrangement.

[0058] As an optional embodiment, the method further includes: acquiring vehicle model samples, state information samples, and environmental state samples of the initial prediction model to be trained; and training the initial prediction model using the vehicle model samples, state information samples, and environmental state samples to obtain candidate prediction models corresponding to the vehicle model samples.

[0059] In this embodiment, vehicle model samples, state information samples, and environmental state samples can be obtained for the initial prediction model to be trained. These vehicle model samples, state information samples, and environmental state samples can then be used to train the initial prediction model, resulting in candidate prediction models corresponding to the vehicle model samples.

[0060] Optionally, a large amount of data samples related to electric vehicle charging can be collected from enterprise cloud or charging history records. These samples include information on different vehicle models, battery status information (such as charging map information, battery capacity, battery cycle count, SOC status, etc.), and environmental status information (such as battery cell temperature, ambient temperature, charging pile type, etc.). This ensures data diversity to cover various possible charging scenarios. The collected data is cleaned and preprocessed, including outlier removal, missing value imputation, data format standardization, and feature engineering, converting non-numerical data into numerical data to facilitate model training. Data preprocessing is a crucial step in ensuring effective model training. The processed dataset is divided into training, validation, and test sets. The training set is used for model training, the validation set for model parameter tuning, and the test set for evaluating the final model performance. This data partitioning helps ensure the model's generalization ability.

[0061] Optionally, the architecture of the initial prediction model is defined. This may include selecting the model type (e.g., neural network, decision tree, regression model), defining model parameters (e.g., learning rate, number of layers, number of nodes), and initializing weights and biases. Using vehicle model samples, state information samples, and environmental state samples as input, and the actual experienced charging time as the output label, the initial prediction model is trained. The model parameters are optimized using the backpropagation algorithm, enabling the model to learn the complex relationship between battery charging time and vehicle model, battery state, and environmental state from the data. During model training, validation set data is used to evaluate model performance, and model parameters are adjusted to improve prediction accuracy. This may involve selecting different optimization algorithms, adjusting the learning rate, increasing or decreasing model complexity, etc., until the model's performance on the validation set meets expectations. After model training and tuning are complete, a final performance evaluation is performed using test set data to ensure that the model not only performs well on the training data but also makes accurate predictions on unseen new data.

[0062] Optionally, if the model performs satisfactorily on the test set, it is saved as a candidate prediction model for the vehicle model sample. After the above steps, the model can be applied to predict the charging time of a specific vehicle model, a specific battery state, and a specific environmental state in real time.

[0063] Optionally, considering the continuous changes in battery characteristics and charging technology, the model can be retrained periodically with new data samples to update the model parameters, so as to maintain the model's prediction accuracy and adapt to new charging scenarios and device characteristics.

[0064] In this embodiment of the application, the above method can train a candidate prediction model that adapts to specific vehicle models, battery states and environmental states using big data, providing more accurate and personalized support for charging time prediction.

[0065] As an optional implementation method, the initial prediction model is trained using vehicle model samples, state information samples, and environmental state samples to obtain candidate prediction models corresponding to the vehicle model samples. This includes: sending the vehicle model samples, state information samples, and environmental state samples to the cloud platform of the vehicle manufacturer corresponding to the vehicle; using the cloud platform, grouping the state information samples and environmental state samples according to the vehicle model samples to obtain grouped state information samples and grouped environmental state samples; denoising the grouped state information samples and grouped environmental state samples respectively to obtain denoised state information samples and denoised environmental state samples; and training the initial prediction model corresponding to the vehicle model samples using the denoised state information samples and denoised environmental state samples to obtain candidate prediction models.

[0066] In this embodiment, during the training of the initial prediction model using vehicle model samples, state information samples, and environmental state samples, the vehicle model samples, state information samples, and environmental information samples can be sent to the cloud platform of the corresponding vehicle manufacturer. The cloud platform can then group the state information samples and environmental state samples according to the vehicle model samples, obtaining grouped state information samples and grouped environmental state samples. Noise reduction processing can be performed on the grouped state information samples and grouped environmental information samples to train the initial prediction model corresponding to the vehicle model samples, thus obtaining candidate prediction models.

[0067] Optionally, information such as vehicle model, battery status (including charging map information, battery capacity, battery cycle count, SOC, etc.), environmental status (such as the highest and lowest battery cell temperatures and ambient temperature), and charging pile information (such as the maximum output current after vehicle-pile matching) are automatically collected during the charging process of each electric vehicle. These data samples are sent to the corresponding automaker's cloud server via the network, forming a large data pool. On the automaker's cloud, the big data system first performs a first-level grouping based on the vehicle model samples. This means that data from all models of the same type are grouped into one large group, while models with different configurations are assigned to the next level of large groups. Further, based on the maximum output current at the charging pile after vehicle-pile matching, the data is divided into specific subgroups to ensure the targeted and effective data processing. The grouped data undergoes preprocessing, primarily including noise reduction. The system identifies and removes unreasonable or abnormal data points, such as extreme values ​​(possibly caused by measurement errors or sudden situations), duplicate data, or invalid information, ensuring the quality of the training dataset and improving the accuracy and reliability of the model.

[0068] Optionally, based on preprocessing, feature selection and engineering processing are performed on the denoised state information samples and environmental state samples. These steps may include extracting key features, data transformation, and creating new features (such as feature crossing) to enhance the model's interpretability of the data and enable it to capture key factors affecting charging time. The processed state information samples and environmental state samples are then used as input features to train the initial prediction model corresponding to the vehicle model samples. This may involve selecting appropriate machine learning algorithms (such as regression analysis, neural networks, random forests, etc.), setting model parameters, and using historical charging data as a training set to iteratively optimize the model, enabling it to learn the complex relationship between charging time and vehicle model, state information, and environmental state.

[0069] Optionally, during model training, model performance is evaluated using validation set data, which may include calculating prediction error and assessing the model's generalization ability. Based on the validation results, model parameters or algorithms may need to be adjusted to improve prediction accuracy. Once the model training and validation achieve satisfactory performance metrics, candidate prediction models corresponding to the vehicle model sample are generated. This marks the completion of the model training process, and the candidate prediction models can be used to predict the charging time of new charging events. The candidate prediction models are stored in a cloud database and associated with specific vehicle model and configuration information. Simultaneously, the system should be designed with a periodic model update mechanism to retrain the model periodically or based on new data to maintain the up-to-dateness and accuracy of the prediction models.

[0070] As an optional implementation method, the vehicle model sample includes a model sample and a configuration sample. Using the vehicle manufacturer's cloud platform, the status information sample and environmental status sample are grouped according to the vehicle model sample to obtain grouped status information samples and grouped environmental status samples. This includes: using the vehicle manufacturer's cloud platform, grouping the status information sample and environmental status sample according to the model sample to obtain a first-layer grouped status information sample and a first-layer grouped environmental status sample; using the vehicle manufacturer's cloud platform, grouping the first-layer status information sample and the first-layer environmental status sample according to the configuration sample to obtain a second-layer grouped status information sample and a second-layer grouped environmental status sample; the method further includes: using the maximum current output of the matching power supply device sample corresponding to the vehicle model sample to group the second-layer status information sample and the second-layer environmental status sample to obtain a third-layer grouped status information sample and a third-layer grouped environmental status sample.

[0071] In this embodiment, during the grouping process of state information samples and environmental state samples, the vehicle manufacturer's cloud platform can be used to group the state information samples and environmental state samples according to model samples, resulting in first-layer state information samples and first-layer environmental state samples. The vehicle manufacturer's cloud platform can also be used to group the first-layer state information samples and first-layer environmental state samples according to configuration samples, resulting in second-layer state information samples and second-layer environmental state samples. Finally, the maximum current output of the matching functional device samples can be used to group the second-layer state information samples and second-layer environmental state samples, resulting in third-layer state information samples and third-layer environmental state samples.

[0072] Optionally, on the vehicle manufacturer's cloud platform, the collected status information samples (such as battery level, charging cycle count, and SOC status) and environmental status samples (such as cell temperature and ambient temperature) are first classified and grouped according to vehicle model samples. This means that data from all vehicles of the same model will be grouped into one large group, forming the first-level status information samples and the first-level environmental status samples after grouping. The above steps aim to extract charging characteristics related to vehicle models from big data, preparing for more refined analysis later.

[0073] Optionally, based on the first-level grouping, the state information samples and environmental state samples are further classified in a more detailed manner according to the vehicle configuration samples (such as battery type, capacity, voltage level, and additional vehicle function configurations). Therefore, the first-level state information samples and environmental state samples will be subdivided into several more specific subgroups, each corresponding to different configuration options, forming the grouped second-level state information samples and second-level environmental state samples. This process enhances the relevance of the data, enabling the model to more accurately learn the impact of different configurations on charging time. Based on this, using the maximum current information output of the vehicle samples corresponding to the vehicle models and the matching power supply equipment (charging piles), the second-level state information samples and second-level environmental state samples are further subdivided. This step involves further dividing the data into smaller groups based on the vehicle-charging pile matching and the maximum output current capability of the charging pile. In this way, the system can capture the actual charging efficiency and characteristics of different vehicle configurations under specific charging conditions (e.g., high-current fast charging, standard-current slow charging), thus obtaining the grouped third-level state information samples and third-level environmental state samples.

[0074] Optionally, deep data cleaning and preprocessing are performed on the grouped data, including noise reduction, outlier detection, and missing data imputation, to ensure the quality of each group of data. Subsequently, feature engineering is implemented to extract the most relevant features for charging time prediction from state information and environmental conditions, such as the current battery state, charging rate, and temperature effects, preparing for model training. One or more candidate prediction models are trained for each group of data. Model selection and optimization are based on the characteristics of the data and may include linear regression, decision trees, deep learning models, etc. Through repeated training and validation, model parameters are adjusted until each model can accurately predict the charging time of its corresponding group of data. A detailed performance evaluation of the trained models is performed, including but not limited to prediction accuracy, model generalization ability, and response speed. Based on the evaluation results, the optimal model is selected, or the advantages of multiple models are integrated through model fusion technology to build a comprehensive prediction system that can flexibly select the most suitable prediction model according to different vehicle models, configurations, and charging conditions.

[0075] In this embodiment, the above method enables deep personalization and refined management of electric vehicle charging time prediction, significantly improving the accuracy and practicality of the prediction and meeting the charging demand prediction needs of different users under different environmental conditions. Simultaneously, the method provides a clear path for subsequent model iteration and optimization, contributing to the continuous improvement of the overall performance and user experience of the prediction system.

[0076] As an optional embodiment, the method further includes: updating the vehicle model sample, the status information sample, and the environmental status sample according to a target period to obtain updated vehicle model samples, updated status information samples, and updated environmental status samples.

[0077] In this embodiment, the vehicle model sample, status information sample, and environmental status sample can be updated according to the target period to obtain the updated vehicle model sample, updated status information sample, and updated environmental status sample.

[0078] Optionally, a large amount of data related to electric vehicle charging is continuously collected during daily operation, including vehicle model samples (vehicle model and configuration information), status information samples (battery SOC, battery temperature, maximum output power after vehicle and charging pile matching, etc.), and environmental status samples (such as ambient temperature during charging). This data is stored in the automaker's cloud database as the basis for subsequent analysis and model training. To ensure the timeliness and accuracy of the prediction model, a target update cycle can be set, such as one week, for regular data updates and model retraining. The purpose of this is to capture and adapt to possible short-term changes in vehicle, battery status, and charging environment. At the end of the target cycle, the system cleans and organizes the data collected in the previous cycle, including removing invalid data, filling in missing values, and standardizing values, to ensure high-quality datasets and prepare for model training.

[0079] Optionally, the latest vehicle model samples, status information samples, and environmental status samples from the previous period are added to the existing dataset, replacing outdated or no longer applicable data, resulting in an updated dataset. This includes updating vehicle model and configuration information, the latest battery status records, and recent environmental condition change data. The prediction model is then retrained using the updated dataset. This may involve fine-tuning or retraining the model to adapt to changes in the dataset, especially when newly collected data reveals charging behavior patterns or environmental change trends not fully covered by the model. The retrained model is then validated to check whether its predictive ability has improved or remained stable, particularly when processing newly collected data. If the model performs poorly or has room for improvement, the system adjusts algorithm parameters, optimizes feature selection, and may even try different model frameworks until satisfactory prediction accuracy is achieved.

[0080] Optionally, a validated and optimized model can be deployed to the system, replacing the original prediction model. Based on the updated model, charging time can now be predicted more accurately, providing users with timely and personalized charging recommendations. Simultaneously, the system retains a history of model updates for future review and data analysis.

[0081] In this embodiment, through the aforementioned series of periodic data updates and model training processes, this embodiment ensures that the prediction model always reflects the latest vehicle status, charging technology, and environmental conditions, thereby continuously providing high-precision charging time prediction services in the rapidly changing electric vehicle market and technological environment.

[0082] As an optional implementation method, the initial prediction model is trained using vehicle model samples, state information samples, and environmental state samples to obtain candidate prediction models corresponding to vehicle model samples. This includes: training the initial prediction model using updated vehicle model samples, updated state information samples, and updated environmental state samples to obtain candidate prediction models.

[0083] In this embodiment, during the training of the initial prediction model using vehicle model samples, state information samples, and environmental state samples, the updated vehicle model samples, updated state information samples, and updated environmental state samples can be used to train the initial prediction model to obtain a candidate prediction model.

[0084] Optionally, at the end of a set target period (e.g., weekly), the system will initiate a data update process. This involves summarizing, cleaning, and organizing all vehicle model samples, status information samples (including battery charge, cycle count, SOC, etc.), and environmental status samples (such as temperature and humidity) collected over the past week. By removing duplicate data and handling outliers and missing values, the accuracy and completeness of the dataset are ensured, providing fresh and high-quality data for model training.

[0085] Optionally, an initial prediction model already exists before the data update, which may be trained based on historical data. After the data update, this initial prediction model is used as a starting point to train and optimize it using the updated dataset. The system retrains the initial prediction model using the updated dataset. This includes adjusting model parameters, optimizing algorithm configuration, and possible feature engineering (e.g., creating new features, performing feature selection). The training process aims to enable the model to learn the latest charging behavior patterns and environmental change trends reflected in the updated dataset, improving the model's predictive ability for new data. During the model training phase, cross-validation is used to evaluate model performance. This involves splitting the updated dataset into training and validation sets, training the model multiple times, and validating it with different subsets of data to ensure that the model not only performs well on the training data but also has high generalization ability on unseen validation data.

[0086] Optionally, after training, a comprehensive performance evaluation is performed on the candidate prediction models to check their accuracy and stability in predicting electric vehicle charging time. Evaluation metrics may include prediction error, model complexity, and computational efficiency. The optimal candidate prediction model is selected based on the evaluation results and prepared for deployment in practical applications. The optimal candidate prediction model is deployed into the system and immediately used to predict the charging time of new charging events. Furthermore, the system maintains an online learning mechanism that continues to collect data during charging, fine-tuning the model in real time to adapt to immediate changes in the charging environment and further improve prediction accuracy.

[0087] In this embodiment, the method described above not only ensures the timeliness of the prediction model, enabling it to reflect the latest market and technological changes in a timely manner, but also improves the model's predictive performance through continuous data-driven optimization. This plays a crucial role in enhancing the charging experience for electric vehicle users, promoting the efficient utilization of charging infrastructure, and driving the development of the electric vehicle industry.

[0088] The technical solutions of the embodiments of the present invention will be illustrated below with reference to preferred embodiments.

[0089] Currently, guided by policies, electric vehicle technology is developing rapidly, and the number of pure electric vehicles on the road is steadily increasing year by year. Considering the significant differences in charging time for pure electric vehicles under different operating conditions, this paper proposes a method for estimating remaining charging time using big data. This method comprehensively considers factors such as user vehicle configuration, charging gun configuration, cell temperature, and battery capacity to more accurately estimate the remaining charging time. The advantages of this application's embodiments are that it allows for real-time estimation and updating, and by utilizing big data to comprehensively consider multiple factors such as vehicle configuration, charging gun configuration, cell temperature, and battery capacity, resulting in greater accuracy.

[0090] The embodiments of the present invention will be further described below.

[0091] Figure 2 This is a schematic diagram of a charging time estimation process based on big data according to an embodiment of the present invention, as shown below. Figure 2As shown, battery information and surrounding environmental information are collected through the BMS. The system obtains current vehicle configuration information, corresponding charging map information, battery capacity, battery cycle count, maximum and minimum battery cell temperatures, battery SOC, and maximum charging power after vehicle-pile matching. After charging, the above information and charging time are sent to the enterprise cloud storage. The big data system groups the real charging data for each charging behavior in the cloud, dividing it into the first large group based on different vehicle models, the next large group based on different configurations, and then further subdividing it into smaller groups based on the maximum current output at the charging pile after vehicle-pile matching. After removing unreasonable data, a training model is established to predict charging time under the influence of multiple factors, based on charging map information, battery capacity, battery cycle count, maximum and minimum battery cell temperatures, and battery SOC parameters. The latest vehicle charging data is updated weekly. Through continuous learning and training, the system can accurately estimate the vehicle charging power curve based on vehicle configuration, battery temperature, ambient temperature, initial SOC, and charging power information. Based on the current charging status and user-set charging limits, the remaining charging time is estimated in real time. According to the present invention, factors such as vehicle configuration, battery capacity, charging station configuration, battery cycle count, battery temperature, and initial state of charge (SOC) can be comprehensively considered to more accurately estimate the remaining charging time.

[0092] For example, under normal temperature conditions, a user needs to charge from 12% to 100% at a high-power charging station. After the vehicle uploads the relevant parameters to the cloud, the big data system automatically matches a charging time prediction model for the same vehicle model and configuration, and estimates the charging time based on the current battery temperature, ambient temperature, and initial state of charge (SOC).

[0093] For another example, in low-temperature environments, users need to charge their batteries from 12% to 80% at high-power charging stations. After the vehicle uploads the relevant parameters to the cloud, the big data system automatically matches a charging time prediction model for the same vehicle model and configuration, estimating the charging time based on the current battery temperature, ambient temperature, initial SOC, and final SOC.

[0094] For another example, in low-temperature environments, a user needs to charge the battery from 12% to 80% at a public charging station. After the vehicle uploads the relevant parameters to the cloud, the big data system automatically matches a charging time prediction model for the same vehicle model and configuration, and estimates the charging time based on the current battery temperature, ambient temperature, initial SOC, and final SOC.

[0095] According to embodiments of the present invention, a device for determining the charging time of an energy storage device in a vehicle is also provided. It should be noted that this device for determining the charging time of an energy storage device in a vehicle can be used to execute the method for determining the charging time of an energy storage device in a vehicle described in the above embodiments.

[0096] Figure 3 This is a schematic diagram of a charging time determination device for an energy storage device in a vehicle according to an embodiment of the present invention, as shown below.Figure 3 As shown, the charging time determination device 300 for the energy storage device in the vehicle may include: an acquisition unit 302, a determination unit 304, and a prediction unit 306.

[0097] The acquisition unit 302 is used to acquire vehicle model information in response to the fact that the vehicle's energy storage device and energy supply device are in a connected state.

[0098] The determining unit 304 is used to determine the target prediction model corresponding to the vehicle model information from multiple candidate prediction models.

[0099] The prediction unit 306 is used to predict the charging time based on the state information of the energy storage device and the charging demand information of the energy storage device using the target prediction model.

[0100] In this embodiment of the invention, the acquisition unit 302 acquires vehicle model information in response to the connection between the vehicle's energy storage device and energy supply device. The determination unit 304 determines the target prediction model corresponding to the vehicle model information from multiple candidate prediction models. The prediction unit 306 uses the target prediction model, based on the state information and charging demand information of the energy storage device, to predict the charging time, thereby solving the technical problem of low accuracy in determining the charging time of energy storage devices in vehicles and achieving the technical effect of improving the accuracy of determining the charging time of energy storage devices in vehicles.

[0101] According to embodiments of the present invention, a computer-readable storage medium is also provided, the storage medium including a stored program, wherein the program executes the methods described in the embodiments of the present invention.

[0102] According to an embodiment of the present invention, a processor is also provided for running a program, wherein the program executes the methods described above in the embodiments of the present invention during runtime.

[0103] According to another aspect of the present invention, an electronic device is also provided. The electronic device includes a memory and a processor, the memory storing a computer program, and the processor being configured to run the computer program to perform the methods described in the embodiments of the present invention.

[0104] According to another aspect of the present invention, a computer program product is also provided. This computer program product includes a computer program that, when executed by a processor, implements the methods described above in the embodiments of the present invention.

[0105] According to another aspect of the present invention, a vehicle is also provided. The vehicle includes a memory and a processor. The memory stores an executable program; the processor runs the program, which, when executed, implements the methods described in the embodiments of the present invention.

[0106] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0107] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0108] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0109] Furthermore, the functional units in the various embodiments of the present invention 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. The integrated unit can be implemented in hardware or as a software functional unit.

[0110] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part 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 the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0111] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for determining the charging time of an energy storage device in a vehicle, characterized in that, The vehicle includes energy storage devices, including: In response to the connection between the energy storage device and the energy supply device, the vehicle model information is obtained; From multiple candidate prediction models, a target prediction model corresponding to the vehicle model information is determined. The accuracy of determining the charging time of the energy storage device using the target prediction model is greater than the accuracy of determining the charging time of the energy storage device using the candidate prediction models other than the target prediction model. Using the target prediction model, based on the state information of the energy storage device and the charging demand information of the energy storage device, the charging duration is predicted, wherein the state information is used to represent the charging state of the energy storage device in the environment where the vehicle is located, and the charging demand information is used to represent the demand for charging the energy storage device using the energy supply device.

2. The method according to claim 1, characterized in that, Using the target prediction model, based on the state information of the energy storage device and the charging demand information of the energy storage device, the charging duration is predicted, including: Using the target prediction model, based on the state information, the power change information of the energy storage device is determined, wherein the power change information is used to represent the degree of change in the charging power of the energy storage device in the charging state; Using the target prediction model, the charging duration is predicted based on the power change information and the charging demand information.

3. The method according to claim 2, characterized in that, The method further includes: Obtain vehicle model samples, state information samples, and environmental state samples for the initial prediction model to be trained; The initial prediction model is trained using the vehicle model samples, the state information samples, and the environmental state samples to obtain the candidate prediction model corresponding to the vehicle model samples.

4. The method according to claim 3, characterized in that, Using the vehicle model samples, the state information samples, and the environmental state samples, the initial prediction model is trained to obtain the candidate prediction model corresponding to the vehicle model samples, including: The vehicle model sample, the status information sample, and the environmental status sample are sent to the cloud of the vehicle manufacturer corresponding to the vehicle. Using the vehicle manufacturer's cloud platform, the status information samples and the environmental status samples are grouped according to the vehicle model samples to obtain the grouped status information samples and the grouped environmental status samples. The grouped state information samples and the grouped environmental state samples are denoised respectively to obtain the denoised state information samples and the denoised environmental state samples. The initial prediction model corresponding to the vehicle model sample is trained using the denoised state information sample and the denoised environment state sample to obtain the candidate prediction model.

5. The method according to claim 4, characterized in that, The vehicle model sample includes a model sample and a configuration sample. Using the vehicle manufacturer's cloud platform, the status information sample and the environmental status sample are grouped according to the vehicle model sample to obtain the grouped status information sample and the grouped environmental status sample, including: Using the vehicle manufacturer's cloud platform, the status information samples and the environmental status samples are grouped according to the model samples to obtain the first layer of the grouped status information samples and the first layer of the grouped environmental status samples. Using the vehicle manufacturer's cloud platform, the first layer of state information samples and the first layer of environmental state samples are grouped according to the configuration sample to obtain the grouped second layer of state information samples and the grouped second layer of environmental state samples. The method further includes: using the maximum current output of the vehicle sample corresponding to the vehicle model sample and the matching power supply device sample, grouping the second layer of state information samples and the second layer of environmental state samples to obtain the grouped third layer of state information samples and the grouped third layer of environmental state samples.

6. The method according to claim 3, characterized in that, The method further includes: According to the target cycle, the vehicle model sample, the status information sample, and the environmental status sample are updated to obtain the updated vehicle model sample, the updated status information sample, and the updated environmental status sample.

7. The method according to claim 6, characterized in that, Using the vehicle model samples, the state information samples, and the environmental state samples, the initial prediction model is trained to obtain the candidate prediction model corresponding to the vehicle model samples, including: The initial prediction model is trained using the updated vehicle model samples, the updated state information samples, and the updated environmental state samples to obtain the candidate prediction model.

8. A device for determining the charging time of an energy storage device in a vehicle, characterized in that, The device includes: The acquisition unit is used to acquire vehicle model information in response to the connection between the vehicle's energy storage device and energy supply device. The determining unit is used to determine the target prediction model corresponding to the vehicle model information from multiple candidate prediction models, wherein the accuracy of determining the charging time of the energy storage device using the target prediction model is greater than the accuracy of determining the charging time of the energy storage device using the candidate prediction models other than the target prediction model from multiple candidate prediction models. The prediction unit is used to predict the charging time based on the state information of the energy storage device and the charging demand information of the energy storage device using the target prediction model. The state information represents the charging state of the energy storage device in the environment where the vehicle is located, and the charging demand information represents the demand for charging the energy storage device using the energy supply device.

9. A processor, characterized in that, The processor is used to run a program, wherein the program executes the method according to any one of claims 1 to 7 when it runs.

10. An electronic device, characterized in that, The method includes a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the method according to any one of claims 1 to 7.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform the method according to any one of claims 1 to 7.

12. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method described in any one of claims 1 to 7.

13. A vehicle, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the method according to any one of claims 1 to 7.

Images