Vehicle energy management control method, system, device, and storage medium

Abstract

The application discloses a vehicle energy management control method, system, device and storage medium, and belongs to the field of vehicle energy management. The method comprises the following steps: obtaining historical driving data of a target vehicle and preprocessing the historical driving data to obtain target driving data; performing data division processing on the target driving data according to a bootstrap method to obtain training set data and verification set data; constructing an initial energy recovery model of the target vehicle and optimizing the initial energy recovery model to obtain a target energy recovery model; obtaining driving parameters of the target vehicle under a driving state every first preset time to match corresponding energy recovery data in the energy recovery model according to the driving parameters; and converting the energy recovery data into vehicle command data by a vehicle controller of the target vehicle, and executing corresponding energy recovery actions according to the vehicle command data. Through the intelligent energy management control method, the vehicle can realize more efficient energy recovery under different driving conditions.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of vehicle energy management, and particularly relates to a vehicle energy management control method, system, device and storage medium. BACKGROUND

[0002] With the rapid development of the automobile industry and the enhancement of environmental protection awareness, vehicle energy management has become one of the key technologies for improving automobile energy efficiency and reducing emissions. Traditional vehicle energy management methods often rely on fixed rules and simple control strategies, and it is difficult to achieve efficient energy recovery under different driving conditions. In order to overcome this limitation, in recent years, vehicle energy management control methods based on big data and artificial intelligence technology have gradually become a research hotspot.

[0003] At present, traditional energy management methods are often based on fixed rules or simple real-time road condition information, which makes it difficult to adapt to complex and variable actual driving needs. SUMMARY

[0004] The purpose of the embodiments of the application is to provide a vehicle energy management control method, system, device and storage medium, which can solve the problem that the existing energy management method is often based on fixed rules or simple real-time road condition information, which makes it difficult to adapt to complex and variable actual driving needs.

[0005] In order to solve the above technical problems, the application is implemented as follows:

[0006] In a first aspect, the embodiments of the application provide a vehicle energy management control method, which comprises:

[0007] Obtaining historical driving data of a target vehicle, preprocessing the historical driving data to obtain target driving data, and transmitting the target driving data to a cloud server for storage;

[0008] According to the self-service method, the target driving data is divided and processed to obtain training set data and validation set data;

[0009] According to the training set data and a deep learning algorithm, an initial energy recovery model of the target vehicle is constructed;

[0010] According to the validation set data, the initial energy recovery model is verified, and the initial energy recovery model is optimized according to the verification result to obtain a target energy recovery model;

[0011] Every first preset time, driving parameters under the driving state of the target vehicle are obtained, and the corresponding energy recovery data in the energy recovery model is matched according to the driving parameters;

[0012] The whole vehicle controller of the target vehicle converts the energy recovery data into vehicle command data, and the target vehicle performs corresponding energy recovery actions according to the vehicle command data.

[0013] As an optional implementation of the first aspect of the application, the historical driving data of the target vehicle is obtained, the historical driving data is preprocessed to obtain target driving data, and the target driving data is transmitted to a cloud server for storage, specifically:

[0014] The historical driving data of the target vehicle is collected by using vehicle-mounted sensors, Internet of Vehicles technology, traffic monitoring cameras and GPS equipment;

[0015] The historical driving data is filtered to delete invalid data in the historical driving data;

[0016] The historical driving data after the filtering is integrated to form a unified data format and storage structure;

[0017] The historical driving data after the data integration is standardized to enable data of different dimensions to be compared and analyzed, and the target driving data is obtained;

[0018] The target driving data is uploaded to the cloud server through a data transmission channel, and a corresponding data directory and database are created on the cloud server to store the target driving data.

[0019] As an optional implementation of the first aspect of the application, the target driving data is divided according to the bootstrap method to obtain training set data and validation set data, specifically:

[0020] The target driving data is evenly divided to obtain a data set containing m data samples;

[0021] The data set is sampled m times with replacement to obtain the training set data;

[0022] The same data in the target driving data as the training set is deleted to obtain the validation set data.

[0023] As an optional implementation of the first aspect of the application, the initial energy recovery model of the target vehicle is constructed according to the training set data and a deep learning algorithm, specifically:

[0024] The training set data is subjected to feature extraction processing to extract energy recovery features in the training set data to construct an energy recovery feature set;

[0025] perform feature screening processing on the energy recovery feature set to screen out key features in the energy recovery feature set that affect energy recovery, to obtain a key feature set;

[0026] According to the deep learning algorithm and the key feature set, the energy recovery model including various energy recovery data is constructed;

[0027] According to the training set, the energy recovery model is trained to obtain the initial energy recovery model.

[0028] As an optional implementation form of the first aspect of the application, according to the verification set data, the initial energy recovery model is verified, and according to the verification result, the initial energy recovery model is optimized to obtain a target energy recovery model, specifically:

[0029] The verification set is input to the initial energy recovery model for verification processing, and the initial energy recovery model outputs a verification result;

[0030] The verification result is compared with the verification set to obtain a difference function between the verification set and the verification result;

[0031] According to the difference function, the difference features in the verification set are analyzed, and the parameters and structure of the initial energy recovery model are adjusted according to the difference features;

[0032] According to the training set, the initial energy recovery model after adjustment is trained to obtain the target energy recovery model.

[0033] As an optional implementation form of the first aspect of the application, the driving parameters of the target vehicle are obtained every first preset time, and the corresponding energy recovery data in the energy recovery model is matched according to the driving parameters, specifically:

[0034] The target vehicle obtains driving parameters in the driving state every first preset time according to the vehicle-mounted sensor;

[0035] The vehicle control unit receives the driving parameters and performs the preprocessing on the driving parameters to obtain target driving parameters;

[0036] The vehicle control unit inputs the target driving parameters to the target energy recovery model, and the target energy recovery model outputs corresponding energy recovery data according to the target driving parameters.

[0037] As an optional implementation of the first aspect of the application, the vehicle controller of the target vehicle converts the energy recovery data into vehicle command data, and the target vehicle performs corresponding energy recovery actions according to the vehicle command data, specifically:

[0038] The vehicle controller converts the energy recovery data output by the target energy recovery model into the vehicle command data that can be recognized and executed by the target vehicle;

[0039] The vehicle controller issues the vehicle command data to the target vehicle through an actuator for execution, and the target vehicle performs corresponding energy recovery actions according to the execution instructions.

[0040] Secondly, the application provides a vehicle energy management control system, which comprises:

[0041] The first acquisition module acquires historical driving data of a target vehicle;

[0042] The data processing module pre-processes the historical driving data to obtain target driving data;

[0043] The data storage module is used for storing the target driving data;

[0044] The data division module performs data division processing on the target driving data according to the bootstrap method to obtain training set data and validation set data;

[0045] The model construction module constructs an initial energy recovery model of the target vehicle according to the training set data and a deep learning algorithm;

[0046] The model optimization module verifies the initial energy recovery model according to the validation set data, optimizes the initial energy recovery model according to the verification result, and obtains a target energy recovery model.

[0047] The second acquisition module acquires driving parameters of the target vehicle at a first preset time interval;

[0048] The data matching module matches the energy recovery data corresponding to the energy recovery model according to the driving parameters;

[0049] The data conversion module converts the energy recovery data into vehicle command data by the vehicle controller of the target vehicle, and the target vehicle performs corresponding energy recovery actions according to the vehicle command data.

[0050] In a third aspect, an electronic device is provided, which includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor, and the program or instructions, when executed by the processor, implement the steps of the method according to the first aspect.

[0051] In a fourth aspect, a readable storage medium is provided, which stores a program or instructions, and the program or instructions, when executed by a processor, implement the steps of the method according to the first aspect.

[0052] In the embodiments of the present application, compared with the prior art, the following technical effects are achieved:

[0053] Through diversified data sources such as vehicle-mounted sensors, Internet of Vehicles technology, traffic monitoring cameras, and GPS devices, historical driving data of the target vehicle is comprehensively collected to ensure the richness and diversity of the data; the historical driving data is filtered, integrated, and standardized to effectively eliminate invalid data, unify the data format and storage structure, and make the data more standardized, easy to analyze and use; the preprocessed target driving data is uploaded to a cloud server for storage, facilitating subsequent data access, analysis, and model training, and improving the efficiency and security of data management. The target driving data is divided by the bootstrap method to obtain a training set and a validation set, ensuring the independence and representativeness of the data in the model training process; through feature extraction and screening processing, key features affecting energy recovery are extracted from the training set data, providing a solid foundation for building an energy recovery model; an initial energy recovery model is constructed using a deep learning algorithm, and is trained using the training set data, so that the model can learn complex energy recovery rules; the initial energy recovery model is verified and optimized through the validation set data, and the parameters and structure of the model are continuously adjusted until a target energy recovery model with high precision and generalization ability is obtained. The target vehicle acquires driving parameters in the driving state at a preset time interval based on the vehicle-mounted sensors to ensure the real-time and accuracy of the data; the vehicle control unit inputs the preprocessed driving parameters to the target energy recovery model, and the model outputs corresponding energy recovery data based on the driving parameters, realizing intelligent energy recovery control; the vehicle control unit converts the energy recovery data into vehicle command data and issues it to the target vehicle through an actuator for execution, so that the vehicle can perform energy recovery actions according to the real-time driving state, improving the efficiency and flexibility of energy recovery. Through the intelligent energy management control method, the vehicle can achieve more efficient energy recovery under different driving conditions, thereby reducing energy waste and pollution emissions; in the long term, this method helps to improve the energy efficiency of the vehicle, reduce operating costs, and reduce the impact on the environment, in line with the concept of sustainable development. BRIEF DESCRIPTION OF DRAWINGS

[0054] Figure 1is a flow chart of a vehicle energy management control method provided by some embodiments of the present application; DETAILED DESCRIPTION

[0055] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0056] The terms "first", "second", and the like in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein. In addition, "and / or" in the specification and claims indicates at least one of the connected objects, and the character " / ", generally indicates that the front and rear associated objects are in a "or" relationship.

[0057] The vehicle energy management control method, system, device and storage medium provided by the embodiments of the present application will be described in detail below with reference to the drawings and specific embodiments and their application scenarios.

[0058] Embodiments

[0059] A vehicle energy management control method comprises the following steps:

[0060] S100: Obtain historical driving data of a target vehicle, preprocess the historical driving data to obtain target driving data, and transmit the target driving data to a cloud server for storage;

[0061] Further, first, historical driving data is obtained from the target vehicle, which includes vehicle speed, acceleration, braking condition, battery status, etc.; then, the raw data is preprocessed to obtain more accurate and usable target driving data; the preprocessed data is transmitted to the cloud server for storage for subsequent analysis and model training.

[0062] It should be noted that in S100, the historical driving data of the target vehicle is obtained, the historical driving data is preprocessed to obtain the target driving data, and the target driving data is transmitted to the cloud server for storage, specifically:

[0063] S110: Collect historical driving data of the target vehicle by using vehicle-mounted sensors, Internet of Vehicles technology, traffic monitoring cameras and GPS devices;

[0064] S120: screening the historical driving data to delete invalid data in the historical driving data;

[0065] S130: data integration on the historical driving data after the screening processing to form a unified data format and storage structure;

[0066] S140: standardization processing on the historical driving data after the data integration to enable comparison and analysis of data of different dimensions to obtain target driving data;

[0067] S150: uploading the target driving data to a cloud server through a data transmission channel, creating a corresponding data directory and database on the cloud server for storing the target driving data.

[0068] Further, the historical driving data of the target vehicle is comprehensively collected by using vehicle-mounted sensors (such as speed sensors, acceleration sensors, etc.), Internet of Vehicles technology (inter-vehicle communication), traffic monitoring cameras (providing external environment information) and GPS devices (positioning and speed information); these data cover various parameters and environmental conditions of vehicle operation. The collected historical driving data is preliminarily screened to remove invalid or redundant data, such as abnormal values caused by sensor failure, repeated records, etc.; this step ensures the data quality and accuracy of subsequent analysis. The screened data is integrated to unify the data format and storage structure; this includes converting data to a standard format, merging data from different sources, filling missing values, etc. to facilitate subsequent data analysis and model training. The integrated data is standardized, i.e. adjusting the dimension and range of the data to enable comparison and analysis of data of different dimensions; this usually includes data normalization, standardization, etc. to ensure that the importance of each feature is relatively balanced during model training. The processed target driving data is uploaded to a cloud server through a secure data transmission channel; a corresponding data directory and database are created on the cloud server for long-term storage and management of these data; this facilitates subsequent data access, analysis and model training.

[0069] S200: data division processing on the target driving data according to the bootstrap method to obtain training set data and validation set data;

[0070] Further, the bootstrap method (a statistical resampling technique) is used to divide the target driving data to obtain training set data and validation set data; the training set data is used to build the model, and the validation set data is used to evaluate the performance of the model.

[0071] It should be noted that in S200, the target driving data is divided according to the bootstrap method to obtain training set data and validation set data, specifically:

[0072] S210: uniformly sampling the target driving data to obtain a data set containing m data samples;

[0073] S220: m times with replacement sampling on the data set to obtain training set data;

[0074] S230: deleting the same data in the target driving data as the training set to obtain validation set data.

[0075] Further, first, the target driving data is uniformly sampled to ensure that the data is uniformly distributed in each dimension; the purpose of this step is to avoid the bias that may occur during data division, and to ensure that the training set and the validation set can fully represent the characteristics of the entire data set; the processed data set contains m data samples, which are the basis for subsequent division. Then, m times with replacement sampling is performed on the data set to construct the training set data. With replacement sampling means that after each sampling, the sampled sample is returned to the data set, so the same sample may be sampled multiple times; this sampling method helps to increase the diversity of the training set, while ensuring the consistency of the training set with the original data set in statistical characteristics; after m times of sampling, the training set data obtained will be used for subsequent model construction. Finally, the data in the target driving data that is the same as the training set is deleted, and the remaining data constitutes the validation set. This step ensures the independence between the training set and the validation set, avoiding the overfitting problem during model validation; the validation set will be used to evaluate the performance of the model, helping to optimize the model parameters and structure.

[0076] S300: constructing an initial energy recovery model of the target vehicle according to the training set data and the deep learning algorithm;

[0077] Further, an initial energy recovery model of the target vehicle is constructed using a deep learning algorithm (such as a neural network) and training set data; this model aims to predict the optimal energy recovery strategy based on the driving state of the vehicle.

[0078] It should be noted that in S300, the initial energy recovery model of the target vehicle is constructed according to the training set data and the deep learning algorithm, specifically:

[0079] S310: performing feature extraction processing on the training set data to extract energy recovery features in the training set data to construct an energy recovery feature set;

[0080] S320: performing feature selection processing on the energy recovery feature set to select key features in the energy recovery feature set that affect energy recovery to obtain a key feature set;

[0081] S330: constructing an energy recovery model including multiple energy recovery data according to the deep learning algorithm and the key feature set;

[0082] S340: training the energy recovery model according to the training set to obtain an initial energy recovery model.

[0083] Further, first, feature extraction processing is performed on the training set data; the purpose of this step is to identify energy recovery-related features from the original data, including vehicle speed, acceleration, brake pressure, battery state, etc.; through feature extraction, an energy recovery feature set can be constructed, providing a basis for subsequent model construction. Then, feature selection processing is performed on the energy recovery feature set. The purpose of feature selection is to identify the key features that have the greatest impact on energy recovery, in order to reduce model complexity and improve model performance; this step involves statistical tests, correlation analysis, recursive feature elimination, etc.; the key feature set after selection will be used for subsequent model construction. According to the deep learning algorithm (such as convolutional neural network, recurrent neural network, deep neural network, etc.) and the key feature set, the energy recovery model is constructed; this model aims to predict the energy recovery potential of the vehicle according to the input key features, including possible energy recovery amount, recovery efficiency, etc.; during model construction, appropriate network structure, loss function and optimization algorithm need to be selected to ensure the accuracy and generalization ability of the model. Finally, the training set is used to train the energy recovery model; during training, the model will continuously adjust its parameters to minimize prediction error; after training is completed, the obtained model is the initial energy recovery model, which will be used for subsequent verification and optimization.

[0084] S400: verifying the initial energy recovery model according to the verification set data, and optimizing the initial energy recovery model according to the verification result to obtain a target energy recovery model.

[0085] Further, the initial energy recovery model is verified using the verification set data to evaluate its prediction performance; according to the verification result, the model is optimized, such as adjusting model parameters, increasing network layers, etc., to obtain a more accurate and robust target energy recovery model.

[0086] It should be noted that in S400, the initial energy recovery model is verified according to the verification set data, and the initial energy recovery model is optimized according to the verification result to obtain a target energy recovery model, specifically:

[0087] S410: inputting the verification set into the initial energy recovery model for verification processing, and the initial energy recovery model outputting a verification result;

[0088] S420: differentiating the verification result from the verification set to obtain a difference function between the verification set and the verification result;

[0089] S430: analyze the difference features in the validation set according to the difference function, and adjust the parameters and structure of the initial energy recovery model according to the difference features;

[0090] S440: train the initial energy recovery model adjusted according to the training set to obtain the target energy recovery model.

[0091] Further, first, the validation set data is input into the initial energy recovery model for verification processing; the purpose of this step is to evaluate the performance of the model on unseen data, i.e. the generalization ability of the model; the model will input the features of the validation set and output the corresponding energy recovery prediction results. Then, the verification results are compared with the real data of the validation set; this step aims to quantify the difference between the model prediction results and the actual data, usually by calculating the error (such as mean square error, mean absolute error, etc.) or constructing a difference function (such as a loss function); the difference function can reflect the prediction performance of the model on different features, providing guidance for subsequent optimization. According to the difference function, analyze the difference features in the validation set, these features may be the key factors that cause the model prediction error to be large; by analyzing these features, we can identify the shortcomings of the model in which aspects, and then adjust the parameters (such as weights, biases, etc.) and structure (such as network layers, number of neurons, etc.) of the initial energy recovery model; the purpose of adjustment is to reduce the prediction error and improve the accuracy and generalization ability of the model. Finally, the training set is used to retrain the initial energy recovery model adjusted; this step aims to consolidate the performance of the model on known data and further improve the generalization ability of the model on unseen data; after retraining, the obtained model is the target energy recovery model, which will be used in the subsequent real-time matching and execution phase.

[0092] S500: obtain the driving parameters of the target vehicle every first preset time to match the corresponding energy recovery data in the energy recovery model according to the driving parameters;

[0093] Further, the driving parameters of the target vehicle are obtained every first preset time (such as every second or every few minutes); according to these driving parameters, the corresponding energy recovery data in the energy recovery model is searched.

[0094] It should be noted that in S500, the driving parameters of the target vehicle are obtained every first preset time to match the corresponding energy recovery data in the energy recovery model according to the driving parameters, specifically:

[0095] S510: the target vehicle obtains the driving parameters of the target vehicle every first preset time according to the vehicle-mounted sensor;

[0096] S520: The vehicle controller receives the driving parameters and pre-processes the driving parameters to obtain target driving parameters;

[0097] S530: The vehicle controller inputs the target driving parameters into the target energy recovery model, and the target energy recovery model outputs corresponding energy recovery data according to the target driving parameters.

[0098] Furthermore, the target vehicle's onboard sensors (such as speed sensors, acceleration sensors, and brake pressure sensors) acquire driving parameters at predetermined intervals (e.g., 1 second, 0.5 seconds, etc., depending on actual needs). These parameters directly reflect the vehicle's current state and include speed, acceleration, brake pressure, and battery status, and are crucial for controlling the energy recovery system. The vehicle control unit (VCU) receives the driving parameters from the onboard sensors and preprocesses them. This preprocessing includes data cleaning (removing noise and outliers), data conversion (such as normalization), and data screening (selecting key parameters that affect energy recovery). These preprocessed parameters are referred to as target driving parameters, which are more accurate and reliable and suitable for subsequent energy recovery model matching. The vehicle controller inputs the target driving parameters into the target energy recovery model. The target energy recovery model is trained and optimized to predict corresponding energy recovery data based on the input driving parameters. This data includes energy recovery amount, recovery efficiency, and recommended recovery strategy. By matching the target energy recovery model, we can obtain the optimal energy recovery solution for the vehicle in its current state in real time, thereby achieving precise control of the energy recovery system.

[0099] S600: The vehicle controller of the target vehicle converts the energy recovery data into vehicle command data, and the target vehicle performs corresponding energy recovery actions according to the vehicle command data.

[0100] Furthermore, the vehicle controller converts these energy recovery data into vehicle command data, such as adjusting the braking system, optimizing the motor working mode, etc.; the target vehicle performs corresponding energy recovery actions according to these vehicle command data, such as braking energy recovery, energy recovery when going downhill, etc.

[0101] It should be noted that in S600, the vehicle controller of the target vehicle converts the energy recovery data into vehicle command data, and the target vehicle performs the corresponding energy recovery action according to the vehicle command data, specifically:

[0102] S610: The vehicle controller converts the energy recovery data output by the target energy recovery model into vehicle command data that can be recognized and executed by the target vehicle;

[0103] S620: The vehicle control unit sends the vehicle command data to the target vehicle through the actuators for execution, and the target vehicle performs the corresponding energy recovery action according to the execution instruction.

[0104] Further, the vehicle control unit (VCU) is responsible for converting the energy recovery data output by the target energy recovery model into vehicle command data that the target vehicle can recognize and execute; this process involves data format conversion, signal amplification or reduction, and necessary adjustments according to the characteristics of the vehicle hardware; the converted vehicle command data should accurately reflect the optimal energy recovery strategy predicted by the energy recovery model, while ensuring compatibility with the vehicle actuators. The vehicle control unit sends the converted vehicle command data to the corresponding actuators through the vehicle's internal communication network (such as CAN bus, LIN bus, etc.); the actuators include the braking system, motor control system, battery management system, etc., which are responsible for executing specific energy recovery actions according to the received command data; the target vehicle adjusts its working state according to the instructions of the actuators, such as adjusting the brake pressure, adjusting the motor working mode or optimizing the battery charging strategy, etc., to achieve effective energy recovery.

[0105] According to the vehicle energy management control method, the historical driving data of the target vehicle is comprehensively collected through diversified data sources such as vehicle-mounted sensors, Internet of Vehicles technology, traffic monitoring cameras and GPS devices, so as to ensure the richness and diversity of the data. The historical driving data is screened, integrated and standardized, invalid data is effectively removed, the data format and storage structure are unified, and the data is more standardized, easy to analyze and use. The preprocessed target driving data is uploaded to a cloud server for storage, which facilitates subsequent data access, analysis and model training, and improves the efficiency and security of data management. The target driving data is divided by using a bootstrap method to obtain a training set and a validation set, so as to ensure the independence and representativeness of the data in the model training process. Key features affecting energy recovery are extracted from the training set data through feature extraction and screening processing, which provides a solid foundation for constructing an energy recovery model. An initial energy recovery model is constructed by using a deep learning algorithm, and is trained by using the training set data, so that the model can learn complex energy recovery rules. The initial energy recovery model is verified and optimized by using the validation set data, and the parameters and structure of the model are continuously adjusted until a target energy recovery model with high precision and generalization ability is obtained. The target vehicle obtains driving parameters in the driving state at a preset time interval according to the vehicle-mounted sensor, so as to ensure the real-time and accuracy of the data. The vehicle controller inputs the driving parameters to the target energy recovery model after preprocessing, and the model outputs corresponding energy recovery data according to the driving parameters, so as to realize intelligent energy recovery control. The vehicle controller converts the energy recovery data into vehicle command data and issues it to the target vehicle through the actuator for execution, so that the vehicle can perform energy recovery actions according to the real-time driving state, and the efficiency and flexibility of energy recovery are improved. Through the intelligent energy management control method, the vehicle can realize more efficient energy recovery under different driving conditions, thereby reducing energy waste and pollution emissions. In the long run, this method helps to improve the energy efficiency of the vehicle, reduce operating costs, reduce the impact on the environment, and conforms to the concept of sustainable development.

[0106] It should be noted that the vehicle energy management control method provided in the embodiments of the present application can be executed by a vehicle energy management control system, or a control module in the vehicle energy management control system for executing the vehicle energy management control method. In the embodiments of the present application, the vehicle energy management control system is taken as an example to execute the vehicle energy management control method, and the vehicle energy management control method provided in the embodiments of the present application is described.

[0107] A vehicle energy management control system comprises:

[0108] The first acquisition module acquires historical driving data of a target vehicle.

[0109] The data processing module: pre-process the historical driving data to obtain target driving data;

[0110] The data storage module: for storing target driving data;

[0111] The data division module: according to the bootstrap method, the target driving data is divided to obtain training set data and validation set data;

[0112] The model construction module: according to the training set data and the deep learning algorithm, the initial energy recovery model of the target vehicle is constructed;

[0113] The model optimization module: according to the validation set data, the initial energy recovery model is verified, and the initial energy recovery model is optimized according to the verification result to obtain the target energy recovery model.

[0114] The second acquisition module: the driving parameters of the target vehicle are acquired every first preset time;

[0115] The data matching module: according to the driving parameters, the corresponding energy recovery data in the energy recovery model is matched;

[0116] The data conversion module: the energy recovery data is converted into vehicle command data by the vehicle controller of the target vehicle, and the target vehicle executes the corresponding energy recovery action according to the vehicle command data.

[0117] The vehicle energy management control system in the embodiment of the application can be a device, a component, an integrated circuit, or a chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. Illustratively, the mobile electronic device can be a mobile phone, a tablet computer, a notebook computer, a palm computer, a vehicle-mounted electronic device, a wearable device, an ultra-mobile personal computer (UMPC), a netbook, or a personal digital assistant (PDA), etc. The non-mobile electronic device can be a server, a network attached storage (NAS), a personal computer (PC), etc. The embodiment of the application is not limited in this regard.

[0118] The vehicle energy management control system in the embodiment of the application can be a device with an operating system. The operating system can be an Android operating system, an ios operating system, or other possible operating systems. The embodiment of the application is not limited in this regard.

[0119] The vehicle energy management control system provided in the embodiment of the application canFigure 1 The processes of the vehicle energy management control method in the method embodiment are not repeated here.

[0120] According to the vehicle energy management control system of the embodiment, the historical driving data of the target vehicle is obtained by the first acquisition module, so that the system can optimize energy management for a specific vehicle instead of using a general strategy, which enhances the personalization and adaptability of the system. The data processing module pre-processes the historical driving data to ensure the quality and consistency of the input data, which helps to improve the accuracy of the subsequently constructed energy recovery model. The data division module divides the data using the bootstrap method to ensure the independence and representativeness of the training set and the validation set, thereby effectively evaluating and optimizing the initial energy recovery model. The model optimization module iteratively optimizes the model based on the feedback of the validation set data, and the final target energy recovery model is more accurate and reliable. The second acquisition module obtains the driving parameters every preset time to ensure that the energy recovery strategy can respond to changes in the vehicle state in real time. The data matching module can quickly find the corresponding energy recovery data according to the current driving parameters, improving the real-time and pertinence of energy recovery decision. The data conversion module converts the energy recovery data into vehicle command data, so that the vehicle controller can directly understand and execute, improving the efficiency and accuracy of energy recovery actions. Through the comprehensive action of the above modules, the system can dynamically adjust the energy recovery strategy according to the actual driving conditions of the target vehicle, effectively improving the energy utilization efficiency of the vehicle, prolonging the cruising range, and reducing the energy consumption cost. The entire vehicle energy management control system is highly automated and intelligent, reducing manual intervention and improving management efficiency and decision quality.

[0121] Optionally, the embodiment of the application further provides an electronic device, including a processor, a memory, a program or instructions stored on the memory and executable on the processor, which implement the processes of the above-mentioned vehicle energy management control method embodiments when executed by the processor, and can achieve the same technical effects, and are not repeated here to avoid repetition.

[0122] The embodiment of the application further provides a readable storage medium, which stores a program or instructions, which implement the processes of the above-mentioned vehicle energy management control method embodiments when executed by the processor, and can achieve the same technical effects, and are not repeated here to avoid repetition.

[0123] The processor is the processor of the electronic device in the above-mentioned embodiments. The readable storage medium includes a computer readable storage medium, such as a computer readable memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a magnetic disk or an optical disk, etc.

[0124] It should be noted that the terms "comprising," "including," and any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises a... " does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element. Additionally, it should be noted that the scope of the methods and apparatus of the present embodiments are not limited by the order of the steps or the sequence for performing the steps, as some steps can occur simultaneously, in other steps can occur sequentially, or in an order different from that described. Furthermore, features described with respect to certain examples can be combined in other examples.

[0125] From the above description of the embodiments, it is apparent that the method of the above embodiments can be realized by software plus necessary universal hardware platforms, and of course can also be realized by hardware, but in many cases the former is a better embodiment. Based on such an understanding, the technical solutions of the present application can be embodied in the form of a software product, and the computer software product is stored in a storage medium (such as a ROM / RAM, a magnetic disk, or an optical disk), and includes a plurality of instructions to make a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) execute the methods described in the various embodiments of the present application.

[0126] The embodiments of the present application are described above in conjunction with the drawings, but the present application is not limited to the above specific embodiments, and the above specific embodiments are merely illustrative rather than restrictive, and a person of ordinary skill in the art can make many forms under the inspiration of the present application without departing from the scope of the present application and the scope protected by the claims.

Claims

1. A vehicle energy management control method, characterized in that: The method comprises: Acquiring historical driving data of a target vehicle, preprocessing the historical driving data to obtain target driving data, and transmitting the target driving data to a cloud server for storage; Performing data segmentation processing on the target driving data according to a bootstrap method to obtain training set data and validation set data; Constructing an initial energy recovery model for the target vehicle based on the training set data and the deep learning algorithm; Verifying the initial energy recovery model according to the verification set data, and optimizing the initial energy recovery model according to the verification result to obtain a target energy recovery model; Acquiring driving parameters of the target vehicle under the driving state once every first preset time, so as to match corresponding energy recovery data in the energy recovery model according to the driving parameters; The vehicle controller of the target vehicle converts the energy recovery data into vehicle command data, and the target vehicle performs a corresponding energy recovery action according to the vehicle command data; The data partitioning process of the target driving data according to the bootstrap method to obtain training set data and validation set data is specifically as follows: uniformly sampling the target driving data to obtain a data set containing m data samples; performing m times of sampling with replacement on the data set to obtain the training set data; and deleting the data in the target driving data that is identical to the data in the training set to obtain the validation set data. The constructing of the initial energy recovery model of the target vehicle based on the training set data and the deep learning algorithm specifically comprises: performing feature extraction processing on the training set data to extract energy recovery features in the training set data to construct an energy recovery feature set; performing feature screening processing on the energy recovery feature set to screen out key features that affect energy recovery in the energy recovery feature set to obtain a key feature set; constructing the energy recovery model including multiple energy recovery data based on the deep learning algorithm and the key feature set; and training the energy recovery model based on the training set to obtain the initial energy recovery model. The initial energy recovery model is verified according to the verification set data, and the initial energy recovery model is optimized according to the verification result to obtain the target energy recovery model, specifically: the verification set is input into the initial energy recovery model for verification processing, and the initial energy recovery model outputs a verification result; the verification result is differentiated and compared with the verification set to obtain a difference function between the verification set and the verification result; the difference features in the verification set are analyzed according to the difference function to adjust the parameters and structure of the initial energy recovery model according to the difference features; the adjusted initial energy recovery model is trained according to the training set to obtain the target energy recovery model.

2. A vehicle energy management control method according to claim 1, characterized in that: The acquiring of historical driving data of the target vehicle, preprocessing of the historical driving data to obtain target driving data, and transmitting the target driving data to a cloud server for storage are specifically as follows: Collecting the historical driving data of the target vehicle using onboard sensors, vehicle networking technology, traffic monitoring cameras, and GPS devices; screening the historical driving data to delete invalid data in the historical driving data; Performing data integration on the historical driving data after the screening process to form a unified data format and storage structure; performing standardization processing on the historical driving data after the data integration, so that data of different dimensions can be compared and analyzed to obtain the target driving data; The target driving data is uploaded to a cloud server through a data transmission channel, and a corresponding data directory and database are created on the cloud server for storing the target driving data.

3. A vehicle energy management control method according to claim 1, characterized in that: The acquiring of the driving parameters of the target vehicle in the driving state once every first preset time, so as to match the corresponding energy recovery data in the energy recovery model according to the driving parameters, is specifically: The target vehicle obtains the driving parameters of the driving state once every first preset time according to the vehicle-mounted sensor; The vehicle controller receives the driving parameters and performs the preprocessing on the driving parameters to obtain target driving parameters; The vehicle controller inputs the target driving parameters into the target energy recovery model, and the target energy recovery model outputs corresponding energy recovery data according to the target driving parameters.

4. A vehicle energy management control method according to claim 3, characterized in that: The vehicle controller of the target vehicle converts the energy recovery data into vehicle command data, and the target vehicle performs a corresponding energy recovery action according to the vehicle command data, specifically: The vehicle controller converts the energy recovery data output by the target energy recovery model into the vehicle command data that can be recognized and executed by the target vehicle; The vehicle controller sends the vehicle command data to the target vehicle through the actuator for execution, and the target vehicle performs the corresponding energy recovery action according to the execution instruction.

5. A vehicle energy management control system, implementing a vehicle energy management control method according to any one of claims 1 to 4, characterized in that: The system comprises: The first acquisition module: acquires the historical driving data of the target vehicle; Data processing module: pre-processing the historical driving data to obtain target driving data; Data storage module: used to store the target driving data; Data partitioning module: performing data partitioning processing on the target driving data according to the bootstrap method to obtain training set data and validation set data; Model construction module: constructing an initial energy recovery model of the target vehicle based on the training set data and the deep learning algorithm; Model optimization module: verifying the initial energy recovery model according to the verification set data, and optimizing the initial energy recovery model according to the verification result to obtain a target energy recovery model; A second acquisition module is configured to acquire the driving parameters of the target vehicle under the driving state once every first preset time; A data matching module is configured to match the corresponding energy recovery data in the energy recovery model according to the driving parameters; Data conversion module: The vehicle controller of the target vehicle converts the energy recovery data into vehicle command data, and the target vehicle performs the corresponding energy recovery action according to the vehicle command data; The data partitioning process of the target driving data according to the bootstrap method to obtain training set data and validation set data is specifically as follows: uniformly sampling the target driving data to obtain a data set containing m data samples; performing m times of sampling with replacement on the data set to obtain the training set data; and deleting the data in the target driving data that is identical to the data in the training set to obtain the validation set data. The constructing of the initial energy recovery model of the target vehicle based on the training set data and the deep learning algorithm specifically comprises: performing feature extraction processing on the training set data to extract energy recovery features in the training set data to construct an energy recovery feature set; performing feature screening processing on the energy recovery feature set to screen out key features that affect energy recovery in the energy recovery feature set to obtain a key feature set; constructing the energy recovery model including multiple energy recovery data based on the deep learning algorithm and the key feature set; and training the energy recovery model based on the training set to obtain the initial energy recovery model. The initial energy recovery model is verified according to the verification set data, and the initial energy recovery model is optimized according to the verification result to obtain the target energy recovery model, specifically: the verification set is input into the initial energy recovery model for verification processing, and the initial energy recovery model outputs a verification result; the verification result is differentiated and compared with the verification set to obtain a difference function between the verification set and the verification result; the difference features in the verification set are analyzed according to the difference function to adjust the parameters and structure of the initial energy recovery model according to the difference features; the adjusted initial energy recovery model is trained according to the training set to obtain the target energy recovery model.

6. An electronic device, characterized in that: It includes a processor, a memory, and a program or instruction stored in the memory and executable on the processor. When the program or instruction is executed by the processor, the steps of a vehicle energy management control method as described in claims 1-4 are implemented.

7. A readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the steps of the vehicle energy management control method as described in claims 1-4 are implemented.

Images