Expressway driving journey identification and energy consumption optimization system and method

Abstract

The invention discloses an expressway driving journey identification and energy consumption optimization system and method. The method comprises the following steps: constructing an expressway energy consumption database by using historical driving data, historical driving environment data and historical energy consumption data of a vehicle in each historical time period; carrying out clustering analysis on the current group of instantaneous vehicle speed sequences of the vehicle through a K-means algorithm, and judging whether the driving place of the vehicle is an expressway or not; when it is judged that the driving place of the vehicle is a highway, historical driving environment data with the highest matching degree with the preprocessed current driving environment data of the vehicle is retrieved from historical driving environment data of all historical time periods in a highway energy consumption database through an RAG model to serve as selected historical driving environment data; the historical energy consumption data and the historical driving data of the time period corresponding to the selected historical driving environment data are screened out from the historical energy consumption data of all the historical time periods, and the RAG model generates the current driving suggestion of the vehicle according to the screened historical energy consumption data and the historical driving data.

Description

Technical Field

[0001] This invention relates to the field of high-speed driving trip recognition for automobiles, specifically to a system and method for highway driving trip recognition and energy consumption optimization. Background Technology

[0002] In the fields of intelligent transportation, vehicle management, and related data analysis, accurately determining whether a vehicle is traveling on a highway or in urban areas is crucial. However, existing methods have several limitations. Traditional methods typically use fixed speed thresholds to distinguish between highway and urban driving. This approach is overly simplistic and fails to consider the complexities of real-world driving. For example, on highways, speeds may fall below the threshold due to traffic congestion or accidents; while on urban expressways, speeds may exceed the threshold. Therefore, fixed-threshold methods are prone to misjudgment and cannot accurately reflect the true driving scenario. Furthermore, vehicle speed is dynamic, and most existing methods only focus on instantaneous speed or average speed over a specific time period, ignoring speed fluctuations. In urban areas, speeds frequently change due to traffic lights, traffic flow, and other factors; on highways, speeds are relatively stable. Without considering the dynamic characteristics of speed changes, it is difficult to accurately distinguish between different driving scenarios. With the generation of massive amounts of vehicle data, efficiently processing and analyzing this data becomes a significant challenge. Existing methods often require significant time and computational resources to process large-scale vehicle speed data, failing to meet real-time requirements. Furthermore, while some complex algorithms can improve accuracy, their high computational complexity limits their practical application. Traffic conditions vary across different regions, including highways and urban areas; for example, speed limits and traffic flow distribution differ in urban areas, and the number of lanes and traffic volume vary across highways. Existing methods are often based on fixed rules or models, lacking adaptability and flexibility, and cannot be adjusted to the specific conditions of different regions, resulting in poor performance in various applications.

[0003] With the widespread adoption of new energy vehicles, highway driving is a common driving scenario for users. However, highway driving conditions differ significantly from urban road conditions. Vehicles on highways typically travel at higher speeds continuously, and frequent overtaking, acceleration, and deceleration, along with prolonged use of equipment such as air conditioning and headlights, all contribute to a substantial increase in energy consumption for new energy vehicles. Furthermore, factors such as terrain variations, weather conditions, and traffic density on different highway sections also significantly impact energy consumption. Currently, while some new energy vehicles possess energy consumption monitoring capabilities, energy consumption optimization suggestions specifically for highway driving scenarios are severely lacking. Existing technologies cannot effectively combine the specific highway environment with the vehicle's real-time status to provide users with accurate and real-time driving energy consumption optimization strategies through effective retrieval and intelligent generation of historical data and professional knowledge. This makes it difficult to meet users' needs for reducing energy consumption and increasing range while driving on highways. Summary of the Invention

[0004] The purpose of this invention is to provide a highway driving trip recognition and energy consumption optimization system and method. This invention can make full use of the vehicle's instantaneous speed data, consider the dynamic changes in vehicle speed, automatically adapt to traffic conditions in different regions, achieve efficient and accurate driving scenario judgment, and utilize RAG technology to deeply mine vehicle data and knowledge in highway driving scenarios, providing users with personalized and real-time driving energy consumption suggestions that fit the actual conditions of highways, effectively reducing the energy consumption of new energy vehicles when driving on highways and extending the vehicle's driving range.

[0005] To achieve this objective, the present invention provides a highway driving trip recognition and energy consumption optimization system, comprising: The resource database creation module is used to build a highway energy consumption database by utilizing historical driving data, historical driving environment data, and historical energy consumption data from various periods of vehicle history. The driving location identification module is used to perform cluster analysis on a set of instantaneous vehicle speeds using the K-means algorithm to determine whether the vehicle is driving on a highway. The energy consumption optimization module is used to select historical driving environment data when the vehicle is determined to be driving on a highway. It uses the RAG model to retrieve the historical driving environment data with the highest matching degree with the preprocessed current driving environment data of the vehicle from the historical driving environment data of various historical time periods in the highway energy consumption database. Then, it filters the historical energy consumption data and historical driving data corresponding to the selected historical driving environment data from the historical energy consumption data of various historical time periods. The RAG model generates the current driving suggestions of the vehicle based on the filtered historical energy consumption data and historical driving data.

[0006] Furthermore, the vehicle's historical driving data includes vehicle speed, acceleration, braking frequency, battery charge, and driving behavior data for various historical time periods; historical driving environment data includes highway section gradient information, traffic flow density, weather conditions, and following distance for various historical time periods; and historical energy consumption data includes driving energy consumption and auxiliary energy consumption.

[0007] Furthermore, the resource database establishment module is used to acquire historical driving data, historical driving environment data, and historical energy consumption data of vehicles. It preprocesses the historical driving data, historical driving environment data, and historical energy consumption data of vehicles, and uses the preprocessed historical driving data, historical driving environment data, and historical energy consumption data of vehicles in various historical periods to construct a highway energy consumption database. The energy consumption optimization module is used to, when the vehicle is determined to be traveling on a highway, use the RAG model to retrieve the historical driving environment data with the highest matching degree with the preprocessed current driving environment data of the vehicle from the historical driving environment data of various historical time periods in the highway energy consumption database as the selected historical driving environment data. Then, it filters the historical energy consumption data and historical driving data corresponding to the selected historical driving environment data from the historical energy consumption data of various historical time periods. The RAG model generates the current driving suggestions for the vehicle based on the filtered historical energy consumption data and historical driving data.

[0008] Furthermore, the method of determining whether a vehicle is traveling on a highway by performing cluster analysis on a set of instantaneous vehicle speed data using the K-means algorithm includes: collecting instantaneous vehicle speed data at set time intervals during vehicle travel, forming an instantaneous speed data series with a set number of instantaneous speeds, dividing the vehicle's travel scenario into urban travel and highway travel, setting the cluster size K value for the K-means algorithm based on the two travel scenarios, and then analyzing the instantaneous speed data series based on the cluster size K value to output the vehicle's travel scenario.

[0009] Furthermore, the K-means algorithm's method for analyzing instantaneous vehicle speed sequences based on the number of clusters, K, includes: The K-means algorithm randomly generates a first and second instantaneous speed centroid based on each vehicle speed data point in the instantaneous speed sequence; it calculates the distance from each vehicle speed data point in the instantaneous speed sequence to the first and second instantaneous speed centroids; and, based on the principle of proximity to the centroid, assigns each vehicle speed data point in the instantaneous speed sequence to the corresponding first or second instantaneous speed centroid, thus forming a first data cluster and a second data cluster. The mean of the second data cluster is used as the new second instantaneous speed centroid, and the mean of the first data cluster is used as the new first instantaneous speed centroid. The distance from each vehicle speed data point in the instantaneous speed sequence to the new second instantaneous speed centroid is then calculated. The distance from each speed data point in the instantaneous speed sequence to the new first instantaneous speed centroid is used to regenerate the updated second data cluster and the updated first data cluster, completing one centroid iteration. This centroid iteration process is repeated until the second instantaneous speed centroid and the first instantaneous speed centroid converge. The converged second instantaneous speed centroid is compared with the converged first instantaneous speed centroid. The speed data points in the data cluster containing the larger speed centroid are identified as highway driving data, and the speed data points in the data cluster containing the smaller speed centroid are identified as urban driving data. When the number of speed data points in the highway driving data is greater than or equal to the number of speed data points in the urban driving data, it is determined that the vehicle is driving on a highway.

[0010] Furthermore, the RAG model generates current driving suggestions for the vehicle based on the selected historical energy consumption data and historical driving data in the following specific way: it generates suggestions on the vehicle operation required when the current energy consumption data of the vehicle reaches the selected historical energy consumption data and the current driving data of the vehicle reaches the selected historical driving data.

[0011] Furthermore, when the vehicle's current driving environment is heavy rain, the historical energy consumption data selected by the RAG model is the lowest historical power battery power consumption of the car in heavy rain, and the historical driving data selected by the RAG model is the historical speed range corresponding to the lowest historical power battery power consumption of the car in heavy rain. The system then generates vehicle operation suggestions to ensure that the vehicle's current energy consumption data reaches the selected lowest historical power battery power consumption and that the vehicle's current driving data reaches the historical speed range corresponding to the selected lowest historical power battery power consumption.

[0012] Furthermore, when the vehicle's current driving environment is hot weather, the historical energy consumption data selected by the RAG model is the lowest historical air conditioning energy consumption of the vehicle's air conditioning in hot weather. The historical driving data selected by the RAG model is the air conditioning temperature and fan speed corresponding to the lowest historical air conditioning energy consumption of the vehicle in hot weather. The system then generates vehicle operation suggestions to ensure that the vehicle's current energy consumption data reaches the selected lowest historical air conditioning energy consumption and that the vehicle's current driving data reaches the air conditioning fan speed and temperature corresponding to the selected lowest historical air conditioning energy consumption. Furthermore, the highway driving trip identification and energy consumption optimization method based on the aforementioned system includes: A highway energy consumption database is constructed using historical driving data, historical driving environment data, and historical energy consumption data from various periods of vehicle history. The K-means algorithm is used to perform cluster analysis on a set of instantaneous vehicle speeds to determine whether the vehicle is traveling on a highway. When the vehicle is determined to be traveling on a highway, the RAG model is used to retrieve the historical driving environment data with the highest matching degree to the vehicle's current driving environment data from the historical driving environment data of various historical time periods in the highway energy consumption database. The selected historical driving environment data is then selected by filtering the historical energy consumption data and historical driving data corresponding to the selected historical driving environment data from the historical energy consumption data of various historical time periods. The RAG model generates the vehicle's current driving suggestions based on the filtered historical energy consumption data and historical driving data.

[0013] The beneficial effects of this invention are as follows: This invention proposes a vehicle driving scenario judgment method based on the K-Means algorithm. This method can fully utilize instantaneous vehicle speed data, consider the dynamic changes in vehicle speed, and automatically adapt to traffic conditions in different regions, achieving efficient and accurate driving scenario judgment. Through this invention, more reliable decision-making basis can be provided for intelligent transportation systems, vehicle management, and related data analysis, improving the efficiency of traffic management and the operational safety of vehicles. After identifying the high-speed driving journey, this invention also provides a highway new energy vehicle driving energy consumption suggestion system and method based on RAG (Rapid Energy Management). Utilizing RAG technology, it deeply mines vehicle data and knowledge under highway driving scenarios, providing users with personalized, real-time driving energy consumption suggestions tailored to actual highway conditions, effectively reducing the energy consumption of new energy vehicles when driving on highways and extending the vehicle's driving range. Attached Figure Description

[0014] Figure 1 This is a flowchart illustrating the overall solution of the present invention; Figure 2 This is a schematic diagram of the structure of the present invention; Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of them. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to represent selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0016] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments: Example 1 like Figure 2 As shown, a highway driving trip recognition and energy consumption optimization system includes: The resource database creation module is used to build a highway energy consumption database by utilizing historical driving data, historical driving environment data, and historical energy consumption data from various periods of vehicle history. The driving location identification module is used to perform cluster analysis on a set of instantaneous vehicle speeds using the K-means algorithm to determine whether the vehicle is driving on a highway. The energy consumption optimization module is used to retrieve the historical driving environment data with the highest matching degree with the vehicle's current driving environment data from the historical driving environment data of various historical time periods in the highway energy consumption database when the vehicle is determined to be driving on a highway. The module then selects the historical energy consumption data and historical driving data corresponding to the selected historical driving environment data from the historical energy consumption data of various historical time periods. The RAG model generates the vehicle's current driving suggestions based on the selected historical energy consumption data and historical driving data.

[0017] like Figure 1 As shown, the process of highway driving trip recognition and energy consumption optimization is as follows: data collection and preprocessing, running the K-Means algorithm to determine the driving scenario corresponding to the cluster, RAG engine module to match historical data and knowledge, suggestion generation module to generate energy consumption suggestions based on highway driving characteristics, and user interaction: visual prompts and voice broadcast.

[0018] In some technical solutions, the vehicle's historical driving data includes vehicle speed, acceleration, braking frequency, battery charge, and driving behavior data for various historical time periods; historical driving environment data includes highway section gradient information, traffic flow density, weather conditions, and following distance for various historical time periods; and historical energy consumption data includes driving energy consumption and auxiliary energy consumption.

[0019] Driving energy consumption refers to the energy consumed directly for vehicle operation, accounting for the majority of total energy consumption. It is related to vehicle speed, load, and road resistance. Auxiliary energy consumption includes the energy consumed by air conditioning, lighting, and onboard equipment. Historical driving data, historical driving environment data, and historical energy consumption data are automatically collected by the T-BOX installed in the new energy vehicle through various types of sensors deployed on the vehicle. In addition to collecting conventional data such as vehicle speed, acceleration, braking frequency, and battery charge, the data focuses on highway scenarios, collecting information such as road slope (obtained through high-precision maps combined with onboard sensors), traffic flow density (using onboard cameras and radar combined with traffic big data analysis), weather conditions (obtained through vehicle-to-everything (V2X) weather information), and following distance. A dedicated highway data sub-database is constructed to store historical energy consumption data, corresponding driving behavior data, and highway-related environmental data for different highway sections, including slope characteristics of each section, traffic flow patterns at different times, and the impact of weather on energy consumption, forming a rich highway energy consumption data resource library.

[0020] In some technical solutions, the resource database establishment module is used to obtain historical driving data, historical driving environment data and historical energy consumption data of vehicles, preprocess the historical driving data, historical driving environment data and historical energy consumption data of vehicles, and use the preprocessed historical driving data, historical driving environment data and historical energy consumption data of vehicles in various historical periods to build a highway energy consumption database. The energy consumption optimization module is used to, when the vehicle is determined to be traveling on a highway, use the RAG model to retrieve the historical driving environment data with the highest matching degree with the preprocessed current driving environment data of the vehicle from the historical driving environment data of various historical time periods in the highway energy consumption database as the selected historical driving environment data. Then, it filters the historical energy consumption data and historical driving data corresponding to the selected historical driving environment data from the historical energy consumption data of various historical time periods. The RAG model generates the current driving suggestions for the vehicle based on the filtered historical energy consumption data and historical driving data.

[0021] Methods for preprocessing historical vehicle driving data, historical driving environment data, and historical energy consumption data include parsing, standardization, and outlier removal to eliminate missing, duplicate, outlier, and redundant data. This improves data reliability. It should be noted that current vehicle driving data, current driving environment data, and current energy consumption data also require preprocessing.

[0022] In some technical solutions, the method of using the K-means algorithm to cluster a set of instantaneous vehicle speed data to determine whether the vehicle is traveling on a highway includes: collecting instantaneous vehicle speed data at set time intervals during vehicle travel, forming an instantaneous speed data series with a set number of instantaneous speeds, dividing the vehicle's travel scenario into urban driving and highway driving, setting the cluster size K value for the K-means algorithm based on the two travel scenarios, and then analyzing the instantaneous speed data series based on the cluster size K value to output the vehicle's travel scenario.

[0023] The K-means algorithm uses cluster analysis of instantaneous vehicle speeds to accurately determine whether a user is driving on a highway or in urban areas. Compared to traditional methods that rely on fixed thresholds, the K-means algorithm takes into account the complexities that may occur during actual driving, thus avoiding the risk of misjudgment. Furthermore, compared to traditional methods that use average vehicle speed to distinguish between highway driving and urban driving, the K-means algorithm considers dynamic changes in vehicle speed, which more realistically reflects the changing characteristics of the driving scenario, improving the reliability and adaptability of the judgment.

[0024] In some technical solutions, the K-means algorithm analyzes instantaneous vehicle speed sequences based on the number of clusters, K. The K-means algorithm randomly generates a first and second instantaneous speed centroid based on each vehicle speed data point in the instantaneous speed sequence. It calculates the distance from each vehicle speed data point in the instantaneous speed sequence to the first and second instantaneous speed centroids. Based on the principle of proximity to the centroid, it assigns each vehicle speed data point in the instantaneous speed sequence to the corresponding first or second instantaneous speed centroid, thus forming a first data cluster and a second data cluster. The mean of the second data cluster is used as the new second instantaneous speed centroid, and the mean of the first data cluster is used as the new first instantaneous speed centroid. The distance from each vehicle speed data point in the instantaneous speed sequence to the new second instantaneous speed centroid is then calculated. The centroid is determined by the distance from each speed data point in the instantaneous speed sequence to the new first instantaneous speed centroid. This process is repeated until the second and first instantaneous speed centroids converge. The converged second and first instantaneous speed centroids are then compared with the converged first instantaneous speed centroid. Speed ​​data points in the data cluster containing the larger centroid are identified as highway driving data, while those in the data cluster containing the smaller centroid are identified as urban driving data. If the number of speed data points in the highway driving data is greater than or equal to the number of speed data points in the urban driving data, the vehicle is determined to be traveling on a highway.

[0025] The centroid nearest principle involves comparing each speed data point in the instantaneous speed sequence with the first and second instantaneous speed centroids, and assigning each speed data point in the instantaneous speed sequence to the nearest centroid to form a data cluster. In some embodiments, when the first instantaneous speed centroid is 45 km / h, the second instantaneous speed centroid is 85 km / h, and a speed data point in the instantaneous speed sequence is 50 km / h, since 50 km / h - 45 km / h is less than 85 km / h - 50 km / h, the speed data point 50 km / h in the instantaneous speed sequence is assigned to the first instantaneous speed centroid 45 km / h to form a data cluster.

[0026] The K-means algorithm analyzes instantaneous vehicle speed sequences based on the number of clusters, K. It randomly generates first and second instantaneous speed centroids, calculates the distance from each speed data point to the centroid, and assigns data points to form first and second data clusters based on the nearest-neighbor principle. The centroids are then iteratively updated until convergence. Next, the two convergent centroids are compared. The data cluster containing the centroid with the larger speed is identified as highway driving data, and the data cluster containing the smaller centroid is identified as urban driving data. By comparing the number of highway driving data points with the number of urban driving data points, the algorithm determines whether a vehicle is traveling on a highway. This improves the accuracy and adaptability of driving scenario judgment, avoiding the misjudgment risk caused by traditional fixed-threshold methods that ignore complex road conditions (such as highway congestion or speeding on urban expressways). It also overcomes the deficiency of average speed methods in responding insufficiently to dynamic changes in vehicle speed (such as frequent starts and stops in urban areas). Through an adaptive clustering mechanism, it automatically learns vehicle speed distribution characteristics, effectively handling instantaneous speed fluctuations. This provides more reliable and real-time data support in intelligent transportation systems and vehicle energy management, optimizing driving safety and energy efficiency.

[0027] In some embodiments, the K-means algorithm is implemented as follows (using Python as an example). import numpy as np from sklearn.cluster import KMeans Def Kmeans_hightway_model(speeds): speeds = np.array([30, 35, 40, 45, 50, 80, 85, 90, 95, 100]).reshape(-1, 1) The sequence of instantaneous vehicle speeds is 30, 35, 40, 45, 50, 80, 85, 90, 95, 100. # Using the K-Means clustering algorithm, the number of clusters K is 2. kmeans = KMeans(n_clusters=2, random_state=42) kmeans.fit(speeds) # Get cluster labels (cluster labels include 0 and 1, where 1 represents highway driving labels and 0 represents urban driving labels) labels = kmeans.labels_ # Determine the clusters corresponding to the highways (mark the data clusters with centroids of larger speeds as 1, and the data clusters with centroids of smaller speeds as 0). cluster_centers = kmeans.cluster_centers highway_cluster = np.argmax(cluster_centers) # Output the clustering result for each sample and whether it is on a highway (output the label corresponding to each instantaneous vehicle speed data in the instantaneous vehicle speed sequence) for i in range(len(speeds)): is_highway = labels[i] == highway_cluster print(f"Vehicle speed: {speeds[i][0]} km / h, Cluster label: {labels[i]}, Whether on highway: {is_highway}") (When the number of instantaneous vehicle speed data with label 1 is greater than or equal to the number of instantaneous vehicle speed data with label 0, the output is "Vehicle is traveling on highway".) In some technical solutions, the RAG model generates current driving suggestions for the vehicle based on the selected historical energy consumption data and historical driving data in the following way: it generates suggestions on the vehicle operation required when the current energy consumption data of the vehicle is to reach the selected historical energy consumption data and the current driving data of the vehicle is to reach the selected historical driving data.

[0028] The RAG engine model is a customized retrieval and generation model developed based on open-source deep learning frameworks (such as TensorFlow). It utilizes natural language processing technology to structure highway-related knowledge and construct a knowledge graph. Through continuous training and optimization of model parameters, it improves retrieval accuracy and the rationality of generated suggestions. Detailed rules for generating highway driving energy consumption suggestions are formulated, and the information output by the RAG engine module is filtered and integrated based on factors such as vehicle type and battery characteristics to generate suggestions that meet actual driving needs. A dedicated highway driving energy consumption suggestion interface is developed for the vehicle's central control system, adopting a simple and clear design style to ensure that the information is clear and easy to read. Advanced speech recognition and synthesis technologies are integrated to achieve natural and fluent voice interaction.

[0029] In some embodiments, when the vehicle's current driving environment is heavy rain, the historical energy consumption data selected by the RAG model represents the vehicle's historical lowest battery consumption during heavy rain, and the historical driving data selected by the RAG model represents the historical speed range corresponding to the vehicle's historical lowest battery consumption during heavy rain. Vehicle operation suggestions are generated to ensure that the vehicle's current energy consumption reaches the selected historical lowest battery consumption and that the vehicle's current driving data falls within the selected historical speed range corresponding to the historical lowest battery consumption. The vehicle operation suggestions involve increasing or decreasing the vehicle's current speed until the current speed is included within the historical speed range corresponding to the historical lowest battery consumption, thus ensuring that the vehicle's current battery consumption reaches the historical lowest battery consumption.

[0030] In some embodiments, when the vehicle's current driving environment is hot weather, the historical energy consumption data selected by the RAG model represents the lowest historical air conditioning energy consumption for the vehicle's air conditioning in hot weather. The historical driving data selected by the RAG model represents the air conditioning temperature and fan speed corresponding to the lowest historical air conditioning energy consumption for the vehicle in hot weather. Vehicle operation suggestions are generated to ensure that the vehicle's current energy consumption reaches the selected lowest historical air conditioning energy consumption, and that the vehicle's current driving data reaches the air conditioning fan speed and temperature corresponding to the selected lowest historical air conditioning energy consumption. The vehicle operation suggestions involve adjusting the vehicle's air conditioning temperature and fan speed to the levels corresponding to the lowest historical air conditioning energy consumption, thereby achieving the lowest historical air conditioning energy consumption for the vehicle's current air conditioning.

[0031] Example 2 The highway driving trip identification and energy consumption optimization method based on the system includes: A highway energy consumption database is constructed using historical driving data, historical driving environment data, and historical energy consumption data from various periods of vehicle history. The K-means algorithm is used to perform cluster analysis on a set of instantaneous vehicle speeds to determine whether the vehicle is traveling on a highway. When the vehicle is determined to be traveling on a highway, the RAG model is used to retrieve the historical driving environment data with the highest matching degree to the vehicle's current driving environment data from the historical driving environment data of various historical time periods in the highway energy consumption database. The selected historical driving environment data is then selected by filtering the historical energy consumption data and historical driving data corresponding to the selected historical driving environment data from the historical energy consumption data of various historical time periods. The RAG model generates the vehicle's current driving suggestions based on the filtered historical energy consumption data and historical driving data.

[0032] Example 3 The present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the method described in Embodiment 2.

[0033] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

Claims

1. A highway driving trip recognition and energy consumption optimization system, characterized in that, It includes: The resource database creation module is used to build a highway energy consumption database by utilizing historical driving data, historical driving environment data, and historical energy consumption data from various periods of vehicle history. The driving location identification module is used to perform cluster analysis on a set of instantaneous vehicle speeds using the K-means algorithm to determine whether the vehicle is driving on a highway. The energy consumption optimization module is used to select historical driving environment data when the vehicle is determined to be driving on a highway. It uses the RAG model to retrieve the historical driving environment data with the highest matching degree with the preprocessed current driving environment data of the vehicle from the historical driving environment data of various historical time periods in the highway energy consumption database. Then, it filters the historical energy consumption data and historical driving data corresponding to the selected historical driving environment data from the historical energy consumption data of various historical time periods. The RAG model generates the current driving suggestions of the vehicle based on the filtered historical energy consumption data and historical driving data.

2. The highway driving trip recognition and energy consumption optimization system according to claim 1, characterized in that: Historical driving data includes vehicle speed, acceleration, braking frequency, battery charge, and driving behavior data for various historical periods; historical driving environment data includes highway gradient information, traffic density, weather conditions, and following distance for various historical periods; historical energy consumption data includes driving energy consumption and auxiliary energy consumption.

3. The highway driving trip recognition and energy consumption optimization system according to claim 1, characterized in that: The resource database creation module is used to acquire historical driving data, historical driving environment data and historical energy consumption data of vehicles. It preprocesses the historical driving data, historical driving environment data and historical energy consumption data of vehicles, and uses the preprocessed historical driving data, historical driving environment data and historical energy consumption data of vehicles in various historical periods to build a highway energy consumption database. The energy consumption optimization module is used to, when the vehicle is determined to be traveling on a highway, use the RAG model to retrieve the historical driving environment data with the highest matching degree with the preprocessed current driving environment data of the vehicle from the historical driving environment data of various historical time periods in the highway energy consumption database as the selected historical driving environment data. Then, it filters the historical energy consumption data and historical driving data corresponding to the selected historical driving environment data from the historical energy consumption data of various historical time periods. The RAG model generates the current driving suggestions for the vehicle based on the filtered historical energy consumption data and historical driving data.

4. The highway driving trip recognition and energy consumption optimization system according to claim 1, characterized in that: The method of determining whether a vehicle is traveling on a highway by performing cluster analysis on a set of instantaneous vehicle speed data using the K-means algorithm includes: collecting instantaneous vehicle speed data at set time intervals during vehicle travel, forming an instantaneous speed data series with a set number of instantaneous speeds, dividing the vehicle's travel scenario into two types: urban driving and highway driving, setting the cluster size K value for the K-means algorithm based on the two travel scenarios, and then analyzing the instantaneous speed data series based on the cluster size K value to output the vehicle's travel scenario.

5. The highway driving trip recognition and energy consumption optimization system according to claim 4, characterized in that: The K-means algorithm analyzes instantaneous vehicle speed sequences based on the number of clusters (K). The algorithm involves randomly generating a first and second instantaneous speed centroid from each speed data point in the sequence. It then calculates the distance from each speed data point to these centroids and assigns the data points to the corresponding centroid based on the principle of proximity. This forms a first and a second data cluster. The mean of the second cluster is used as the new second instantaneous speed centroid, and the mean of the first cluster is used as the new first instantaneous speed centroid. Finally, the algorithm calculates the distance from each speed data point to the new second instantaneous speed centroid and the distance from the first cluster to the second instantaneous speed centroid. The distance from each vehicle speed data point in the speed-time sequence to the new first instantaneous velocity centroid is used to regenerate the updated second data cluster and the updated first data cluster, completing one centroid iteration. This centroid iteration process is repeated until the second instantaneous velocity centroid and the first instantaneous velocity centroid converge. The converged second instantaneous velocity centroid is compared with the converged first instantaneous velocity centroid. Each vehicle speed data point in the data cluster containing the larger centroid is identified as highway driving data, and each vehicle speed data point in the data cluster containing the smaller centroid is identified as urban driving data. When the number of vehicle speed data points in the highway driving data is greater than or equal to the number of vehicle speed data points in the urban driving data, it is determined that the vehicle is traveling on a highway.

6. The highway driving trip recognition and energy consumption optimization system according to claim 1, characterized in that: The RAG model generates current driving suggestions for the vehicle based on the selected historical energy consumption data and historical driving data as follows: it generates suggestions on the vehicle operation required when the current energy consumption data of the vehicle reaches the selected historical energy consumption data and the current driving data of the vehicle reaches the selected historical driving data.

7. The highway driving trip recognition and energy consumption optimization system according to claim 6, characterized in that: When the vehicle's current driving environment is heavy rain, the historical energy consumption data selected by the RAG model is the lowest historical power battery power consumption of the car in heavy rain. The historical driving data selected by the RAG model is the historical speed range corresponding to the lowest historical power battery power consumption of the car in heavy rain. The system generates vehicle operation suggestions to ensure that the current energy consumption data of the vehicle reaches the selected lowest historical power battery power consumption and the current driving data reaches the historical speed range corresponding to the selected lowest historical power battery power consumption.

8. The highway driving trip recognition and energy consumption optimization system according to claim 6, characterized in that: When the vehicle's current driving environment is hot weather, the historical energy consumption data selected by the RAG model is the lowest historical air conditioning energy consumption of the vehicle's air conditioning in hot weather. The historical driving data selected by the RAG model is the air conditioning temperature and fan speed corresponding to the lowest historical air conditioning energy consumption of the vehicle in hot weather. The system generates vehicle operation suggestions that require the vehicle's current energy consumption data to reach the selected lowest historical air conditioning energy consumption and the vehicle's current driving data to reach the air conditioning fan speed and temperature corresponding to the selected lowest historical air conditioning energy consumption.

9. A method for highway driving trip identification and energy consumption optimization based on the system described in claim 1, characterized in that, It includes: A highway energy consumption database is constructed using historical driving data, historical driving environment data, and historical energy consumption data from various periods of vehicle history. The K-means algorithm is used to perform cluster analysis on a set of instantaneous vehicle speeds to determine whether the vehicle is traveling on a highway. When the vehicle is determined to be traveling on a highway, the RAG model is used to retrieve the historical driving environment data with the highest matching degree to the vehicle's current driving environment data from the historical driving environment data of various historical time periods in the highway energy consumption database. The selected historical driving environment data is then selected by filtering the historical energy consumption data and historical driving data corresponding to the selected historical driving environment data from the historical energy consumption data of various historical time periods. The RAG model generates the vehicle's current driving suggestions based on the filtered historical energy consumption data and historical driving data.

10. A computer program product comprising a computer program / instructions that, when executed by a processor, implement the steps of the method described in claim 9.

Images