A method and system for evaluating the satisfaction of charging infrastructure based on big data

Abstract

The present invention provides a method and system for evaluating the satisfaction of charging infrastructure based on big data, including: collecting the operation monitoring data of electric vehicles, combining with location data to obtain a regional data set, obtaining the charging station locations and the number of charging stations based on the density clustering method, obtaining the charging station service capacity and the regional charging capacity, constructing a charging demand prediction model and training it to obtain charging demand data, inputting the trained charging demand prediction model to obtain the regional charging demand, dividing the region into full coverage, screening out the areas that meet the preset accuracy, combining the service capacity of the charging stations, the regional charging capacity and the regional charging demand, judging the satisfaction of the charging service capacity within the area, and comprehensively evaluating whether the regional charging service capacity meets the user's needs based on the judgment results of all areas. The present invention can evaluate the satisfaction of the charging infrastructure, which is convenient for targeted construction of charging infrastructure in areas where the charging demand is not met.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicle charging infrastructure, and in particular, to a method and system for evaluating the satisfaction of charging infrastructure based on big data. Background Art

[0002] In the initial stage of the development of charging infrastructure, it is unreasonable and imperfect, and the data between operators are not interoperable, resulting in an unclear overall situation and operation status of the national charging infrastructure, which is not conducive to the later construction and operation management. In addition, scientific and reasonable planning of charging facilities will directly affect the construction cost, operation cost, user satisfaction, local power grid, etc. Therefore, there is an urgent need for a method to evaluate the satisfaction of electric vehicle charging infrastructure. Summary of the Invention

[0003] Based on this, in view of the above technical problems, it is necessary to provide a method and system for evaluating the satisfaction of charging infrastructure based on big data.

[0004] A method for evaluating the satisfaction of charging infrastructure based on big data includes the following steps: collecting operation monitoring data of electric vehicles, where the operation monitoring data includes single-trip data and single-charge data; processing the operation monitoring data to obtain valid data, and obtaining location data of the operation of electric vehicles, combining the valid data with the location data to obtain a regional data set; based on the regional data set, obtaining the location and quantity of charging stations based on the density clustering method, and calculating the number of charging piles of the charging stations based on the maximum time overlap degree, and evaluating the service capacity of the charging stations and the regional charging capacity according to the number of charging piles; obtaining the historical dynamic data of electric vehicles, constructing a charging demand prediction model based on a machine learning algorithm, training the charging demand prediction model according to the historical dynamic data, obtaining charging demand data according to the regional data set, inputting the charging demand data into the charging demand prediction model to obtain the regional charging demand; using the Uber H3 algorithm to divide the region into full coverage to obtain several slices, screening out the slices that meet the preset accuracy, combining the service capacity of the charging stations, the regional charging capacity and the regional charging demand, evaluating whether the charging service capacity in the slice meets the charging demand of users, and comprehensively judging whether the charging service capacity in the region meets the charging demand of users based on the evaluation results of all slices in the region.

[0005] In one embodiment, the single-trip data includes the vehicle identification number, trip start time, trip end time, trip duration, trip start longitude, trip start latitude, trip end longitude, trip end latitude, trip start state of charge, and trip end state of charge; the single-charge data includes the vehicle unique identifier, charge start time, charge end time, charge duration, charge start longitude, charge start latitude, and charge start state of charge.

[0006] In one embodiment, processing the operation monitoring data to obtain valid data, obtaining the location data of the electric vehicle operation, and combining the valid data with the location data to obtain a regional data set specifically includes: intercepting the date parts of the trip start time and charge start time of the electric vehicle to obtain the trip start date and charge start date; processing the dates according to month, day of the week, and whether it is a holiday to obtain month, week information, and holiday information; obtaining the position information where the segment starts according to the trip start longitude, trip start latitude, charge start longitude, and charge start latitude, and obtaining weather and temperature information according to the position information; sorting the single-trip data and single-charge data based on time to obtain the previous dynamic segment information, where the previous dynamic segment information includes whether the previous dynamic segment is a trip or a charge, the previous charge end time, the previous segment end time, the state of charge at the end of the previous segment, the time interval since the previous charge ended, the weather of the previous segment, and the temperature of the previous segment; constructing and obtaining the regional data set according to the trip start date, charge start date, month, week information, holiday information, weather, temperature information, and the previous dynamic segment information.

[0007] In one embodiment, based on the regional data set, obtaining the charging station location and the number of charging stations using the density clustering method, calculating the number of charging piles of the charging station based on the maximum time overlap, and evaluating the service capacity of the charging station and the regional charging capacity according to the number of charging piles, specifically including: obtaining the charging location according to the charge start longitude and charge start latitude, and counting the total number of charges at the charging location. When the total number of charges corresponding to the charging location exceeds the preset number of charges, the charging location is determined as a public charging point; using the distance between public charging points as the clustering basis, clustering the public charging points using the density clustering method to obtain the charging station location and the number of charging stations; calculating the maximum number of vehicles charging at the charging station simultaneously based on the maximum time overlap to obtain the number of charging piles in the charging station; evaluating the service capacity of the charging station according to the number of charging piles, and evaluating the regional charging capacity according to the number of charging piles corresponding to all charging stations in the region.

[0008] In one embodiment, the method for obtaining the historical dynamic data of the electric vehicle, constructing a charging demand prediction model based on a machine learning algorithm, training the charging demand prediction model according to the historical dynamic data, obtaining charging demand data according to a regional data set, and inputting the charging demand data into the charging demand prediction model to obtain the regional charging demand specifically includes: selecting the historical dynamic data of the electric vehicle up to time T-1, where the historical dynamic data includes the state of charge at the start of charging, the start time of charging, the longitude and latitude at the start of charging, the state of charge at the start of driving, the state of charge at the end of driving, and the segment information of the previous dynamic; judging whether charging occurred at time T-2 according to the regional data set and adding the judgment result to the historical dynamic data to obtain a historical data set; dividing the historical data set into a test group and an experimental group according to a preset ratio by SMOTE sampling; constructing a charging demand prediction model based on a machine learning algorithm and training the charging demand prediction model through the test group and the experimental group; obtaining charging demand data according to the regional data set, inputting the charging demand data into the charging demand prediction model to obtain the regional charging demand, where the regional charging demand includes the number of potential users, the distribution of charging demand time periods, and the longitude and latitude of charging.

[0009] In one embodiment, the method for using the Uber H3 algorithm to conduct a full-coverage division of a region, obtaining several sub-regions, screening out the sub-regions that meet the preset accuracy, combining the service capacity of the charging station, the regional charging capacity, and the regional charging demand, judging whether the charging service capacity within the sub-region meets the charging demand of users, and comprehensively evaluating whether the charging service capacity within the region meets the charging demand of users according to the judgment results of all sub-regions within the region specifically includes: using the Uber H3 algorithm to conduct a full-coverage division of the region to obtain several sub-regions, and screening out the sub-regions with an H3 accuracy of 7 from the several sub-regions; calculating the corresponding charging capacity and charging demand of the sub-region according to the service capacity of the charging station, the regional charging capacity, and the regional charging demand; judging the maximum number of simultaneous chargings according to the distribution of the charging demand time period and the number of potential users in the sub-region, comparing the maximum number of simultaneous chargings with the number of charging piles within the sub-region, judging whether the charging service capacity within the sub-region meets the charging demand of users according to the comparison result, and comprehensively evaluating whether the charging service capacity within the region meets the charging demand of users according to the judgment results of all sub-regions within the region.

[0010] In one embodiment, the method for comprehensively evaluating whether the charging service capacity within the region meets the charging demand of users according to the judgment results of all sub-regions within the region specifically includes: when the charging service capacity of all sub-regions within the region meets the charging demand of users, it is determined that the charging service capacity within the region meets the charging demand of users; when the charging service capacity of at least one sub-region does not meet the charging demand of users, it is determined that the charging service capacity within the region does not meet the charging demand of users.

[0011] A satisfaction evaluation system for charging infrastructure based on big data, comprising: an operation monitoring data acquisition module for acquiring operation monitoring data of electric vehicles, where the operation monitoring data includes single-trip data and single-charge data; a regional data set acquisition module for processing the operation monitoring data to obtain valid data, and obtaining location data of the operation of electric vehicles, combining the valid data with the location data to obtain a regional data set; a charging capacity evaluation module for obtaining the location and quantity of charging stations based on the density clustering method according to the regional data set, calculating the number of charging piles of the charging stations based on the maximum time overlap degree, and evaluating the service capacity of the charging stations and the regional charging capacity according to the number of charging piles; a charging demand prediction module for obtaining the historical dynamic data of electric vehicles, constructing a charging demand prediction model based on a machine learning algorithm, training the charging demand prediction model according to the historical dynamic data, obtaining charging demand data from the regional data set, and inputting the charging demand data into the trained charging demand prediction model to obtain the regional charging demand; a charging demand judgment module for using the Uber H3 algorithm to divide the region into full coverage to obtain several areas, screening out the areas that meet the preset accuracy, combining the service capacity of the charging stations, the regional charging capacity and the regional charging demand, judging whether the charging service capacity in the area meets the charging demand of users, and comprehensively evaluating whether the charging service capacity in the region meets the charging demand of users based on the judgment results of all areas in the region.

[0012] Compared with the prior art, the advantages and beneficial effects of the present invention are as follows: The present invention can obtain a regional data set by combining operation monitoring data of electric vehicles with location data, obtain the location and quantity of charging stations based on the density clustering method, calculate the number of charging piles of the charging stations based on the maximum time overlap degree, obtain the service capacity of the charging stations and the regional charging capacity, construct a charging demand prediction model based on a machine learning algorithm, train the model according to historical dynamic data, obtain charging demand data from the regional data set, and input it into the trained charging demand prediction model to obtain the regional charging demand. Use the Uber H3 algorithm to divide the region into full coverage to obtain several areas, screen out the areas that meet the preset accuracy, combine the service capacity of the charging stations, the regional charging capacity and the regional charging demand, judge the charging service capacity in the area, and comprehensively evaluate whether the charging service capacity in the region meets the needs of users based on the judgment results of all areas in the region, so as to determine the satisfaction of the charging infrastructure in the region, facilitate the construction of charging infrastructure in the areas where the charging demand is not met, thereby alleviating the charging anxiety of users and improving the satisfaction of users' charging demand. Description of the Drawings

[0013] Figure 1 It is a schematic flowchart of a method for evaluating the satisfaction of charging infrastructure based on big data in an embodiment;

[0014] Figure 2 It is a schematic structural diagram of an evaluation system for the satisfaction of charging infrastructure based on big data in an embodiment. Specific implementation manners

[0015] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below through specific implementation manners in combination with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0016] In one embodiment, as Figure 1 shown, a method for evaluating the satisfaction of charging infrastructure based on big data is provided, including the following steps:

[0017] Step S101, collect the operation monitoring data of electric vehicles, where the operation monitoring data includes single-trip data and single-charging data.

[0018] Specifically, the electric vehicles described in this application are private electric passenger vehicles. In order to estimate the vehicles with charging requirements at the current moment in the area from the perspective of electric vehicles and determine whether the charging requirements are met in the area, the static data and dynamically updated operation data of electric vehicles are required. Therefore, collect the operation monitoring data of electric vehicles for a period of time, including single-trip data and single-charging data.

[0019] Among them, the single-trip data includes the vehicle identification number, trip start time, trip end time, trip duration, trip start longitude, trip start latitude, trip end longitude, trip end latitude, trip start state of charge and trip end state of charge; the single-charging data includes the vehicle unique identifier, charging start time, charging end time, charging duration, charging start longitude, charging start latitude and charging start state of charge.

[0020] Step S102, process the operation monitoring data to obtain valid data, and obtain the position data of the operation of the electric vehicle, and combine the valid data with the position data to obtain a regional data set.

[0021] Specifically, process the operation monitoring data, extract the valid data in the operation monitoring data, for example, the single-trip start time, the longitude and latitude information where the segment starts, the state of charge at the end of the segment, temperature, weather, etc., and obtain the position data of the operation of the electric vehicle according to the longitude and latitude information, and construct a regional data set according to the valid data and the position data.

[0022] Step S103: Based on the regional dataset, obtain the charging station locations and the number of charging stations using the density clustering method, calculate the number of charging piles at the charging stations based on the maximum time overlap, and evaluate the service capacity of the charging stations and the regional charging capacity according to the number of charging piles.

[0023] Specifically, based on the regional dataset, obtain the longitude and latitude information when the electric vehicle starts charging, determine whether the charging point is a public charging point based on the number of charging times at the charging point, and process the public charging points using the density clustering method to obtain the charging station locations and the number of charging stations. Then, based on the maximum time overlap, determine the maximum number of simultaneous chargings at the charging station as the number of charging piles at the charging station, which is the service capacity of the charging station. Count the total number of charging piles in the region as the regional charging capacity.

[0024] Step S104: Obtain the historical dynamic data of the electric vehicles, build a charging demand prediction model based on the machine learning algorithm, train the charging demand prediction model according to the historical dynamic data, obtain the charging demand data based on the regional dataset, and input the charging demand data into the charging demand prediction model to obtain the regional charging demand.

[0025] Specifically, build a charging demand prediction model using the machine learning algorithm, obtain the historical dynamic data of the electric vehicles, divide the historical dynamic data into historical charging demand data and historical charging demands, and train and validate the charging demand prediction model using the historical charging demand data and historical charging demands to obtain the final charging demand prediction model. Then, obtain the corresponding charging demand data based on the regional dataset, and input the charging demand data into the final charging demand prediction model to obtain the regional charging demand.

[0026] Step S105: Use the Uber H3 algorithm to fully cover and divide the region to obtain several sub - regions, screen out the sub - regions that meet the preset accuracy, and combine the service capacity of the charging stations, the regional charging capacity, and the regional charging demand to determine whether the charging service capacity within the sub - region meets the charging demands of users, and comprehensively evaluate whether the charging service capacity within the region meets the charging demands of users based on the judgment results of all sub - regions within the region.

[0027] Specifically, the Uber H3 algorithm is used to divide the area for full coverage, obtaining several sub-areas, and screening out the sub-areas with an H3 accuracy meeting the preset accuracy. Combining the service capacity of charging stations, the regional charging capacity, and the regional charging demand, it is determined whether the charging service capacity within the sub-area meets the charging needs of users, and the charging service capacity within the area is evaluated based on the judgment results of all sub-areas within the area to determine whether it meets the charging needs of users. For example, if an area is divided into 10 sub-areas, when the charging service capacities of all 10 sub-areas meet the charging needs of users, it is determined that the regional charging service capacity meets the charging needs of users; if there is at least one sub-area that does not meet the charging needs of users, it is determined that the regional charging service capacity does not meet the charging needs of users, and charging infrastructure can be added or adjusted for the sub-areas that do not meet the needs as needed to relieve the charging pressure and improve the satisfaction of users' charging needs.

[0028] Among them, H3 is a spatial division and spatial indexing system for the Earth. The H3 geospatial indexing system is a discrete global grid system composed of spherical multi-precision hexagonal tiles with a hierarchical structure index. A hexagonal grid system is created on the plane of the spherical circumscribed icosahedron, and then the grid cells are projected onto the surface of the sphere using a polyhedron structure projection centered on the reverse side.

[0029] In this embodiment, through the operation monitoring data of electric vehicles and combined with location data, a regional data set is obtained. Based on the density clustering method, the location and number of charging stations are obtained, and the number of charging piles of the charging stations is calculated based on the maximum time overlap. The service capacity of the charging stations and the regional charging capacity are obtained. A charging demand prediction model is constructed based on a machine learning algorithm and trained with historical dynamic data to obtain the charging demand data in the regional data set, which is then input into the trained charging demand prediction model to obtain the regional charging demand. The Uber H3 algorithm is used to divide the area for full coverage, obtaining several sub-areas, screening out the sub-areas that meet the preset accuracy, combining the service capacity of the charging stations, the regional charging capacity, and the regional charging demand, determining the charging service capacity within the sub-area, and comprehensively evaluating whether the charging service capacity within the area meets the needs of users based on the judgment results of all sub-areas within the area, which can determine the satisfaction of the charging infrastructure within the area, facilitating the construction of charging infrastructure for areas that do not meet the charging needs, thereby relieving users' charging anxiety and improving the satisfaction of users' charging needs.

[0030] Among them, step S102 specifically includes: intercepting the driving start time and charging start time of a single electric vehicle trip, obtaining the driving start date and charging start date; processing the dates according to month, day of the week, and whether it is a holiday to obtain month, week information, and holiday information; obtaining the position information where the segment starts based on the driving start longitude, driving start latitude, charging start longitude, and charging start latitude, and obtaining weather and temperature information based on the position information; sorting the single-trip driving data and single-trip charging data based on time to obtain the information of the previous dynamic segment, where the information of the previous dynamic segment includes whether the previous dynamic segment is driving or charging, the previous charging end time, the previous segment end time, the state of charge at the end of the previous segment, the time interval since the previous charging ended, the weather of the previous segment, and the temperature of the previous segment; constructing a dataset for the acquisition area based on the driving start date, charging start date, month, week information, holiday information, weather, temperature information, and the information of the previous dynamic segment.

[0031] Specifically, according to the single-trip driving data and single-trip charging data, intercept the date parts of the driving start time and charging start time of a single electric vehicle segment, and extract them in the format of year-month-day YYYYMMDD to obtain the driving start date and charging start date; process the date data according to month, day of the week, and whether it is a holiday, represent the month as 1 to 12 to obtain the month, represent the day of the week as 1 to 7 to obtain the week information, record holidays as 1, and non-holidays as 0 to obtain the holiday information.

[0032] Based on the longitude and latitude information at the start of driving and charging, obtain the position information where the segment starts, obtain the corresponding weather and temperature through the network, and select the average of the highest temperature and the lowest temperature in the start time date as the temperature information.

[0033] Sort the single-trip driving data and single-trip charging data based on time, which can be in ascending or descending order, to obtain the information of the previous dynamic segment, including whether the previous dynamic is driving or charging, the previous charging end time, the previous segment end time, the state of charge at the end of the previous segment, the time interval since the previous charging ended, the weather of the previous segment, and the temperature of the previous segment. Among them, the time interval since the previous charging ended is obtained by calculating the time interval from the previous charging end time to the data statistics time.

[0034] Construct a dataset for the acquisition area through all the above-obtained data information.

[0035] Among them, step S103 specifically includes: obtaining the charging location according to the starting longitude and latitude of charging, and counting the total number of charging times at the charging location. When the total number of charging times exceeds the preset number of charging times, the charging location is identified as a public charging point; taking the distance between public charging points as the clustering basis, using the density clustering method to cluster the public charging points to obtain the charging station location and the number of charging stations; calculating the maximum number of vehicles charging simultaneously at the charging station based on the maximum time overlap, and obtaining the number of charging piles in the charging station; evaluating the service capacity of the charging station according to the number of charging piles, and evaluating the regional charging capacity according to the number of charging piles corresponding to all charging stations in the region.

[0036] Specifically, obtain the charging location according to the longitude and latitude information at the start of charging, and count the total number of charging times corresponding to the charging location. When the total number of charging times exceeds the preset number of charging times, the charging location is identified as a public charging point. The preset number of charging times can be set according to the time interval between the first charging and the last charging at the charging location, that is, the number of working days of the charging pile at the charging location. For example, if the time interval is 50 days, the preset number of charging times can be set to 200. When the number of charging times at this charging location exceeds 200 times, this charging location is identified as a public charging point.

[0037] Through the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) density clustering method, taking the distance between charging locations as the clustering basis, obtain the charging station location and the number of charging stations. Based on the maximum time overlap, count the maximum number of vehicles charging simultaneously at the charging station, obtain the number of charging piles in the charging station, evaluate the service capacity of the charging station according to the number of charging piles, and evaluate the regional charging capacity according to the number of charging piles corresponding to all charging stations in the region.

[0038] Among them, step S104 specifically includes: selecting the historical dynamic data of electric vehicles up to time T-1. The historical dynamic data includes the starting state of charge at charging, the starting time of charging, the starting longitude and latitude of charging, the starting state of charge during driving, the ending state of charge during driving, and the segment information of the previous dynamic; judging whether charging occurred at time T-2 according to the regional data set, and adding the judgment result to the historical dynamic data to obtain the historical data set; dividing the historical data set into a test group and an experimental group according to a preset ratio by SMOTE sampling; constructing a charging demand prediction model based on a machine learning algorithm, and training the charging demand prediction model through the test group and the experimental group; obtaining the charging demand data according to the regional data set, inputting the charging demand data into the charging demand prediction model to obtain the regional charging demand. The regional charging demand includes the number of potential users, the distribution of charging demand time periods, and the charging longitude and latitude.

[0039] Specifically, select the historical dynamic data of electric vehicles in the operation monitoring data up to time T-1, such as the historical dynamic data in the one year before time T-1; determine whether charging has been carried out within time T-2 according to the regional data set, and add the judgment result to the historical dynamic data to obtain the historical data set; divide the historical data set into a test group and an experimental group according to a preset ratio by SMOTE sampling, such as a ratio of test group: experimental group = 2:8, to ensure the balance of the data set; build a charging demand prediction model based on machine learning algorithms, such as random forest, xgboost, etc., and train the charging demand prediction model through the test group and the experimental group; obtain the charging demand data according to the regional data set, such as whether the user has charged on day T-2, the charging time period, the charging location, etc., and input the charging demand data set into the trained charging demand prediction model to obtain the regional charging demand.

[0040] Among them, the charging demand prediction model is based on the characteristic variables of day T-1 during the prediction process, predicts whether there is a charging demand for users on day T+1, screens out users with charging demands, calculates the confidence level of users charging in each time period on the next day according to the time period in which each historical charging start time is located, selects the time period with the highest confidence level as the current charging demand time period of the user, and clusters according to the charging locations of users in the natural week to estimate the charging locations of users with charging demands, and counts the charging demands of all users, so as to obtain the regional charging demand.

[0041] Among them, step S105 specifically includes: using the Uber H3 algorithm to conduct full-coverage division of the region to obtain several areas, and screening out the areas with an H3 precision of 7 among the several areas; calculating the corresponding area charging capacity and area charging demand according to the service capacity of the charging station, the regional charging capacity and the regional charging demand; judging the maximum simultaneous charging number according to the distribution of the area charging demand time period and the number of potential users, comparing the maximum simultaneous charging number with the number of charging piles in the area, judging whether the charging service capacity in the area meets the charging needs of users according to the comparison result, and comprehensively evaluating whether the charging service capacity in the region meets the charging needs of users according to the judgment results of all areas in the region.

[0042] Specifically, the Uber H3 hexagonal hierarchical indexing grid system is adopted to fully cover and divide the area, obtaining several hexagonal areas. Among these areas, the areas with H3 precision = 7 (each single hexagon covers an area of 5.16 square kilometers) are screened out; in combination with the service capacity of charging stations, the regional charging capacity, and the regional charging demand, the corresponding charging capacity and charging demand of the areas are calculated; based on the distribution of the charging demand time periods and the number of potential users in the areas, the maximum simultaneous charging number is judged, and the maximum simultaneous charging number is compared with the number of charging piles in the areas. If the number of charging piles in the area is less than the maximum simultaneous charging number, it is determined that the charging service capacity in the area does not meet the user's needs; if the number of charging piles in the area is greater than or equal to the maximum simultaneous charging number, it is determined that the charging service capacity in the area meets the user's charging needs, and the judgment results of all areas in the region are comprehensively considered to evaluate whether the charging service capacity in the region meets the user's charging needs, so as to be able to judge whether the charging infrastructure in the region meets the user's charging needs, and it is convenient to carry out corresponding infrastructure construction on the areas with insufficient charging service capacity when the charging infrastructure does not meet the charging needs.

[0043] Specifically, when the charging service capacity of all areas in the region meets the user's charging needs, it is determined that the charging service capacity in the region meets the user's charging needs; when there is at least one area where the charging service capacity does not meet the user's charging needs, it is determined that the charging service capacity in the region does not meet the user's charging needs.

[0044] As Figure 2 shown, a satisfaction evaluation system 20 for charging infrastructure based on big data is provided, including: an operation monitoring data acquisition module 21, a regional data set acquisition module 22, a charging capacity evaluation module 23, a charging demand prediction module 24, and a charging service capacity evaluation module 25, where:

[0045] The operation monitoring data acquisition module 21 is used to acquire the operation monitoring data of electric vehicles, and the operation monitoring data includes single-trip data and single-charging data;

[0046] The regional data set acquisition module 22 is used to process the operation monitoring data, obtain valid data, and obtain the location data of the operation of electric vehicles, and combine the valid data with the location data to obtain a regional data set;

[0047] The charging capacity evaluation module 23 is used to obtain the charging station locations and the number of charging stations based on the density clustering method according to the regional data set, calculate the number of charging piles of the charging stations based on the maximum time overlap degree, and evaluate the service capacity of the charging stations and the regional charging capacity according to the number of charging piles;

[0048] The charging demand prediction module 24 is used to obtain the historical dynamic data of the electric vehicle, construct a charging demand prediction model based on the machine learning algorithm, train the charging demand prediction model according to the historical dynamic data, obtain the charging demand data according to the regional data set, input the charging demand data into the charging demand prediction model, and obtain the regional charging demand;

[0049] The charging service capacity evaluation module 25 is used to divide the area completely by using the Uber H3 algorithm, obtain several areas, screen out the areas that meet the preset accuracy, combine the service capacity of the charging station, the regional charging capacity and the regional charging demand, judge whether the charging service capacity in the area meets the charging demand of users, and comprehensively evaluate whether the charging service capacity in the area meets the charging demand of users according to the judgment results of all areas in the region.

[0050] In one embodiment, the regional data set acquisition module 22 is specifically used to: intercept the driving start time and charging start time of the electric vehicle once, and obtain the driving start date and charging start date; process the date according to the month, day of the week and whether it is a holiday, and obtain the month, week information and holiday information; obtain the position information at the start of the segment according to the driving start longitude, driving start latitude, charging start longitude and charging start latitude, and obtain the weather and temperature information according to the position information; sort the single driving data and single charging data based on time to obtain the segment information of the previous dynamic, and the segment information of the previous dynamic includes whether the previous dynamic segment is driving or charging, the previous charging end time, the previous segment end time, the state of charge at the end of the previous segment, the time interval since the previous charging ended, the weather of the previous segment and the temperature of the previous segment; construct and obtain the regional data set according to the driving start date, charging start date, month, week information, holiday information, weather, temperature information and the segment information of the previous dynamic.

[0051] In one embodiment, the charging capacity evaluation module 23 is specifically used to: obtain the charging location according to the charging start longitude and charging start latitude, and count the total number of charging times at the charging location. When the total number of charging times exceeds the preset number of charging times, the charging location is determined as a public charging point; use the distance between public charging points as the clustering basis, and cluster the public charging points by using the density clustering method to obtain the charging station location and the number of charging stations; calculate the maximum number of vehicles charging at the charging station simultaneously based on the maximum time overlap degree, and obtain the number of charging piles in the charging station; evaluate the service capacity of the charging station according to the number of charging piles, and evaluate the regional charging capacity according to the number of charging piles corresponding to all charging stations in the region.

[0052] In one embodiment, the charging demand prediction module 24 is specifically configured to: select the historical dynamic data of the electric vehicle up to time T-1, where the historical dynamic data includes the starting state of charge for charging, the starting time of charging, the starting longitude and latitude for charging, the starting state of charge for driving, the ending state of charge for driving, and the segment information of the previous dynamic; determine whether charging occurred at time T-2 according to the regional data set, and add the determination result to the historical dynamic data to obtain the historical data set; divide the historical data set into a test group and an experimental group according to a preset ratio by SMOTE sampling; construct a charging demand prediction model based on a machine learning algorithm, and train the charging demand prediction model through the test group and the experimental group; obtain the charging demand data according to the regional data set, input the charging demand data into the charging demand prediction model, and obtain the regional charging demand, where the regional charging demand includes the number of potential users, the distribution of charging demand time periods, and the charging longitude and latitude.

[0053] In one embodiment, the charging service capacity evaluation module 25 is specifically configured to: use the Uber H3 algorithm to perform full-coverage division on the region to obtain several sub-regions, and screen out the sub-regions with an H3 precision of 7 from the several sub-regions; calculate the corresponding sub-region charging capacity and sub-region charging demand according to the service capacity of the charging station, the regional charging capacity, and the regional charging demand; determine the maximum simultaneous charging number according to the distribution of the charging demand time period in the sub-region and the number of potential users, compare the maximum simultaneous charging number with the number of charging piles in the sub-region, and determine whether the charging service capacity in the sub-region meets the charging needs of users according to the comparison result, and comprehensively evaluate whether the charging service capacity in the region meets the charging needs of users based on the determination results of all sub-regions in the region.

[0054] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above methods. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc.

[0055] Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the present invention can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. Optionally, they can be implemented by program codes executable by the computing device. Thus, they can be stored in a computer storage medium (ROM / RAM, magnetic disk, optical disk) and executed by the computing device. And in some cases, the steps shown or described can be executed in a sequence different from that here, or they can be separately fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module for implementation. Therefore, the present invention is not limited to any specific combination of hardware and software.

[0056] The above content is a further detailed description of the present invention in combination with specific implementation manners. It cannot be determined that the specific implementation of the present invention is only limited to these descriptions. For those of ordinary skill in the technical field to which the present invention pertains, without departing from the concept of the present invention, several simple deductions or substitutions can still be made, and all should be regarded as belonging to the protection scope of the present invention.

Claims

1. A method for evaluating the satisfaction of charging infrastructure based on big data, characterized in that, It includes the following steps: Collect the operation monitoring data of the electric vehicle. The operation monitoring data includes single-trip data and single-charge data. The single-trip data includes the vehicle identification number, trip start time, trip end time, trip duration, trip start longitude, trip start latitude, trip end longitude, trip end latitude, trip start state of charge, and trip end state of charge. The single-charge data includes the vehicle unique identifier, charge start time, charge end time, charge duration, charge start longitude, charge start latitude, and charge start state of charge; Process the operation monitoring data to obtain valid data, and obtain the location data of the electric vehicle operation. Combine the valid data with the location data to obtain a regional data set; Based on the regional data set, obtain the charging station locations and the number of charging stations using the density clustering method, and calculate the number of charging piles at the charging stations based on the maximum time overlap. Evaluate the service capacity of the charging stations and the regional charging capacity according to the number of charging piles; Obtain the historical dynamic data of the electric vehicle, construct a charging demand prediction model based on the machine learning algorithm, train the charging demand prediction model according to the historical dynamic data, obtain the charging demand data according to the regional data set, and input the charging demand data into the charging demand prediction model to obtain the regional charging demand; Use the Uber H3 algorithm to divide the region into full coverage to obtain several sub-regions, screen out the sub-regions that meet the preset accuracy, and combine the service capacity of the charging stations, the regional charging capacity, and the regional charging demand to determine whether the charging service capacity within the sub-region meets the charging demand of users, and comprehensively evaluate whether the charging service capacity within the region meets the charging demand of users based on the judgment results of all sub-regions within the region.

2. The method for evaluating the satisfaction of charging infrastructure based on big data according to claim 1, characterized in that, The processing of the operation monitoring data to obtain valid data, and obtaining the location data of the electric vehicle operation, and combining the valid data with the location data to obtain a regional data set specifically includes: Intercept the date part of the trip start time and charge start time of a single electric vehicle to obtain the trip start date and charge start date; Process the date according to the month, day of the week, and whether it is a holiday to obtain the month, week information, and holiday information; Obtain the location information at the start of the segment according to the trip start longitude, trip start latitude, charge start longitude, and charge start latitude, and obtain the weather and temperature information according to the location information; Sort the single-trip data and single-charge data based on time to obtain the segment information of the previous dynamic. The segment information of the previous dynamic includes whether the previous dynamic segment is a trip or a charge, the previous charge end time, the previous segment end time, the state of charge at the end of the previous segment, the time interval since the previous charge ended, the weather of the previous segment, and the temperature of the previous segment; Construct and obtain a regional data set according to the trip start date, charge start date, month, week information, holiday information, weather, temperature information, and segment information of the previous dynamic.

3. The method for evaluating the satisfaction of charging infrastructure based on big data according to claim 1, characterized in that, Based on the regional dataset, the charging station locations and the number of charging stations are obtained using a density clustering method, and the number of charging piles at the charging stations is calculated based on the maximum time overlap. The service capacity of the charging stations and the regional charging capacity are evaluated according to the number of charging piles, which specifically includes: The charging location is obtained based on the charging start longitude and latitude, and the total number of charging times at the charging location is counted. When the total number of charging times corresponding to the charging location exceeds the preset number of charging times, the charging location is determined as a public charging point; Taking the distance between public charging points as the clustering basis, the density clustering method is used to cluster the public charging points to obtain the charging station locations and the number of charging stations; Based on the maximum time overlap, the maximum number of vehicles charging at the charging station simultaneously is calculated to obtain the number of charging piles in the charging station; The service capacity of the charging station is evaluated according to the number of charging piles, and the regional charging capacity is evaluated according to the number of charging piles corresponding to all charging stations in the region.

4. The method for evaluating the satisfaction of charging infrastructure based on big data according to claim 2, characterized in that, The historical dynamic data of the electric vehicle is obtained, a charging demand prediction model is constructed based on a machine learning algorithm, the charging demand prediction model is trained according to the historical dynamic data, the charging demand data is obtained according to the regional dataset, and the charging demand data is input into the charging demand prediction model to obtain the regional charging demand, which specifically includes: Select the historical dynamic data of electric vehicles up to time T - 1, where the historical dynamic data includes the charging start state of charge, the charging start time, the charging start longitude and latitude, the driving start state of charge, the driving end state of charge, and the segment information of the previous dynamic; Judge whether charging occurred at time T - 2 according to the regional dataset, and add the judgment result to the historical dynamic data to obtain the historical dataset; The historical dataset is divided into a test group and an experimental group according to a preset ratio by SMOTE sampling; A charging demand prediction model is constructed based on a machine learning algorithm, and the charging demand prediction model is trained by the test group and the experimental group; The charging demand data is obtained according to the regional dataset, and the charging demand data is input into the charging demand prediction model to obtain the regional charging demand, where the regional charging demand includes the number of potential users, the distribution of charging demand time periods, and the charging longitude and latitude.

5. The method for evaluating the satisfaction of charging infrastructure based on big data according to claim 4, characterized in that, The Uber H3 algorithm is used to divide the region into full coverage to obtain several sub - regions, and the sub - regions that meet the preset accuracy are selected. Combining the service capacity of the charging stations, the regional charging capacity, and the regional charging demand, it is judged whether the charging service capacity within the sub - region meets the charging needs of users, and the judgment results of all sub - regions within the region are comprehensively evaluated to determine whether the charging service capacity within the region meets the charging needs of users, which specifically includes: The Uber H3 algorithm is used to divide the region into full coverage to obtain several sub - regions, and the sub - regions with an H3 accuracy of 7 are selected from the several sub - regions; According to the service capacity of the charging stations, the regional charging capacity, and the regional charging demand, the corresponding sub - region charging capacity and sub - region charging demand are calculated; Determine the maximum simultaneous charging number according to the distribution of charging demand time periods and the number of potential users in the area, compare the maximum simultaneous charging number with the number of charging piles in the area, and judge whether the charging service capacity in the area meets the charging needs of users according to the comparison result, and comprehensively evaluate whether the charging service capacity in the area meets the charging needs of users based on the judgment results of all areas in the region.

6. The method for evaluating the satisfaction of charging infrastructure based on big data according to claim 5, characterized in that, The comprehensive evaluation of whether the charging service capacity in the region meets the charging needs of users based on the judgment results of all areas in the region specifically includes: When the charging service capacity of all areas in the region meets the charging needs of users, it is determined that the charging service capacity in the region meets the charging needs of users; When the charging service capacity of at least one area does not meet the charging needs of users, it is determined that the charging service capacity in the region does not meet the charging needs of users.

7. A satisfaction evaluation system for charging infrastructure based on big data, characterized in that, It includes: An operation monitoring data acquisition module for acquiring operation monitoring data of electric vehicles. The operation monitoring data includes single-trip data and single-charging data. The single-trip data includes the vehicle identification number, driving start time, driving end time, driving duration, driving start longitude, driving start latitude, driving end longitude, driving end latitude, driving start state of charge, and driving end state of charge. The single-charging data includes the vehicle unique identifier, charging start time, charging end time, charging duration, charging start longitude, charging start latitude, and charging start state of charge; A regional data set acquisition module for processing the operation monitoring data to obtain valid data, obtaining the position data of the operation of the electric vehicle, combining the valid data with the position data to obtain a regional data set; A charging capacity evaluation module for obtaining the charging station location and the number of charging stations based on the density clustering method according to the regional data set, calculating the number of charging piles of the charging station based on the maximum time overlap degree, and evaluating the service capacity of the charging station and the regional charging capacity according to the number of charging piles; A charging demand prediction module for obtaining the historical dynamic data of electric vehicles, constructing a charging demand prediction model based on a machine learning algorithm, training the charging demand prediction model according to the historical dynamic data, obtaining charging demand data according to the regional data set, and inputting the charging demand data into the charging demand prediction model to obtain the regional charging demand; A charging service capacity evaluation module for using the Uber H3 algorithm to divide the region for full coverage, obtaining several areas, screening out the areas that meet the preset accuracy, combining the service capacity of the charging station, the regional charging capacity, and the regional charging demand, judging whether the charging service capacity in the area meets the charging needs of users, and comprehensively evaluating whether the charging service capacity in the region meets the charging needs of users based on the judgment results of all areas in the region.

Images