A method for site selection based on charging demand

By constructing historical charging and parking behavior data of electric vehicles, using machine learning to predict charging demand, and optimizing the location selection of charging stations for new energy vehicles, the problem of unreasonable layout of charging facilities is solved, and the utilization rate and economic benefits of charging facilities are improved.

CN117541310BActive Publication Date: 2024-11-26CHONGQING INNOVATION CENTER OF BEIJING INSTITUTE OF TECHNOLOGY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311679956.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-12-08
Publication Date
2024-11-26
Estimated Expiration
2043-12-08

AI Technical Summary

Technical Problem

In the new energy vehicle industry, public charging scenarios have increased and charging pile utilization rates have polarized differences. Enterprises do not consider market demand and passenger flow density factors when building websites, resulting in unreasonable site selection and resulting in repeated investment and waste of resources.

Method used

Using a site selection method based on charging needs, the parking-charging network relationship model of the AOI area is constructed by obtaining the historical charging and parking behavior data of electric vehicles, and using machine learning algorithms to predict charging urgency and power demand to optimize the charging station. Site selection and construction scale.

Benefits of technology

Provide data to support enterprises in charging network layout and site selection, reduce repeated investment and resource waste, and improve the utilization rate and economic benefits of charging facilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117541310B_ABST
    Figure CN117541310B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for site selection based on charging demand, including: obtaining the historical charging behavior and parking behavior of pure electric vehicles and determining the corresponding charging position and parking position; obtaining the AOI area distribution and charging station distribution of each city through offline map analysis; matching the parking position and charging position with the AOI area, and matching the charging position with the charging station position to obtain the parking and charging situation of each AOI area; constructing a parking-charging network relationship model of the AOI area, and predicting the charging urgency and charging power demand of the AOI area according to a machine learning algorithm to obtain a prediction result; determining the site selection distribution of the charging station and the construction scale of each charging station according to the prediction result. The present invention can provide data support for enterprises to carry out charging network layout and site selection and construction, thereby alleviating the phenomenon of repeated investment and excessive resource consumption of electric vehicle charging stations during the actual site selection period.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of site selection optimization, and in particular relates to a site selection method for station construction based on charging demand. Background Art

[0002] Under the dual influence of policies and markets, my country's new energy vehicle industry has entered a new stage of rapid development with large-scale and high quality, and the penetration rate has increased rapidly. In terms of scale, my country has become the world's largest new energy vehicle market, and its production and sales volume has ranked first in the world for several consecutive years. New energy vehicle charging piles are an important logistical part of maintaining the energy supply of new energy electric vehicles and ensuring the convenient and smooth travel of new energy vehicles. In order to promote the healthy, stable and sustainable development of my country's new energy vehicle industry, the energy supply facilities supporting new energy electric vehicles have become the focus of domestic and foreign commercial capital. Setting up reasonable charging stations in shopping malls, residential communities, commercial buildings and other places will be conducive to the promotion and growth of new energy vehicles, stimulate the development of the new energy vehicle industry in the sinking market, and at the same time promote the use of clean energy and energy conservation.

[0003] On the one hand, given the increasing number of parking and charging scenarios in the public charging field, the problems of "how to provide a convenient charging network for new energy vehicle owners" and "how to solve the difficulty of charging in cities" remain prominent. On the other hand, many companies did not consider factors such as market demand and passenger flow density when building stations and charging piles in the early stage, resulting in an unbalanced layout and polarized differences in the utilization rate of many public charging piles. The site selection of charging stations is an important factor affecting the operating efficiency of the stations. As the competition in the charging pile market becomes more and more fierce, how to select the site for charging station construction has become a core issue of concern to companies. Summary of the invention

[0004] In view of this, the purpose of the present invention is to propose a site selection method based on charging demand, which is used to provide data support for enterprises to carry out charging network layout and site selection and construction, thereby alleviating the phenomenon of repeated investment and excessive consumption of resources during the actual site selection of electric vehicle charging stations.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] The present invention provides a method for selecting a site for a charging station based on charging requirements, comprising:

[0007] Step S1, obtaining the historical charging behavior and parking behavior of the pure electric vehicle and determining the corresponding charging position and parking position;

[0008] Step S2: Obtain the AOI area distribution and charging station distribution of each city through offline map analysis;

[0009] Step S3: Match the parking position and charging position with the AOI area, and match the charging position with the charging station position to obtain the parking and charging conditions of each AOI area;

[0010] Step S4: constructing a parking-charging network relationship model of the AOI area based on the parking and charging conditions of the AOI area, and predicting the charging urgency and charging power demand of the AOI area according to a machine learning algorithm to obtain a prediction result;

[0011] Step S5: Determine the location distribution of charging stations according to the prediction results, and determine the construction scale of each charging station.

[0012] Preferably, step S1 comprises:

[0013] Extract static and dynamically updated status data of the electric vehicle from a designated data platform, and determine charging data and parking data of the electric vehicle based on the status data;

[0014] The charging data and parking data include single driving data and single charging data of the electric vehicle, wherein the single driving data includes 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, driving end state of charge, single charging data includes the electric vehicle unique identifier, charging start time, charging end time, charging duration, charging start longitude, charging start latitude, charging start state of charge, charging end state of charge;

[0015] A parking position of the electric vehicle is determined based on the single driving data, and a charging position of the electric vehicle is determined based on the single charging data.

[0016] Preferably, determining the charging data and parking data of the electric vehicle according to the status data includes:

[0017] The charging data is filtered and preprocessed according to the status data to obtain the charging data, wherein the preprocessing process includes: removing the charging data whose charging start longitude and latitude are not in the same area as the charging end longitude and latitude;

[0018] All coordinate information in the remaining charging-related data is uniformly converted to the WSG-84 coordinate system to obtain charging data;

[0019] The parking data is filtered according to the status data and preprocessed to obtain the parking data. The preprocessing process includes: sorting the status data of the electric vehicle according to the start time and determining the driving segment, taking the end time of the previous driving segment of the current driving segment as the start time of the parking segment, taking the start time of the current driving segment as the end time of the parking segment, and obtaining the start time and end time of the parking segment;

[0020] The end longitude and latitude of the previous driving segment of the current driving segment are used as the start longitude and latitude of the parking segment, and the start longitude and latitude of the current driving segment are used as the end longitude and latitude of the parking segment;

[0021] Eliminate abnormal parking segments with short parking duration and those whose parking start and end latitudes and longitudes are not in the same area;

[0022] The coordinate system of the parking start longitude and latitude and the parking end longitude in the data corresponding to the remaining parking segments is converted into the WSG-84 coordinate system to obtain the parking data.

[0023] Preferably, step S2 comprises:

[0024] Obtaining the original data of the offline map, and determining the first basic information of the required AOI based on the original data, wherein the first basic information includes the AOI name, AOI type, center point longitude and latitude, borders, province, city, and district and county;

[0025] The AOI name is standardized to obtain a standardized name format, and is mapped and merged according to the first basic information to obtain a primary label of the AOI type, wherein the primary label includes: residential housing, shopping mall, company enterprise, and industrial park;

[0026] The coordinates and longitude and latitude information in the first basic information are converted into the WSG-84 coordinate system to obtain the AOI area distribution;

[0027] Determine the second basic information of the charging station in the city based on the original data, wherein the second basic information includes the station name, station operator, station ID, station latitude and longitude, province, city, and district and county;

[0028] The coordinates and longitude and latitude information in the second basic information are converted to the WSG-84 coordinate system to obtain the distribution of charging stations.

[0029] Preferably, step S3 comprises:

[0030] The distance is calculated based on the latitude and longitude information of the start of each charging segment and the latitude and longitude information of the charging station. If the distance is less than or equal to a preset first distance, the charging segment is marked with a label unique to the charging station; if there are several stations whose distances are less than or equal to the first distance at the same time, the charging segment is marked with a label unique to the nearest charging station;

[0031] Determine whether the charging and parking segments fall within the boundary of the AOI area according to their latitude and longitude. If they do, the corresponding segments are labeled with a unique label corresponding to the AOI.

[0032] The parking and charging conditions of each AOI area are determined based on the tag information corresponding to the charging segment and the parking segment.

[0033] Preferably, the process of constructing the parking-charging network relationship model includes:

[0034] Count the cumulative number of parking times and cumulative parking duration in the AOI area;

[0035] According to the monthly average charging start SOC of the single vehicle, the cumulative number of charging needs in the current AOI area is judged. Specifically, if the SOC value of the electric vehicle when it is parked is equal to or less than the monthly average charging start SOC, it is determined that the electric vehicle has a charging need for this parking behavior;

[0036] According to the charging station tag of the charging event, determine whether there are charging conditions around the AOI area. If there is a matching charging station tag in the charging event, it is considered that there are charging conditions around the AOI area;

[0037] Calculate the cumulative number of charging times and cumulative charging time in the AOI area;

[0038] The charging urgency is calculated based on the actual charging time and parking time with charging demand, so as to reflect the satisfaction degree of the charging service capacity in the area;

[0039] The fields corresponding to the cumulative number of parking times in the AOI area, the cumulative parking time in the AOI area, the cumulative number of charging needs in the AOI area, whether there are charging conditions around the AOI area, the cumulative number of charging times in the AOI area, the cumulative charging time in the AOI area, and the maximum number of simultaneously available charging positions are used as feature values, and the field corresponding to the charging urgency is used as the attribute value to construct a parking-charging network relationship model.

[0040] Preferably, the prediction results obtained by predicting the charging urgency and charging power demand of the AOI area according to the machine learning algorithm include:

[0041] The first type of prediction results are obtained by statistically predicting the charging urgency and charging power demand of the AOI area through machine learning algorithms;

[0042] The charging urgency and charging power requirements of the AOI area are divided into multiple levels in advance;

[0043] According to the first type of prediction results, the charging urgency of each level and the proportion of the duration of each charging power demand to the total duration are counted and sorted to obtain the probability density function of the charging urgency of each level and the charging power demand in a normally distributed state;

[0044] The prediction results are calculated based on the probability density function and the preset percentage standard.

[0045] Preferably, the site selection of charging stations in the AOI area is performed by constructing an optimization model, and the process includes:

[0046] Determine the number and scale of charging stations in the AOI area based on the forecast results, and determine the number of charging piles, land occupation, and equipment requirements based on the number and scale of charging stations;

[0047] Construct a construction cost calculation model based on the number of charging piles, land occupation and equipment demand information:

[0048]

[0049] Among them, i represents the distribution of charging demand points, j represents the candidate site for station construction, Indicates the total number of candidate sites for site selection. is the decision variable. If you choose to go to the station to charge, it is 1. If you do not go to the station to charge, it is 0. The decision is made based on the shortest distance. is the fixed cost of construction, is the unit price of charging pile, is the equivalent investment coefficient of the equipment, is the number of charging piles at the candidate location for station construction, n is the expected service life of the charging station, is the discount rate;

[0050] The accumulated charging demand in the AOI area and the electric vehicle location distribution when charging demand is generated each time are counted, and a distance cost model for users to go to charging stations for charging is constructed:

[0051]

[0052] in, Represents the total number of charging demand points, which is determined by weighted average of statistical results. Unit time cost, is the demand quantity of the demand point, is the path distance from the demand point to the candidate site for station construction, is the congestion coefficient, is the average speed of electric vehicles;

[0053] Based on the number of charging piles in the candidate site selection, a waiting cost model for users to queue up for charging is constructed:

[0054]

[0055]

[0056]

[0057] In the formula, Indicates the waiting time. For the charging pile service intensity, is the average number of electric vehicles arriving at the charging station per hour, is the number of charging pile services per unit time, is the idle probability of the charging station, k is the kth charging pile, T is the maximum waiting time, and W is the waiting cost coefficient;

[0058] Construct a site selection optimization model based on the cost calculation model, distance cost model and waiting cost model:

[0059]

[0060] in, , , Respectively represent three preset correction coefficients;

[0061] Set the constraints of the site selection optimization model:

[0062]

[0063] in, Indicates the minimum number of charging piles in a charging station. Indicates the maximum number of charging piles in the charging station;

[0064] Taking the lowest cost as the optimization goal, calculations are performed based on constraints and the station site selection optimization model, and the site selection and charging pile scale of the charging station in the AOI area are determined according to the calculation results.

[0065] The present invention has achieved at least the following beneficial effects:

[0066] 1. It can provide data support for enterprises to carry out charging network layout and station site selection, thereby alleviating the phenomenon of repeated investment and excessive consumption of resources during the actual site selection of electric vehicle charging stations.

[0067] 2. It can extract charging data and parking data and filter out abnormal data, and unify the coordinate-related information obtained using different coordinate systems on different platforms into the same coordinate system, which is convenient for subsequent related positioning calculations, and realizes the recognition of parking segments. Based on the recognition method of parking segments, similar recognition work can be performed on charging segments.

[0068] Other advantages, objectives and features of the present invention will be described in the following description and will be apparent to those skilled in the art to some extent, or those skilled in the art may be taught from the practice of the present invention. The objectives and other advantages of the present invention may be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to make the purpose, technical solution and beneficial effects of the present invention clearer, the present invention provides the following drawings for illustration:

[0070] Figure 1 This is a flowchart of the steps of a method for selecting a site for a charging station based on charging requirements in an embodiment of the present invention;

[0071] Figure 2 This is a flow chart of the steps of determining charging data and parking data of an electric vehicle according to status data in an embodiment of the present invention. DETAILED DESCRIPTION

[0072] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0073] AOI (Area of ​​Interest), as the name suggests, refers to the area of ​​interest in the Internet electronic map. It also contains four basic information and is mainly used to express regional geographical entities in the map, such as a residential area, a university, an office building, an industrial park, a comprehensive shopping mall, a hospital, a scenic spot or a stadium, etc.

[0074] The present invention provides a method for selecting a site for a charging station based on charging demand, referring to Figure 1 ,include:

[0075] Step S1, obtaining the historical charging behavior and parking behavior of the pure electric vehicle and determining the corresponding charging position and parking position;

[0076] Step S2: Obtain the AOI area distribution and charging station distribution of each city through offline map analysis;

[0077] Step S3: Match the parking position and charging position with the AOI area, and match the charging position with the charging station position to obtain the parking and charging conditions of each AOI area;

[0078] Step S4: constructing a parking-charging network relationship model of the AOI area based on the parking and charging conditions of the AOI area, and predicting the charging urgency and charging power demand of the AOI area according to a machine learning algorithm to obtain a prediction result;

[0079] Step S5: Determine the location distribution of charging stations according to the prediction results, and determine the construction scale of each charging station.

[0080] The working principle and beneficial effects of the above technical solution are as follows: the pure electric passenger car operation charging data collected in the new energy vehicle big data platform is used as the analysis sample. First, the historical charging behavior (charging time, charging power, charging power, charging location, etc.) and parking behavior (parking time, parking location, etc.) of pure electric vehicles are extracted; secondly, the AOI distribution (including AOI name, AOI type AOI center point, AOI boundary, etc.) and charging station distribution (including station name, station address, station latitude and longitude, etc.) of each city are obtained through offline map analysis; finally, the parking location and charging location are matched with the AOI location, and the charging location is matched with the charging station location to obtain the parking and charging situation of each AOI area. Through the above data, the number of charging times, parking times, charging urgency, etc. of the AOI area are analyzed, so as to construct a parking & charging network relationship model, understand user habits, and master the travel heat map of new energy vehicles. Based on the parking & charging network relationship model, according to the machine learning algorithm, the charging urgency and charging power demand of the AOI area in the future are predicted. The above methods can provide data support for enterprises to carry out charging network layout and station site selection and construction, thereby alleviating the phenomenon of repeated investment and excessive resource consumption during the actual site selection of electric vehicle charging stations, so that the constructed charging piles can not only meet the charging needs of local electric vehicles, but also improve the economic benefits of related enterprises.

[0081] In a preferred embodiment, step S1 comprises:

[0082] Extract static and dynamically updated status data of the electric vehicle from a designated data platform, and determine charging data and parking data of the electric vehicle based on the status data;

[0083] The charging data and parking data include single driving data and single charging data of the electric vehicle, wherein the single driving data includes 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, driving end state of charge, single charging data includes the electric vehicle unique identifier, charging start time, charging end time, charging duration, charging start longitude, charging start latitude, charging start state of charge, charging end state of charge;

[0084] A parking position of the electric vehicle is determined based on the single driving data, and a charging position of the electric vehicle is determined based on the single charging data.

[0085] The working principle and beneficial effects of the above technical solution are as follows: the electric vehicle charging pile site selection method proposed in the embodiment of the present invention is based on the actual vehicle status data. In order to estimate the travel conditions in the area from the perspective of electric vehicles and determine whether the current charging piles in the area meet the charging needs, it is necessary to extract the static data of electric vehicles and the dynamically updated status data from the national platform.

[0086] Extract the single driving data of pure electric passenger vehicles, including the following fields:

[0087] vin: vehicle frame number;

[0088] run_start_time: driving start time;

[0089] run_end_time: the end time of the run;

[0090] run_time_lengths: driving duration

[0091] run_start_lon: the longitude at the start of the run;

[0092] run_start_lat: running start latitude;

[0093] run_end_lon: the longitude at the end of the run;

[0094] run_end_lat: the latitude at the end of the run;

[0095] run_start_soc: state of charge at the start of driving;

[0096] run_end_soc: state of charge at the end of running.

[0097] Extract the single charge data of pure electric passenger vehicles, including the following fields:

[0098] vin: unique identifier of electric vehicle;

[0099] charge_start_time: charging start time;

[0100] charge_end_time: charging end time;

[0101] charge_time_lengths: Charging time

[0102] charge_start_lon: charging start longitude;

[0103] charge_start_lat: charging start latitude;

[0104] charge_start_soc: charging start state of charge;

[0105] charge_end_soc: charging start state of charge.

[0106] In a preferred embodiment, referring to Figure 2 , the charging data and parking data of the electric vehicle are determined according to the status data, including:

[0107] Step S61, screening charging-related data according to the status data and preprocessing to obtain charging data, the preprocessing process includes: eliminating charging-related data whose charging start longitude and latitude and charging end longitude are not in the same area (the distance between the two is greater than a preset first threshold);

[0108] Step S62, converting all coordinate information in the remaining charging-related data into the WSG-84 coordinate system to obtain charging data;

[0109] Step S63, filtering parking-related data according to the status data and performing preprocessing to obtain parking data, the preprocessing process comprising: sorting the status data of the electric vehicle according to the start time and determining the driving segment, taking the end time of the previous driving segment of the current driving segment as the start time of the parking segment, taking the start time of the current driving segment as the end time of the parking segment, and obtaining the start time and end time of the parking segment;

[0110] Step S64, taking the end longitude and latitude of the previous driving segment of the current driving segment as the start longitude and latitude of the parking segment, and taking the start longitude and latitude of the current driving segment as the end longitude and latitude of the parking segment;

[0111] Step S65, eliminating abnormal parking segments with a short parking duration (shorter than a preset first time length) and abnormal parking segments whose parking start longitude and longitude are not within the same area as the parking end longitude and longitude (the distance between the two is greater than a preset second threshold);

[0112] Step S66, converting the coordinate system of the parking start longitude and latitude and the parking end longitude in the data corresponding to the remaining parking segments into the WSG-84 coordinate system to obtain parking data.

[0113] The beneficial effects of the above technical solution are: realizing the extraction of charging data and parking data and the screening of abnormal data, and unifying the coordinate-related information obtained using different coordinate systems on different platforms into the same coordinate system, so as to facilitate subsequent related positioning calculation work, and realize the recognition of parking segments. Based on the recognition method of parking segments, similar recognition work can be performed on charging segments.

[0114] In a preferred embodiment, step S2 comprises:

[0115] Obtaining the original data of the offline map, and determining the first basic information of the required AOI based on the original data, wherein the first basic information includes the AOI name, AOI type, center point longitude and latitude, borders, province, city, and district and county;

[0116] The AOI name is standardized to obtain a standardized name format, and is mapped and merged according to the first basic information to obtain a primary label of the AOI type, wherein the primary label includes: residential housing, shopping mall, company enterprise, and industrial park;

[0117] The coordinates and longitude and latitude information in the first basic information are converted into the WSG-84 coordinate system to obtain the AOI area distribution;

[0118] Determine the second basic information of the charging station in the city based on the original data, wherein the second basic information includes the station name, station operator, station ID, station latitude and longitude, province, city, and district and county;

[0119] The coordinates and longitude and latitude information in the second basic information are converted to the WSG-84 coordinate system to obtain the distribution of charging stations.

[0120] The beneficial effects of the above technical solution are: realizing the cleaning and processing of AOI-related information, and determining the location information mapped to the AOI area and the station name, station operator, station ID, station longitude and latitude, province, city and district and county mapped corresponding to the charging station based on the original data, so as to facilitate the rapid determination of the corresponding AOI area and charging station according to the mapping results.

[0121] In a preferred embodiment, step S3 comprises:

[0122] The distance is calculated based on the latitude and longitude information of the start of each charging segment and the latitude and longitude information of the charging station. If the distance is less than or equal to a preset first distance, the charging segment is marked with a label unique to the charging station; if there are several stations whose distances are less than or equal to the first distance at the same time, the charging segment is marked with a label unique to the nearest charging station;

[0123] Determine whether the charging and parking segments fall within the boundary of the AOI area according to their latitude and longitude. If they do, the corresponding segments are labeled with a unique label corresponding to the AOI.

[0124] The parking and charging conditions of each AOI area are determined based on the tag information corresponding to the charging segment and the parking segment.

[0125] The working principle and beneficial effects of the above technical solution are as follows: by establishing a logical relationship between the charging segment and the charging station through the calculation results of the longitude and latitude between the charging segment and the charging station, the classification analysis of the charging location is realized, wherein the preset first distance can be appropriately adjusted according to the accuracy of the device positioning, wherein the charging segment corresponds to a variety of information, including but not limited to location coordinates, charging time, charging status, charging maintenance duration, etc., and the positioning of the area where the charging event and the parking event occur is realized based on the charging segment and the parking segment, so as to determine the AOI area corresponding to the charging and parking segments and establish a corresponding mapping relationship, thereby facilitating the rapid search and statistics of relevant data samples based on the AOI area.

[0126] In a preferred embodiment, the process of constructing the parking-charging network relationship model includes:

[0127] Count the cumulative number of parking times and cumulative parking duration in the AOI area;

[0128] According to the monthly average charging start SOC of the single vehicle, the cumulative number of charging needs in the current AOI area is judged. Specifically, if the SOC value of the electric vehicle when it is parked is equal to or less than the monthly average charging start SOC, it is determined that the electric vehicle has a charging need for this parking behavior;

[0129] According to the charging station tag of the charging event, determine whether there are charging conditions around the AOI area. If there is a matching charging station tag in the charging event, it is considered that there are charging conditions around the AOI area;

[0130] Calculate the cumulative number of charging times and cumulative charging time in the AOI area;

[0131] The charging urgency is calculated based on the actual charging time and parking time with charging demand, so as to reflect the satisfaction degree of the charging service capacity in the area;

[0132] The fields corresponding to the cumulative number of parking times in the AOI area, the cumulative parking time in the AOI area, the cumulative number of charging needs in the AOI area, whether there are charging conditions around the AOI area, the cumulative number of charging times in the AOI area, the cumulative charging time in the AOI area, and the maximum number of simultaneously available charging positions are used as feature values, and the field corresponding to the charging urgency is used as the attribute value to construct a parking-charging network relationship model.

[0133] The working principle and beneficial effects of the above technical solution are as follows: by calculating the cumulative number of parking times park_nums and the cumulative parking time park_lengths in the AOI area; judging the cumulative number of charging demands chargedemandcount in the current AOI area according to the monthly average charging starting SOC of the single vehicle, specifically as follows: if the SOC value of the electric vehicle when it is parked is equal to or less than the monthly average charging starting SOC, it is judged that the electric vehicle has a charging demand for this parking behavior; judging whether there is a charging condition is_charge around the AOI area according to the station tag of the charging event, if there is a matching station tag in the charging event, it is considered that there is a charging condition around the AOI area; calculating the cumulative charging times in the AOI area The number of charges charge_nums and the cumulative charging time charge_lengths; the charging urgency charge_urgency is calculated based on the actual charging time and the parking time with charging demand, so as to reflect the satisfaction degree of the charging service capacity in the area; the park_nums, park_lengths, chargedemandcount, is_charge, charge_nums, charge_lengths, max_station_num fields are used as feature values, and the charge_urgency field is used as the attribute value to build the parking & charging network relationship model and grasp the travel heat map of new energy vehicles. Based on the parking & charging network relationship model, the machine learning algorithm is used to predict the charging urgency of the area in the future, so as to provide data support for enterprises to carry out charging network layout planning and station site selection and construction.

[0134] In a preferred embodiment, based on the parking & charging network relationship model, enterprises can be guided on where to build stations, that is, the appropriate location for building charging stations. However, what to build, that is, whether to choose DC piles or AC piles, and how many to build, that is, the number of DC piles and AC piles built in different regions, will also greatly affect the user's charging experience. According to the average monthly charging starting SOC of a single vehicle, the number of cumulative charging needs in the current AOI area is judged as follows:

[0135] If the SOC value of an electric vehicle when it is parked is equal to or less than the average charging starting SOC of the month, it is determined that the electric vehicle has a charging demand for this parking behavior; the historical charging habits of electric vehicles with charging needs in the AOI area are calculated, including the characteristics of fast and slow charging, the power usage characteristics of charging facilities, the selection characteristics of charging sites, etc., and these characteristics are combined to use machine learning algorithms to predict the charging power requirements of specific models, providing data support for enterprises to choose to build DC / AC charging piles; the charging time is segmented based on the shortest dispatch cycle of the power grid, and the segmentation rule is to divide the 24 hours of a day into 15-minute time lengths. A segment is divided into intervals from [00:00, 00:15) to [23:45, 24:00). If the charging start time to the charging end time spans multiple intervals, it is considered that there is charging behavior in the corresponding intervals. Based on this, the following are calculated: 1) The number of electric vehicles in each interval reflects the maximum number of electric vehicles charging in the AOI area, and the maximum value is considered to be the number of charging piles in the area, reflecting the maximum charging service capacity of the charging station in the AOI area; 2) The charging power distribution of each interval reflects the number of DC piles and AC piles in the area. The parking behavior with charging demand is processed by time segmentation. The segmentation rule is to divide the 24 hours of a day into intervals from [00:00, 00:05) to [23:55, 24:00) according to the length of 5 minutes. If the parking start time to the parking end time spans multiple intervals, it is considered that there is parking behavior in the corresponding interval. Based on this, the charging demand of each time period in the area is calculated to guide the subsequent construction of charging piles.

[0136] The beneficial effects of the above technical solution are: the above technology can reflect the number of existing charging piles and the number of DC / AC piles in the area, as well as the maximum number of charging piles that may be required in the area at a certain moment. Based on this, it can guide enterprises to choose how many piles to build and the distribution ratio of DC / AC piles, so as to provide users with better charging services.

[0137] In a preferred embodiment, the prediction results of the charging urgency and charging power demand of the AOI area are predicted according to the machine learning algorithm, including:

[0138] The first type of prediction results are obtained by statistically predicting the charging urgency and charging power demand of the AOI area through machine learning algorithms;

[0139] The charging urgency and charging power requirements of the AOI area are divided into multiple levels in advance;

[0140] According to the first type of prediction results, the charging urgency of each level and the proportion of the duration of each charging power demand to the total duration are counted and sorted to obtain the probability density function of the charging urgency of each level and the charging power demand in a normally distributed state;

[0141] The prediction results are calculated based on the probability density function and the preset percentage standard.

[0142] The beneficial effects of the above technical solution are: through the machine learning algorithm, it is possible to quickly make statistical predictions on the charging urgency and charging power demand in the AOI area, and construct probability density functions of the charging urgency and charging power demand of each level of normal distribution according to the first type of prediction results, so as to facilitate the rapid determination of the prediction results of the charging demand in the AOI area that meets most of the charging demand according to the probability density function. According to the prediction results, it is possible to determine how many additional charging stations need to be built in the AOI area or how large the total scale of charging stations should be to meet the charging demand most of the time.

[0143] In a preferred embodiment, the site selection of charging stations in the AOI area is performed by constructing an optimization model, and the process includes:

[0144] Determine the number and scale of charging stations in the AOI area based on the forecast results, and determine the number of charging piles, land occupation, and equipment requirements based on the number and scale of charging stations;

[0145] Construct a construction cost calculation model based on the number of charging piles, land occupation and equipment demand information:

[0146]

[0147] Among them, i represents the distribution of charging demand points, j represents the candidate site for station construction, Indicates the total number of candidate sites for site selection. is the decision variable. If you choose to go to the station to charge, it is 1. If you do not go to the station to charge, it is 0. The decision is made based on the shortest distance. is the fixed cost of construction, is the unit price of charging pile, is the equivalent investment coefficient of the equipment, is the number of charging piles at the candidate location for station construction, n is the expected service life of the charging station, is the discount rate;

[0148] The accumulated charging demand in the AOI area and the electric vehicle location distribution when charging demand is generated each time are counted, and a distance cost model for users to go to charging stations for charging is constructed:

[0149]

[0150] in, Represents the total number of charging demand points, which is determined by weighted average of statistical results. Unit time cost, is the demand quantity of the demand point, is the path distance from the demand point to the candidate site for station construction, is the congestion coefficient, is the average speed of electric vehicles;

[0151] Based on the number of charging piles in the candidate site selection, a waiting cost model for users to queue up for charging is constructed:

[0152]

[0153]

[0154]

[0155] In the formula, Indicates the waiting time. For the charging pile service intensity, is the average number of electric vehicles arriving at the charging station per hour, is the number of charging pile services per unit time, is the idle probability of the charging station, k is the kth charging pile, T is the maximum waiting time, and W is the waiting cost coefficient;

[0156] Construct a site selection optimization model based on the cost calculation model, distance cost model and waiting cost model:

[0157]

[0158] in, , , Respectively represent three preset correction coefficients;

[0159] Set the constraints of the site selection optimization model:

[0160]

[0161] in, Indicates the minimum number of charging piles in a charging station. Indicates the maximum number of charging piles in the charging station;

[0162] Taking the lowest cost as the optimization goal, calculations are performed based on constraints and the station site selection optimization model, and the site selection and charging pile scale of the charging station in the AOI area are determined according to the calculation results.

[0163] The beneficial effect of the above technical solution is: by constructing an optimization model, the site selection of charging stations and the scale of charging piles in the AOI area are optimized based on the construction demand for the scale of charging stations in the determined AOI area. According to the optimization calculation results, the construction cost and public cost of the site selection of each charging station can be calculated, so as to facilitate the determination of the optimal site selection and scale of charging piles according to the calculation results.

[0164] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail through the above preferred embodiments, those skilled in the art should understand that various changes can be made in form and details without departing from the scope defined by the claims of the present invention.

Claims

1. A site selection method based on charging demand. It is characterized in that include: Step S1, obtaining the historical charging behavior and parking behavior of the pure electric vehicle and determining the corresponding charging position and parking position; Step S2: Obtain the AOI area distribution and charging station distribution of each city through offline map analysis; Step S3: Match the parking position and charging position with the AOI area, and match the charging position with the charging station position to obtain the parking and charging conditions of each AOI area; Step S4: constructing a parking-charging network relationship model of the AOI area based on the parking and charging conditions of the AOI area, and predicting the charging urgency and charging power demand of the AOI area according to a machine learning algorithm to obtain a prediction result; Step S5: Determine the location distribution of charging stations according to the prediction results, and determine the construction scale of each charging station; The construction process of the parking-charging network relationship model includes: Count the cumulative number of parking times and cumulative parking duration in the AOI area; According to the monthly average charging start SOC of the single vehicle, the cumulative number of charging needs in the current AOI area is judged. Specifically, if the SOC value of the electric vehicle when it is parked is equal to or less than the monthly average charging start SOC, it is determined that the electric vehicle has a charging need for this parking behavior; According to the charging station tag of the charging event, determine whether there are charging conditions around the AOI area. If there is a matching charging station tag in the charging event, it is considered that there are charging conditions around the AOI area; Calculate the cumulative number of charging times and cumulative charging time in the AOI area; The charging urgency is calculated based on the actual charging time and parking time with charging demand, so as to reflect the satisfaction degree of the charging service capacity in the area; The fields corresponding to the cumulative number of parking times in the AOI area, the cumulative parking time in the AOI area, the cumulative number of charging needs in the AOI area, whether there are charging conditions around the AOI area, the cumulative number of charging times in the AOI area, the cumulative charging time in the AOI area, and the maximum number of simultaneously available charging positions are used as feature values, and the field corresponding to the charging urgency is used as the attribute value to construct a parking-charging network relationship model; The prediction results obtained by predicting the charging urgency and charging power demand of the AOI area according to the machine learning algorithm include: The first type of prediction results are obtained by statistically predicting the charging urgency and charging power demand of the AOI area through machine learning algorithms; The charging urgency and charging power requirements of the AOI area are divided into multiple levels in advance; According to the first type of prediction results, the charging urgency of each level and the proportion of the duration of each charging power demand to the total duration are counted and sorted to obtain the probability density function of the charging urgency of each level and the charging power demand in a normally distributed state; The prediction results are calculated based on the probability density function and the preset percentage standard.

2. A method for selecting a site for a charging station based on charging requirements according to claim 1, It is characterized in that The step S1 comprises: Extract static and dynamically updated status data of the electric vehicle from a designated data platform, and determine charging data and parking data of the electric vehicle based on the status data; The charging data and parking data include single driving data and single charging data of the electric vehicle, wherein the single driving data includes 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, driving end state of charge, single charging data includes the electric vehicle unique identifier, charging start time, charging end time, charging duration, charging start longitude, charging start latitude, charging start state of charge, charging end state of charge; A parking position of the electric vehicle is determined based on the single driving data, and a charging position of the electric vehicle is determined based on the single charging data.

3. A method for selecting a site for a charging station based on charging demand according to claim 2, It is characterized in that Determining the charging data and parking data of the electric vehicle according to the status data includes: The charging data is filtered and preprocessed according to the status data to obtain the charging data, wherein the preprocessing process includes: removing the charging data whose charging start longitude and latitude are not in the same area as the charging end longitude and latitude; All coordinate information in the remaining charging-related data is uniformly converted to the WSG-84 coordinate system to obtain charging data; The parking data is filtered according to the status data and preprocessed to obtain the parking data. The preprocessing process includes: sorting the status data of the electric vehicle according to the start time and determining the driving segment, taking the end time of the previous driving segment of the current driving segment as the start time of the parking segment, taking the start time of the current driving segment as the end time of the parking segment, and obtaining the start time and end time of the parking segment; The end longitude and latitude of the previous driving segment of the current driving segment are used as the start longitude and latitude of the parking segment, and the start longitude and latitude of the current driving segment are used as the end longitude and latitude of the parking segment; Eliminate abnormal parking segments with short parking duration and those whose parking start and end latitudes and longitudes are not in the same area; The coordinate system of the parking start longitude and latitude and the parking end longitude in the data corresponding to the remaining parking segments is converted into the WSG-84 coordinate system to obtain the parking data.

4. A method for selecting a site for a charging station based on charging requirements according to claim 1, It is characterized in that The step S2 comprises: Obtaining the original data of the offline map, and determining the first basic information of the required AOI based on the original data, wherein the first basic information includes the AOI name, AOI type, center point longitude and latitude, borders, province, city, and district and county; The AOI name is standardized to obtain a standardized name format, and is mapped and merged according to the first basic information to obtain a primary label of the AOI type, wherein the primary label includes: residential housing, shopping mall, company enterprise, and industrial park; The coordinates and longitude and latitude information in the first basic information are converted into the WSG-84 coordinate system to obtain the AOI area distribution; Determine the second basic information of the charging station in the city based on the original data, wherein the second basic information includes the station name, station operator, station ID, station latitude and longitude, province, city, and district and county; The coordinates and longitude and latitude information in the second basic information are converted to the WSG-84 coordinate system to obtain the distribution of charging stations.

5. A method for selecting a site for a charging station based on charging demand according to claim 1, It is characterized in that The step S3 comprises: The distance is calculated based on the latitude and longitude information of the start of each charging segment and the latitude and longitude information of the charging station. If the distance is less than or equal to a preset first distance, the charging segment is marked with a label unique to the charging station; if there are several stations whose distances are less than or equal to the first distance at the same time, the charging segment is marked with a label unique to the nearest charging station; Determine whether the charging and parking segments fall within the boundary of the AOI area according to their latitude and longitude. If they do, the corresponding segments are labeled with a unique label corresponding to the AOI. The parking and charging conditions of each AOI area are determined based on the tag information corresponding to the charging segment and the parking segment.

6. A method for selecting a site for a charging station based on charging demand according to claim 1, It is characterized in that The site selection of charging stations in the AOI area is carried out by constructing an optimization model, and the process includes: Determine the number and scale of charging stations in the AOI area based on the forecast results, and determine the number of charging piles, land occupation, and equipment requirements based on the number and scale of charging stations; Construct a construction cost calculation model based on the number of charging piles, land occupation and equipment demand information: Among them, i represents the distribution of charging demand points, j represents the candidate site for station construction, Indicates the total number of candidate sites for site selection. is the decision variable. If you choose to go to the station to charge, it is 1. If you do not go to the station to charge, it is 0. The decision is made based on the shortest distance. is the fixed cost of construction, is the unit price of charging pile, is the equivalent investment coefficient of the equipment, is the number of charging piles at the candidate location for station construction, n is the expected service life of the charging station, is the discount rate; The accumulated charging demand in the AOI area and the electric vehicle location distribution when charging demand is generated each time are counted, and a distance cost model for users to go to charging stations for charging is constructed: in, Represents the total number of charging demand points, which is determined by weighted average of statistical results. Unit time cost, is the demand quantity of the demand point, is the path distance from the demand point to the candidate site for station construction, is the congestion coefficient, is the average speed of electric vehicles; Based on the number of charging piles in the candidate site selection, a waiting cost model for users to queue up for charging is constructed: In the formula, Indicates the waiting time. For the charging pile service intensity, is the average number of electric vehicles arriving at the charging station per hour, is the number of charging pile services per unit time, is the idle probability of the charging station, k is the kth charging pile, T is the maximum waiting time, and W is the waiting cost coefficient; Construct a site selection optimization model based on the cost calculation model, distance cost model and waiting cost model: in, , , Respectively represent three preset correction coefficients; Set the constraints of the site selection optimization model: in, Indicates the minimum number of charging piles in a charging station. Indicates the maximum number of charging piles in the charging station; Taking the lowest cost as the optimization goal, calculations are performed based on constraints and the station site selection optimization model, and the site selection and charging pile scale of the charging station in the AOI area are determined according to the calculation results.