A method and system for predicting the spatiotemporal distribution of charging demand for electric taxis

Abstract

The present invention discloses a method and system for predicting the spatiotemporal distribution of charging demand for electric taxis, and belongs to the technical field of electric taxi charging. Specifically, it includes: dividing the charging demand prediction area into multiple traffic areas; based on the traffic areas, linearizing the regional road network, determining the travel starting point and travel path set of the electric taxi, and determining the initial geographical location of the electric taxi; based on the initial geographical location, considering the shift demand and charging randomness of the electric taxi, simulating the travel and charging process of the electric taxi within a single day, and determining the three key parameters of the spatiotemporal distribution of the charging demand of the electric taxi; based on the key parameters, calculating the spatiotemporal distribution prediction data of the charging demand of the electric taxi; based on the predicted data, performing time period data verification. The disclosed embodiments of the present invention can solve the problem of predicting the spatiotemporal distribution of charging demand in the area where electric taxis are initially deployed, and provide basic planning data for the construction of charging facilities in the area.

Description

Technical Field

[0001] The present application relates to the technical field of electric taxi charging demand, and in particular to a method and system for predicting the spatiotemporal distribution of electric taxi charging demand. Background Art

[0002] The prediction of electric vehicle charging demand is an important basis for the construction of charging infrastructure and a key prerequisite for ensuring that the charging infrastructure meets the charging needs of electric vehicle users and the quality of charging services.

[0003] Existing charging demand prediction methods have the problem of requiring a large amount of historical charging data. In newly built large-scale event areas or industrial parks, the number of electric taxis deployed is unclear and there is a lack of historical charging data.

[0004] For example, CN115829124A discloses a charging pile site selection method, which is characterized in that the method includes: using the data points of a preset charging application to obtain the charging data of electric vehicles in the target area, and according to the charging data, dividing the electric vehicles into daytime charging type, nighttime charging type and irregular charging type, and counting the number of vehicles of each type; according to the preset intersection and union strategy and the number of vehicles of each type, obtaining the vehicle set when the charging demand in the target area is the largest, and constructing the maximum power consumption vehicle distribution map according to the permanent address information of each vehicle in the vehicle set in the charging application and the administrative area layer of the target area; according to the maximum power consumption vehicle distribution map, obtaining the charging records of each existing charging site, and according to the preset clustering strategy and the permanent address information of each vehicle in the charging record, obtaining the loyal user distribution range of each charging site; using The distribution range of loyal users of each charging station and the preset distance ratio-level configuration rules are used to configure the layers of the maximum power consumption vehicle distribution map to obtain a user charging convenience map; the user charging convenience map is eroded according to a preset threshold to obtain a charging difficulty area, and based on the vehicle distribution density in the charging difficulty area, the cluster center of the charging difficulty area is obtained, and the human geographical environment type set within the preset range of the cluster center is obtained; using the pre-trained charging area scale prediction model, the charging area scale is predicted for the range size of the charging difficulty area and the human geographical environment type set to obtain the predicted charging area scale under each human geographical environment type in the human geographical environment type set; the location corresponding to the human geographical environment type with the smallest predicted charging area scale is selected as the charging pile site selection area. The above-mentioned prior art has the problem of requiring a large amount of historical charging data.

[0005] At the same time, the existing charging demand prediction methods only predict the overall charging load curve but do not clearly define the temporal and spatial distribution of the charging demand. This leads to unclear timing and size of the charging demand that needs to be met at specific charging facility construction sites, making it inconvenient to plan the charging infrastructure and improve the charging service quality at specific charging facility construction sites. Summary of the invention

[0006] In view of this, an embodiment of the present application provides a method and system for predicting the spatiotemporal distribution of charging demand for electric taxis, so as to solve the technical problems existing in the related art that the prediction of charging demand for electric taxis requires a large amount of historical charging data and only predicts the overall charging load curve without clarifying the spatiotemporal distribution of charging demand.

[0007] According to a first aspect of an embodiment of the present application, a method is provided, including:

[0008] S1: Divide the charging demand prediction area into multiple traffic zones;

[0009] S2: Based on the traffic zone, linearize the regional road network to determine the travel starting point and travel path set of the electric taxi;

[0010] S3: Determine the initial geographical location of the electric taxi based on the travel starting point and travel path set of the electric taxi;

[0011] S4: Based on the initial geographical location, taking into account the shift demand and charging randomness of the electric taxi, simulating the travel and charging process of the electric taxi in a single day, determining the time, location and remaining power of the electric taxi when charging demand is generated, that is, the three key parameters of the spatiotemporal distribution of the charging demand of the electric taxi;

[0012] S5: Determine the prediction data of the spatiotemporal distribution of the charging demand of the electric taxi based on the three key parameters of the spatiotemporal distribution of the charging demand;

[0013] S6: Based on the predicted data of the spatiotemporal distribution of the charging demand of the electric taxi, the predicted data is verified by time period to determine the rationality of the predicted data.

[0014] The above-mentioned method and system for predicting the spatiotemporal distribution of charging demand for electric taxis is characterized in that the charging demand prediction area is divided into a plurality of traffic zones, including:

[0015] S11: Clarify the land use attributes of each part in the charging demand prediction area and divide the traffic zones according to the land use attributes;

[0016] S12: Based on the gravity model method, the travel probabilities of electric taxis in the loaded state and the empty state between each traffic zone are calculated to obtain the loaded probability matrix and the empty probability matrix of electric taxis traveling between traffic zones. The elements of the loaded probability matrix are the loaded travel probabilities of electric taxis between each traffic zone, and the elements of the empty probability matrix are the empty travel probabilities of electric taxis between each traffic zone.

[0017] The above-mentioned method and system for predicting the spatiotemporal distribution of charging demand for electric taxis is characterized in that, based on the traffic zone, the regional road network of the charging demand prediction area is linearized to determine the travel starting point and travel path set of the electric taxis, including:

[0018] S21: Linearize the regional road network and simplify it into a vector road network consisting of road network nodes and line segments, where the coordinates of each road network node correspond to the center position of the traffic zone;

[0019] The above-mentioned method and system for predicting the spatiotemporal distribution of charging demand of electric taxis is characterized in that the initial geographical location of the electric taxi trip is determined based on the set of the trip starting point and the trip path of the electric taxi, including:

[0020] S31: Calculate the number of electric taxis in the charging demand prediction area based on the empty driving rate method;

[0021] S32: obtaining the initial operating state quantity of electric taxis whose quantity is equal to the ownership quantity based on Monte Carlo sampling, the initial operating state quantity including the initial power quantity and the initial operation time, and allocating the initial geographical location of the electric taxis based on the parking lot data of each traffic community;

[0022] The above-mentioned method and system for predicting the spatiotemporal distribution of charging demand for electric taxis are characterized by simulating the travel and charging process of electric taxis in a single day based on the initial geographical location, taking into account the shift demand and charging randomness of electric taxis, and determining the time, location and remaining power of the charging demand of electric taxis, that is, the three key parameters of the spatiotemporal distribution of charging demand of electric taxis, including:

[0023] S41: for a simulated single electric taxi, based on the charging demand prediction area's passenger carrying probability matrix and empty driving probability matrix, determine whether the electric taxi is in a passenger carrying travel path or an empty driving travel path during the travel process;

[0024] S42: It is clear that the electric taxi will continuously switch between the three states of passenger travel path, empty travel path and charging path during driving. The travel destination of the electric taxi is determined according to the travel path type of the electric taxi:

[0025] 1) Passenger travel path: When an electric taxi carries passengers, the travel destination is the road network node with the highest probability of passenger travel relative to the current road network node;

[0026] 2) Empty travel path: When the electric taxi is empty, the travel destination is the road network node with the highest probability of empty travel relative to the current road network node;

[0027] 3) Charging route: After the electric taxi has completed a passenger-carrying or empty-running route, determine whether it needs to be charged. If it needs to be charged, record the time, geographical location of the electric taxi, and remaining power, and designate the electric taxi to go to the nearest charging facility for recharging. After charging is completed within a certain period of time, it will set off again to enter the passenger-carrying route or empty-running route;

[0028] S43: Based on the travel destination of the travel path of a single electric taxi, combined with the information of the current vector road network, a path with the shortest travel time is determined according to the Floyd shortest path algorithm, and the electric taxi is designated to travel to the travel destination along this path in the simulation;

[0029] S44: Clarify the mathematical relationship between geographic location, driving speed, and road network conditions during the simulation of electric taxi travel and charging behavior;

[0030] S45: Clarify the mathematical relationship between the remaining power of electric taxis and the driving speed, driving range, charging power and battery capacity;

[0031] S46: During the simulation of the travel and charging behavior of the electric taxi, after each travel route (passenger-carrying, empty travel or charging) is completed, it is determined whether it is close to lunch, shift change or end of operation time. If so, the electric taxi is designated to return to the initial geographical location, otherwise it continues to enter the passenger travel route, empty travel route or charging route;

[0032] S47: Setting a remaining power warning value. When the remaining power of the electric taxi after the current route (passenger-carrying, empty or charging) is lower than the warning value, the electric taxi is designated to enter the charging route and complete the supplementary power. At the same time, a step of generating a random number is set. When the generated random number is greater than 0.5, even if the remaining power of the electric taxi is not lower than the power warning value, the electric taxi is designated to enter the charging route and complete the supplementary power, simulating the randomness of the electric taxi charging.

[0033] The above-mentioned method and system for predicting the spatiotemporal distribution of charging demand for electric taxis is characterized in that the prediction data of the spatiotemporal distribution of charging demand for electric taxis is calculated based on three key parameters of the spatiotemporal distribution of charging demand, including:

[0034] S51: Based on the number of electric taxis in the charging demand prediction area, simulate the travel and charging process of all electric taxis in a single day, determine the time, location and remaining power of all electric taxis in the charging demand prediction area when they generate several charging demands in a single day, and summarize and record them in the charging demand spatiotemporal distribution matrix, that is, calculate and obtain the prediction data of the spatiotemporal distribution of charging demand of electric taxis;

[0035] The above-mentioned method and system for predicting the spatiotemporal distribution of charging demand for electric taxis is characterized in that, based on the predicted data of the spatiotemporal distribution of charging demand for electric taxis, the predicted data is verified by time periods to determine the rationality of the predicted data, including:

[0036] S61: Clarify the mathematical relationship between the charging time of an electric taxi and the initial remaining power of the electric taxi when it is recharged;

[0037] S62: Based on the kstest function in MATLAB, the predicted data is subjected to a KS (Kolmogorov-Smirnov) normality test, the charging interval of electric taxis within a single day is subjected to an exponential distribution test for different time periods, and the charging time of electric taxis within a single day is subjected to a normal distribution test for different time periods. When the exponential distribution test and the normal distribution test are passed, it is determined that the predicted data on the spatiotemporal distribution of the charging demand for electric taxis is reasonable.

[0038] The above-mentioned method and system for predicting the spatiotemporal distribution of charging demand for electric taxis are characterized by designing a software system for the method for predicting the spatiotemporal distribution of charging demand for electric taxis, and the software system implements the steps of the above-mentioned method when executed by a processor:

[0039] A division module, used to divide the charging demand prediction area into multiple traffic zones;

[0040] The linearization module is used to perform linearization processing on the regional road network according to the traffic zone division results, and determine the travel starting point and travel path set of the electric taxi;

[0041] A determination module, used to determine the initial geographical location of the electric taxi trip based on the trip starting point and the trip path set of the electric taxi;

[0042] The simulation module is used to simulate the travel and charging process of electric taxis in a single day based on the initial geographical location of electric taxis, taking into account the shift demand and charging randomness of electric taxis, and determine the time, location and remaining power of electric taxis when charging demand is generated, that is, the three key parameters of the spatiotemporal distribution of charging demand of electric taxis;

[0043] A calculation module, used to calculate the prediction data of the time-space distribution of the charging demand of the electric taxi based on three key parameters of the time-space distribution of the charging demand;

[0044] The verification module is used to verify the predicted data by time period based on the predicted data of the spatiotemporal distribution of the charging demand of electric taxis, so as to determine the rationality of the predicted data.

[0045] The technical solution provided by the embodiments of the present application may include the following beneficial effects: the method and system for predicting the spatiotemporal distribution of charging demand for electric taxis provided by the present application divide the charging demand prediction area into multiple traffic zones; linearize the regional road network to determine the travel starting point and travel route set of the electric taxis; determine the initial geographical location of the electric taxis; consider the shift demand and charging randomness of the electric taxis, simulate the travel and charging process of the electric taxis within a single day, determine the time, location and remaining power when the electric taxis generate charging demand, that is, the three key parameters of the spatiotemporal distribution of the charging demand of the electric taxis; determine the predicted data of the spatiotemporal distribution of the charging demand of the electric taxis; perform data verification on the predicted data by time period to determine the rationality of the predicted data. This application takes into account the road network characteristics and the travel and charging characteristics of electric taxis in the charging demand prediction area, and clarifies the relationship between the time, location, remaining power and initial state, driving path and driving speed of electric taxis in the area when charging demand is generated; considering the random factors of electric taxi shifts and charging, the spatiotemporal distribution of charging demand for electric taxis in the area is reasonably predicted and verified; it is suitable for predicting the spatiotemporal distribution of charging demand for electric taxis in newly built large-scale event areas or industrial parks where historical data is lacking, and can provide data reference for the construction of charging facilities in the area, avoiding long charging queues and uneven charging services caused by the blind construction of charging facilities.

[0046] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0048] Figure 1 It is an implementation flow chart of a method and system for predicting the spatiotemporal distribution of charging demand for electric taxis according to an exemplary embodiment.

[0049] Figure 2 The diagram is a schematic diagram of dividing traffic zones in an area for predicting charging demand of electric taxis according to an exemplary embodiment.

[0050] Figure 3 The figure is a linearized schematic diagram of a regional road network for predicting charging demand for electric taxis according to an exemplary embodiment.

[0051] Figure 4 It is a program flow chart of a simulation of the travel and charging process of an electric taxi according to an exemplary embodiment, in which SOC (State of Charge) represents the remaining power of the battery of the electric taxi.

[0052] Figure 5 The present invention is a diagram showing the distribution pattern verification of the charging time and charging interval of an electric taxi according to an exemplary embodiment.

[0053] Figure 6 It is a block diagram of a software system for predicting the spatiotemporal distribution of charging demand for electric taxis according to an exemplary embodiment. DETAILED DESCRIPTION

[0054] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0055] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0056] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0057] Figure 1 is a flowchart of an implementation of a method and system for predicting the spatiotemporal distribution of charging demand for electric taxis according to an exemplary embodiment. Figure 1 As shown, in a specific embodiment, the method includes the following steps:

[0058] S1: Divide the charging demand prediction area into multiple traffic zones;

[0059] In a specific embodiment, the charging demand prediction area may be a newly built large-scale event area or industrial park, where electric taxis are planned to be deployed; the traffic area is a number of areas divided by the charging demand prediction area according to land use attributes and road network structure;

[0060] S11: clarifying the land use attributes of each part in the charging demand prediction area, and dividing the traffic districts according to the land use attributes;

[0061] In a specific embodiment, Figure 2 As shown, it is a schematic diagram of the traffic subdivision division of the electric taxi charging demand prediction area. Based on the land use attributes of the charging demand prediction area, the charging demand prediction area is divided into several traffic subdivisions;

[0062] S12: Based on the gravity model method, the travel probabilities of the electric taxi in the passenger-carrying state and the empty state between the traffic zones are calculated to obtain the passenger-carrying probability matrix and the empty-running probability matrix of the electric taxi traveling between the traffic zones, wherein the elements of the passenger-carrying probability matrix are the passenger-carrying travel probabilities of the electric taxi between the traffic zones, and the elements of the empty-running probability matrix are the empty-running travel probabilities of the electric taxi between the traffic zones;

[0063] In a specific embodiment, the passenger carrying probability matrix and empty driving probability matrix of the electric taxi traveling between traffic zones can be expressed as N 载 and N 空 :

[0064]

[0065] Where F is the total number of traffic zones divided in the area; N 载ij (i=1…F,j=1…F) is the probability that a passenger electric taxi in traffic zone i chooses traffic zone j as its travel destination; N 空ij is the probability that an unloaded electric taxi in traffic zone i chooses traffic zone j as its travel destination;

[0066] It can be seen from the above embodiments that the present application, based on the traffic zone division and gravity model method, can clarify the travel probability between the potential stops of various electric taxis in the charging demand prediction area, that is, clarify the travel possibility of electric taxis moving with passengers and moving empty between various traffic zones;

[0067] S2: Based on the traffic zone, linearize the regional road network to determine the travel starting point and travel path set of the electric taxi;

[0068] S21: Linearizing the regional road network to simplify the regional road network into a vector road network composed of road network nodes and line segments, wherein the coordinates of each road network node correspond to the central position of the traffic community, each road network node in the vector road network may be the initial geographical location or travel destination of a certain trip of the electric taxi, and each line segment in the vector road network may be a component of a certain trip path of the electric taxi;

[0069] In a specific embodiment, Figure 3 The figure shows a linearized schematic diagram of the regional road network for predicting the charging demand of electric taxis. The traffic zone is simplified into electric taxi stops replaced by the coordinates of the center point, and the road network is simplified into the travel paths between the stops. That is, the regional road network is linearized into a vector road network combining points and lines. The vector road network is the set of travel starting points and travel paths of electric taxis.

[0070] It can be seen from the above embodiments that the present application can simplify the travel starting point and travel path of the electric taxi into coordinates and path movement between coordinates by linearizing the road network in the charging demand prediction area, so as to facilitate calculation and recording of the time, location and remaining power of the electric taxi during the entire travel and charging process, and monitor whether the electric taxi needs to be recharged;

[0071] S3: Determine the initial geographical location of the electric taxi based on the travel starting point and travel path set of the electric taxi;

[0072] In a specific embodiment, the electric taxis are a group of electric taxis with a number of N; the initial geographical location of the electric taxi trip may be the corresponding coordinates of the electric taxi's location in the charging demand prediction area in the vector road network;

[0073] S31: Calculating the number of electric taxis in the charging demand prediction area based on the empty driving rate method;

[0074] In a specific embodiment, the number of electric taxis N is the number of electric taxis to be deployed in the charging demand prediction area, which can be calculated based on the number of fixed population and floating population in the area;

[0075] S32: obtaining the initial operating state quantity of the electric taxis whose number is equal to the number of the electric taxis based on Monte Carlo sampling, wherein the initial operating state quantity includes the initial power quantity and the initial operation time, and allocating the initial geographical location of the electric taxis based on the parking lot data of each of the traffic communities;

[0076] In a specific embodiment, after the initial geographical location of the electric taxis is obtained based on the parking lot data of each traffic zone, the number of electric taxis with each traffic zone as the initial geographical location can be described as:

[0077]

[0078] Where F is the total number of traffic zones divided in the area; N i is the number of electric taxis in the initial geographical location of traffic zone i (i=1…F); N is the number of electric taxis in the region; C i is the number of parking spaces in the traffic zone i in the area.

[0079] It can be seen from the above embodiments that the present application calculates the number of electric taxis in the area where electric taxis are planned to be deployed, allocates the number of electric taxis with initial geographical locations to each traffic zone according to the parking lot data of each traffic zone, and realizes the initial position determination of the electric taxi travel and charging process simulation.

[0080] S4: Based on the initial geographical location, taking into account the shift demand and charging randomness of the electric taxi, simulating the travel and charging process of the electric taxi in a single day, determining the time, location and remaining power of the electric taxi when charging demand is generated, that is, the three key parameters of the spatiotemporal distribution of the charging demand of the electric taxi;

[0081] In a specific embodiment, the shift demand of the electric taxi may be a two-shift system, and the charging randomness may generate a random number of 0 to 1 during the simulation process, and the random number value is compared with 0.5 to simulate the randomness of the electric taxi charging; Figure 4 As shown, it is a program flow chart of the simulation of the travel and charging process of an electric taxi. The specific simulation of the travel and charging process of an electric taxi includes the following steps:

[0082] S41: for the simulated single electric taxi, based on the passenger-carrying probability matrix and the empty-running probability matrix of the electric taxi in the charging demand prediction area, determining whether the electric taxi is in a passenger-carrying travel path or an empty-running travel path during the travel process;

[0083] S42: It is clear that the electric taxi in motion will continuously switch between the three states of the passenger travel path, the empty travel path and the charging path, and the travel destination of the electric taxi is determined according to the travel path type of the electric taxi:

[0084] 1) Passenger travel path: When the electric taxi carries passengers, the travel destination is the road network node with the highest probability of passenger travel relative to the current road network node;

[0085] 2) Empty travel path: When the electric taxi is empty, the travel destination is the road network node with the highest probability of empty travel relative to the current road network node;

[0086] 3) Charging route: after the electric taxi has completed a passenger-carrying or empty-running route, determine whether it needs to be charged. If it needs to be charged, record the time, the geographical location of the electric taxi and the remaining power, and designate the electric taxi to go to the nearest charging facility for recharging. After charging is completed within a certain period of time, the electric taxi will set out again to enter the passenger-carrying route or the empty-running route;

[0087] S43: Based on the travel destination of the travel path of a single electric taxi, combined with the information of the current vector road network, a path with the shortest travel time is determined according to the Floyd shortest path algorithm, and the electric taxi is designated to travel to the travel destination along this path in the simulation;

[0088] S44: clarifying the mathematical calculation relationship between the geographical location and the driving speed and road network conditions during the simulation of the travel and charging behavior of the electric taxi;

[0089] In a specific embodiment, the mathematical calculation relationship between the geographical location, the driving speed, and the road network conditions during the simulation of the electric taxi travel and charging behavior can be expressed as follows:

[0090]

[0091] in, They represent the planar geographical location of the electric taxi k on the road l at time t; flag is the running direction coefficient of the electric taxi (flag = 1 indicates the positive direction, flag = -1 indicates the reverse direction); v k is the driving speed of electric taxi k; t l is the end time of the electric taxi k traveling on road l; path The number of roads contained in the path;

[0092] S45: clarifying the mathematical calculation relationship between the remaining power of the electric taxi and the driving speed, cruising range, charging power, and battery capacity;

[0093] In a specific embodiment, the mathematical calculation relationship between the remaining power of the electric taxi and the driving speed, cruising range, charging power, and battery capacity can be expressed as follows:

[0094]

[0095] in, represents the remaining power of electric taxi k on road l at time t; M is the range of the electric taxi (km); P is the charging power of the electric taxi (kW); S E is the battery capacity of the electric taxi (kW·h);

[0096] S46: During the simulation of the travel and charging behavior of the electric taxi, after each travel route (passenger-carrying, empty travel or charging) is completed, it is determined whether it is close to lunch, shift change or end of operation time. If so, the electric taxi is designated to return to the initial geographical location, otherwise it continues to enter the passenger travel route, empty travel route or charging route;

[0097] S47: Setting a remaining power warning value. When the remaining power of the electric taxi after the current route (passenger-carrying, empty or charging) is lower than the warning value, the electric taxi is designated to enter the charging route and complete the supplementary power. At the same time, a step of generating a random number is set. When the generated random number is greater than 0.5, even if the remaining power of the electric taxi is not lower than the power warning value, the electric taxi is designated to enter the charging route and complete the supplementary power, simulating the randomness of the charging of the electric taxi.

[0098] It can be seen from the above embodiments that the present application simulates the travel and charging process of all electric taxis in the charging demand prediction area in a single day, and records the time, location and remaining power of each electric taxi in real time during the simulation process. The shift demand and charging randomness of electric taxis are taken into account during the simulation process, and the travel and charging behavior of electric taxis in the area can be simulated more comprehensively, and the time, location and size of charging demand in the area can be predicted;

[0099] S5: Calculating the predicted data of the temporal and spatial distribution of the charging demand of the electric taxi based on the three key parameters of the temporal and spatial distribution of the charging demand;

[0100] In a specific embodiment, the prediction data of the spatiotemporal distribution of charging demand is the time, location and magnitude of charging demand generated by all electric taxis in the charging demand prediction area, wherein the location generating the charging demand is represented by the coordinates of the location of the electric taxi, and the magnitude of the charging demand is represented by the remaining power of the electric taxi;

[0101] S51: According to the number of electric taxis in the charging demand prediction area, simulate the travel and charging process of all the electric taxis in a single day, determine the time, location and remaining power of all the electric taxis in the charging demand prediction area when they generate several charging demands in a single day, and summarize and record them in the charging demand spatiotemporal distribution matrix, that is, calculate and obtain the prediction data of the spatiotemporal distribution of the charging demand of the electric taxis;

[0102] In a specific embodiment, the above-mentioned time-space distribution matrix of charging demand can be expressed as matrix D EV :

[0103]

[0104] Among them, t i 、x i (t), y i (t), SOC i (i=1…n) represents the time, geographic location abscissa, geographic location ordinate and remaining power of the electric taxi for the i-th time of charging demand in a single day; n is the total number of times all electric taxis in the predicted area generate charging demand in a single day;

[0105] It can be seen from the above embodiments that the present application records the time, location and remaining power of the electric taxis during the travel and charging simulation of the electric taxis, and summarizes the charging demand of all electric taxis in the area within a single day based on the charging demand prediction into a charging demand spatiotemporal distribution matrix containing time, location and size information, so as to facilitate further verification and use of the charging demand spatiotemporal distribution data;

[0106] S6: Based on the predicted data of the spatiotemporal distribution of the charging demand of the electric taxi, the predicted data is verified by time period to determine the rationality of the predicted data.

[0107] In a specific embodiment, the above-mentioned time division is to divide the operation period of electric taxis into multiple time periods according to the shift demand of electric taxis and the travel peak and valley characteristics between traffic zones; then, for all electric taxis arriving at a specific charging infrastructure point in each time period, the charging interval and charging duration data are verified respectively, wherein the charging interval is the interval duration between the time when the electric taxis arrive at the charging facility point in each time period, and the charging duration is the charging duration calculated based on the power of the charging facility and the remaining power of the electric taxi when it arrives at the station. The specific data verification process includes:

[0108] S61: clarifying the mathematical calculation relationship between the charging time of the electric taxi and the initial remaining power of the electric taxi when it is supplemented with electricity;

[0109] S62: Based on the kstest function in MATLAB, a KS (Kolmogorov-Smirnov) normality test is performed on the predicted data, an exponential distribution test is performed on the charging interval of the electric taxi within a single day, and a normal distribution test is performed on the charging time of the electric taxi within a single day. When the exponential distribution test and the normal distribution test are passed, it is determined that the predicted data of the spatiotemporal distribution of the electric taxi charging demand is reasonable.

[0110] In a specific embodiment, the mathematical distribution check of charging demand data is performed based on the charging demand spatiotemporal distribution matrix obtained by simulating the travel and charging behavior of electric taxis, such as Figure 5 As shown, it is a verification diagram of the distribution law of charging time and charging interval of electric taxis. The charging intervals of electric taxis in two time periods in a single day have passed the exponential distribution test, and the charging time of electric taxis in two time periods in a single day have passed the normal distribution test.

[0111] It can be seen from the above embodiments that the present application can further calculate the charging interval and charging duration of the electric taxis arriving at the specific charging facility points in each time period according to the time when the electric taxis arrive at the charging facility points and the remaining power, and then perform data verification on the charging interval and charging duration in each time period to verify the validity of the simulated charging demand spatiotemporal distribution prediction data.

[0112] In summary, the method for predicting the spatiotemporal distribution of charging demand for electric taxis proposed in the embodiment of the present invention takes into account the regional road network and travel characteristics of electric taxis in the charging demand prediction area, and clarifies the relationship between the time, location, remaining power and initial state and driving speed of electric taxis in the region when charging demand is generated; considers the shift change and charging randomness factors of electric taxis, and makes reasonable predictions and verifications on the spatiotemporal distribution of charging demand for electric taxis in the region; solves the problem of predicting the spatiotemporal distribution of charging demand for electric taxis in areas where charging facilities are to be built in the absence of historical charging data, and avoids the problem of unbalanced charging services caused by blind construction.

[0113] Corresponding to the aforementioned embodiment of the method for predicting the spatiotemporal distribution of charging demand for electric taxis, the present application also provides an embodiment of a software system for predicting the spatiotemporal distribution of charging demand for electric taxis.

[0114] Figure 6 is a software system block diagram showing a prediction of the spatiotemporal distribution of charging demand for electric taxis according to an exemplary embodiment. Figure 6 When the software system is executed by the processor, the steps of any one of the aforementioned methods for predicting the spatiotemporal distribution of charging demand for electric taxis are implemented:

[0115] A division module 1 is used to divide the charging demand prediction area into multiple traffic zones;

[0116] Linearization module 2, used to perform linearization processing on the regional road network according to the traffic zone division result, and determine the travel starting point and travel path set of the electric taxi;

[0117] Determination module 3, used to determine the initial geographical location of the electric taxi based on the travel starting point and travel path set of the electric taxi;

[0118] Simulation module 4 is used to simulate the travel and charging process of electric taxis in a single day based on the initial geographical location of electric taxis, taking into account the shift demand and charging randomness of electric taxis, and determine the time, location and remaining power of electric taxis when charging demand is generated, that is, the three key parameters of the spatiotemporal distribution of charging demand of electric taxis;

[0119] A calculation module 5, for calculating the prediction data of the temporal and spatial distribution of the charging demand of the electric taxi based on the three key parameters of the temporal and spatial distribution of the charging demand;

[0120] The verification module 6 is used to verify the predicted data by time period based on the predicted data of the spatiotemporal distribution of the charging demand of the electric taxis, so as to determine the rationality of the predicted data.

[0121] Those skilled in the art will readily appreciate other embodiments of the present application after considering the description and practicing the contents disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The description and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the claims.

[0122] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for predicting the spatiotemporal distribution of charging demand for electric taxis, characterized in that: include: S1: Divide the charging demand prediction area into multiple traffic zones; S2: Based on the traffic zone, the regional road network is linearized to determine the travel starting point and travel path set of the electric taxi; S3: Determine the initial geographical location of the electric taxi trip based on the set of the electric taxi's travel starting point and travel path; S4: Based on the initial geographical location, considering the shift demand and charging randomness of electric taxis, the travel and charging process of electric taxis in a single day is simulated to determine the time, location and remaining power of the electric taxis when they need to charge, i.e., the three key parameters of the spatiotemporal distribution of the charging demand of electric taxis; S5: Based on the three key parameters of the spatiotemporal distribution of charging demand, calculate the predicted data of the spatiotemporal distribution of charging demand for electric taxis; S6: Based on the forecast data of the spatiotemporal distribution of the charging demand of electric taxis, the forecast data is verified by time period to determine the rationality of the forecast data; The charging demand forecast area is divided into multiple traffic zones, including: S11: Clarify the land use attributes of each part in the charging demand prediction area and divide the traffic zones according to the land use attributes; S12: Based on the gravity model method, the travel probabilities of electric taxis in the loaded state and the empty state between each traffic zone are calculated to obtain the loaded probability matrix and the empty probability matrix of electric taxis traveling between traffic zones. The elements of the loaded probability matrix are the loaded travel probabilities of electric taxis between each traffic zone, and the elements of the empty probability matrix are the empty travel probabilities of electric taxis between each traffic zone. The loaded probability matrix and the empty probability matrix of electric taxis traveling between traffic zones are expressed as N 载 and N 空 : Where F is the total number of traffic zones divided in the area; N 载ij (i=1…F,j=1…F) is the probability that a passenger electric taxi in traffic zone i chooses traffic zone j as its travel destination; N 空ij is the probability that an unloaded electric taxi in traffic zone i chooses traffic zone j as the travel destination; based on the traffic zone, the regional road network in the charging demand prediction area is linearized to determine the travel starting point and travel path set of the electric taxi, including: S21: Linearize the regional road network and simplify it into a vector road network consisting of road network nodes and line segments, where the coordinates of each road network node correspond to the center position of the traffic zone; Based on the set of the travel starting point and travel path of the electric taxi, the initial geographical location of the electric taxi is determined, including: S31: Calculate the number of electric taxis in the charging demand prediction area based on the empty driving rate method; after the initial geographical location of the electric taxis is obtained based on the parking lot data of each traffic zone, the number of electric taxis with each traffic zone as the initial geographical location is described as: Where F is the total number of traffic zones divided in the area; N i is the number of electric taxis in the initial geographical location of traffic zone i (i=1…F); N is the number of electric taxis in the region; C i is the number of parking spaces in the traffic zone i in the area; S32: obtaining the initial operating state quantity of electric taxis whose quantity is equal to the ownership quantity based on Monte Carlo sampling, the initial operating state quantity including the initial power quantity and the initial operation time, and allocating the initial geographical location of the electric taxis based on the parking lot data of each traffic community; Based on the initial geographical location, considering the shift demand and charging randomness of electric taxis, the travel and charging process of electric taxis in a single day is simulated to determine the time, location and remaining power of the charging demand of electric taxis, that is, the three key parameters of the spatiotemporal distribution of the charging demand of electric taxis, including: S41: for a simulated single electric taxi, based on the charging demand prediction area's passenger carrying probability matrix and empty driving probability matrix, determine whether the electric taxi is in a passenger carrying travel path or an empty driving travel path during the travel process; S42: It is clear that the electric taxi will continuously switch between the three states of passenger travel path, empty travel path and charging path during driving. The travel destination of the electric taxi is determined according to the travel path type of the electric taxi: 1) Passenger travel path: When an electric taxi carries passengers, the travel destination is the road network node with the highest probability of passenger travel relative to the current road network node; 2) Empty travel path: When the electric taxi is empty, the travel destination is the road network node with the highest probability of empty travel relative to the current road network node; 3) Charging route: After the electric taxi has completed a passenger-carrying or empty-running route, determine whether it needs to be charged. If it needs to be charged, record the time, geographical location of the electric taxi, and remaining power, and designate the electric taxi to go to the nearest charging facility for recharging. After charging is completed within a certain period of time, it will set off again to enter the passenger-carrying route or empty-running route; S43: Based on the travel destination of the travel path of a single electric taxi, combined with the information of the current vector road network, a path with the shortest travel time is determined according to the Floyd shortest path algorithm, and the electric taxi is designated to travel to the travel destination along this path in the simulation; S44: The mathematical relationship between the geographical location, driving speed and road network conditions during the simulation of electric taxi travel and charging behavior is clarified; the mathematical relationship between the geographical location, driving speed and road network conditions during the simulation of electric taxi travel and charging behavior is expressed as follows: in, They represent the planar geographical location of the electric taxi k on the road l at time t; flag is the running direction coefficient of the electric taxi, flag = 1 represents the positive direction, flag = -1 represents the reverse direction; v k is the driving speed of electric taxi k; t l is the end time of the electric taxi k traveling on road l; path The number of roads contained in the path; S45: Clarify the mathematical calculation relationship between the remaining power of an electric taxi and its driving speed, cruising range, charging power, and battery capacity; the mathematical calculation relationship between the remaining power of an electric taxi and its driving speed, cruising range, charging power, and battery capacity is expressed as follows: in, represents the remaining power of electric taxi k on road l at time t; M is the range of the electric taxi (km); P is the charging power of the electric taxi (kW); S E is the battery capacity of the electric taxi (kW·h); S46: During the simulation of the travel and charging behavior of the electric taxi, after each travel path is completed, including the time statistics of carrying passengers, idling or charging, it is determined whether it is close to lunch, shift change or end of operation time. If so, the electric taxi is designated to return to the initial geographical location, otherwise it continues to enter the passenger travel path, idling travel path or charging path; S47: Setting a remaining power warning value. When the remaining power of the electric taxi after the current route ends is lower than the warning value, and the current route includes passenger-carrying, idling or charging time, the electric taxi is designated to enter the charging route and complete the supplementary power. At the same time, a step of generating a random number is set. When the generated random number is greater than 0.5, even if the remaining power of the electric taxi is not lower than the power warning value, the electric taxi is designated to enter the charging route and complete the supplementary power, simulating the randomness of the electric taxi charging. Based on the three key parameters of the spatiotemporal distribution of charging demand, the forecast data of the spatiotemporal distribution of charging demand for electric taxis is calculated, including: S51: Based on the number of electric taxis in the charging demand prediction area, simulate the travel and charging process of all electric taxis in a single day, determine the time, location and remaining power of all electric taxis in the charging demand prediction area when they generate several charging demands in a single day, and summarize and record them in the charging demand spatiotemporal distribution matrix, that is, calculate and obtain the prediction data of the spatiotemporal distribution of the charging demand of electric taxis; the spatiotemporal distribution matrix of the charging demand is expressed as matrix D EV : Among them, t i 、x i (t), y i (t), SOC i (i=1…n) represents the time when the i-th charging demand is generated in a single day, the horizontal coordinate of the geographical location, the vertical coordinate of the geographical location and the remaining power of the electric taxi; n is the total number of times that all electric taxis in the forecast area generate charging demand in a single day; based on the forecast data of the spatiotemporal distribution of the charging demand of electric taxis, the forecast data is verified by time period to determine the rationality of the forecast data, including: S61: Clarify the mathematical relationship between the charging time of an electric taxi and the initial remaining power of the electric taxi when it is recharged; S62: Based on the kstest function in MATLAB, a KS normality test is performed on the predicted data, an exponential distribution test is performed on the charging interval of the electric taxis in a single day in different time periods, and a normal distribution test is performed on the charging time of the electric taxis in a single day in different time periods. When the exponential distribution test and the normal distribution test are passed, it is determined that the predicted data of the spatiotemporal distribution of the charging demand of electric taxis is reasonable.

2. The method for predicting the spatiotemporal distribution of charging demand for electric taxis according to claim 1, characterized in that: A software system for predicting the spatiotemporal distribution of charging demand for electric taxis is designed. When the software system is executed by a processor, the steps of the method as claimed in any one of claim 1 are implemented: A division module, used to divide the charging demand prediction area into multiple traffic zones; The linearization module is used to perform linearization processing on the regional road network according to the traffic zone division results, and determine the travel starting point and travel path set of the electric taxi; A determination module, used to determine the initial geographical location of the electric taxi trip based on the trip starting point and the trip path set of the electric taxi; The simulation module is used to simulate the travel and charging process of electric taxis in a single day based on the initial geographical location of electric taxis, taking into account the shift demand and charging randomness of electric taxis, and determine the time, location and remaining power of electric taxis when charging demand is generated, that is, the three key parameters of the spatiotemporal distribution of charging demand of electric taxis; A calculation module, used to calculate the prediction data of the time-space distribution of the charging demand of the electric taxi based on three key parameters of the time-space distribution of the charging demand; The verification module is used to verify the predicted data by time period based on the predicted data of the spatiotemporal distribution of the charging demand of electric taxis, so as to determine the rationality of the predicted data.

Images