Method for determining a charging request probability of electric vehicles, and occupancy level prediction of charging stations

Abstract

The invention relates to a method for determining a charging request probability of approaching electric vehicles (50) for at least one charging station (100-n), the method comprising the steps of: detecting the geo-position of the at least one charging station (100-n); defining a vicinity (130-n) around the geo-position of the at least one charging station (100-n); detecting a number per time interval of electric vehicles (50) entering the vicinity (130-n) of the at least one charging station (100-n); detecting a number of charging processes per time interval; determining a charging request probability based on a quotient formed by dividing the number of charging processes per time interval by the number of electric vehicles (50) that have entered the vicinity (130-n) in the same time interval. The invention also relates to a method for predicting an occupancy level of a plurality of charging stations (100-n) based on the determined charging request probabilities for an electric vehicle (50) approaching a charging station.

Description

[0001] Description

[0002] Method for determining the charging request probability of electric vehicles and predicting the utilization of charging points

[0003] The invention relates to a method for determining the probability of a charging request from approaching electric vehicles for at least one charging point. Furthermore, the invention relates to a method for predicting the utilization of a plurality of charging points.

[0004] For owners of electric vehicles who need to charge electrical energy into an electric vehicle's electrical storage device before starting a journey and / or during journey breaks, it is advantageous to be able to determine the utilisation of publicly accessible charging points as accurately as possible.

[0005] A charging point is defined here as charging infrastructure at a specific geolocation. A charging point can comprise multiple charging stations for electric vehicles, where electrical energy can be simultaneously charged into an electric vehicle's storage unit. The infrastructure unit that supplies a charging station is also referred to as a charging station. A charging point can therefore comprise multiple charging stations.

[0006] CN 114954129 discloses a method and apparatus for recommending charging station information. The method includes the steps of: responding to a recommendation request input by a user and determining the first time and the first remaining electric power of a current vehicle traveling to each charging station that matches a travel route; acquiring charging queue information of each charging station at the first time from a cloud server; and sorting the charging stations according to the first remaining electric power and the charging queue information and recommending charging station information according to a sorting result. According to the technical scheme of the described embodiment, reasonable travel charging suggestions can be provided by integrating various factors, and a user's charging experience can be improved.

[0007] US20170276503A1 relates to a big data-based method or system for recommending charging stations for electric vehicles with the shortest charging waiting time, comprising a travel information receiving unit for receiving a charging station search request and information about the estimated discharge time and a current location of an electric vehicle; a charging station information receiving unit for receiving real-time power consumption data from charging stations; a charging waiting time calculating unit for receiving the power consumption data from the charging station information receiving unit and calculating the charging waiting time of the charging stations in response to a charging station search request based on their power consumption;and an optimal charging station providing unit for receiving the information about the estimated discharge time and the current location from the travel information receiving unit and for receiving the charging waiting time from the charging waiting time calculating unit, and for providing information about at least one optimal charging station that can be reached within the estimated discharge time based on the current location and can recharge the vehicle in the shortest time.;

[0008] KR 20190102589 A discloses a mobile terminal and a method for guiding a charging station of the mobile terminal. According to one embodiment, the method for guiding a charging station of the mobile terminal comprises the steps of: detecting a connection with a vehicle in response to boarding the vehicle; detecting a charging station guide request; receiving information about chargeable charging stations from a server in response to a charging station guide request; and displaying at least one filtered and searched charging station information item on a navigation screen based on a learned user behavior pattern and information related to the vehicle.

[0009] DE 102022 115 122 A1 describes a method for determining the number of electric vehicles waiting for a free charging point at a charging infrastructure with at least one charging point, wherein the charging infrastructure is located within a geofence area delimited by a geofence. In the method, an electric vehicle with a geofence function is automatically classified as waiting for a charging point if all charging points are occupied, at least one charge state of a drive battery of this electric vehicle fulfills an associated charge state criterion, and at least one movement parameter of this electric vehicle fulfills an associated movement criterion. Also described is a charging infrastructure with at least one charging point for charging electric vehicles located within a geofence area delimited by a geofence, wherein the charging infrastructure is configured to carry out the method.Implementing the methods described in the state of the art often requires the exchange of very large amounts of data and very specific information about a large number of electric vehicles. However, there are situations in which these requirements are not met, and the institutions from which predictions for charging point utilization are to be made. Rather, only limited information is available, for example, from a manufacturer's electric vehicles. It is desirable to be able to make statements about charging point utilization even in such situations. Furthermore, charging points are generally not installed away from other infrastructure. For example, charging points are located in parking lots of retail stores or restaurants, etc.In order to be able to make statements about the possible utilization of the charging points, it is desirable to know how likely it is that an approaching electric vehicle wants to charge at the charging point.

[0010] The invention is based on the object of enabling a statement on the probability of a charging request from an electric vehicle approaching a charging point, based on as little information as possible for at least one, preferably a plurality of, charging points, as well as a method for predicting the utilization of a plurality of charging points, even when only limited data is available for individual charging points.

[0011] The invention is achieved by a method having the features of patent claim 1 and a method having the features of patent claim 10. Advantageous embodiments emerge from the subclaims.

[0012] The invention is based on the idea of ​​evaluating data from a vehicle fleet, for example, data from a manufacturer's electric vehicles, in order to derive a probability that electric vehicles, when approaching a charging point, will actually want to charge their batteries at that particular charging point. According to one aspect of the invention, a charging request probability is thus determined for electric vehicles approaching a specific charging point.

[0013] According to a second aspect of the invention, charging points can be classified depending on their environment. For example, characteristics such as proximity to a motorway or expressway, proximity to certain retail outlets, the type of retail outlet, the type, number, and size of restaurants, charging capacity in terms of the number of available charging stations where a charging process can be carried out simultaneously, etc. can play a role. Charging points can be classified based on these characteristics. Using this classification, it is possible to transfer the charging request probability of approaching electric vehicles determined for a specific charging point to another charging point with the same classification. Together with the currently determined number of electric vehicles approaching in a given time interval, the utilization of the charging point can be determined.

[0014] In particular, a method for determining a charging request probability of approaching electric vehicles for at least one charging point is provided, comprising the steps:

[0015] Recording the geo-position of at least one charging point;

[0016] Defining a proximity area (also called a capture area) around the geo-position of at least one charging point;

[0017] Recording a number of electric vehicles entering the vicinity of at least one charging point per time interval;

[0018] Recording a number of charging processes per time interval,

[0019] Determining a charging request probability based on a quotient of the number of charging processes per time interval divided by the number of electric vehicles driven in during the same time interval.

[0020] The locations of various public charging points can be recorded from a variety of sources. Typically, the charging point operators make this information publicly available, for example, on the internet.

[0021] A near area, also known as a capture zone, is defined around such a geolocation of a charging point. Electric vehicles that enter the near area or

[0022] Vehicles entering the capture zone should be recorded wherever possible. An automobile manufacturer can achieve this, for example, by having vehicle computers or control units linked to the vehicle navigation system and its position detection system iteratively check the current position of the electric vehicle to determine whether the electric vehicle enters a specified proximity area around one of the charging points. If this is the case, this information is transmitted to the automobile manufacturer.

[0023] Since modern automobiles are equipped with a mobile radio unit, a data packet can be sent to the vehicle manufacturer via a mobile radio connection, indicating that the electric vehicle has entered the charging point's coverage area. Preferably, the corresponding charging point is identified via an identifier that is transmitted along with the data. In addition, further data about the electric vehicle and its status can be transmitted. Furthermore, a time information is preferably transmitted, indicating the time or period at which the electric vehicle entered the vicinity of the charging point.

[0024] The entry of an electric vehicle into the near field can also be detected using sensors, which may include cameras, induction loops, light barriers, etc.

[0025] Furthermore, the electric vehicles are preferably configured to transmit a data packet to the vehicle manufacturer when they initiate a charging process. This preferably also includes an identifier for the charging point at which the electric vehicle charges electrical energy into its energy storage device. In addition, a time or period at which the electric vehicle began a charging process at the charging point is preferably also transmitted.

[0026] The vehicle manufacturer records the information transmitted to it regarding entry into the capture zone and actual charging. For both pieces of information, the times of transmission—preferably a time of day and preferably also the day of the week—are preferably recorded and assigned to the data. Alternatively or additionally, the times can be included in the data itself.

[0027] For data protection reasons, the electric vehicle may include time information in the data packet about entering the capture zone and / or the data packet containing information about actual charging. This information includes both the time of day and / or the corresponding day of the week. However, this information does not specify the exact time, but rather assigns it to a time period. This can ensure a certain degree of anonymity of the transmitted data.

[0028] In both cases, the vehicle manufacturer is able to use the recorded data to determine a charging request probability for electric vehicles approaching the charging point by calculating a quotient of the number of electric vehicles actually charged in a time interval and the number of electric vehicles that have entered the nearby area. As already indicated, the charging request probability for an electric vehicle approaching the charging point differs depending on the time of day and / or the day of the week. Therefore, a preferred embodiment provides for the charging request probability to be determined for a plurality of time periods at different times of day and / or on different days of the week, so that the charging probability for an approaching electric vehicle is determined depending on the time of day and / or the day of the week.In many cases, it is not the exact day of the week that matters, but rather the type of weekday, for example, weekday or Sunday or public holiday.

[0029] A conclusion can be improved by additionally recording whether the charging point is or was fully utilized during the time interval. If all charging stations at a charging point are occupied, it cannot be ruled out that some of the electric vehicles entering the vicinity requested charging but were unable to do so because none of the charging stations were available.

[0030] To take this into account, an uncertainty measure is assigned to the charging request probability, which provides information about how reliably the charging request probability could be determined. If information about the charging point utilization is available and this charging point is not fully utilized, the determined charging request probability has a high degree of reliability. If it is known that the charging point was fully utilized, the charging request probability has a low degree of reliability. If the charging point utilization is not known, a medium uncertainty value can be assumed. The uncertainty can also be determined by the number of charging stations in relation to the actual charging processes in the time interval, taking into account a proportion factor, whereby the proportion factor indicates a proportion of the electric vehicles in the fleet or the manufacturer to the total number of electric vehicles.For example, the manufacturer's market share of electric vehicles sold in relation to the total number of electric vehicles sold can be used.

[0031] To increase the reliability of the statements, separately recorded charging request probabilities are merged. In the simplest variant, geometric mean values ​​are calculated. Here, preferably only charging request probabilities for the same time of day and the same day of the week are merged. Likewise, time-dependent charging request probabilities can be determined separately for weekdays, Saturdays and Sundays. In this case, time-dependent summaries are also preferably made. The grading can be selected and / or specified depending on the amount and quality of the available data. If only a small amount of data is available, charging request probabilities can be determined for longer time periods, such as for the time periods from 6:00 a.m. to 9:00 a.m., 9:00 a.m. to 12:00 p.m., 12:00 p.m. to 4:00 p.m., 4:00 p.m. to 9:00 p.m., and 9:00 p.m. to 6:00 a.m.It is clear to the person skilled in the art that this is an arbitrarily chosen classification.

[0032] If reliability information is available, it will be taken into account in the merger.

[0033] It has proven particularly advantageous to identify environmental characteristics for charging points. Environmental characteristics are defined as all features that describe and characterize the environment and surroundings of the charging point, as well as the charging point itself. These include, for example, proximity to a motorway or other expressway, proximity to retail outlets, type of retail outlets, size of retail outlets, type, number, and size of restaurants, type and number of recreational facilities such as sports facilities, cinemas, or theaters, number of vehicle parking spaces, charging capacity in terms of the available charging points, etc. This list is not intended to be exhaustive.

[0034] These environmental features can be used to classify charging points. Therefore, one embodiment provides for environmental features to be recorded for the charging point, and the charging point to be classified based on its environmental features.

[0035] The quality of the information, especially when only a small amount of data is available, can be increased by merging the charging request probabilities of similarly classified charging points when merging charging request probabilities.

[0036] It is particularly preferable to obtain a collection of charging request probabilities that are determined depending on the time of day and day of the week and that can be retrieved, for example, from a memory and are also available for differently classified charging points.

[0037] The information obtained in this way can now be used to forecast the utilization of a charging point. To do so, it is only necessary to determine the number of electric vehicles that have entered the corresponding vicinity of the charging point. Together with the corresponding charging request probability for electric vehicles approaching the charging point for the correspondingly classified charging point, which is preferably available based on the time of day and day of the week or retrieved from a collection, for example, a database, the current utilization of the charging point can be determined, taking into account the charging capacity of the charging point, for example, the number of available charging spaces.

[0038] A method for forecasting the utilization of a plurality of charging points provides that for each of the charging points the environmental characteristics are recorded; a number of its charging stations is recorded; a classification is carried out based on its recorded environmental characteristics; a charging request probability for this or a similarly classified charging point is retrieved from a storage device; its geoposition is recorded and its local area is determined; a number of electric vehicles entering its local area per time interval is recorded and / or measured and multiplied by the retrieved charging request probability and divided by a number of its charging stations in order to determine a utilization value, and a utilization value determined in this way is provided, wherein for at least one of the charging points the retrieved charging request probability was not determined for the corresponding charging point itself.

[0039] The procedure thus makes it possible to provide a reliable forecast even for charging points for which no charging request probabilities have been recorded.

[0040] Furthermore, conclusions can be drawn about the overall behavior of all electric vehicles from a sample of electric vehicles, namely the electric vehicles of one manufacturer. Here, it is assumed that the behavior of electric vehicle users with regard to charging behavior is approximately the same regardless of the manufacturer. This is at least true if the charging request probability is determined based on electric vehicles in a fleet that includes a good cross-section of the various available electric vehicles in terms of range, storage size, comfort, etc. It is therefore only necessary to determine the charging request probability for a subset of the electric vehicles approaching a charging point. This information can then be used to predict the utilization of charging points once the number of electric vehicles in the local area has been determined.

[0041] This determination can be made, for example, using sensors that use cameras, for example, to detect electric vehicles entering a nearby area. However, the data can also be collected by the electric vehicles themselves, which transmit this information to a fleet management system, such as an automobile manufacturer, as in the case of determining the charging request probability. With knowledge of the market share of the automobile manufacturer's electric vehicles and, if applicable, further knowledge of driving behavior compared to electric vehicles from other manufacturers, which is obtained through traffic monitoring measurements, it is possible to determine the total number of electric vehicles approaching a nearby charging point. Together with the determined charging request probability for the corresponding charging point or a similarly classified charging point, the utilization can be determined.This capacity information can be provided, for example, by the car manufacturer and used by that car manufacturer's electric vehicles or, if applicable, all electric vehicles. This can improve the selection of the charging point being used or to be used, with a view to minimizing waiting times.

[0042] The invention is explained in more detail below with reference to a drawing:

[0043] Fig. 1 is a schematic representation of a plurality of charging points for which a charging request probability is determined and / or a utilization is forecast.

[0044] Figure 1 schematically illustrates a plurality of charging points 100-n with their respective surroundings 150-n. The insertion "-n" indicates an index that is chosen differently for similar features in order to uniquely identify them. The index takes on integer values, for example, 1, 2, 3, .... Each of the charging points 100-n comprises one or more charging stations 110, each of which corresponds to a charging station 120.

[0045] An environment 150-n is described and characterized by so-called environmental features. These include information about private and public facilities, infrastructure, etc. An important environmental feature is proximity to a highway 310 or a dual carriageway / rural road 320. Other essential features can include the number, type, and size of restaurants 220, the number, type, and size of retail stores 210, the number, type, and size of leisure facilities 230, the number of parking spaces 240, the number, type, and size of service establishments 250, and so on, to name a few. The charging capacity of the charging point in terms of its available charging spaces is also an environmental feature. It is expressly pointed out that this list is only exemplary and not exhaustive.

[0046] Electric vehicles 50 approach the charging points 100-n. In doing so, they enter the respective proximity zone 130-n. This can be detected using sensors 140, which are configured, for example, as cameras, or also using induction sensors or other sensors. The proximity zones 130-n, also referred to as capture zones, are defined around the geopositions of the charging points 100-n. The geopositions of the charging points 100-n are published and made public, for example, by charging point operators 700.

[0047] In one embodiment, a fleet management system 500, which is, for example, an automobile manufacturer, defines the proximity ranges 130-n using a capture range definition device 530 and transmits these via a communication device 510 to vehicle communication devices 58 of the electric vehicles 50 in its fleet. An electric vehicle in the fleet is understood here to be, for example, any electric vehicle manufactured by the manufacturer.

[0048] The electric vehicles 50 each additionally have a navigation device 52 that enables them to determine their own geoposition. This can be done, for example, via a so-called satellite navigation unit (not shown), which evaluates signals from a satellite navigation system, such as the Global Positioning System (GPS). A control unit 54 in the electric vehicle 50 is coupled to the navigation device 52 and monitors whether the electric vehicle 50 enters a defined capture zone or near range 130-n transmitted by the fleet management system 500.

[0049] The information about the nearby areas 130-n can also be stored in a memory during the manufacture of the electric vehicle 50. As soon as the electric vehicle 50 detects that it is entering the nearby area 130-n of one of the charging points 100-n, it registers this. The control unit 54 then initiates the transmission of a data packet to the fleet management system 500. This transmission can occur promptly or with a time delay. In addition to the geoposition or another identifier of the corresponding charging point that the electric vehicle 50 is approaching, time information can also be transmitted, which preferably also includes a date or day of the week in addition to the time of day. For data protection reasons, the electric vehicle 50 can assign the time (i.e., the time of entering the nearby area 130-n) to a time period and only transmit the corresponding time period in which the arrival or entry time into the nearby area 130-n falls.

[0050] An electric vehicle 50 behaves similarly when it begins a charging process at one of the charging stations 120 of charging point 100-n. The control unit 54 transmits a data packet indicating the charging of the electric vehicle. This data packet also preferably includes, in addition to the geolocation and / or identifier of charging point 100-n, a time indication, which includes a time or a time range, as well as preferably a day of the week or the complete date.

[0051] The fleet management system 500 uses an evaluation device 520 to determine a charging request probability for electric vehicles approaching a specific charging point 100-n. For this purpose, the number of electric vehicles 50 beginning a charging process at the corresponding charging point 100-n per time interval is divided by the number of electric vehicles 50 that have entered the vicinity 130-n of the corresponding charging point 100-n during the corresponding time interval. This detection is preferably performed depending on the time of day and the day of the week. The corresponding results are stored in a data memory 550.

[0052] If a charging point operator 700 provides utilization information, this information is taken into account by the evaluation device 520 when determining the charging request probabilities. If a charging point 100-n is fully utilized in a time interval with regard to its charging capacity, in the sense that all charging stations 120 are occupied, the determined number of charging processes in the time interval is subject to uncertainty, since it cannot be guaranteed that every electric vehicle 50 that requested charging was actually able to begin a charging process. If, however, one of the charging stations 120 was available at any time in the time interval, the determined number of charging processes has no or only a very low uncertainty. The uncertainties thus determined are taken into account, in particular when merging determined charging request probabilities.In addition, the fleet management system 500 records environmental features for the surroundings 150-n of the charging points 100-n. Based on these environmental features, the individual charging points 100-n are classified. In the schematically simplified example shown, for example, charging points 100-1 and 100-2 have similar environments. Both are accessed via a highway 310 and each have a restaurant 220, a retail business 210, and a leisure facility 230 in the surroundings 150-n. Again, please note that a highly simplified representation has been chosen here. The other charging point 100-3 shown, in contrast, only has a restaurant 210 and is accessible via a country road 320.

[0053] If the charging request probability for electric vehicles 50 is determined for the charging point 100-1, this charging request probability can be transferred to an equally classified charging point 100-2.

[0054] It is therefore sufficient if the charging points are classified according to their environmental characteristics and, in addition, the charging request probabilities for approaching electric vehicles are determined for a limited number of charging points.

[0055] If the number of electric vehicles approaching a charging point or located in its vicinity is determined, together with knowledge of the charging point capacity with regard to the available charging spaces and the probability of charging being requested, the utilisation of the charging point can be easily predicted.

[0056] To determine charging point utilization, it is only necessary to create a collection of charging request probabilities for approaching electric vehicles at a charging point for charging points classified according to environmental characteristics. Then, for the charging points for which utilization information is to be predicted, the environmental characteristics and their charging capacity must be recorded, at least with regard to the available charging spaces, and the charging points must be classified. To determine current utilization, only the number of electric vehicles entering the nearby areas must be determined. This allows the utilization to be calculated for the corresponding charging points. For this purpose, the number of electric vehicles in the nearby area or that have entered the nearby area per time interval is measured or determined based on data packets transmitted by the electric vehicles.If only a subset of electric vehicles transmit data packets when entering a charging point, the total number of entering electric vehicles must be predicted. This can be done, for example, based on the ratio of electric vehicles transmitting data packets to the total number of electric vehicles. If only one-tenth of the electric vehicles transmit data packets, the total number of entering electric vehicles is ten times the number determined based on the number of received data packets.

[0057] Multiplying the number of electric vehicles by the corresponding charging request probability retrieved from the data storage 550 and dividing this by the number of charging stations at the charging point yields a statement about the utilization. The utilization rates are determined in a prediction device 560 of the fleet management system 500.

[0058] The capacity utilizations thus determined are provided via an output device 570. This can be configured, for example, as a database server from which the capacity utilization statements can be retrieved. Electric vehicles can use these for route planning.

[0059] Charging point operators can particularly use the charging request probabilities to predict the utilization of existing charging points and predict the utilization of new charging points.

[0060] List of reference symbols

[0061] Electric vehicles

[0062] Navigation device

[0063] control unit

[0064] Data storage

[0065] Communication device

[0066] Charging point

[0067] Charging station

[0068] Loading bays

[0069] Close range

[0070] Sensors

[0071] Vicinity

[0072] trading business

[0073] catering business

[0074] Leisure facility

[0075] Parking spaces

[0076] Service companies

[0077] Highway

[0078] country road

[0079] Fleet management system

[0080] Communication device

[0081] Evaluation device

[0082] Catch area definition device

[0083] Classification facility

[0084] Storage Prediction device Output device Charging point operator

Claims

Patent claims 1. Method for determining a charging request probability of approximating Electric vehicles (50) for at least one charging point (100-n) comprising the steps of: detecting the geo-position of the at least one charging point (100-n); Defining a proximity range (130-n) around the geo-position of the at least one charging point (100-n); Detecting a number per time interval of electric vehicles (50) entering the vicinity (130-n) of the at least one charging point (100-n); Recording a number of charging processes per time interval, Determining a charging request probability based on a quotient of the number of charging processes per time interval divided by the number of electric vehicles (50) entering the local area (130- n) in the same time interval.

2. Method according to claim 1, characterized in that the charging request probability is determined for a plurality of time periods at different times of day and / or different days of the week, so that the charging probability for an approaching electric vehicle (50) is determined as a function of the time of day and / or the day of the week.

3. Method according to claim 1 or 2, characterized in that a utilization of the charging point (100-n) is detected and the determined charging request probability is assigned an uncertainty measure which is dependent on the utilization of charging stations (120) of the charging point (100-n) in the time interval.

4. Method according to one of the preceding claims, characterized in that the charging request probability for time periods in which the at least one charging point (100-n) is fully utilized is marked as uncertain.

5. Method according to claim 4, characterized in that the charging request probability is fused depending on the time of day and / or the day of the week, whereby their uncertainty is taken into account.

6. Method according to one of the preceding claims, characterized in that environmental features are recorded for the at least one charging point (100-n), and the charging point (100-n) is classified based on its environmental features.

7. The method according to claim 6, characterized in that the environmental features of the charging point (100-n) include a proximity to a motorway (300), a number of parking spaces (240) located in an environment (150), a number, type and / or size of commercial stores (210), a number, type and / or size of catering establishments (220), a number, type and / or size of leisure facilities (230).

8. Method according to one of claims six or seven, characterized in that the charging request probability is determined for a plurality of charging points (100-n) which are classified on the basis of the environmental features, 9. Method according to claim 8, characterized in that the charging request probabilities for equally classified charging points (100-n) are averaged.

10. A method for predicting the utilization of a plurality of charging points (100-n), wherein for each of the charging points (100-n) the environmental features are recorded; a number of its charging locations (120) is recorded; a classification is carried out based on its recorded environmental features, charging request probability information for this or a similarly classified charging point (100-n) is retrieved from a storage device; its geoposition is recorded and its proximity (130) is determined;a number of electric vehicles (50) entering its immediate vicinity (130) per time interval is detected and / or measured and multiplied by the retrieved charging request probability and divided by a number of its charging stations (120), and the utilization value thus determined is provided, wherein for at least one of the charging points (100-2) the retrieved charging request probability was not determined for this corresponding charging point (100-2) itself;

Images