METHOD AND SYSTEM FOR DETERMINING EFFECTIVE WIND VERSIONS FOR MOTOR VEHICLES

Description

[0001] The invention relates to a method for predicting wind speeds for motor vehicles. Electric vehicles (EVs) have a limited range, making energy consumption estimation or range prediction very important for the driver. For an accurate range prediction, knowledge of all factors relevant to energy consumption is necessary. Due to aerodynamic drag, wind speed influences the drive energy requirement and must be taken into account, especially for long-distance journeys.

[0002] Literature sources indicate the influence of wind speed at up to 16.4% (Source: Viehl, A., Cakar E., Engler M., Köhler. S: Weather data in the range forecast for electric vehicles, ATZ 05 / 2016, pp. 28-34). The energy consumption forecast requires detailed and current data from weather services with high resolution (e.g., wind speed every 5 km). The weather data (wind speed and direction) are made available to the vehicles via a backend connection. The wind speed acting on the vehicle's approach surface is determined based on the vehicle's direction of travel.

[0003] The wind speed acting on an electric vehicle can only be inadequately determined from precise weather data. Weather data provide the average wind speed and direction for a specific area, which offers a good forecast, especially for flat terrain. Topography (valleys, mountains) significantly changes wind speed and direction. Microclimatic effects (e.g., evaporation) also influence wind speed and are difficult to forecast. In addition, buildings and vegetation (e.g., forests) along the route have a significant impact on the effective wind speed acting on the vehicle. In addition, roads are often embanked on both sides for noise protection reasons, so that forecasts from weather services are not accurate for the conditions on the road.

[0004] DE 102012001649 A1 discloses a method and device for determining the speed of the crosswind acting on a motor vehicle using in-vehicle sensors. DE 102014211273 A1 discloses a vehicle mass estimation method.

[0005] US 2013 / 319096 A1 discloses a method for determining the speed of a wind flowing toward a vehicle based on a coolant mass flow. WO 2014 / 037986 A1 discloses the determination of air resistance from the difference between a driving force and driving resistance.

[0006] Against this background, the forecast of the effective wind speed acting on the vehicles should be improved.

[0007] The object is achieved by a method according to the independent claims. Advantageous further developments are specified in the dependent claims. The invention relates to a method for forecasting wind speeds, comprising the steps: Determining an effective wind speed to which a motor vehicle is exposed during a journey, based on an effective drive power determined during the journey, for a plurality of vehicles; transmitting the respectively determined effective wind speed (W) via a wireless communication connection (41) from the plurality of vehicles to a backend (40); collecting data on the effective wind speeds, which were determined based on drive powers recorded by the plurality of vehicles, in the backend and storing this data in a database; and evaluating the stored data on the effective wind speeds, which were determined based on drive powers recorded by a plurality of vehicles, with the aid of statistical methods and / or artificial intelligence methods, wherein the evaluation of the collected data on wind speeds comprises a forecast of future wind speeds..

[0008] The invention further comprises a corresponding method for forecasting wind speeds in a backend, wherein data on effective wind speeds transmitted from a plurality of vehicles via a wireless communication connection are received in a backend, collected there, and stored in a database. The effective wind speeds are determined based on the effective propulsion power of the vehicle determined during the journey. The evaluation is carried out according to the method described above.

[0009] The method is preferably used in electric vehicles, i.e. vehicles that have an electric drive motor. In electric vehicles, the effective mechanical drive power P mcan be precisely determined by measuring the current and voltage at the electric motor, provided the efficiency map of the electric motor and the losses in the vehicle's drivetrain (e.g., in the transmission, etc.) are known. This is usually the case for series-produced vehicles.

[0010] Since the drive train with an electric drive motor usually does not have to be interrupted by a multi-speed transmission, there is the further advantage that the drive power of the electric motor can generally be determined continuously and can be constantly recorded by on-board sensors and control devices while driving.

[0011] In general, the effective mechanical drive power P m the necessary power to overcome the driving resistance: P m = F a + F l + F st + F r * v Driving resistances include acceleration resistance F a(depending on vehicle mass, mass factor and acceleration), the gradient resistance First (depending on vehicle mass and road gradient) and the rolling resistance F r (depending on vehicle mass and rolling resistance coefficient) as well as the air resistance F l = 1 2 * ρ * c w * A * v − W 2 .

[0012] The air resistance F l depends on the vehicle characteristic cw * A, the vehicle speed v and the effective wind speed W A is the effective vehicle cross-sectional area (perpendicular to the direction of travel). ρ is the air density.

[0013] In a preferred embodiment of the method according to the invention, the wind speed effectively acting on the vehicle W out the effective air resistance F l The air resistance is calculated by subtracting the effective drive power P m known and / or estimated driving resistances are deducted.

[0014] The effective wind speed acting on the vehicle W , which can be determined using the method described above, refers in particular to the wind in the direction of travel or speed. Wind components transverse (i.e., perpendicular) to the direction of travel, so-called crosswinds, are not included.

[0015] Known driving resistances are those mentioned above, whose input variables are fixed and known or measurable. Some of the input variables mentioned are variable and cannot be easily measured, such as the vehicle mass, which changes depending on the payload and the number of passengers. All non-measurable vehicle parameters can be estimated using known methods. For this purpose, the vehicle mass can be estimated using a known mass estimator, the rolling resistance coefficient, and the vehicle characteristic value (which changes, for example, with open windows or roof luggage).

[0016] Suitable estimation methods for the individual parameters include filter methods such as Kalman filters and least-squares estimation methods. The input variables for the estimation methods are the measured driving dynamics variables (speed, acceleration), the vehicle parameters, and estimated values ​​from past points in time. The corresponding driving resistances can then be estimated from the estimated vehicle parameters. The estimation methods also provide information about the quality of the estimated value.

[0017] To calculate air resistance, data from internal vehicle sensors, vehicle environmental sensors, vehicle position data, data from digital maps, infrastructure data, data received from other vehicles and / or data received from a backend are preferably used.

[0018] If precise gradient data is available, for example, from digital maps combined with vehicle position data, all input parameters for the driving resistance are known, except for wind speed. This allows the wind speed to be calculated or estimated. A suitable filtering approach (2nd B, Kalman filter, or least-squares estimator) is also suitable for this.

[0019] The other estimators (mass, rolling resistance coefficient, and vehicle characteristic) do not change as dynamically as the wind speed, so the wind speed estimator's adjustment speed is faster than that of the other filters. The individual filters can be combined with each other, so that, for example, if the estimated mass value changes, the quality of the determined wind speed is reduced and possibly marked as invalid.

[0020] According to an advantageous variant of the method according to the invention, a quality factor of the determined air resistance or the determined effective wind speed is additionally determined based on the data used. The quality factor indicates how reliable the determined value is.

[0021] According to a further preferred embodiment, overtaking maneuvers and / or vehicles driving ahead of the motor vehicle are detected by means of environmental sensors, vehicle-internal sensors and / or vehicle-to-vehicle communication.

[0022] For example, radar sensors can be used to determine whether the vehicle is traveling behind another vehicle, particularly a truck. Overtaking maneuvers can also be detected using in-vehicle sensors. Certain steering movements, indicator activation, and acceleration values, for example, can indicate an overtaking maneuver. The quality of the wind speed determination during overtaking maneuvers and when following behind trucks is preferably reduced, as the determined values ​​for the effective wind speed are of limited use in these driving situations.

[0023] The method according to the invention is used to determine the effective wind speed (on the road) acting on a motor vehicle, preferably on a location-based basis. This means that the method according to the invention determines the effective wind speed at a specific location. Since the method is preferably carried out continuously while driving, the wind speed and its changes along a traveled route can be determined. In doing so, the influence of the wind speed that is relevant for the energy consumption forecast is determined. The effective wind speed therefore relates to the respective wind component in the direction of travel or against the direction of travel. A separate evaluation of the wind direction is not absolutely necessary, since the effective wind speed is georeferenced and applies to a specific road section.

[0024] Crosswinds, i.e., wind components perpendicular to the vehicle's direction of travel, can preferably also be detected using appropriate sensors on the vehicle, such as ultrasonic sensors. Steering signals, such as those from a steering angle sensor, can also be used to detect crosswinds, for example, when strong crosswinds require constant counter-steering.

[0025] The actual calculation of the effective wind speed W can be carried out by a computing unit in the vehicle based on the data determined, estimated, or received in the vehicle. Advantageously, the effective wind speeds determined in the vehicle according to the method according to the invention are transmitted from the vehicle to a backend via a wireless communication connection. Preferably, the location of the vehicle and / or a quality and / or time information are also transmitted to the backend. This means that the determined effective wind speeds are georeferenced and sent to the backend with a timestamp. The backend can be formed, for example, by a central server, but also by a cloud.

[0026] According to an alternative variant, only the specific vehicle data required to calculate the effective wind speeds (sensor data, drive power, geoposition) are sent from the vehicle to the backend. The actual calculation of the effective wind speed can then also be performed by a computing unit in the backend. Any necessary estimation procedures can also be executed in the backend.

[0027] According to the invention, data on wind speeds, which were determined taking into account the propulsion power collected by a plurality of vehicles during the journey, are collected and stored in a database. According to the invention, the wind speed data correspond to the effective wind speeds calculated according to the method described above. In addition, data on crosswinds recorded by suitable sensors on the vehicle can also be collected and stored. Just like a determined effective wind speed, a recorded crosswind speed with geoposition and timestamp can be sent from the vehicle to the backend.

[0028] In the backend, the collected wind speed values, determined based on the propulsion power recorded by a number of vehicles, are evaluated using statistical methods and artificial intelligence techniques. The effective wind speeds are preferably treated separately by direction of travel.

[0029] This allows for the determination of average values ​​or a most probable value for effective wind speed and crosswind speed at a specific location. These average values ​​or most probable values ​​can be further validated using data from weather services. In addition to the average values ​​and most probable values, locations with strong winds, especially those with increased crosswinds, can also be identified.

[0030] In addition, the evaluation of wind speeds can also be a basis for a forecast of future wind speeds.

[0031] Another example, which does not fall within the scope of the appended claims, is a database containing wind speed data determined, collected, and / or evaluated using a variant of the method according to the invention. The database is preferably implemented in the backend and preferably contains georeferenced wind speed values.

[0032] In addition to the collected and stored wind speed values, the database can also contain average values ​​or most probable values ​​for wind speeds at a specific location, obtained by evaluating data from multiple networked vehicles. These are preferably values ​​for the effective wind speed (in the direction of travel or opposite), determined based on the vehicle's propulsion power. Additionally or alternatively, georeferenced values ​​for crosswind intensity can also be collected in the database.

[0033] The database can also be implemented in the form of a digital map, with the collected and / or evaluated wind speed values ​​georeferenced to locations in a road network. Such a digital map can be accessed via wireless communication, for example, by navigation devices in vehicles.

[0034] Accordingly, the information stored in the database or in the digital map can be made available to networked vehicles via a wireless communication connection for various services, such as range estimation and energy consumption forecasting of electric vehicles.

[0035] Information on crosswind occurrences (particularly in locations with increased crosswind occurrences) stored in the database or the digital map can advantageously be made available to all networked vehicles (including conventional, combustion engine-powered vehicles), for example, to issue warnings to the driver or for the parameterization of a lane keeping assistance system or to be taken into account in the lane planning of autonomous or semi-autonomous vehicles.

[0036] A system for determining wind speeds to which a motor vehicle is exposed during a journey comprises, for example, a sensor arrangement for recording environmental data, vehicle data, and / or crosswind occurrence; a communication device for receiving position data, infrastructure data, data from a digital map and / or data transmitted from other vehicles or from a backend; a communication device for transmitting wind speeds calculated in the vehicle or data recorded in the motor vehicle for calculating a wind speed to a backend; and a computing unit for calculating an effective wind speed based on the data recorded and / or received in the motor vehicle, wherein the computing unit can be arranged in the motor vehicle or in the backend.

[0037] The system for determining wind speeds to which a motor vehicle is exposed during a journey is preferably designed to carry out the method according to the invention for determining an effective wind speed.

[0038] The invention enables the determination of the wind speed effectively acting on the vehicles, independent of the average values ​​of the weather forecasts. This information about the wind speeds effectively acting on the vehicles can be collected and evaluated in a cloud or in a backend. The data thus collected and processed can then be made available to other vehicles. Thus, the invention is suitable for enabling improved energy consumption and range forecasts for electric or hybrid vehicles. Furthermore, the information about wind speeds and, in particular, crosswinds can be transmitted to driver assistance systems such as for exampleLane keeping assistants or autonomous and semi-autonomous vehicles, thus increasing driving safety.

[0039] In principle, the method according to the invention for determining the effective wind speed is also conceivable for vehicles powered by internal combustion engines or hybrid vehicles. However, its application is particularly advantageous for electric vehicles, as the effective drive power can be determined particularly accurately. The data obtained can be used beneficially for all types of motor vehicles.

[0040] In the following, the invention will be described with reference to Figures 1 and 2 will be explained further by way of example.

[0041] They show schematically: Figure 1: an overview of the individual method steps for determining the effective wind speed according to a preferred embodiment of the invention; and Figure 2: an overview of a preferred application of the wind speeds determined using the method according to the invention and the components involved.

[0042] Figure 1 shows an overview of the process steps for determining an effective wind speed using the example of an electric vehicle.

[0043] In step 10, the vehicle's drive power is determined based on signals relating to the vehicle's longitudinal dynamics. In particular, the drive power provided by the electric motor is determined based on the current and voltage in the electric motor. This power is preferably corrected for the power loss in the drivetrain.

[0044] In addition to the vehicle's drive power, the driving resistances must be determined or estimated (step 20). For this purpose, in step 11, the vehicle mass is estimated using a mass estimator based on signals from the longitudinal and lateral dynamics, speed, and acceleration values. From these signals, rolling resistance can also be estimated in step 12, and a vehicle characteristic value can be estimated in step 13. In addition, in step 14, overtaking maneuvers can be detected based on signals from the steering, vehicle pedals, turn signals, speed, and acceleration values.

[0045] The determined or estimated values ​​for vehicle mass, rolling resistance, and vehicle characteristic are used to calculate the driving resistances in step 20. Since the values ​​estimated in steps 11 to 13 are subject to a certain degree of uncertainty, the driving resistances calculated from them also contain a certain degree of error or exhibit a certain degree of quality 21.

[0046] Based on the drive power determined in step 10 and the driving resistances calculated in step 20, the effective wind speed W in the direction of travel or in the opposite direction can now be estimated in step 30. Depending on the quality of the input data and driving resistances 21, the estimate of the effective wind speed 30 also has a certain quality 31. Information from a radar sensor, whether a vehicle ahead (e.g. a truck) has been detected, is also included in the quality of the effective wind speed. Any overtaking maneuvers detected in step 14 also affect the quality 31 of the estimated wind speed and reduce it. In addition, crosswinds perpendicular to the direction of travel can be detected in step 32 using ultrasonic sensors on the vehicle or steering signals.

[0047] The values ​​thus determined for the effective wind speed and the quality of the determined value as well as any crosswinds detected are then sent together with a geoposition (e.g. determined via GPS or another navigation satellite system) and a time indication / time stamp via wireless communication 41 to a backend 40.

[0048] Figure 2 shows the data collection of effective wind speeds (and crosswind speeds, if applicable) by a number of networked vehicles 50.

[0049] The wind speeds determined using the method described above are sent to the backend 40 via a wireless communication connection 41. There, the data is stored in a database 42 along with the associated location data. A computing unit in the backend 40 evaluates the received and stored data using statistical or artificial intelligence methods. In particular, average and most probable values ​​can be determined, and forecasts of future wind speeds can be created. The evaluated wind speeds can be validated using weather data 43 received from a weather service.

[0050] The wind speeds thus determined and verified (effective wind and crosswind) are then entered into a digital map 44 and can thus be assigned to a road network.

[0051] If new wind speed data from connected vehicles are continuously sent to the backend, the database and especially the digital map can be updated continuously or regularly.

[0052] The location-based information on wind speeds thus obtained and stored in the digital map can then be made available to other networked vehicles 50. These vehicles can send a query to the backend or a digital service via wireless communication 41, wherein the query preferably contains location or route information. Based on the received location-based query, the backend can then send location-based information on the wind speeds stored in the digital map to the networked vehicle 50.

[0053] In the connected vehicles 50, the information thus received about wind speeds along the route can be used for various functions. Examples of functions include driver assistance systems, lane departure warnings, driver warnings regarding wind speeds, energy consumption forecasts, range forecasts, and others.

[0054] In the context of the present invention, GPS is intended to represent all global navigation satellite systems (GNSS), such as GPS, Galileo, GLONASS, Compass, or IRNSS. The vehicle's position can also be determined via cell positioning. This is particularly useful when using GSM, UMTS, or 4G networks.

Claims

1. Method for forecasting wind speeds, comprising the following steps: - determining (30) an effective wind speed (W) to which a respective vehicle (50) is exposed while driving, for a plurality of vehicles; - transmitting the respectively determined effective wind speed (W) from the plurality of vehicles to a backend (40) via a wireless communication connection (41), - collecting data in relation to the effective wind speeds (W) that have been determined on the basis of drive powers acquired from the plurality of vehicles in the backend (40) and storing them in a database, characterized in that the effective wind speed (W) is determined in each case on the basis of an active drive power (Pm) of the vehicle (50) as determined while driving, and by the following step: - evaluating the stored data in relation to the effective wind speeds with the aid of statistical methods and / or using artificial intelligence methods, wherein evaluating the collected data in relation to wind speeds comprises forecasting future wind speeds.

2. Method for forecasting wind speeds, comprising the following steps: - receiving data in relation to effective wind speeds (W) to which a respective vehicle (50) is exposed while driving, for a plurality of vehicles, in a backend (40); wherein the data in relation to effective wind speeds (W) are transmitted from the plurality of vehicles to the backend (40) via a wireless communication connection (41); - collecting the data in relation to the effective wind speeds (W) that have been determined on the basis of drive powers acquired from the plurality of vehicles in the backend (40) and storing them in a database, characterized in that the effective wind speeds (W) are determined in each case on the basis of an active drive power (Pm) of the vehicle (50) as determined while driving; and by the following step: - evaluating the stored data in relation to the effective wind speeds with the aid of statistical methods and / or using artificial intelligence methods, wherein evaluating the collected data in relation to wind speeds comprises forecasting future wind speeds.

3. Method according to Claim 1 or 2, wherein a vehicle (50) from the plurality of vehicles in each case has an electric drive motor, and the active mechanical drive power (Pm) while driving is determined by measuring current and voltage on the electric drive motor driving the vehicle.

4. Method according to one of the preceding claims, wherein the wind speed (W) effectively acting on a vehicle from the plurality of vehicles is calculated from an air resistance, which is calculated (20) by subtracting known and / or estimated driving resistances from the active drive power (10).

5. Method according to Claim 4, wherein data from a sensor system internal to the vehicle, surroundings sensor system of the vehicle, position data of the vehicle, data from digital maps, infrastructure data, data received from other vehicles and / or data received from a backend are used to calculate the air resistance (20).

6. Method according to one of the preceding claims, wherein overtaking manoeuvres and / or vehicles driving in front are detected (14) by way of a surroundings sensor system, a sensor system internal to the vehicle and / or vehicle-to-vehicle communication.

7. Method according to one of the preceding claims, wherein a quality (31) of the determined air resistance (21) or of the determined effective wind speed (W) is determined on the basis of the data that are used.