Electric vehicle energy consumption prediction method and system

Abstract

The invention provides an electric vehicle energy consumption prediction method and system. The method comprises the following steps: dividing a planned travel of a target vehicle into a plurality of planned travel segments, determining the type of each planned travel segment, matching the type of each planned travel segment with a reference travel segment library, and determining a characteristic value of a driving condition of each planned travel segment; according to the characteristic value of the driving condition of each planned travel segment, obtaining a predicted driving condition of the planned travel; selecting a target vehicle energy consumption model corresponding to the vehicle type of the target vehicle from a vehicle model library; and inputting the predicted driving condition of the planned travel and the total mass of the target vehicle into the target vehicle energy consumption model to obtain predicted energy consumption output by the target vehicle energy consumption model. The method can improve the prediction accuracy of the endurance mileage of the vehicle, and can assist in optimizing a motorcade distribution plan, a charging plan, a motorcade deployment plan and a charging and battery swap station deployment plan of a logistics enterprise.

Description

Technical Field

[0001] The present application relates to the technical field of data analysis and processing, and specifically to a method and system for predicting energy consumption of electric vehicles. Background Art

[0002] Promoting the electrification of passenger and commercial vehicles is a key measure for reducing carbon emissions in the transportation sector. Electric commercial vehicles, especially those used in logistics and transportation, have short ranges and require frequent charging or battery swapping. Therefore, the deployment of electric commercial vehicles, charging and battery swapping plans, and the deployment of charging and battery swapping stations all rely on accurate assessments of the vehicles' payload capacity and range.

[0003] A prior art big data collection and processing system is used to estimate the range of electric vehicles. The system integrates on-board sensor information and real-time road, traffic, weather and other information to estimate the future driving status and range of electric vehicles. Specifically, this prior art updates the original vehicle dynamics model and battery model based on a linear regression model, which cannot accurately fit the actual nonlinear system. In addition, the big data processing system provided by this prior art requires real-time collection and upload of a large amount of on-board data from the vehicle's controller area network (CAN) bus, which is difficult for most fleet management companies to achieve. In addition, this prior art also does not consider the impact of the vehicle's cargo weight on the range.

[0004] Therefore, there is an urgent need for an electric vehicle energy consumption prediction method that can improve the accuracy of electric vehicle energy consumption prediction to solve the problem of vehicle electrification in logistics companies. Summary of the Invention

[0005] At least one embodiment of the present application provides a method and system for predicting energy consumption of electric vehicles, which can improve the accuracy of energy consumption prediction of electric vehicles and solve the problem of vehicle electrification in logistics companies.

[0006] According to the first aspect of the present application, at least one embodiment provides a method for predicting energy consumption of an electric vehicle, comprising:

[0007] The target vehicle's planned trip is divided into multiple planned trip segments, the type of each planned trip segment is determined, and the segments are matched with the reference trip segment library to determine the characteristic value of the driving condition of each planned trip segment; based on the characteristic value of the driving condition of each planned trip segment, a predicted driving condition of the planned trip is obtained; wherein the reference trip segment library stores characteristic values of driving conditions of different types of reference trip segments, and the driving condition includes at least one of the following parameters: aerodynamic speed v aero squared average speed Travel distance, characteristic acceleration

[0008] Selecting a target vehicle energy consumption model corresponding to the target vehicle model from a vehicle model library, wherein the vehicle model library stores vehicle energy consumption models of different vehicle models, the vehicle energy consumption model inputs being the driving condition and gross vehicle mass of the trip, and outputting the predicted energy consumption of the trip;

[0009] The predicted driving conditions of the planned trip and the total mass of the target vehicle are input into the target vehicle energy consumption model to obtain the predicted energy consumption output by the target vehicle energy consumption model.

[0010] Optionally, the method further includes: establishing a reference itinerary segment library, wherein establishing the reference itinerary segment library specifically includes:

[0011] Collect historical travel data of various vehicle models and establish a travel database for each vehicle model. The travel database includes a vehicle model parameter table and a historical travel table. The historical travel table includes driving data of multiple historical trips. The driving data includes at least one of the following: trip number, vehicle model, vehicle load, gross vehicle weight, driving distance, driving time, battery capacity, battery health SOH, power consumption per kilometer, average speed, average and the average characteristic acceleration;

[0012] Segmenting the historical trips in the trip database into historical trip segments, determining map segments and time attributes associated with the historical trip segments, and determining the types of the historical trip segments based on the map segments and time attributes associated with the historical trip segments;

[0013] Based on the driving data of the historical trips, characteristic values of the driving conditions of each historical driving segment of the historical trips are determined, and the characteristic values of the driving conditions of each historical driving segment of the same type are averaged to obtain the characteristic values of the driving conditions of a reference driving segment of that type; and / or, characteristic values of the driving conditions of each type of standard driving segment in an external standard driving condition library are obtained as the characteristic values of the driving conditions of the reference driving segment of that type.

[0014] Optionally, determine the type of each planned trip segment, including:

[0015] For each planned trip segment, a map segment and a time attribute associated with the planned trip segment are determined, and a type of the planned trip segment is determined based on the map segment and the time attribute associated with the planned trip segment.

[0016] Optionally, the reference trip segment library is matched to determine the characteristic values of the driving conditions of each planned trip segment, including:

[0017] For each planned trip segment, a reference trip segment matching the type of the planned trip segment is searched in the reference trip segment library, and the characteristic value of the driving condition of the found reference trip segment is used as the characteristic value of the driving condition of the planned trip segment.

[0018] Optionally, also include:

[0019] Based on historical travel data of the same vehicle model, a vehicle energy consumption model is trained for each vehicle model. The vehicle energy consumption model includes: a first functional relationship between the total energy consumption of the vehicle in a travel segment and first, second, and third types of energy; and a second functional relationship between the battery energy consumed or compensated by vehicle travel in each travel segment and the vehicle model parameters and driving conditions. The first type of energy is the sum of the battery energy consumed by vehicle travel in the travel segments of the travel, the second type of energy is the sum of the battery energy compensated by vehicle travel in the travel segments of the travel, and the third type of energy is energy consumption unrelated to vehicle travel. The travel is divided into multiple travel segments with a preset time interval of Δt. The vehicle model parameters include at least one of the following: the vehicle's drag coefficient, aerodynamic frontal area, rolling resistance coefficient, and gross vehicle mass.

[0020] Optionally, a vehicle energy consumption model is trained for each vehicle type based on historical travel data of the same vehicle type, specifically including:

[0021] For each vehicle model, obtain the characteristic values of the driving conditions of the historical travel segments of the vehicle model and the actual energy consumption value of each historical travel segment as training data;

[0022] The training data is input into the vehicle energy consumption model to obtain the predicted energy consumption output by the vehicle energy consumption model, the difference between the predicted energy consumption output by the vehicle energy consumption model and the actual energy consumption value is calculated using a loss function, and a random algorithm or a population evolution algorithm is used to solve the optimal solution of the model parameters of the vehicle energy consumption model that minimizes the difference value to obtain the vehicle energy consumption model.

[0023] Optionally, the driving data further includes temperature; the vehicle energy consumption model of the same vehicle model includes multiple vehicle energy consumption models corresponding to different temperature ranges, wherein the vehicle energy consumption model of a vehicle model corresponding to a temperature range is obtained by training using the training data of the vehicle model in the temperature range;

[0024] A target vehicle energy consumption model corresponding to the model of the target vehicle is selected from a vehicle model library, specifically: determining the temperature range of the planned trip; and selecting a vehicle energy consumption model that matches the temperature range of the planned trip from multiple vehicle energy consumption models corresponding to the model of the target vehicle as the target vehicle energy consumption model.

[0025] Optionally, collect historical travel data for various types of vehicles, including:

[0026] Receive the vehicle's historical travel data collected and reported by the on-board IoT terminal.

[0027] Optionally, also include:

[0028] Calculating the remaining power of the battery based on the nominal power of the battery of the target vehicle, the battery health, the percentage of remaining power of the battery, and the temperature coefficient corresponding to the planned driving time of the planned trip;

[0029] When the remaining power of the battery is greater than or equal to the predicted energy consumption output by the energy consumption model of the target vehicle, determining that the remaining power of the target vehicle can meet the needs of the planned trip;

[0030] When the remaining power of the battery is less than the predicted energy consumption output by the target vehicle energy consumption model, the maximum drivable distance of the target vehicle is calculated based on the predicted power consumption per unit distance of the target vehicle and the remaining power.

[0031] According to a second aspect of the present application, at least one embodiment provides an electric vehicle energy consumption prediction system, comprising:

[0032] The driving condition prediction module is used to divide the planned trip of the target vehicle into multiple planned trip segments, determine the type of each planned trip segment, match it with the reference trip segment library, and determine the characteristic value of the driving condition of each planned trip segment; according to the characteristic value of the driving condition of each planned trip segment, the predicted driving condition of the planned trip is obtained; wherein the reference trip segment library stores the characteristic values of the driving conditions of different types of reference trip segments, and the driving condition includes at least one of the following parameters: aerodynamic speed v aero squared average speed Travel distance, characteristic acceleration

[0033] a model selection module, configured to select a target vehicle energy consumption model corresponding to the target vehicle model from a vehicle model library, wherein the vehicle model library stores vehicle energy consumption models of different vehicle models, wherein the input of the vehicle energy consumption model is the driving condition and gross vehicle mass of the trip, and the output is the predicted energy consumption of the trip;

[0034] The vehicle energy consumption prediction module is used to input the predicted driving conditions of the planned trip and the total mass of the target vehicle into the target vehicle energy consumption model to obtain the predicted energy consumption output by the target vehicle energy consumption model.

[0035] According to the third aspect of the present application, at least one embodiment provides an electric vehicle energy consumption prediction system, comprising a processor, a memory, and a program or instruction stored on the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the method described in any one of the first aspects.

[0036] According to the fourth aspect of the present application, at least one embodiment provides a computer-readable storage medium having a program stored thereon, and when the program is executed by a processor, the steps of any method of the first aspect are implemented.

[0037] According to a fifth aspect of the present application, at least one embodiment provides a computer program product, comprising computer instructions, which, when executed by a processor, implement the steps of the method described in any one of the first aspects.

[0038] Compared to existing technologies, the electric vehicle energy consumption prediction method and system provided in the embodiments of this application can accurately predict the effective load capacity and range of electric vehicles based on dynamic load and vehicle driving conditions, thereby improving the accuracy of vehicle energy consumption prediction. In addition, the embodiments of this application can collect historical vehicle travel data through an onboard IoT terminal without requiring access to the vehicle's CAN system, reducing the complexity of data collection and improving the practicality of the solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present application. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0040] Figure 1 A schematic flow chart of a method for predicting energy consumption of an electric vehicle according to an embodiment of the present application;

[0041] Figure 2 This is an example diagram of the architecture of the electric vehicle energy consumption prediction system according to an embodiment of the present application;

[0042] Figure 3 A module structure diagram of an electric vehicle energy consumption prediction system according to an embodiment of the present application;

[0043] Figure 4 This is an example process of the electric vehicle energy consumption prediction method according to an embodiment of the present application;

[0044] Figure 5 This is an example diagram of a total energy consumption calculation model for a vehicle according to an embodiment of the present application;

[0045] Figure 6 A schematic diagram of the structure of an electric vehicle energy consumption prediction system according to an embodiment of the present application;

[0046] Figure 7 This is another structural diagram of the electric vehicle energy consumption prediction system according to an embodiment of the present application. DETAILED DESCRIPTION

[0047] The following describes exemplary embodiments of the present application in more detail with reference to the accompanying drawings. Although exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0048] It should be understood that references throughout this specification to "one embodiment" or "an embodiment" mean that a particular feature, structure, or characteristic associated with the embodiment is included in at least one embodiment of the present application. Therefore, the appearance of "in one embodiment" or "in an embodiment" throughout this specification does not necessarily refer to the same embodiment. Furthermore, these particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. The terms "first," "second," and so on, used in the specification and claims of this application are used to distinguish similar items and are not necessarily used to describe a particular order or sequential sequence. It should be understood that such usage is interchangeable where appropriate, such that the embodiments of the present application described herein can, for example, be implemented in an order other than that illustrated or described herein. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, product, or apparatus. The term "and / or" used in the specification and claims refers to at least one of the connected items.

[0049] In the various embodiments of the present application, it should be understood that the size of the serial numbers of the following processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0050] The following description provides examples and does not limit the scope, applicability, or configuration set forth in the claims. Changes may be made to the function and arrangement of the elements discussed without departing from the spirit and scope of this disclosure. The various examples may appropriately omit, substitute, or add various procedures or components. For example, the described methods may be performed in an order different from that described, and various steps may be added, omitted, or combined. In addition, features described with reference to certain examples may be combined in other examples.

[0051] The electric vehicle energy consumption prediction method and electric vehicle energy consumption prediction system provided in the embodiments of the present application collect historical travel data of vehicles of different models in advance, establish a travel database for each model based on the historical travel data, and then establish a reference travel segment library. Here, the reference travel segment library stores characteristic values of driving conditions of different types of reference travel segments, and the driving conditions include at least one of the following parameters: aerodynamic speed v aero squared average speed Travel distance, characteristic acceleration

[0052] In addition, the present embodiment also establishes a vehicle model library that includes vehicle energy consumption models for various vehicle types. Thus, based on the reference trip segment library and the vehicle model library, energy consumption can be predicted for the planned trip of a target vehicle (referred to as the target vehicle for ease of description).

[0053] The establishment of the reference itinerary fragment library is described below.

[0054] The electric vehicle energy consumption prediction system of the present application embodiment can collect historical travel data for various vehicle models. For example, the vehicle can collect and report historical travel data, and the electric vehicle energy consumption prediction system receives the historical travel data reported by the vehicle-mounted Internet of Things terminal. In this way, by collecting and reporting historical travel data through the vehicle-mounted Internet of Things terminal, the vehicle-mounted Internet of Things terminal does not need to be connected to the vehicle's CAN bus, which can reduce the complexity of data collection and improve the practicality of the solution.

[0055] In this way, the electric vehicle energy consumption prediction system can establish a travel database for each vehicle model based on the received historical travel data of various vehicle models, wherein the travel database includes a vehicle model parameter table and a historical travel table, wherein the historical travel table includes driving data of multiple historical trips, wherein the driving data includes at least one of the following: trip number (unique identifier of the trip), vehicle model, vehicle load (i.e., the weight or mass of the load carried by the vehicle), gross vehicle weight (i.e., the total weight or total mass of the vehicle), driving distance (total distance of the trip), driving time (total driving time of the trip, which may include the start time, end time and total driving time), battery capacity, battery health (State Of Health, SOH), power consumption per kilometer, average speed, square of aerodynamic speed The mean (average Here, v aero represents the aerodynamic speed) and the average characteristic acceleration.

[0056] Then, the historical trips in the trip database are segmented into historical trip segments. Specifically, the historical trips can be segmented into multiple historical trip segments, each of which is sequentially connected and non-overlapping. These segments are sequentially combined to form the historical trip. Specifically, the duration of each historical trip segment can be a preset time interval. If the duration of a historical trip is not an integer multiple of the time interval, the remaining portion of the trip that is less than one time interval can be ignored through approximation.

[0057] After segmenting the historical travel segments, the map segments and time attributes associated with the historical travel segments are determined, and the type of the historical travel segments is determined based on the map segments and time attributes associated with the historical travel segments. Here, the map segments are segments defined in the map that correspond to actual roads, and the time attributes refer to the attributes of the travel time of the travel segments. The time attributes can be pre-defined based on the time interval in which the travel time of the travel segments falls. Generally, multiple time attributes can be defined based on the characteristics of traffic conditions in different time intervals. For example, multiple time attributes can be pre-defined, including morning rush hour time intervals, evening rush hour time intervals, off-peak time intervals, holiday time intervals, rainy and snowy weather periods, etc.

[0058] Then, based on the driving data of the historical trip, the characteristic value of the driving condition of each historical driving segment of the historical trip is determined, and then the characteristic values of the driving conditions of each historical driving segment of the same type are averaged to obtain the characteristic value of the driving condition of the reference driving segment of the same type. The driving condition of each historical driving segment also includes the aerodynamic speed v aero squared average speed Travel distance, characteristic acceleration

[0059] Among them, the aerodynamic speed v of a certain travel segment aero squared The calculation of the characteristic value of can refer to the following formula 10. It can be seen that the aerodynamic speed v aero squared is the ratio of the first product to the travel distance of the trip segment, wherein the first product is a cubic function of the speed of the trip segment The duration of the trip segment (Δt j,j+1 ), the speed cubic function of the travel segment is the first speed at the starting time point of the travel segment (such as v j ) and the second speed of the end time period (such as v j+1 ), and the specific calculation method can refer to Formula 5. For example, the speed cubic function of the travel segment is the average of four products, and the four products are: the cube of the first speed, the product of the square of the first speed and the second speed, the product of the first speed and the square of the second speed, and the cube of the second speed.

[0060] Characteristic acceleration of a travel segment The calculation of the characteristic value can refer to the following formula 11. It can be seen that the characteristic acceleration is the sum of the first value and the second value, where the first value is The second value is Here, v j 、v j+1 are the vehicle speeds at the start and end time points of the trip segment, Δd j,j+1 is the distance traveled by the vehicle in the trip segment, h j 、h j+1 Respectively represent the height of the vehicle at the start time point and the end time point of the trip segment.

[0061] Through the above method, the characteristic values of the driving conditions of various types of reference travel segments can be obtained, thereby obtaining the reference travel segment library.

[0062] Of course, embodiments of the present application may also obtain characteristic values of driving conditions for each type of standard trip segment from an external standard driving condition library and use them as characteristic values of driving conditions for reference trip segments of that type, thereby generating the reference trip segment library. For example, when the amount of historical trip data is relatively small, embodiments of the present application may generate the reference trip segment library based on the standard driving condition library.

[0063] The vehicle energy consumption model and its training are explained below.

[0064] In an embodiment of the present application, a vehicle energy consumption model is pre-trained for each vehicle model based on historical travel data for the same vehicle model. The vehicle energy consumption model includes: a first functional relationship between the vehicle's total energy consumption during a trip and first, second, and third types of energy; and a second functional relationship between the battery energy consumed or compensated by vehicle travel in each trip segment and the vehicle model parameters and driving conditions. The first type of energy is the sum of the battery energy consumed by vehicle travel in the trip segments of the trip, the second type of energy is the sum of the battery energy compensated by vehicle travel in the trip segments of the trip, and the third type of energy is energy consumption unrelated to vehicle travel. The trip is divided into multiple trip segments with a preset time interval Δt. The vehicle model parameters include at least one of the following: the vehicle's drag coefficient, aerodynamic frontal area, rolling resistance coefficient, and gross vehicle mass. Specifically, the model parameters of the vehicle energy consumption model include: coefficients corresponding to the first, second, and third types of energy, respectively; and coefficients corresponding to the vehicle model parameters and / or driving conditions.

[0065] When training a vehicle energy consumption model for each vehicle model based on historical travel data for the same vehicle model, embodiments of the present application can obtain, for each vehicle model, characteristic values of the driving conditions for each historical travel segment of the vehicle model, as well as the actual energy consumption value for each historical travel segment, as training data. This training data is then input into the vehicle energy consumption model to obtain the predicted energy consumption output by the vehicle energy consumption model. A loss function is used to calculate the difference between the predicted energy consumption output by the vehicle energy consumption model and the actual energy consumption value. A randomized algorithm or a population evolutionary algorithm is then used to determine the optimal solution for the model parameters of the vehicle energy consumption model that minimizes this difference, thereby obtaining the vehicle energy consumption model.

[0066] Please refer to Figure 1 The electric vehicle energy consumption prediction method provided in the embodiment of the present application includes the following steps:

[0067] Step 11: Divide the planned trip of the target vehicle into multiple planned trip segments, determine the type of each planned trip segment, and match it with the reference trip segment library to determine the characteristic value of the driving condition of each planned trip segment; obtain the predicted driving condition of the planned trip based on the characteristic value of the driving condition of each planned trip segment; wherein the reference trip segment library stores the characteristic values of the driving conditions of different types of reference trip segments, and the driving condition includes at least one of the following parameters: aerodynamic speed v aero squared average speed Travel distance, characteristic acceleration

[0068] Here, first, the planned itinerary of the target vehicle is obtained. The planned itinerary is the route that the target vehicle will travel within the planned time. For example, a path planning algorithm can be used to formulate the route that the target vehicle will travel within a predetermined time period, thereby obtaining the above-mentioned planned itinerary. In an embodiment of the present application, the planned itinerary can be divided into a plurality of planned itinerary segments. These planned itinerary segments are sequentially connected and do not overlap with each other. Combining them together in order constitutes the planned itinerary. Specifically, the duration of each planned itinerary segment can be a preset time interval. When the entire travel time of the planned itinerary is not an integer multiple of the time interval, the remaining travel portion that is less than one time interval can be ignored through approximate processing.

[0069] After segmenting the planned journey segments, the embodiment of the present application also determines the map segments and time attributes associated with each planned journey segment. Here, the map segments are segments defined in various maps that correspond to actual roads, and the time attributes refer to the attributes of the travel time of the journey segments. The time attributes can be pre-defined based on the characteristics of the time interval and / or traffic conditions. For example, the same map segment has different traffic conditions under different time attributes. The embodiment of the present application can pre-define multiple time attributes, for example, including: morning rush hour time interval, evening rush hour time interval, off-peak time interval, holiday time interval, rainy and snowy weather period, etc.

[0070] Then, based on the time interval in which the travel time of the planned trip segment falls, the time attribute associated with the planned trip segment is determined, and then based on the map segment and time attribute associated with the planned trip segment, the type of the planned trip segment is determined. For example, a time attribute can correspond to a type of trip segment. For another example, a map segment can correspond to a type of trip segment. For another example, each combination of a map segment and a time attribute corresponds to a type of trip segment. In this way, for each planned trip segment, a reference trip segment that matches the type of the planned trip segment can be searched in the reference trip segment library, and the characteristic value of the driving condition of the reference trip segment found can be used as the characteristic value of the driving condition of the planned trip segment.

[0071] After obtaining the characteristic values of the driving conditions of each planned trip segment, a predicted driving condition of the planned trip may be generated based on the characteristic values of the driving conditions of each planned trip segment.

[0072] Step 12: Select a target vehicle energy consumption model corresponding to the target vehicle model from a vehicle model library, wherein the vehicle model library stores vehicle energy consumption models of different models, the input of the vehicle energy consumption model is the driving conditions and vehicle load data of the trip, and the output is the predicted energy consumption of the trip.

[0073] Here, according to the model of the target vehicle, the target vehicle energy consumption model corresponding to the model is selected for subsequent energy consumption prediction. The vehicle load data can be the total weight of the vehicle.

[0074] Step 13: Input the predicted driving conditions of the planned trip and the load data of the target vehicle into the target vehicle energy consumption model to obtain the predicted energy consumption output by the target vehicle energy consumption model.

[0075] Here, the load data of the target vehicle may be data such as the total mass, load weight, etc. The target vehicle energy consumption model predicts the energy consumption required for the target vehicle to execute the planned trip based on the input predicted driving conditions of the planned trip and the load data of the target vehicle.

[0076] Through the above steps, the embodiment of the present application generates a predicted driving condition for a planned trip based on historical trip segments, and then uses a pre-trained vehicle energy consumption model to predict energy consumption. In the above method, the driving condition information obtained from the historical trip is fully utilized for prediction, which can improve the accuracy of the vehicle energy consumption prediction. In addition, the vehicle energy consumption model of the embodiment of the present application is generated based on data such as the total mass of the vehicle. The input parameters of the model include adjustable vehicle load data. In addition, the predicted driving condition can be a driving condition predicted based on a selected time period and route. Therefore, the embodiment of the present application can accurately predict the effective load capacity and cruising range of electric vehicles based on dynamic loads and predicted driving conditions of planned trips, thereby improving the accuracy of vehicle cruising range prediction and assisting in optimizing logistics companies' fleet distribution plans, charging plans, fleet deployment plans, and charging and swapping station deployment plans.

[0077] During model training, the embodiments of the present application utilize a randomized algorithm or a population evolution algorithm to find the optimal solution, thereby improving the performance of the trained model. Furthermore, the embodiments of the present application utilize an onboard IoT terminal to collect historical vehicle travel data, eliminating the need for access to the vehicle's CAN system. This reduces the complexity of data collection and improves the practicality of the solution.

[0078] After step 13, the embodiment of the present application may further calculate the remaining battery capacity of the target vehicle based on the nominal battery capacity, battery health, percentage of remaining battery capacity, and the temperature coefficient corresponding to the planned driving time of the planned trip. Then, based on the relationship between the remaining battery capacity and the predicted energy consumption output by the target vehicle energy consumption model, it is determined whether the target vehicle can complete the planned trip.

[0079] For example, when the remaining battery power is greater than or equal to the predicted energy consumption output by the target vehicle energy consumption model, it is determined that the remaining power of the target vehicle can meet the needs of the planned trip. If the remaining battery power is less than the predicted energy consumption output by the target vehicle energy consumption model, it is determined that the remaining power of the target vehicle cannot meet the needs of the planned trip. In this case, the maximum drivable distance of the target vehicle can be calculated based on the predicted power consumption per unit distance of the target vehicle and the remaining power. The predicted power consumption per unit distance can be an average value obtained based on the historical distance and historical energy consumption statistics of the target vehicle, and this embodiment of the application does not specifically limit this.

[0080] Taking into account the important influence of air temperature on the working efficiency of vehicle batteries, when establishing a vehicle energy consumption model, the embodiment of the present application can also establish multiple vehicle energy consumption models corresponding to different temperature ranges for vehicles of the same model. At this time, when collecting the driving data of the historical trip, the air temperature corresponding to the historical trip is also collected. The vehicle energy consumption model corresponding to a certain temperature range of a certain model is obtained by training using the training data of the model in the temperature range. In this way, in step 12, when selecting the target vehicle energy consumption model, the embodiment of the present application can first determine the temperature range of the planned trip, and then, from the multiple vehicle energy consumption models corresponding to the model of the target vehicle, select the vehicle energy consumption model that matches the temperature range of the planned trip as the target vehicle energy consumption model.

[0081] The following further illustrates the electric vehicle energy consumption prediction method and system according to the embodiment of the present application through more specific examples.

[0082] Figure 2 This is an example diagram of the architecture of the electric vehicle energy consumption prediction system provided in an embodiment of the present application. Figure 3 This is the module structure diagram of the electric vehicle energy consumption prediction system. Figures 2 and 3 As shown in the figure, the electric vehicle energy consumption prediction system consists of a general Internet of Things (IoT) terminal (herein referred to as the acquisition terminal or IoT terminal) and a cloud platform. The acquisition terminal collects vehicle travel data, such as location, speed, acceleration, remaining battery power, and payload weight. The cloud platform receives and stores this data from the acquisition terminal and stores it in a historical travel database. The vehicle energy consumption model parameter estimation module, driving condition prediction module, and vehicle carrying capacity prediction module run on the cloud platform and are responsible for evaluating the vehicle's carrying capacity and range based on the input parameters.

[0083] This example also provides a range prediction algorithm that takes load and driving conditions into account. The algorithm calculates and optimizes the model parameters of the vehicle energy consumption model for each vehicle model based on historical travel data. The model parameters include the load of the adjustable vehicle. The driving condition prediction module can predict driving conditions based on a planned or selected time period and route. After selecting the model parameters of the vehicle model, inputting the vehicle's dynamic load and predicted driving conditions, the vehicle energy consumption prediction module can calculate the predicted energy consumption, thereby determining the maximum achievable mileage under the dynamic load. This example improves the accuracy of vehicle range prediction and can be used to optimize logistics companies' fleet distribution plans, charging plans, fleet deployment plans, and charging and swapping station deployment plans.

[0084] The electric vehicle energy consumption prediction system of this example can predict the maximum mileage of the vehicle based on the dynamic load of the electric vehicle, the ambient temperature (air temperature) and the driving road conditions.

[0085] The inputs of the system include vehicle position, speed, optional acceleration, remaining power (remaining power at the start of the trip and remaining power at the end of the trip), weight, and dynamic load from a general IoT terminal; battery SOC sequence from an optional on-board terminal (the current SOC is collected once at each time interval); temperature forecast and optional road information (such as road construction conditions, congestion status, etc.) from an external data interface; vehicle model and parameters (including basic vehicle parameters such as drag coefficient, frontal area, rolling resistance, etc.), battery model and parameters (including battery nominal capacity, weight, etc.) from an external interface; vehicle delivery time period, delivery route, and trip dynamic load from an external interface; and battery SOH and SOC of the current vehicle from an external interface.

[0086] The system's outputs include a range energy consumption forecast. If the target vehicle's battery's nominal charge, current SOC, and SOH are known, the system can also output a predicted range under a specified load.

[0087] The IoT data acquisition module collects driving data from a common onboard IoT terminal, including GPS location, speed, optional acceleration, dynamic load, mileage, remaining battery life at the start and end of a trip, time, and temperature. The optional onboard terminal can be directly connected to the vehicle bus to collect real-time timing data such as battery and motor data.

[0088] The data collected by the IoT terminal is stored in the historical travel database after necessary data processing. The stored content includes vehicle type, driving conditions, load, power consumption, route, driving distance, travel time, etc.

[0089] The vehicle energy consumption model parameter estimation module establishes a dynamic energy consumption model for each vehicle model, uses the data in the historical travel database to optimize the model parameters, and stores the optimized model parameters in the vehicle model database.

[0090] The driving condition prediction module generates a predicted driving condition based on the historical driving database and external input information including driving route, driving time, and road information.

[0091] The vehicle energy consumption prediction module selects the corresponding vehicle energy consumption model based on external input vehicle model information, temperature forecast, etc., and then calculates the power consumption of the planned trip based on the selected vehicle energy consumption model and the predicted driving conditions and dynamic load, thereby predicting the maximum mileage of the electric vehicle under the dynamic load, driving route, and driving time period.

[0092] The system's output of predicted energy consumption and drivable mileage can be fed back to the human-machine interface or the Transportation Management System (TMS) through an external interface, thereby guiding fleet planning, vehicle dispatch plan management, and charging and battery replacement management.

[0093] Figure 4 An example process of an electric vehicle energy consumption prediction method is provided, including:

[0094] 1. Establish a preliminary vehicle energy consumption model based on the vehicle's nominal parameters.

[0095] 2. Process the travel data collected by the IoT terminal and generate a historical travel database.

[0096] 3. Use the historical travel database and optimization algorithm to optimize the parameters of the estimated vehicle energy consumption model. Because temperature has a significant impact, this example can also optimize the vehicle energy consumption model for different temperature ranges.

[0097] 4. Predict driving conditions based on historical travel data and future driving routes, time periods, etc.

[0098] 5. Select the corresponding vehicle energy consumption model according to the vehicle model and temperature.

[0099] 6. Input the predicted driving conditions and load into the vehicle energy consumption prediction module to obtain the energy consumption prediction value.

[0100] 7. Predict the maximum mileage based on the vehicle's remaining power.

[0101] The following explains them separately.

[0102] 1. Vehicle energy consumption model modeling steps

[0103] The total energy consumption calculation model of the vehicle in this example is as follows Figure 5 As shown, road energy consumption is mainly composed of wind resistance, rolling resistance, uphill and downhill work, and acceleration and deceleration work. Inputting the speed sequence (speed is collected once at each time interval), acceleration sequence (acceleration is collected once at each time interval or calculated based on speed difference), height difference sequence (road height difference is collected once at each time interval or calculated based on road inclination), and actual load into the road energy consumption demand model can calculate the required electric vehicle motor work value or energy feedback value. Inputting the motor work demand or motor energy feedback demand into the power system model can calculate the required output energy or input energy. Inputting the output value of the power system model into the battery model to obtain the total energy consumption E FUEL In order to simplify the calculation, this example combines the power system model and the battery model into an average transmission efficiency and average energy feedback efficiency Because the operating sequence is derived from actual operating data, scenarios where the power demand exceeds the maximum motor power and the maximum battery power are not considered to simplify the calculation. The battery's state of health (SOH) and state of charge (SOC) are primarily considered when calculating the battery's total energy consumption and remaining charge.

[0104] 2. Establish a preliminary vehicle energy consumption model based on the vehicle’s nominal parameters.

[0105]

[0106] First, establish the vehicle's instantaneous power demand model as shown in Formula 1. Where P t is the power requirement of the electric vehicle. m is the total mass of the vehicle, v is the instantaneous speed of the vehicle, θ is the angle between the road surface and the horizontal direction, ρ is the density of the air, and C D is the vehicle's drag coefficient, A f is the aerodynamic frontal area, C RR is the rolling resistance coefficient, and g is the acceleration due to gravity.

[0107]

[0108]

[0109]

[0110]

[0111]

[0112] Δt j,j+1 =t j+1 -t j (6)

[0113] Δh j,j+i =hj+1 -h j (7)

[0114] Then calculate the vehicle's j,j+1 Energy demand E for the time period Δt,j,j+1 , as shown in Formula 2. To simplify the calculation, it is assumed here that Δt j,j+1 The speed changes linearly within the time period, as determined by v j The uniform speed change is v j+1 , then the average speed is The acceleration is Then calculate to get formula 3. Where Δh j,j+1 is Δt j,j+1 The height difference between the vehicle climbing or descending during a time period. Here, Δt j,j+1 Indicates time point t j At time t j+1 time period, Δt represents the time period Δt j,j+1 duration;v j 、v j+1 Represents time t j , t j+1 Vehicle speed at time h j 、h j+1 Represents time point t j , t j+1 The height of the vehicle at the time (such as altitude, etc.).

[0115]

[0116] C rolling =C RR g (9)

[0117]

[0118]

[0119]

[0120]

[0121] Then define C aero As shown in formula 8, C rolling As shown in formula 9. is Δt j,j+1 The square of the aerodynamic speed during the time period, is Δt j,j+1 Characteristic acceleration during the time period. aero , C rolling Only related to the parameters of the vehicle, Only relevant to driving conditions. Δd j,j+1is Δt j,j+1 The distance traveled by the vehicle in the time period. Substituting the above formula into formula 3 yields formula 13. From formula 13, we can see that the car j,j+1 The relationship between energy consumption within a time period and vehicle parameters, weight, and driving conditions.

[0122]

[0123]

[0124]

[0125] Since the direction of energy demand is different when the vehicle accelerates or decelerates, when E Δt,j,j+1 >0, energy is input by the vehicle, when E Δt,j,j+1 When <0, the energy is consumed by the braking system or returned to the power battery through the energy feedback system. Formula 14 is all E Δt,j,j+1 >0, the energy accumulation sum, formula 15 is all E Δt,j,j+1 <0. Formula 16 is the total energy consumption after considering the vehicle conversion efficiency. It is the average efficiency of the power system, which mainly includes the efficiency product of the electric motor, gearbox, and wheel drive shaft. is the average efficiency of energy feedback, mainly considering the energy feedback strength, motor efficiency, battery charging efficiency, etc. other Other energy consumption of the vehicle, such as the air conditioning system.

[0126]

[0127]

[0128] The vehicle-related parameters in formula 16, such as drag coefficient, rolling resistance coefficient, frontal area, etc., can be determined based on the vehicle's nominal values or empirical parameters. It can be filled in as the maximum value, such as 0.95. other You can fill in the value by multiplying the air conditioning power by the total time or an empirical value. Finally, add the coefficient X = [x1, x2, x3, x4, x5, x6] used for adjustment and optimization to Formula 13 and Formula 16 to obtain Formula 17 and Formula 18.

[0129] 3. Collect travel data (IoT data collection module)

[0130] Input data, such as the operating conditions of trip 1 (Trip1) to trip n, may include time, vehicle speed, height difference (altitude difference), load, trip number, etc. Each operating condition is shown in Table 1.

[0131] Table 1

[0132] time Speed height difference load Trip number …

[0133] In this way, by summarizing each trip, the trip summary information shown in Table 2 can be obtained.

[0134] Table 2

[0135] Trip number Model load Power consumption distance … Trip1 No.1 1000 10kwh 30km … … … … … … Trip1 No.n 1500 20kwh 40km

[0136] To accurately calculate model parameters, we first collect operating data for each trip and store it in a historical trip database. This data includes time, speed, optional elevation difference, and trip number. Because speed is approximated to increase or decrease linearly within each time interval, it's best to collect data at least once per second to ensure accurate calculations. For each trip, we also record the power consumption and payload. Finally, we aggregate the data for n trips, recording the vehicle type, payload, power consumption, and distance for each trip.

[0137] 4. Processing travel data (data preprocessing module)

[0138] One form of the itinerary database is shown in Table 3 and Table 4.

[0139] Table 3

[0140]

[0141] Table 4

[0142]

[0143]

[0144] After collecting the data of each trip, the data needs to be processed. In the above two trip databases, Table 3 shows the trip summary table, and Table 4 shows the vehicle model table. The trip summary table contains trip number, vehicle model, load, total weight, driving distance, driving time, battery capacity, battery SOH, power consumption per kilometer, average speed, average and the average characteristic acceleration The vehicle model table contains relatively fixed parameters of a specific vehicle model, including vehicle model, battery model, empty weight, battery capacity, motor power, motor efficiency curve, battery charging power, energy feedback efficiency, energy feedback level, battery weight, C aero , C rolling and other information.

[0145] Then calculate the driving condition matrix, including the square value matrix of aerodynamic speed Characteristic acceleration matrix Velocity Matrix Distance matrix ΔD. Each column in these matrices represents a trip, assuming there are m trips in total, and each row represents a trip segment, for example, corresponding to Δt j,j+1 Δt can be 1 second or other duration. Taking formula 19 as an example, represents the journey i from time point t j At time t j+1 The square of the aerodynamic speed of the travel segment. For Equation 20, represents the journey i from time point t j At time t j+1 The characteristic acceleration of . For Formula 21, represents the journey i from time point t j At time t j+1 The average speed of the travel segment. For Equation 22, Δd i,j,j+1 represents the journey i from time point t j At time t j+1 The driving distance of the trip segment.

[0146]

[0147]

[0148]

[0149]

[0150] 5. Optimize and estimate vehicle energy consumption model parameters (vehicle energy consumption model parameter estimation module)

[0151]

[0152]

[0153] In formula 23, f cost (X) is the loss function, where X is the vector [x1,x2,x3,x4,x5,x6]. The optimization objective is as shown in Formula 24, and we find the value that makes f cost (X) Minimum X value. Due to the influence of energy feedback braking, f cost (X) is quite special. It is not a first-order or second-order smooth function and has multiple local extreme points. Conventional direct search methods or gradient descent methods are difficult to find the optimal solution. In this example, a randomized algorithm or a population evolution algorithm can effectively converge to the optimal solution. The following example uses a population evolution algorithm as an example.

[0154] 1) First, generate a uniformly distributed hyperplane set U of m samples.

[0155] U=[X1…X m ]

[0156] 2) Calculate the f corresponding to each sample X separately cost (X), a single sample is calculated as follows:

[0157] The sample X i Substitute into formula 17, and then according to formulas 19, 20, 21, 22 and vehicle parameter C aero , C rolling Calculate the energy consumption demand matrix E ROAD , each column represents a trip, for example, m trips, and each row represents a time segment, for example, 1 second:

[0158]

[0159] Then for E ROAD Apply Equation 14 and Equation 15 to each column of the matrix to obtain E ROAD,POS and E ROAD,NEG .

[0160] E ROAD,POS =[E road,pos,1 …E road,pos,m ] (26)

[0161] For Formula 26, E road，pos，i Represents the total energy consumption value of the i-th trip.

[0162] E ROAD，NEG =[E road，neg，1 …E road，neg，m ] (27)

[0163] For Formula 27, E road，neg，i Represents the total energy feedback value of the i-th trip.

[0164] Substitute Formula 26 and Formula 27 into Formula 18 to obtain E FUEL :

[0165] E FUEL =[E fuel，1 …E fuel，m ] (28)

[0166] For Formula 28, E fuel，i represents the total battery energy consumption value of the i-th trip. Then substitute formula 28 into formula 23 to obtain f cost (X).

[0167] 3) Select the first N smallest f cost (X) The corresponding X set is used as the parent sample. [X1, X2…X N ]

[0168] 4) Randomly select another sample from each parent sample to pair with, such as [X i , X j ].

[0169] 5) Cross each sample pair in the parent sample pair set, and randomly select the crossover position. After the crossover is completed, a sub-sample set is generated. The sample pair crossover method is as follows: Sample pair X i =[x i1 , x i2 , x i3 , x i4 , x i5 , x i6 ] X j =[x j1 , x j2 , x j3 , x j4 , x j5 , x j6 If the randomly chosen intersection point selects the first two terms of the first vector and the remaining terms come from the second vector, then the subvector after the intersection is X ij =[x i1 , x i2 , x j3 , x j4 , x j5 , x j6 ].

[0170] 6) Mutate each sub-vector in the sub-vector set, for example: X ij =X ij *σλ where λ = [λ1, λ2, λ3, λ4, λ5, λ6] is a vector of normal distribution and σ takes a value between 0 and 1.

[0171] 7) Return to step 2 for the subvector set and calculate the f corresponding to each sample X cost (X), and then continue the above process K times, or until (X) < ε, ε is a very small value.

[0172] 8) Sort the final set, and then test the first m samples using the validation set to find the optimal X value.

[0173] 6. Driving condition prediction (driving condition prediction module)

[0174] 1) Split the trips in the historical trip database into sub-trip segments and match them with the actual road segments on the map. Then classify them according to time types, such as morning rush hour, evening rush hour, off-peak hours, holidays, rainy and snowy weather, etc. Calculate the feature values of each sub-trip segment separately, such as These feature values are classified according to different time periods and the mean is calculated.

[0175] 2) The planned travel route is also divided into sub-trip segments, and matched with the sub-trip segments of the same type in the historical travel database, and the feature values of each sub-trip are matched according to the planned travel time of each sub-trip.

[0176] 3) The driving condition characteristic values in the standard driving condition library To match the characteristic values of each of the above sub-trips, select the matching operating condition fragments in the standard driving condition library to fill each of the above sub-trips.

[0177] 4) Connect each sub-trip to generate a complete predicted driving condition.

[0178] 7. Predicting power consumption (vehicle energy consumption prediction module)

[0179] Select the corresponding model from the model library based on the vehicle model and temperature forecast. Input the predicted driving conditions and vehicle load into the model. Calculate ΔD and Each line is the calculated value of a time interval, usually 1 second. According to formula 14, formula 15, formula 17, and formula 18, the predicted power consumption E is calculated. fuel,predict .

[0180]

[0181]

[0182]

[0183]

[0184] 8. Predict maximum mileage (vehicle energy consumption prediction module)

[0185] The maximum driving range is calculated based on the calculated power consumption and the remaining battery power.

[0186] The calculation method is as follows:

[0187] E remain =SOC*SOH*T*ESS (29)

[0188] Among them E remain is the remaining capacity of the battery, SOC is the current charge percentage of the battery, SOH is the health of the battery, T is the temperature coefficient, and ESS is the nominal capacity of the battery.

[0189] If E predict >E remain , then the power cannot meet the planned route, E predict <Eremain ,The remaining power can meet the planned route.,The maximum drivable distance can also be calculated based on the,predicted power consumption per unit distance.

[0190] Please refer to Figure 6 , an embodiment of the present application provides a structure of an electric vehicle energy consumption prediction system, including:

[0191] The driving condition prediction module 601 is used to divide the planned trip of the target vehicle into multiple planned trip segments, determine the type of each planned trip segment, match it with the reference trip segment library, and determine the characteristic value of the driving condition of each planned trip segment; according to the characteristic value of the driving condition of each planned trip segment, the predicted driving condition of the planned trip is obtained; wherein the reference trip segment library stores the characteristic values of the driving conditions of different types of reference trip segments, and the driving condition includes at least one of the following parameters: aerodynamic speed v aero squared average speed Travel distance, characteristic acceleration

[0192] A model selection module 602 is configured to select a target vehicle energy consumption model corresponding to the target vehicle model from a vehicle model library, wherein the vehicle model library stores vehicle energy consumption models of different vehicle models, wherein the input of the vehicle energy consumption model is the driving condition and gross vehicle mass of the trip, and the output is the predicted energy consumption of the trip;

[0193] The vehicle energy consumption prediction module 603 is used to input the predicted driving conditions of the planned trip and the total mass of the target vehicle into the target vehicle energy consumption model to obtain the predicted energy consumption output by the target vehicle energy consumption model.

[0194] Through the above modules, the embodiments of the present application can improve the accuracy of energy consumption prediction of electric vehicles and solve the problem of vehicle electrification of logistics companies.

[0195] Optionally, the above system further includes:

[0196] The first establishing module is used to establish a reference itinerary segment library. The establishment of the reference itinerary segment library specifically includes:

[0197] Collect historical travel data of various vehicle models and establish a travel database for each vehicle model. The travel database includes a vehicle model parameter table and a historical travel table. The historical travel table includes driving data of multiple historical trips. The driving data includes at least one of the following: trip number, vehicle model, vehicle load, gross vehicle weight, driving distance, driving time, battery capacity, battery health SOH, power consumption per kilometer, average speed, average and the average characteristic acceleration;

[0198] Segmenting the historical trips in the trip database into historical trip segments, determining map segments and time attributes associated with the historical trip segments, and determining the types of the historical trip segments based on the map segments and time attributes associated with the historical trip segments;

[0199] Based on the driving data of the historical trips, characteristic values of the driving conditions of each historical driving segment of the historical trips are determined, and the characteristic values of the driving conditions of each historical driving segment of the same type are averaged to obtain the characteristic values of the driving conditions of a reference driving segment of that type; and / or, characteristic values of the driving conditions of each type of standard driving segment in an external standard driving condition library are obtained as the characteristic values of the driving conditions of the reference driving segment of that type.

[0200] Optionally, the driving condition prediction module 601 is further configured to:

[0201] For each planned trip segment, a map segment and a time attribute associated with the planned trip segment are determined, and a type of the planned trip segment is determined based on the map segment and the time attribute associated with the planned trip segment.

[0202] Optionally, the driving condition prediction module 601 is further configured to:

[0203] For each planned trip segment, a reference trip segment matching the type of the planned trip segment is searched in the reference trip segment library, and the characteristic value of the driving condition of the found reference trip segment is used as the characteristic value of the driving condition of the planned trip segment.

[0204] Optionally, the system further includes:

[0205] A training module is configured to train a vehicle energy consumption model for each vehicle model based on historical travel data of the same vehicle model, the vehicle energy consumption model comprising: a first functional relationship between the total energy consumption of the vehicle in a travel segment and first, second, and third types of energy; and a second functional relationship between the battery energy consumed or compensated by vehicle travel in each travel segment and the vehicle model parameters and driving conditions; wherein the first type of energy is the sum of the battery energy consumed by vehicle travel in the travel segments of the travel segment, the second type of energy is the sum of the battery energy compensated by vehicle travel in the travel segments of the travel segment, and the third type of energy is energy consumption unrelated to vehicle travel; the travel segment is divided into a plurality of travel segments with a preset time interval Δt; and the vehicle model parameters comprise at least one of the following: the vehicle's drag coefficient, aerodynamic frontal area, rolling resistance coefficient, and gross vehicle mass.

[0206] Optionally, the training module is further used to:

[0207] For each vehicle model, obtain the characteristic values of the driving conditions of the historical travel segments of the vehicle model and the actual energy consumption value of each historical travel segment as training data;

[0208] The training data is input into the vehicle energy consumption model to obtain the predicted energy consumption output by the vehicle energy consumption model, the difference between the predicted energy consumption output by the vehicle energy consumption model and the actual energy consumption value is calculated using a loss function, and a random algorithm or a population evolution algorithm is used to solve the optimal solution of the model parameters of the vehicle energy consumption model that minimizes the difference value to obtain the vehicle energy consumption model.

[0209] Optionally, the driving data further includes temperature; the vehicle energy consumption model of the same vehicle model includes multiple vehicle energy consumption models corresponding to different temperature ranges, wherein the vehicle energy consumption model of a vehicle model corresponding to a temperature range is obtained by training using the training data of the vehicle model in the temperature range;

[0210] The model selection module 602 is further used to: determine the temperature range of the planned trip; and select a vehicle energy consumption model that matches the temperature range of the planned trip from multiple vehicle energy consumption models corresponding to the model of the target vehicle as the target vehicle energy consumption model.

[0211] Optionally, the first establishing module is further used to receive historical travel data of the vehicle collected and reported by the on-board Internet of Things terminal.

[0212] Optionally, the vehicle energy consumption prediction module is further used to:

[0213] Calculating the remaining power of the battery based on the nominal power of the battery of the target vehicle, the battery health, the percentage of remaining power of the battery, and the temperature coefficient corresponding to the planned driving time of the planned trip;

[0214] When the remaining power of the battery is greater than or equal to the predicted energy consumption output by the energy consumption model of the target vehicle, determining that the remaining power of the target vehicle can meet the needs of the planned trip;

[0215] When the remaining power of the battery is less than the predicted energy consumption output by the target vehicle energy consumption model, the maximum drivable distance of the target vehicle is calculated based on the predicted power consumption per unit distance of the target vehicle and the remaining power.

[0216] It should be noted that the various systems provided in the above embodiments are devices corresponding to the above-mentioned electric vehicle energy consumption prediction method. The implementation methods in the above-mentioned embodiments are applicable to the embodiments of the device and can achieve the same technical effects. The above-mentioned device provided in the embodiments of the present application can implement all the method steps implemented in the above-mentioned method embodiments and can achieve the same technical effects. The parts and beneficial effects of this embodiment that are the same as those in the method embodiments will not be detailed here.

[0217] Please refer to Figure 7 , a structural diagram of another electric vehicle energy consumption prediction system provided in an embodiment of the present application, the device includes: a processor 701, a transceiver 702, a memory 703, a user interface 704 and a bus interface.

[0218] In the embodiment of the present application, the device further includes: a program stored in the memory 703 and executable on the processor 701 .

[0219] The transceiver 702 is configured to send and receive data under the control of the processor;

[0220] The processor 701 is configured to read the computer program in the memory and perform the following operations:

[0221] The target vehicle's planned trip is divided into multiple planned trip segments, the type of each planned trip segment is determined, and the segments are matched with the reference trip segment library to determine the characteristic value of the driving condition of each planned trip segment; based on the characteristic value of the driving condition of each planned trip segment, a predicted driving condition of the planned trip is obtained; wherein the reference trip segment library stores characteristic values of driving conditions of different types of reference trip segments, and the driving condition includes at least one of the following parameters: aerodynamic speed v aero squared average speed Travel distance, characteristic acceleration

[0222] Selecting a target vehicle energy consumption model corresponding to the target vehicle model from a vehicle model library, wherein the vehicle model library stores vehicle energy consumption models of different vehicle models, the vehicle energy consumption model inputs being the driving condition and gross vehicle mass of the trip, and outputting the predicted energy consumption of the trip;

[0223] The predicted driving conditions of the planned trip and the total mass of the target vehicle are input into the target vehicle energy consumption model to obtain the predicted energy consumption output by the target vehicle energy consumption model.

[0224] Optionally, the processor is further configured to: establish a reference itinerary segment library, wherein establishing the reference itinerary segment library specifically includes:

[0225] Collect historical travel data of various vehicle models and establish a travel database for each vehicle model. The travel database includes a vehicle model parameter table and a historical travel table. The historical travel table includes driving data of multiple historical trips. The driving data includes at least one of the following: trip number, vehicle model, vehicle load, gross vehicle weight, driving distance, driving time, battery capacity, battery health SOH, power consumption per kilometer, average speed, average and the average characteristic acceleration;

[0226] Segmenting the historical trips in the trip database into historical trip segments, determining map segments and time attributes associated with the historical trip segments, and determining the types of the historical trip segments based on the map segments and time attributes associated with the historical trip segments;

[0227] Based on the driving data of the historical trips, characteristic values of the driving conditions of each historical driving segment of the historical trips are determined, and the characteristic values of the driving conditions of each historical driving segment of the same type are averaged to obtain the characteristic values of the driving conditions of a reference driving segment of that type; and / or, characteristic values of the driving conditions of each type of standard driving segment in an external standard driving condition library are obtained as the characteristic values of the driving conditions of the reference driving segment of that type.

[0228] Optionally, the processor is further configured to:

[0229] For each planned trip segment, a map segment and a time attribute associated with the planned trip segment are determined, and a type of the planned trip segment is determined based on the map segment and the time attribute associated with the planned trip segment.

[0230] Optionally, the processor is further used to: for each planned trip segment, search the reference trip segment library for a reference trip segment that matches the type of the planned trip segment, and use the characteristic value of the driving condition of the found reference trip segment as the characteristic value of the driving condition of the planned trip segment.

[0231] Optionally, the processor is further used to: train a vehicle energy consumption model for each vehicle model based on historical travel data of the same vehicle model, the vehicle energy consumption model including: a first functional relationship between the total energy consumption of the vehicle in a journey and the first type of energy, the second type of energy, and the third type of energy; a second functional relationship between the battery energy consumed or compensated by the vehicle in each journey segment and the vehicle model parameters and driving conditions; wherein, the first type of energy is the sum of the battery energy consumed by the vehicle in the journey segments of the journey, the second type of energy is the sum of the battery energy compensated by the vehicle in the journey segments of the journey, and the third type of energy is energy consumption unrelated to the vehicle's travel; the journey is divided into multiple journey segments with a preset time interval of Δt; the vehicle model parameters include at least one of the following: the vehicle's drag coefficient, aerodynamic frontal area, rolling resistance coefficient, and total vehicle mass.

[0232] Optionally, the processor is further configured to: for each vehicle model, obtain, as training data, characteristic values of driving conditions of historical travel segments of the vehicle model and actual energy consumption values of each historical travel segment;

[0233] The training data is input into the vehicle energy consumption model to obtain the predicted energy consumption output by the vehicle energy consumption model, the difference between the predicted energy consumption output by the vehicle energy consumption model and the actual energy consumption value is calculated using a loss function, and a random algorithm or a population evolution algorithm is used to solve the optimal solution of the model parameters of the vehicle energy consumption model that minimizes the difference value to obtain the vehicle energy consumption model.

[0234] Optionally, the driving data further includes temperature; the vehicle energy consumption model of the same vehicle model includes multiple vehicle energy consumption models corresponding to different temperature ranges, wherein the vehicle energy consumption model of a vehicle model corresponding to a temperature range is obtained by training using the training data of the vehicle model in the temperature range;

[0235] Optionally, the processor is further used to: determine the temperature range of the planned trip; and select a vehicle energy consumption model that matches the temperature range of the planned trip from multiple vehicle energy consumption models corresponding to the model of the target vehicle as the target vehicle energy consumption model.

[0236] Optionally, the processor is also used to receive historical travel data of the vehicle collected and reported by the on-board Internet of Things terminal.

[0237] Optionally, the processor is further configured to calculate the remaining power of the battery according to the nominal power of the battery of the target vehicle, the battery health, the percentage of remaining power of the battery, and the temperature coefficient corresponding to the planned driving time of the planned trip;

[0238] When the remaining power of the battery is greater than or equal to the predicted energy consumption output by the energy consumption model of the target vehicle, determining that the remaining power of the target vehicle can meet the needs of the planned trip;

[0239] When the remaining power of the battery is less than the predicted energy consumption output by the target vehicle energy consumption model, the maximum drivable distance of the target vehicle is calculated based on the predicted power consumption per unit distance of the target vehicle and the remaining power.

[0240] It is understandable that in the embodiment of the present application, when the computer program is executed by the processor 701, each process of the above-mentioned electric vehicle energy consumption prediction method embodiment can be implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0241] exist Figure 7 In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically linking together various circuits of one or more processors represented by processor 701 and memory represented by memory 703. The bus architecture may also link together various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. The bus interface provides an interface. The transceiver 702 may be a plurality of components, i.e., a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium. For different user devices, the user interface 704 may also be an interface capable of connecting external or internal devices as required, and the connected devices include but are not limited to a keypad, a display, a speaker, a microphone, a joystick, etc.

[0242] The processor 701 is responsible for managing the bus architecture and general processing, and the memory 703 can store data used by the processor 701 when performing operations.

[0243] It should be noted that the device in this embodiment is a device corresponding to the above-mentioned electric vehicle energy consumption prediction method. The implementation methods in the above-mentioned embodiments are all applicable to the embodiments of this device and can achieve the same technical effects. In this device, the transceiver 702 and the memory 703, as well as the transceiver 702 and the processor 701 can be connected through a bus interface for communication. The functions of the processor 701 can also be implemented by the transceiver 702, and the functions of the transceiver 702 can also be implemented by the processor 701. It should be noted that the above-mentioned device provided in the embodiment of the present application can implement all the method steps implemented in the above-mentioned method embodiment and can achieve the same technical effects. The parts and beneficial effects that are the same as those in the method embodiment will not be specifically described here.

[0244] In some embodiments of the present application, a computer-readable storage medium is further provided, on which a program is stored. When the program is executed by a processor, the following steps are implemented:

[0245] The target vehicle's planned trip is divided into multiple planned trip segments, the type of each planned trip segment is determined, and the segments are matched with the reference trip segment library to determine the characteristic value of the driving condition of each planned trip segment; based on the characteristic value of the driving condition of each planned trip segment, a predicted driving condition of the planned trip is obtained; wherein the reference trip segment library stores characteristic values of driving conditions of different types of reference trip segments, and the driving condition includes at least one of the following parameters: aerodynamic speed v aero squared average speed Travel distance, characteristic acceleration

[0246] Selecting a target vehicle energy consumption model corresponding to the target vehicle model from a vehicle model library, wherein the vehicle model library stores vehicle energy consumption models of different vehicle models, the vehicle energy consumption model inputs being the driving condition and gross vehicle mass of the trip, and outputting the predicted energy consumption of the trip;

[0247] The predicted driving conditions of the planned trip and the total mass of the target vehicle are input into the target vehicle energy consumption model to obtain the predicted energy consumption output by the target vehicle energy consumption model.

[0248] When the program is executed by the processor, it can implement all the implementation methods of the above-mentioned electric vehicle energy consumption prediction method and achieve the same technical effect. To avoid repetition, it will not be described here.

[0249] An embodiment of the present application also provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, the various processes of the above-mentioned electric vehicle energy consumption prediction method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, they will not be repeated here.

[0250] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0251] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0252] In the embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0253] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0254] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0255] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

[0256] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A method for predicting energy consumption of electric vehicles, characterized in that: include: The target vehicle's planned trip is divided into multiple planned trip segments, the type of each planned trip segment is determined, and the segments are matched with the reference trip segment library to determine the characteristic value of the driving condition of each planned trip segment; based on the characteristic value of the driving condition of each planned trip segment, a predicted driving condition of the planned trip is obtained; wherein the reference trip segment library stores characteristic values of driving conditions of different types of reference trip segments, and the driving condition includes at least one of the following parameters: aerodynamic speed v aero squared average speed Travel distance, characteristic acceleration Selecting a target vehicle energy consumption model corresponding to the target vehicle model from a vehicle model library, wherein the vehicle model library stores vehicle energy consumption models of different vehicle models, the vehicle energy consumption model inputs being the driving condition and gross vehicle mass of the trip, and outputting the predicted energy consumption of the trip; The predicted driving conditions of the planned trip and the total mass of the target vehicle are input into the target vehicle energy consumption model to obtain the predicted energy consumption output by the target vehicle energy consumption model.

2. The method according to claim 1, wherein The method further includes: establishing a reference itinerary segment library, wherein establishing the reference itinerary segment library specifically includes: Collect historical travel data of various vehicle models and establish a travel database for each vehicle model. The travel database includes a vehicle model parameter table and a historical travel table. The historical travel table includes driving data of multiple historical trips. The driving data includes at least one of the following: trip number, vehicle model, vehicle load, gross vehicle weight, driving distance, driving time, battery capacity, battery health SOH, power consumption per kilometer, average speed, average and the average characteristic acceleration; Segmenting the historical trips in the trip database into historical trip segments, determining map segments and time attributes associated with the historical trip segments, and determining the types of the historical trip segments based on the map segments and time attributes associated with the historical trip segments; Based on the driving data of the historical trips, characteristic values of the driving conditions of each historical driving segment of the historical trips are determined, and the characteristic values of the driving conditions of each historical driving segment of the same type are averaged to obtain the characteristic values of the driving conditions of a reference driving segment of that type; and / or, characteristic values of the driving conditions of each type of standard driving segment in an external standard driving condition library are obtained as the characteristic values of the driving conditions of the reference driving segment of that type.

3. The method according to claim 2, wherein Identify the type of each planned trip segment, including: For each planned trip segment, a map segment and a time attribute associated with the planned trip segment are determined, and a type of the planned trip segment is determined based on the map segment and the time attribute associated with the planned trip segment.

4. The method according to claim 3, wherein Matching with the reference trip segment library to determine the characteristic values of the driving conditions of each planned trip segment, including: For each planned trip segment, a reference trip segment matching the type of the planned trip segment is searched in the reference trip segment library, and the characteristic value of the driving condition of the found reference trip segment is used as the characteristic value of the driving condition of the planned trip segment.

5. The method according to any one of claims 2 to 4, characterized in that Also includes: Based on historical travel data of the same vehicle model, a vehicle energy consumption model is trained for each vehicle model. The vehicle energy consumption model includes: a first functional relationship between the total energy consumption of the vehicle in a travel segment and first, second, and third types of energy; and a second functional relationship between the battery energy consumed or compensated by vehicle travel in each travel segment and the vehicle model parameters and driving conditions. The first type of energy is the sum of the battery energy consumed by vehicle travel in the travel segments of the travel, the second type of energy is the sum of the battery energy compensated by vehicle travel in the travel segments of the travel, and the third type of energy is energy consumption unrelated to vehicle travel. The travel is divided into multiple travel segments with a preset time interval of Δt. The vehicle model parameters include at least one of the following: the vehicle's drag coefficient, aerodynamic frontal area, rolling resistance coefficient, and gross vehicle mass.

6. The method according to claim 5, wherein Based on the historical travel data of the same vehicle model, a vehicle energy consumption model is trained for each vehicle model, including: For each vehicle model, obtain the characteristic values of the driving conditions of the historical travel segments of the vehicle model and the actual energy consumption value of each historical travel segment as training data; The training data is input into the vehicle energy consumption model to obtain the predicted energy consumption output by the vehicle energy consumption model, the difference between the predicted energy consumption output by the vehicle energy consumption model and the actual energy consumption value is calculated using a loss function, and a random algorithm or a population evolution algorithm is used to solve the optimal solution of the model parameters of the vehicle energy consumption model that minimizes the difference value to obtain the vehicle energy consumption model.

7. The method according to any one of claims 2 to 4, characterized in that The driving data also includes temperature; the vehicle energy consumption model of the same vehicle model includes multiple vehicle energy consumption models corresponding to different temperature ranges, wherein the vehicle energy consumption model corresponding to a temperature range of a vehicle model is trained using the training data of the vehicle model in the temperature range; A target vehicle energy consumption model corresponding to the model of the target vehicle is selected from a vehicle model library, specifically: determining the temperature range of the planned trip; and selecting a vehicle energy consumption model that matches the temperature range of the planned trip from multiple vehicle energy consumption models corresponding to the model of the target vehicle as the target vehicle energy consumption model.

8. The method according to any one of claims 2 to 4, characterized in that Collect historical travel data of various types of vehicles, including: Receive the vehicle's historical travel data collected and reported by the on-board IoT terminal.

9. The method according to claim 1, wherein Also includes: Calculating the remaining power of the battery based on the nominal power of the battery of the target vehicle, the battery health, the percentage of remaining power of the battery, and the temperature coefficient corresponding to the planned driving time of the planned trip; When the remaining power of the battery is greater than or equal to the predicted energy consumption output by the energy consumption model of the target vehicle, determining that the remaining power of the target vehicle can meet the needs of the planned trip; When the remaining power of the battery is less than the predicted energy consumption output by the target vehicle energy consumption model, the maximum drivable distance of the target vehicle is calculated based on the predicted power consumption per unit distance of the target vehicle and the remaining power.

10. An electric vehicle energy consumption prediction system, characterized in that: include: The driving condition prediction module is used to divide the planned trip of the target vehicle into multiple planned trip segments, determine the type of each planned trip segment, match it with the reference trip segment library, and determine the characteristic value of the driving condition of each planned trip segment; according to the characteristic value of the driving condition of each planned trip segment, the predicted driving condition of the planned trip is obtained; wherein the reference trip segment library stores the characteristic values of the driving conditions of different types of reference trip segments, and the driving condition includes at least one of the following parameters: aerodynamic speed v aero squared average speed Travel distance, characteristic acceleration a model selection module, configured to select a target vehicle energy consumption model corresponding to the target vehicle model from a vehicle model library, wherein the vehicle model library stores vehicle energy consumption models of different vehicle models, wherein the input of the vehicle energy consumption model is the driving condition and gross vehicle mass of the trip, and the output is the predicted energy consumption of the trip; The vehicle energy consumption prediction module is used to input the predicted driving conditions of the planned trip and the total mass of the target vehicle into the target vehicle energy consumption model to obtain the predicted energy consumption output by the target vehicle energy consumption model.

11. A computer program product, characterized in that The method comprises computer instructions, which, when executed by a processor, implement the steps of the method according to any one of claims 1 to 9.

Images