A method and device for predicting vehicle travel energy consumption

Abstract

The present invention provides a method and device for predicting vehicle travel energy consumption. The method includes: training a vehicle-section energy consumption prediction model for each vehicle on each road section and a vehicle model-section energy consumption prediction model for each vehicle model on each road section according to the travel records of multiple vehicles in a preset area collected in advance; generating a default road section energy consumption prediction model for the vehicle according to the default energy consumption of the vehicle or the vehicle model to which the vehicle belongs; connecting the above models on adjacent road sections to respectively generate vehicle, vehicle model, and default energy consumption prediction grids; based on the above grids, using at least one of the vehicle, vehicle model, and default road section energy consumption prediction models of the target vehicle (preferably the one with a higher priority), predicting the total energy of the target vehicle on the candidate travel path. The present invention comprehensively considers personal driving habits, road condition information, and travel conditions, can improve the accuracy of energy consumption prediction results, is easy to implement in engineering, and has the advantages of a small amount of calculation and a low implementation cost.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicles. Specifically, the present invention relates to a method and device for predicting the energy consumption of a vehicle during a journey. Background Art

[0002] Regarding the prediction of vehicle journey energy consumption in the prior art, the following solutions mainly exist:

[0003] One solution is: using the experimental data collected by the vehicle under standard road conditions to make a comparison table of the state of charge (SOC) and the remaining mileage. When it is necessary to calculate the remaining mileage that the remaining power of the vehicle can travel, the predicted value of the remaining mileage is obtained by looking up the table according to the current SOC value. Another solution is: using the average energy consumption of the vehicle in the past several kilometers (such as 100 kilometers) as the reference energy consumption for predicting the remaining mileage, and calculating the remaining mileage that the remaining power of the vehicle can travel based on this reference energy consumption. For the above two solutions, the implementation difficulty is small and the cost is low. However, since neither of them considers the actual road conditions during vehicle driving, the accuracy of the prediction results is not high.

[0004] There are also some other solutions in the prior art for predicting the vehicle journey energy consumption. By adding more information, such as real-time road conditions, or introducing complex models, such as machine learning methods like neural networks or random forests, for energy consumption prediction. The above solutions require collecting real-time road condition information, and the introduced complex models require the device to have powerful computing capabilities. To support the above solutions, usually more software and hardware costs need to be invested, which is not conducive to engineering implementation. Summary of the Invention

[0005] The technical problem to be solved by the embodiments of the present invention is to provide a method and device for predicting the energy consumption of a vehicle during a journey, improve the accuracy of the energy consumption prediction result, and reduce the implementation cost of the energy consumption prediction.

[0006] To solve the above technical problem, according to one aspect of the present invention, a method for predicting the energy consumption of a vehicle during a journey is provided, including:

[0007] Based on the travel records of multiple vehicles in a preset area collected in advance, a vehicle section energy consumption prediction model for each vehicle in each section and a vehicle model section energy consumption prediction model for each vehicle model in each section are trained, where the travel records include the travel time, travel conditions, vehicle identifier, vehicle model identifier, geographical location and energy consumption data of the vehicle at each sampling time point;

[0008] Generating a default section energy consumption prediction model of the vehicle according to the default energy consumption of the vehicle or the vehicle model to which the vehicle belongs and the length of the section;

[0009] Predict the total energy required for the target vehicle to travel on the candidate travel route by using at least one of the vehicle section energy consumption prediction model, the vehicle model section energy consumption prediction model, and the default section energy consumption prediction model of the target vehicle.

[0010] In addition, according to at least one embodiment of the present invention, before the step of predicting the total energy required for the target vehicle to travel on the candidate travel route, the method further includes:

[0011] For each vehicle, generate the energy consumption grid of the preset area respectively. The energy consumption grid includes the energy consumption prediction model associated with each section in the preset area, where:

[0012] When there is a vehicle section energy consumption prediction model of the vehicle in this section, the energy consumption prediction model associated with this section is the vehicle section energy consumption prediction model of the vehicle in this section;

[0013] When there is no vehicle section energy consumption prediction model of the vehicle in this section, but there is a vehicle model section energy consumption prediction model of the vehicle, the energy consumption prediction model associated with this section is the product of the vehicle model section energy consumption prediction model of the vehicle in this section and the first ratio;

[0014] When there is only a default section energy consumption prediction model of the vehicle, the energy consumption prediction model associated with this section is the product of the default section energy consumption prediction model of the vehicle and the second ratio;

[0015] Wherein, the first ratio is the ratio of the average energy consumption of the vehicle on all sections to the average energy consumption of the vehicles of the vehicle model to which the vehicle belongs on all sections; the second ratio is the ratio of the average energy consumption of the vehicle on all sections to the default energy consumption of the vehicle or the vehicle model to which the vehicle belongs.

[0016] In addition, according to at least one embodiment of the present invention, the step of predicting the total energy required for the target vehicle to travel on the candidate travel route includes:

[0017] According to each section passed by the candidate travel route, search for the energy consumption grid of the target vehicle, obtain the energy consumption prediction model associated with each section of the target vehicle, and use the obtained energy consumption prediction model to calculate and accumulate the energy consumption requirements of each section to obtain the total energy required for the target vehicle to pass through the candidate travel route.

[0018] In addition, according to at least one embodiment of the present invention, the step of training the vehicle section energy consumption prediction model of each vehicle in each section includes:

[0019] Split each trip record into sub-trips of each section of the trip;

[0020] For each sub-trip of each road segment, based on the average value of the energy consumption data at each sampling time point on the road segment during the sub-trip and the length of the road segment, calculate the energy required by the corresponding vehicle on the road segment during the sub-trip, and generate a sub-trip record. The sub-trip record includes travel time, travel conditions, vehicle identification, vehicle type identification, and the energy required by the vehicle during the sub-trip.

[0021] Group the sub-trip records according to the vehicle identification and road segment identification to obtain a plurality of first groups. Each first group includes sub-trip records with the same vehicle identification and road segment identification.

[0022] Respectively use each first group to train a decision tree model to obtain a vehicle-road segment energy consumption prediction model for the corresponding vehicle on the corresponding road segment for each first group.

[0023] In addition, according to at least one embodiment of the present invention, the steps of training a vehicle type-road segment energy consumption prediction model for each vehicle type on each road segment include:

[0024] Group the sub-trip records according to the vehicle type identification and road segment identification to obtain a plurality of second groups. Each second group includes sub-trip records with the same vehicle type identification and road segment identification.

[0025] Respectively use each second group to train a decision tree model to obtain a vehicle type-road segment energy consumption prediction model for the vehicle of the corresponding vehicle type on the corresponding road segment for each first group.

[0026] In addition, according to at least one embodiment of the present invention, the method further includes: calculating the first ratio and the second ratio in the following manner:

[0027] For each vehicle, respectively calculate the first average energy consumption of the vehicle based on all the sub-trip records of the vehicle; calculate the second average energy consumption of the vehicle type to which the vehicle belongs based on all the sub-trip records of the vehicle type to which the vehicle belongs; calculate the ratio of the first average energy consumption to the second average energy consumption to obtain the first ratio.

[0028] And calculate the ratio of the first average energy consumption to the default energy consumption of the vehicle or the vehicle type to which the vehicle belongs to obtain the second ratio.

[0029] In addition, according to at least one embodiment of the present invention, the method further includes:

[0030] According to the total energy required for the target vehicle to travel on at least one candidate travel path predicted based on the energy consumption grid, select a target travel path that meets the preset path selection strategy from the at least one candidate travel path, and prompt the total energy required for the target travel path.

[0031] In addition, according to at least one embodiment of the present invention, the travel conditions include at least one of air temperature and vehicle tire pressure.

[0032] In addition, according to at least one embodiment of the present invention, the road section is a section of road with travel records, and no road fork is found in this road section according to the travel records.

[0033] According to another aspect of the present invention, there is also provided a prediction device for vehicle travel energy consumption, including:

[0034] A first model generation unit, configured to train a vehicle road section energy consumption prediction model for each vehicle on each road section and a vehicle model road section energy consumption prediction model for each vehicle model on each road section according to the travel records of multiple vehicles in a preset area collected in advance, where the travel records include the travel time, travel conditions, vehicle identification, vehicle model identification, geographical location of the vehicle at each sampling time point, and energy consumption data of each trip;

[0035] A second model generation unit, configured to generate a default road section energy consumption prediction model of the vehicle according to the default energy consumption of the vehicle or the vehicle model to which the vehicle belongs and the length of the road section;

[0036] An energy consumption prediction unit, configured to predict the total energy required for the target vehicle to travel on a candidate travel path by using at least one of the vehicle road section energy consumption prediction model, the vehicle model road section energy consumption prediction model, and the default road section energy consumption prediction model of the target vehicle.

[0037] In addition, according to at least one embodiment of the present invention, the prediction device further includes:

[0038] An energy consumption grid generation unit, configured to generate an energy consumption grid of the preset area for each vehicle, where the energy consumption grid includes the energy consumption prediction models associated with each road section in the preset area, where:

[0039] When there is a vehicle road section energy consumption prediction model of the vehicle in this road section, the energy consumption prediction model associated with this road section is the vehicle road section energy consumption prediction model of the vehicle in this road section;

[0040] When there is no vehicle road section energy consumption prediction model of the vehicle in this road section, but there is a vehicle model road section energy consumption prediction model of the vehicle, the energy consumption prediction model associated with this road section is the product of the vehicle model road section energy consumption prediction model of the vehicle in this road section and a first ratio;

[0041] When only the default road section energy consumption prediction model of the vehicle exists, the energy consumption prediction model associated with this road section is the product of the default road section energy consumption prediction model of the vehicle and a second ratio;

[0042] Wherein, the first ratio is the ratio of the average energy consumption of the vehicle on all sections to the average energy consumption of vehicles of the same model as the vehicle on all sections; the second ratio is the ratio of the average energy consumption of the vehicle on all sections to the default energy consumption of the vehicle or the vehicle model.

[0043] In addition, according to at least one embodiment of the present invention, the first model generation unit includes:

[0044] A trip record splitting unit for splitting each trip record into sub-trips of each section of the trip;

[0045] A sub-trip processing unit for, for each sub-trip of each section, calculating the energy required by the corresponding vehicle on the section in the sub-trip according to the average value of the energy consumption data at each sampling time point on the section and the length of the section, and generating a sub-trip record, the sub-trip record including travel time, travel conditions, vehicle identification, vehicle model identification, and the energy required by the vehicle in the sub-trip;

[0046] A first grouping unit for grouping the sub-trip records according to vehicle identification and section identification to obtain a plurality of first groupings, each first grouping including sub-trip records with the same vehicle identification and section identification;

[0047] A first training unit for respectively using each first grouping to train a decision tree model to obtain a vehicle-section energy consumption prediction model of the vehicle corresponding to each first grouping on the corresponding section.

[0048] In addition, according to at least one embodiment of the present invention, the first model generation unit further includes:

[0049] A second grouping unit for grouping the sub-trip records according to vehicle model identification and section identification to obtain a plurality of second groupings, each second grouping including sub-trip records with the same vehicle model identification and section identification;

[0050] A second training unit for respectively using each second grouping to train a decision tree model to obtain a vehicle-section energy consumption prediction model of the vehicles of the vehicle model corresponding to each first grouping on the corresponding section.

[0051] In addition, according to at least one embodiment of the present invention, the prediction device further includes:

[0052] A path recommendation unit for, according to the total energy required for the target vehicle to travel on at least one candidate travel path predicted based on the energy consumption grid, selecting a target travel path that meets a preset path selection strategy from the at least one candidate travel path, and prompting the total energy required for the target travel path.

[0053] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for predicting vehicle trip energy consumption as described above are implemented.

[0054] Compared with the prior art, the method and device for predicting vehicle trip energy consumption provided by the embodiments of the present invention have at least the following beneficial effects: The prediction method of the embodiments of the present invention comprehensively considers personal driving habits, road condition information, and travel conditions, can improve the accuracy of energy consumption prediction results, is easy to be implemented in engineering, and has advantages such as a small amount of calculation and low implementation cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required to be used in the description of the embodiments of the present invention. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0056] Figure 1 It is a system block diagram of a device for predicting vehicle trip energy consumption provided by an embodiment of the present invention;

[0057] Figure 2 It is a flowchart of a method for predicting vehicle trip energy consumption provided by an embodiment of the present invention;

[0058] Figure 3 It is a distribution diagram of the prediction errors between the prediction method of the embodiments of the present invention and the traditional method;

[0059] Figure 4 It is a structural diagram of a device for predicting vehicle trip energy consumption provided by an embodiment of the present invention;

[0060] Figure 5 It is another structural diagram of a device for predicting vehicle trip energy consumption provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0061] In order to make the technical problems, technical solutions, and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the drawings and specific embodiments. In the following description, providing specific details such as specific configurations and components is only to help comprehensively understand the embodiments of the present invention. Therefore, those skilled in the art should clearly understand that various changes and modifications can be made to the embodiments described here without departing from the scope and spirit of the present invention. In addition, descriptions of known functions and structures are omitted for clarity and conciseness.

[0062] It should be understood that the "one embodiment" or "an embodiment" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present invention. Therefore, the "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner.

[0063] In various embodiments of the present invention, it should be understood that the magnitudes of the serial numbers of the following processes do not mean the sequence of execution. The execution sequence of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0064] As described in the background art, in the prior art vehicle travel energy consumption prediction methods, generally, there are problems that the schemes with low implementation cost have poor prediction result accuracy, while the schemes with higher prediction accuracy have high implementation cost. To solve at least one of the above problems, an embodiment of the present invention provides a vehicle travel energy consumption prediction method, which makes a trade-off between the implementation cost and the accuracy of the prediction result, can improve the accuracy of the energy consumption prediction result, and reduce the implementation cost of the energy consumption prediction.

[0065] Figure 1 It is a schematic diagram of a system block diagram of a vehicle travel energy consumption prediction device according to an embodiment of the present invention. Figure 1 The shown vehicle travel energy consumption prediction device 10 includes an online module 101 and an offline module 102. Among them, the online module 101 further includes a data collection unit, a path recommendation unit, an energy consumption prediction unit and a user interface, and the offline module 102 includes a data processing unit, a model generation unit, an energy consumption grid generation unit and a system management unit.

[0066] In Figure 1 :

[0067] The data collection module is used to collect the travel records of multiple vehicles within a preset area (such as a certain target city or a certain target area), and store them in the data storage unit ( Figure 1 Data Storage in

[0068] The data processing unit is used to preprocess these data, and the preprocessing can include data cleaning and noise removal, etc.

[0069] The model generation unit is configured to generate three energy consumption prediction models of the embodiments of the present invention according to the data processed by the data processing unit. It should be noted that the energy prediction models are generated separately for each road segment, that is, 3 energy consumption prediction models can be generated separately for each road segment within the preset area. In addition, the road segment described in the embodiments of the present invention refers to a section of road with travel records and no road forks are found in the road according to the travel records.

[0070] The energy consumption grid generation unit then generates an energy consumption grid of the preset area for each vehicle according to the energy consumption prediction model of each road segment. The energy consumption grid includes the energy consumption prediction models associated with each road segment within the preset area. The energy consumption grid generated by the energy consumption grid generation unit can be stored in the grid storage unit ( Figure 1 GridStorage in it), and the embodiments of the present invention can execute the above-mentioned various modules regularly or irregularly according to the update period of the travel records to update the energy consumption grid.

[0071] After generating the energy consumption grid, the embodiments of the present invention can receive the user's energy consumption prediction requirement or path recommendation request through the user interface, and provide corresponding services for the user through the energy consumption prediction unit and the path recommendation unit. Among them, the energy consumption prediction unit can predict the total energy required for a specific vehicle to travel on a certain path according to the energy consumption grid, and the path recommendation unit can select the target travel path with the best cost for the user from multiple candidate travel paths according to the preset path selection strategy and display it to the user through the user interface.

[0072] Through the analysis of the energy consumption data of various vehicles, the inventor found that there are many factors affecting the energy consumption of a trip, usually including 4 main influencing factors. First is the vehicle model. Different models of vehicles usually have different energy consumptions. Second is the route and road condition factors. The length and road conditions of different routes (such as the elevation change of the road) require different energy consumptions. Thirdly, driving habits (such as the driving habit of frequently accelerating and braking suddenly, or the driving habit of driving smoothly and evenly at a constant speed) also affect energy consumption. Finally, there are travel conditions, such as travel time, vehicle load, air temperature, etc. Usually, the heavier the vehicle load, the greater the energy consumption (the embodiments of the present invention use tire pressure to approximately represent the vehicle load); the air temperature also affects the engine and battery efficiency, and thus affects the vehicle energy consumption. In addition to the above factors, there may be some other minor factors, such as: a) the energy consumption of accessories such as air conditioners and multimedia playback devices, b) traffic conditions, etc. For the sake of simplicity in processing, the energy consumption changes caused by these minor factors are regarded as a kind of random disturbance and not considered in this article.

[0073] Such as Figure 2As shown, a method for predicting vehicle travel energy consumption provided by an embodiment of the present invention includes:

[0074] Step 21: Based on the travel records of multiple vehicles in a preset area collected in advance, train a vehicle-section energy consumption prediction model for each vehicle on each section and a vehicle-model-section energy consumption prediction model for each vehicle model on each section. Among them, the travel records include the travel time, travel conditions, vehicle identification, vehicle model identification, geographical location of the vehicle at each sampling time point, and energy consumption data of each trip.

[0075] Here, the preset area is the target area where route planning is required. For example, a certain city or a certain region. Embodiments of the present invention can use the travel records of various vehicles in the preset area to train multiple prediction models. A travel record is the relevant data collected during a trip of a specific vehicle in the preset area. Usually, the vehicle can report data at each sampling time point according to a preset sampling period. These data can specifically include the travel time, travel conditions, vehicle identification, vehicle model identification, geographical location of the vehicle at each sampling time point, and energy consumption data of the trip.

[0076] Among them, the travel time can include the specific date and time. The travel conditions specifically include at least one of the current temperature and vehicle tire pressure. The vehicle identification is used to uniquely identify a specific vehicle, and the vehicle model identification is the vehicle model to which the vehicle belongs, specifically, it can be the brand and model of a certain vehicle. The geographical location can be longitude and latitude coordinates, and the energy consumption data can be the current real-time energy consumption, such as the average power consumption per 100 kilometers currently. Based on the energy consumption data, the energy required for the vehicle to pass through a certain section in a certain travel record can be calculated. For example, the average value of the real-time energy consumption collected in this section of this travel record can be calculated, and then multiplied by the length of this section to obtain the energy required for this section.

[0077] Step 22: Generate a default-section energy consumption prediction model for the vehicle according to the default energy consumption of the vehicle or the vehicle model to which the vehicle belongs.

[0078] Here, embodiments of the present invention also generate a default-section energy consumption prediction model for each vehicle according to the default energy consumption of the vehicle or the vehicle model to which the vehicle belongs. For example, usually each vehicle or each vehicle model has a default energy consumption data set before leaving the factory, which represents the default energy consumption of the vehicle or vehicle model under a certain road condition. Embodiments of the present invention can directly use this default energy consumption to generate a default-section energy consumption prediction model for the vehicle. This default-section energy consumption prediction model is used to represent the default energy consumption of the vehicle on any section. For the specific energy consumption of a specific section, it can be the default energy consumption multiplied by the length of the specific section.

[0079] Step 23: Predict the total energy required for the target vehicle to travel on the candidate travel route by using at least one of the vehicle section energy consumption prediction model, the vehicle model section energy consumption prediction model, and the default section energy consumption prediction model of the target vehicle.

[0080] After obtaining the energy consumption prediction models in Steps 21-22, embodiments of the present invention can use the above models to predict the total energy required for the target vehicle to travel on a certain candidate travel route. For example, calculate the energy required for each section on the candidate travel route respectively, and then accumulate to obtain the total energy.

[0081] As an implementation manner, for any section, the priority of the vehicle section energy consumption prediction model, the vehicle model section energy consumption prediction model, and the default section energy consumption prediction model of the vehicle on this section decreases in turn. When calculating the energy consumption of each section on the candidate travel route, use the energy consumption prediction model with the highest priority existing on this section to calculate the energy consumption of this section.

[0082] Through the above steps, embodiments of the present invention generate multiple energy consumption prediction models by using the pre-collected trip records. Since the trip records include travel conditions and the driving data of the driver on specific sections, personal driving habits, road conditions information, and travel conditions are comprehensively considered in the models, thereby improving the accuracy of the model prediction results. In addition, when training the model and predicting the energy consumption requirements of the candidate travel route, embodiments of the present invention do not need to consider real-time road conditions information, and a relatively simple algorithm model can be used to implement the above solution, thereby reducing the computational complexity and implementation cost of the solution and making the solution easy to implement in engineering.

[0083] In embodiments of the present invention, after obtaining the relevant energy consumption prediction models of the vehicle, an energy consumption grid of the preset area can be generated for each vehicle respectively. Here, the energy consumption grid includes the energy consumption prediction models associated with each section in the preset area, where:

[0084] 1) When the vehicle section energy consumption prediction model of the vehicle exists on this section, the energy consumption prediction model associated with this section is the vehicle section energy consumption prediction model of the vehicle on this section;

[0085] 2) When the vehicle section energy consumption prediction model of the vehicle does not exist on this section, but the vehicle model section energy consumption prediction model of the vehicle exists, the energy consumption prediction model associated with this section is the product of the vehicle model section energy consumption prediction model of the vehicle on this section and the first ratio;

[0086] 3) When only the default section energy consumption prediction model of the vehicle exists, the energy consumption prediction model associated with this section is the product of the default section energy consumption prediction model of the vehicle and the second ratio.

[0087] Wherein, the first ratio is the ratio of the average energy consumption of the vehicle on all road segments to the average energy consumption of vehicles of the same model as the vehicle on all road segments; the second ratio is the ratio of the average energy consumption of the vehicle on all road segments to the default energy consumption of the vehicle or the vehicle model.

[0088] In this way, in step 23, according to each road segment passed by the candidate travel route, the energy consumption grid of the target vehicle can be searched to obtain the energy consumption prediction model associated with the target vehicle on each road segment. Using the obtained energy consumption prediction model and the travel conditions corresponding to the candidate travel route, the energy consumption requirements for each road segment are calculated and accumulated to obtain the total energy required for the target vehicle to pass through the candidate travel route.

[0089] As an implementation, the energy consumption grid can be generated by the server and sent to each vehicle, and each vehicle saves the energy consumption grid of its own vehicle locally. In this way, when the driver hopes to obtain the energy consumption of a certain candidate travel route or requests a recommended route between a certain starting point and ending point, the energy consumption of each candidate travel route can be calculated according to the locally saved energy consumption grid of the vehicle.

[0090] According to at least one embodiment of the present invention, in the above step 21, the step of training the vehicle-road segment energy consumption prediction model for each vehicle on each road segment may specifically include:

[0091] A) Split each trip record into sub-trips for each road segment of the trip.

[0092] B) For each sub-trip of each road segment, according to the average value of the energy consumption data at each sampling time point on the road segment and the length of the road segment in the sub-trip, calculate the energy required for the corresponding vehicle on the road segment in the sub-trip, and generate a sub-trip record. The sub-trip record includes travel time, travel conditions, vehicle identification, vehicle model identification, and the energy required for the vehicle in the sub-trip.

[0093] C) Group the sub-trip records according to vehicle identification and road segment identification to obtain a plurality of first groups. Each first group includes sub-trip records with the same vehicle identification and road segment identification.

[0094] D) Use each first group to train a decision tree model respectively to obtain the vehicle-road segment energy consumption prediction model for the vehicle corresponding to each first group on the corresponding road segment.

[0095] Through the above steps, the embodiment of the present invention generates a vehicle-road segment energy consumption prediction model for a specific vehicle on a specific road segment. By inputting parameters such as travel conditions and travel time into the model, the energy required for the specific vehicle to pass through the specific road segment can be obtained.

[0096] In addition, before the above-mentioned step A, the embodiments of the present invention can also filter the data of the trip records. For example, for each trip and each road segment, delete the records lacking relevant information (such as travel time, date, vehicle identifier (vehicle_id), vehicle model identifier or geographical location information), and then delete the trips whose record numbers are in the first N or last N positions of the trip record number list, that is, eliminate the abnormal trips with too many or too few records. Here, N is a preset integer.

[0097] To reduce the computational complexity and training time required for model training, in the above-mentioned step B, the embodiments of the present invention can also discretize the sub-trip records of each road segment according to the preset time segments to which the travel time belongs and different gears of travel conditions, and merge the sub-trip records of each vehicle on each road segment belonging to each appearance time period and each travel condition gear to reduce the number of sub-trip records. For example, for each sub-trip record, divide the travel time into one gear every five minutes, divide the temperature into one gear every 2 degrees Celsius, divide the tire pressure into one gear every 0.1 Kpa, and so on.

[0098] According to at least one embodiment of the present invention, in the above-mentioned step 21, the step of training the vehicle model for predicting energy consumption of a vehicle model on each road segment includes:

[0099] E) Group the sub-trip records obtained in the above-mentioned step B according to the vehicle model identifier and the road segment identifier to obtain a plurality of second groups, and each second group includes sub-trip records with the same vehicle model identifier and road segment identifier;

[0100] F) Use each second group to train a decision tree model respectively to obtain the vehicle model for predicting energy consumption of a vehicle model on the corresponding road segment corresponding to each first group.

[0101] Through the above steps, the embodiments of the present invention generate a vehicle model for predicting energy consumption of a vehicle model on a specific road segment. By inputting parameters such as travel conditions and travel time into the model, the energy required for the vehicle of the specific vehicle model to pass through the specific road segment can be obtained. For a specific vehicle, the energy can be further multiplied by the first ratio corresponding to the specific vehicle, so that the energy required for the specific vehicle to pass through the specific road segment can be obtained by using the vehicle model for predicting energy consumption of a vehicle model on a specific road segment.

[0102] In addition, it should be noted that in the above steps D and F, a decision tree model is used for training. In the embodiments of the present invention, other algorithms or models can also be used, such as random forest model, general linear model, linear regression model, and support vector regression model, etc. Examples are not given one by one here.

[0103] Similarly, when there is only a default road section energy consumption prediction model for a specific vehicle on a certain road section, the energy output by the default road section energy consumption prediction model can be multiplied by the second ratio corresponding to the specific vehicle, so that the energy required for the specific vehicle to pass through the specific road section can be obtained by using the default road section energy consumption prediction model.

[0104] In addition, in the embodiments of the present invention, the first ratio and the second ratio can also be calculated in the following manner: for each vehicle, calculate the first average energy consumption of the vehicle respectively according to all sub-trip records of the vehicle; calculate the second average energy consumption of the vehicle model to which the vehicle belongs according to all sub-trip records of the vehicle model to which the vehicle belongs; calculate the ratio of the first average energy consumption to the second average energy consumption to obtain the first ratio; and calculate the ratio of the first average energy consumption to the default energy consumption of the vehicle or the vehicle model to which the vehicle belongs to obtain the second ratio.

[0105] As can be seen from the above, the vehicle trip energy consumption prediction method provided by the embodiments of the present invention comprehensively considers personal driving habits, road conditions information, and travel conditions, can improve the accuracy of energy consumption prediction results, is easy to be implemented in engineering, and has the advantages of small calculation amount and low implementation cost.

[0106] Based on the above method, in the embodiments of the present invention, a target travel route that meets the preset route selection strategy can also be recommended to the driver. For example, the total energy required for the target vehicle to travel on at least one candidate travel route can be predicted by using the above method, and then, from the at least one candidate travel route, a target travel route that meets the preset route selection strategy is selected, and the driver is prompted with the total energy required for the target travel route.

[0107] Based on the collected trip records, by comparing the solution of the present application with the traditional energy consumption prediction solution based on the average energy consumption per 100 kilometers through simulation, it can be found that the error of the prediction result in the embodiments of the present invention, whether it is the mean square error or the worst result, is smaller than that of the traditional method, as shown in Table 1 specifically.

[0108]

[0109] Table 1

[0110] Figure 3 The distribution of the prediction errors between the present application and the traditional solution is provided, where the abscissa represents the prediction error, and the closer it is to 0, the better the error; the ordinate represents the number of occurrences of the prediction error within this range. From Figure 3 it can be seen that in this simulation example, the worst result of the present application is better than the best result of the traditional method.

[0111] Based on the collected trip records, the solution of the present application is simulated and compared with the energy consumption prediction solutions using machine learning algorithms such as neural networks or random forests in the prior art. It can be found that the errors of the prediction results of the embodiments of the present invention, whether the mean square error or the variance after cross - validation, are smaller than the corresponding values of the existing methods (the smaller the value, the more accurate the prediction), as shown in Table 2 specifically.

[0112]

[0113] Table 2

[0114] Based on the above - mentioned vehicle trip energy consumption prediction method, the embodiments of the present invention further provide a device for implementing the above - mentioned method.

[0115] Please refer to Figure 4 , a vehicle trip energy consumption prediction device 40 provided by the embodiments of the present invention includes:

[0116] A first model generation unit, configured to train a vehicle - section energy consumption prediction model for each vehicle on each road section and a vehicle - model - section energy consumption prediction model for each vehicle model on each road section according to the trip records of multiple vehicles in a preset area collected in advance, where the trip records include the travel time, travel conditions, vehicle identifier, vehicle model identifier, geographical location, and energy consumption data of the vehicle at each sampling time point of each trip;

[0117] A second model generation unit, configured to generate a default - section energy consumption prediction model of the vehicle according to the default energy consumption of the vehicle or the vehicle model to which the vehicle belongs and the length of the road section;

[0118] An energy consumption prediction unit, configured to predict the total energy required for the target vehicle to travel on a candidate travel path by using at least one of the vehicle - section energy consumption prediction model, the vehicle - model - section energy consumption prediction model, and the default - section energy consumption prediction model of the target vehicle.

[0119] Through the above - mentioned modules, the prediction device of the embodiments of the present invention comprehensively considers personal driving habits, road condition information, and travel conditions, can improve the accuracy of the energy consumption prediction result, is easy to be implemented in engineering, and has advantages such as a small amount of calculation and a low implementation cost.

[0120] According to at least one embodiment of the present invention, the prediction device further includes the following modules (not shown in the figure):

[0121] An energy consumption grid generation unit, configured to generate an energy consumption grid of the preset area for each vehicle respectively, where the energy consumption grid includes an energy consumption prediction model associated with each road section within the preset area, and:

[0122] When there is a vehicle road section energy consumption prediction model of the vehicle on the road section, the energy consumption prediction model associated with the road section is the vehicle road section energy consumption prediction model of the vehicle on the road section;

[0123] When there is no vehicle road section energy consumption prediction model of the vehicle on the road section, but there is a vehicle model road section energy consumption prediction model of the vehicle, the energy consumption prediction model associated with the road section is the product of the vehicle model road section energy consumption prediction model of the vehicle on the road section and a first ratio;

[0124] When there is only a default road section energy consumption prediction model of the vehicle, the energy consumption prediction model associated with the road section is the product of the default road section energy consumption prediction model of the vehicle and a second ratio;

[0125] Wherein, the first ratio is the ratio of the average energy consumption of the vehicle on all road sections to the average energy consumption of the vehicles of the vehicle model to which the vehicle belongs on all road sections; the second ratio is the ratio of the average energy consumption of the vehicle on all road sections to the default energy consumption of the vehicle or the vehicle model to which the vehicle belongs.

[0126] According to at least one embodiment of the present invention, the energy consumption prediction unit is further configured to, according to each road section passed by the candidate travel route, search for the energy consumption grid of the target vehicle, obtain the energy consumption prediction models associated with the target vehicle on each road section, and use the obtained energy consumption prediction models to calculate the energy consumption requirements of each road section and accumulate them to obtain the total energy required for the target vehicle to pass through the candidate travel route.

[0127] According to at least one embodiment of the present invention, the first model generation unit includes:

[0128] A trip record splitting unit, configured to split each trip record into sub-trips of each road section of the trip;

[0129] A sub-trip processing unit, configured to, for each sub-trip of each road section, calculate the energy required for the corresponding vehicle on the road section according to the average value of the energy consumption data at each sampling time point on the road section in the sub-trip and the length of the road section, and generate a sub-trip record, where the sub-trip record includes travel time, travel conditions, vehicle identification, vehicle model identification, and the energy required for the vehicle in the sub-trip;

[0130] A first grouping unit, configured to group the sub-trip records according to vehicle identification and road section identification, to obtain a plurality of first groupings, where each first grouping includes sub-trip records with the same vehicle identification and road section identification;

[0131] A first training unit, configured to respectively use each first grouping to train a decision tree model, to obtain a vehicle-road section energy consumption prediction model for the vehicle corresponding to each first grouping on the corresponding road section.

[0132] According to at least one embodiment of the present invention, the first model generation unit further includes:

[0133] A second grouping unit, configured to group the sub-trip records according to vehicle type identification and road section identification, to obtain a plurality of second groupings, where each second grouping includes sub-trip records with the same vehicle type identification and road section identification;

[0134] A second training unit, configured to respectively use each second grouping to train a decision tree model, to obtain a vehicle type-road section energy consumption prediction model for the vehicle of the vehicle type corresponding to each first grouping on the corresponding road section.

[0135] According to at least one embodiment of the present invention, the prediction device further includes the following modules (not shown in the figure):

[0136] A ratio calculation unit, configured to, for each vehicle, respectively calculate a first average energy consumption of the vehicle according to all sub-trip records of the vehicle; calculate a second average energy consumption of the vehicle type to which the vehicle belongs according to all sub-trip records of the vehicle type to which the vehicle belongs; calculate a ratio of the first average energy consumption to the second average energy consumption to obtain the first ratio; and calculate a ratio of the first average energy consumption to a default energy consumption of the vehicle or the vehicle type to which the vehicle belongs to obtain the second ratio.

[0137] According to at least one embodiment of the present invention, the prediction device further includes the following modules (not shown in the figure):

[0138] A path recommendation unit, configured to select a target travel path that meets a preset path selection strategy from the at least one candidate travel path according to the total energy required for the target vehicle to travel on at least one candidate travel path predicted based on the energy consumption grid, and prompt the total energy required for the target travel path.

[0139] According to at least one embodiment of the present invention, the travel conditions include at least one of temperature and vehicle tire pressure. The road section is a section of road with travel records, and no road fork is found in the road according to the travel records.

[0140] Such as Figure 5As shown in the figure, an embodiment of the present invention further provides another prediction device 50 for vehicle travel energy consumption. The prediction device 50 for vehicle travel energy consumption specifically includes a processor 51, a memory 52, a bus system 53, a receiver 54, and a transmitter 55. Among them, the processor 51, the memory 52, the receiver 54, and the transmitter 55 are connected through the bus system 53. The memory 52 is used to store instructions, and the processor 51 is used to execute the instructions stored in the memory 52 to control the receiver 54 to receive signals and control the transmitter 55 to send signals;

[0141] Among them, the processor 51 is used to read the program in the memory and execute the following processes:

[0142] According to the travel records of multiple vehicles in a preset area collected in advance, train a vehicle section energy consumption prediction model for each vehicle on each section and a vehicle model section energy consumption prediction model for each vehicle model on each section. Among them, the travel records include the travel time, travel conditions, vehicle identification, vehicle model identification, geographical location, and energy consumption data of the vehicle at each sampling time point;

[0143] Generate a default section energy consumption prediction model for the vehicle according to the default energy consumption of the vehicle or the vehicle model to which the vehicle belongs and the length of the section;

[0144] Use at least one of the vehicle section energy consumption prediction model, the vehicle model section energy consumption prediction model, and the default section energy consumption prediction model of the target vehicle to predict the total energy required for the target vehicle to travel on the candidate travel path.

[0145] It can be understood that in the embodiment of the present invention, when the program is executed by the processor 51, it can implement the above Figure 2 shown in the method embodiment of each process, and can achieve the same technical effect. To avoid repetition, it will not be described here again.

[0146] It should be understood that in the embodiment of the present invention, the processor 51 may be a central processing unit (Central Processing Unit, abbreviated as "CPU"). The processor 51 may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0147] The memory 52 may include a read-only memory and a random access memory, and provide instructions and data to the processor 51. A part of the memory 52 may also include a non-volatile random access memory. For example, the memory 52 may also store information about the device type.

[0148] In addition to the data bus, the bus system 53 may further include a power bus, a control bus, a status signal bus, etc. However, for the sake of clear illustration, various buses are labeled as the bus system 53 in the figure.

[0149] In the implementation process, each step of the above method can be completed by the integrated logic circuit in hardware in the processor 51 or instructions in the form of software. The steps of the method disclosed in combination with the embodiments of the present invention can be directly embodied as being executed and completed by the hardware processor, or executed and completed by a combination of the hardware and software modules in the processor. The software module can be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. This storage medium is located in the memory 52, and the processor 51 reads the information in the memory 52 and combines its hardware to complete the steps of the above method. To avoid repetition, it will not be described in detail here.

[0150] In some embodiments of the present invention, a computer-readable storage medium is further provided, on which a program is stored, and when the program is executed by a processor, the following steps can be implemented:

[0151] According to the travel records of multiple vehicles in a preset area collected in advance, a vehicle-segment energy consumption prediction model for each vehicle on each road segment and a vehicle-model-segment energy consumption prediction model for each vehicle model on each road segment are trained, where the travel records include the travel time, travel conditions, vehicle identification, vehicle model identification, geographical location of the vehicle at each sampling time point, and energy consumption data of each trip;

[0152] Generate a default segment energy consumption prediction model for the vehicle according to the default energy consumption of the vehicle or the vehicle model to which the vehicle belongs and the length of the road segment;

[0153] Use at least one of the vehicle-segment energy consumption prediction model, the vehicle-model-segment energy consumption prediction model, and the default segment energy consumption prediction model of the target vehicle to predict the total energy required for the target vehicle to travel on the candidate travel path.

[0154] When the program is executed by the processor, it can implement Figure 2 All implementation manners in the prediction method of vehicle travel energy consumption shown, and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0155] Those of ordinary skill in the art will realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled professionals can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present invention.

[0156] Those skilled in the art can 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 foregoing method embodiments, and will not be elaborated herein.

[0157] In the embodiments provided in the present 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 illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, 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 displayed or discussed mutual coupling, direct coupling, or communication connection can be through some interfaces. The indirect coupling or communication connection of the devices or units can be electrical, mechanical, or other forms.

[0158] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiments of the present invention.

[0159] In addition, the functional units in various embodiments of the present invention can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0160] When the above-mentioned 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 invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0161] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for predicting the energy consumption of a vehicle's journey, characterized in that, Including: Based on the travel records of multiple vehicles in a preset area collected in advance, a vehicle-segment energy consumption prediction model for each vehicle on each road segment and a vehicle model-segment energy consumption prediction model for each vehicle model on each road segment are trained, where the travel records include the travel time, travel conditions, vehicle identification, vehicle model identification, geographical locations of the vehicle at each sampling time point, and energy consumption data for each trip; Generating a default segment energy consumption prediction model for a vehicle according to the default energy consumption of the vehicle or the vehicle model to which the vehicle belongs and the length of the road segment; Using at least one of the vehicle-segment energy consumption prediction model, the vehicle model-segment energy consumption prediction model, and the default segment energy consumption prediction model of the target vehicle to predict the total energy required for the target vehicle to travel on the candidate travel path, where, for any road segment, the priorities of the vehicle-segment energy consumption prediction model, the vehicle model-segment energy consumption prediction model, and the default segment energy consumption prediction model of the vehicle on this road segment decrease in sequence; when calculating the energy consumption of each road segment on the candidate travel path, the energy consumption prediction model with the highest priority existing on this road segment is used to calculate the energy consumption of this road segment.

2. The prediction method according to claim 1, wherein Before the step of predicting the total energy required for the target vehicle to travel on the candidate travel path, the method further includes: For each vehicle, an energy consumption grid of the preset area is respectively generated, and the energy consumption grid includes the energy consumption prediction models associated with each road segment in the preset area, where: When the vehicle-segment energy consumption prediction model of the vehicle exists on this road segment, the energy consumption prediction model associated with this road segment is the vehicle-segment energy consumption prediction model of the vehicle on this road segment; When the vehicle-segment energy consumption prediction model of the vehicle does not exist on this road segment but the vehicle model-segment energy consumption prediction model of the vehicle exists, the energy consumption prediction model associated with this road segment is the product of the vehicle model-segment energy consumption prediction model of the vehicle on this road segment and the first ratio; When only the default segment energy consumption prediction model of the vehicle exists, the energy consumption prediction model associated with this road segment is the product of the default segment energy consumption prediction model of the vehicle and the second ratio; where, the first ratio is the ratio of the average energy consumption of the vehicle on all road segments to the average energy consumption of the vehicles of the vehicle model to which the vehicle belongs on all road segments; the second ratio is the ratio of the average energy consumption of the vehicle on all road segments to the default energy consumption of the vehicle or the vehicle model to which the vehicle belongs.

3. The prediction method according to claim 2, wherein, The step of predicting the total energy required for the target vehicle to travel on the candidate travel path includes: According to each road segment passed by the candidate travel path, searching for the energy consumption grid of the target vehicle, obtaining the energy consumption prediction models associated with the target vehicle on each road segment, and using the obtained energy consumption prediction models to calculate the energy consumption requirements of each road segment and accumulate them to obtain the total energy required for the target vehicle to pass through the candidate travel path.

4. The prediction method according to claim 2, wherein The step of training the vehicle-segment energy consumption prediction model for each vehicle on each road segment includes: Splitting each travel record into sub-trips of each road segment of this trip; For each sub-trip of each road segment, based on the average value of the energy consumption data at each sampling time point on the road segment in the sub-trip and the length of the road segment, calculate the energy required for the corresponding vehicle on the road segment in the sub-trip, and generate a sub-trip record, where the sub-trip record includes travel time, travel conditions, vehicle identification, vehicle type identification, and the energy required for the vehicle in the sub-trip; Group the sub-trip records according to the vehicle identification and road segment identification to obtain a plurality of first groups, and each first group includes sub-trip records with the same vehicle identification and road segment identification; Respectively use each first group to train a decision tree model to obtain a vehicle-road segment energy consumption prediction model for the vehicle corresponding to each first group on the corresponding road segment.

5. The prediction method according to claim 4, wherein The steps of training a vehicle-road segment energy consumption prediction model for each vehicle type on each road segment include: Group the sub-trip records according to the vehicle type identification and road segment identification to obtain a plurality of second groups, and each second group includes sub-trip records with the same vehicle type identification and road segment identification; Respectively use each second group to train a decision tree model to obtain a vehicle-road segment energy consumption prediction model for the vehicle of the corresponding vehicle type in the corresponding road segment for each first group.

6. The prediction method according to claim 5, characterized in that, It further includes: Calculate the first ratio and the second ratio in the following manner: For each vehicle, respectively calculate the first average energy consumption of the vehicle based on all the sub-trip records of the vehicle; calculate the second average energy consumption of the vehicle type to which the vehicle belongs based on all the sub-trip records of the vehicle type to which the vehicle belongs; Calculate the ratio of the first average energy consumption to the second average energy consumption to obtain the first ratio; And calculate the ratio of the first average energy consumption to the default energy consumption of the vehicle or the vehicle type to which the vehicle belongs to obtain the second ratio.

7. The prediction method according to claim 2, wherein It further includes: According to the total energy required for the target vehicle to travel on at least one candidate travel path predicted based on the energy consumption grid, select a target travel path that meets the preset path selection strategy from the at least one candidate travel path, and prompt the total energy required for the target travel path.

8. The prediction method according to claim 1, wherein, The travel conditions include at least one of temperature and vehicle tire pressure.

9. The prediction method according to claim 1, wherein, The road segment is a section of road with travel records and no road forks are found in the road according to the travel records.

10. A prediction device for vehicle travel energy consumption, characterized in that, It includes: A first model generation unit for training a vehicle-road segment energy consumption prediction model for each vehicle on each road segment and a vehicle-road segment energy consumption prediction model for each vehicle type on each road segment based on the travel records of a plurality of vehicles in a preset area collected in advance, where the travel records include travel time, travel conditions, vehicle identification, vehicle type identification, geographical location, and energy consumption data of the vehicle at each sampling time point; A second model generation unit for generating a default road segment energy consumption prediction model of the vehicle according to the default energy consumption of the vehicle or the vehicle type to which the vehicle belongs and the length of the road segment; An energy consumption prediction unit, configured to predict the total energy required for the target vehicle to travel on a candidate travel route by using at least one of a vehicle section energy consumption prediction model, a vehicle model section energy consumption prediction model, and a default section energy consumption prediction model of the target vehicle. Wherein, for any section, the priorities of the vehicle section energy consumption prediction model, the vehicle model section energy consumption prediction model, and the default section energy consumption prediction model of the vehicle on this section decrease in turn; when calculating the energy consumption of each section on the candidate travel route, the energy consumption prediction model with the highest priority existing on this section is used to calculate the energy consumption of this section.

11. The prediction device according to claim 10, characterized in that, It further includes: An energy consumption grid generation unit, configured to generate an energy consumption grid of the preset area for each vehicle respectively, where the energy consumption grid includes the energy consumption prediction model associated with each section in the preset area, and: When the vehicle section energy consumption prediction model of the vehicle exists on this section, the energy consumption prediction model associated with this section is the vehicle section energy consumption prediction model of the vehicle on this section; When the vehicle section energy consumption prediction model of the vehicle does not exist on this section, but the vehicle model section energy consumption prediction model of the vehicle exists, the energy consumption prediction model associated with this section is the product of the vehicle model section energy consumption prediction model of the vehicle on this section and the first ratio; When only the default section energy consumption prediction model of the vehicle exists, the energy consumption prediction model associated with this section is the product of the default section energy consumption prediction model of the vehicle and the second ratio; Wherein, the first ratio is the ratio of the average energy consumption of the vehicle on all sections to the average energy consumption of the vehicles of the vehicle model to which the vehicle belongs on all sections; the second ratio is the ratio of the average energy consumption of the vehicle on all sections to the default energy consumption of the vehicle or the vehicle model to which the vehicle belongs.

12. The prediction device according to claim 11, wherein The first model generation unit includes: A trip record splitting unit, configured to split each trip record into sub-trips of each section of this trip; A sub-trip processing unit, configured to, for each sub-trip of each section, calculate the energy required for the corresponding vehicle on this section in this sub-trip according to the average value of the energy consumption data at each sampling time point on this section and the length of this section, and generate a sub-trip record, where the sub-trip record includes travel time, travel conditions, vehicle identification, vehicle model identification, and the energy required for the vehicle in this sub-trip; A first grouping unit, configured to group the sub-trip records according to vehicle identification and section identification to obtain a plurality of first groupings, and each first grouping includes sub-trip records with the same vehicle identification and section identification; A first training unit, configured to respectively use each first grouping to train a decision tree model to obtain the vehicle section energy consumption prediction model of the corresponding vehicle on the corresponding section for each first grouping.

13. The prediction device according to claim 12, characterized in that, The first model generation unit further includes: A second grouping unit, configured to group the sub-trip records according to vehicle model identification and section identification to obtain a plurality of second groupings, and each second grouping includes sub-trip records with the same vehicle model identification and section identification; A second training unit, configured to respectively use each second group to train a decision tree model, so as to obtain a vehicle model-road section energy consumption prediction model for vehicles of the vehicle model corresponding to each first group on the corresponding road section.

14. The prediction device according to claim 11, wherein It further includes: A path recommendation unit, configured to select, from the at least one candidate travel path, a target travel path that meets a preset path selection strategy according to the total energy required for the target vehicle to travel on at least one candidate travel path predicted based on the energy consumption grid, and prompt the total energy required for the target travel path.

Images