CN112406875B

CN112406875B - Vehicle energy consumption analysis method and device

Info

Publication number: CN112406875B
Application number: CN202011385839.3A
Authority: CN
Inventors: 龙荣深; 邓俊松; 何锐邦
Original assignee: Guangzhou Xiaopeng Motors Technology Co Ltd; Guangzhou Chengxingzhidong Automotive Technology Co., Ltd
Current assignee: Guangzhou Xiaopeng Motors Technology Co Ltd; Guangzhou Chengxingzhidong Automotive Technology Co., Ltd
Priority date: 2020-11-30
Filing date: 2020-11-30
Publication date: 2023-08-29
Anticipated expiration: 2040-11-30
Also published as: CN112406875A

Abstract

The embodiment of the invention provides a vehicle energy consumption analysis method and device, which are characterized in that vehicle driving information of a vehicle target journey is obtained, the vehicle driving information comprises target road condition information and vehicle energy consumption, then a similar journey for the target journey is obtained according to the target road condition information and a journey query model, then an optimized journey is generated according to sample driving information and the target road condition information of the similar journey, the sample driving information and the target road condition information of the optimized journey are input into an energy consumption estimation model, the sample energy consumption of the optimized journey is generated, energy consumption comparison is carried out, an energy consumption suggestion for the target journey is output according to an energy consumption comparison result between the optimized journey and the vehicle target journey, and the similar journey is searched through the road condition information, so that the obtained similar journey is a journey executed by a real vehicle, the feasibility of energy consumption prediction is improved, the accuracy of the prediction is effectively improved based on the energy consumption prediction of the road condition information and the driving information, and the method is applicable to different vehicles and has high universality.

Description

Vehicle energy consumption analysis method and device

Technical Field

The invention relates to the technical field of vehicles, in particular to a vehicle energy consumption analysis method and a vehicle energy consumption analysis device.

Background

With the development of new energy technology, new energy automobiles are becoming popular, and the new energy automobiles have obvious advantages in energy consumption and emission compared with traditional internal combustion engine automobiles, such as good dynamic property, small driving noise, energy conservation, zero emission and the like. However, the range of new energy vehicles is also short and the charging time is long due to the limitations of battery technology development. Pure new energy automobile drivers are concerned about whether they can reach a destination with the current remaining energy, which is known as "mileage anxiety", which is one of the main factors currently limiting the acceptance of new energy automobiles. It is apparent that installing a large capacity battery, rapidly charging and establishing more charging stations are effective means of effectively alleviating and solving the "mileage anxiety", but these methods still require a long time to be realized due to the limitations of the state of the art and the funding conditions. Therefore, if the energy consumption of the vehicle can be predicted, the user can adjust the driving mode according to the predicted energy consumption result so that the vehicle can reach a farther place. However, the current energy consumption prediction method of the vehicle has the problems of low executable performance, inaccurate prediction and low universality.

Disclosure of Invention

The embodiment of the invention provides a vehicle energy consumption analysis method, a vehicle energy consumption analysis device, a vehicle and a computer readable storage medium, which are used for solving or partially solving the problems of low vehicle energy consumption prediction performability, inaccurate prediction and low universality in the prior art.

The embodiment of the invention discloses a vehicle energy consumption analysis method, which comprises the following steps:

acquiring vehicle driving information of a vehicle target journey, wherein the vehicle driving information at least comprises target road condition information and vehicle energy consumption;

obtaining at least one similar journey aiming at the target journey according to the target road condition information and a preset journey query model;

generating at least one optimized journey for the target journey according to the sample driving information of at least one similar journey and the target road condition information;

inputting the sample driving information corresponding to the optimized journey and the target road condition information into a preset energy consumption estimation model to generate sample energy consumption aiming at the optimized journey;

and comparing the sample energy consumption of at least one optimized journey with the vehicle energy consumption respectively, and outputting an energy consumption suggestion for the target journey according to a comparison result.

Optionally, the obtaining at least one similar trip for the target trip according to the target road condition information and a preset trip query model includes:

inputting the target road condition information into a preset travel query model to obtain at least one travel sample aiming at the target travel;

comparing the travel sample with the target travel by position points to obtain travel similarity between the travel sample and the target travel;

and taking a travel sample with travel similarity greater than or equal to a preset similarity threshold as a similar travel for the target travel.

Optionally, the generating at least one optimized trip for the target trip according to the sample driving information of at least one similar trip and the target road condition information includes:

and carrying out information combination on the target road condition information and sample driving information corresponding to the similar journey respectively to generate at least one optimized journey aiming at the target journey.

Optionally, the comparing the sample energy consumption of at least one similar sample with the vehicle energy consumption, and outputting an energy consumption suggestion for the target trip according to the comparison result, includes:

Acquiring an energy consumption coefficient aiming at the energy consumption of the vehicle;

generating target vehicle energy consumption by adopting the energy consumption coefficient and the vehicle energy consumption;

selecting an optimized journey of which the energy consumption of the sample is less than or equal to that of the target vehicle as a target optimized journey, and acquiring target vehicle driving information of the target optimized journey;

and outputting an energy consumption suggestion matched with the comparison result according to the comparison result between the vehicle driving information of the target journey and the target vehicle driving information of the target optimization journey.

Optionally, the vehicle driving information further includes target driving information, and the outputting the energy consumption advice for the target trip according to a comparison result between the vehicle driving information of the target trip and the target vehicle driving information of the target optimized trip includes:

comparing the target driving information of the target journey with the sample driving information of the target optimization journey to generate a comparison result aiming at the target driving information;

and outputting the target road condition information, the target driving information and the energy consumption advice matched with the comparison result.

Optionally, the travel query model is generated by:

Acquiring training sample data, wherein the training sample data comprises historical trips of different vehicles and historical driving information of the historical trips, and the historical driving information at least comprises historical road condition information;

inputting the historical road condition information into a preset initial travel query model for iteration to generate a corresponding first predicted value;

comparing the first predicted value with a preset first reference value, and performing reverse training on the initial travel query model according to the comparison result to generate the travel query model.

Optionally, the historical driving information further includes a trip identifier of the historical trip, and the step of inputting the historical road condition information into a preset initial trip query model for iteration to generate a corresponding first predicted value includes:

vectorizing the journey identification and the historical road condition information to generate a first training vector;

inputting the first training vector into a preset initial forming query model for iteration, and calculating a plurality of first loss functions of the initial travel query model after each iteration;

comparing the first predicted value with a preset first reference value, and performing reverse training on the initial travel query model according to a comparison result to generate the travel query model, wherein the method comprises the following steps:

And stopping iteration when the first loss functions of the iterated initial travel query model are all minimized, and generating a travel query model.

Optionally, the historical driving information further includes historical driving information and a travel distance of the historical travel, and the energy consumption pre-estimation model is generated by the following method:

dividing the historical road condition information and the historical driving information of the historical journey in the training sample data into a training sample set and a verification sample set by adopting the journey distance;

inputting the training sample set into a preset initial energy consumption estimation model for iteration to generate a corresponding second predicted value;

comparing the second predicted value with a preset second reference value, and performing reverse training on the initial energy consumption estimated model according to the comparison result to generate a trained energy consumption estimated model;

and performing cross verification on the trained energy consumption estimation model according to the verification sample set to generate an energy consumption estimation model.

Optionally, the inputting the training sample set into a preset initial energy consumption estimation model for iteration to generate a corresponding second predicted value includes:

vectorizing the historical road condition information and the historical driving information of the training sample set to generate a second training vector;

Inputting the second training vector into a preset initial energy consumption estimation model for iteration, and calculating a plurality of second loss functions of the initial travel query model after each iteration;

comparing the second predicted value with a preset second reference value, and performing reverse training on the initial energy consumption estimation model according to the comparison result to generate a trained energy consumption estimation model, wherein the method comprises the following steps of:

and stopping iteration when a plurality of second loss functions of the iterated initial energy consumption pre-estimated model are minimized, and generating a first travel query model.

Optionally, the cross-verifying the trained energy consumption pre-estimation model according to the verification sample set, to generate the energy consumption pre-estimation model, including:

vectorizing the historical road condition information and the historical driving information of the verification sample set to generate a verification feature vector;

inputting the verification feature vector into the first travel inquiry model for cross verification, and calculating a plurality of verification error values of the verified first travel inquiry model;

judging whether the verification error values meet a preset error threshold value or not;

if yes, taking the first journey query model meeting the preset error threshold as the journey query model.

The embodiment of the invention also discloses a device for analyzing the energy consumption of the vehicle, which comprises the following components:

the vehicle driving information acquisition module is used for acquiring vehicle driving information of a vehicle target journey, wherein the vehicle driving information at least comprises target road condition information and vehicle energy consumption;

the similar journey determining module is used for obtaining at least one similar journey aiming at the target journey according to the target road condition information and a preset journey query model;

the optimal journey generation module is used for generating at least one optimal journey for the target journey according to the sample driving information of at least one similar journey and the target road condition information;

the sample energy consumption determining module is used for inputting the sample driving information corresponding to the optimized journey and the target road condition information into a preset energy consumption estimating model to generate sample energy consumption aiming at the optimized journey;

and the energy consumption suggestion output module is used for respectively comparing the sample energy consumption of at least one optimized journey with the vehicle energy consumption and outputting an energy consumption suggestion for the target journey according to a comparison result.

Optionally, the similar travel determination module includes:

the travel sample determining sub-module is used for inputting the target road condition information into a preset travel query model to obtain at least one travel sample aiming at the target travel;

The similarity obtaining sub-module is used for comparing the travel sample with the target travel by position points to obtain travel similarity between the travel sample and the target travel;

and the similar stroke determining sub-module is used for taking a stroke sample with the stroke similarity being greater than or equal to a preset similarity threshold as a similar stroke for the target stroke.

Optionally, the optimization journey generation module is specifically configured to:

Optionally, the sample energy consumption determining module includes:

the energy consumption coefficient acquisition sub-module is used for acquiring an energy consumption coefficient aiming at the energy consumption of the vehicle;

the energy consumption calculation sub-module is used for generating target vehicle energy consumption by adopting the energy consumption coefficient and the vehicle energy consumption;

the energy consumption comparison sub-module is used for selecting an optimized journey of which the sample energy consumption is smaller than or equal to the target vehicle energy consumption as a target optimized journey and acquiring target vehicle driving information of the target optimized journey;

and the energy consumption suggestion output sub-module is used for outputting the energy consumption suggestion matched with the comparison result according to the comparison result between the vehicle driving information of the target journey and the target vehicle driving information of the target optimization journey.

Optionally, the vehicle driving information further includes target driving information, and the energy consumption advice output submodule is specifically configured to:

Optionally, the journey query model is generated by:

the system comprises a training sample data acquisition module, a data processing module and a data processing module, wherein the training sample data is used for acquiring training sample data, the training sample data comprises historical trips of different vehicles and historical driving information of the historical trips, and the historical driving information at least comprises historical road condition information;

the first predicted value generation module is used for inputting the historical road condition information into a preset initial travel query model for iteration to generate a corresponding first predicted value;

the first model training module is used for comparing the first predicted value with a preset first reference value, and carrying out reverse training on the initial travel query model according to the comparison result to generate the travel query model.

Optionally, the historical driving information further includes a trip identifier of the historical trip, and the first predicted value generating module is specifically configured to:

the first model training module is specifically configured to:

Optionally, the historical driving information further includes historical driving information and a travel distance of the historical travel, and the energy consumption pre-estimation model is generated by the following modules:

the sample set determining module is used for dividing the historical road condition information and the historical driving information of the historical journey in the training sample data into a training sample set and a verification sample set by adopting the journey distance;

the second predicted value generation module is used for inputting the training sample set into a preset initial energy consumption estimation model for iteration to generate a corresponding second predicted value;

The second model training module is used for comparing the second predicted value with a preset second reference value, and carrying out reverse training on the initial energy consumption estimated model according to the comparison result to generate a trained energy consumption estimated model;

and the model verification module is used for carrying out cross verification on the trained energy consumption estimated model according to the verification sample set to generate an energy consumption estimated model.

Optionally, the second predicted value generating module is specifically configured to:

the second model training module is specifically configured to:

Optionally, the model verification module is specifically configured to:

The embodiment of the invention also discloses a vehicle, which comprises:

one or more processors; and

a computer readable storage medium having instructions stored thereon, which when executed by the one or more processors, cause the vehicle to perform the method as described above.

Embodiments of the invention also disclose a computer-readable storage medium having instructions stored thereon, which when executed by one or more processors, cause the processors to perform the method as described above.

The embodiment of the invention has the following advantages:

in the embodiment of the invention, the vehicle driving information of the target travel of the vehicle can be obtained, wherein the vehicle driving information at least comprises the target road condition information and the vehicle energy consumption, at least one similar travel for the target travel can be obtained according to the target road condition information and the travel query model, then at least one optimized travel is generated according to the sample driving information of the at least one similar travel and the target road condition information, then the sample driving information corresponding to the optimized travel and the target road condition information are input into the energy consumption prediction model to generate the sample energy consumption of the optimized travel, and then the energy consumption comparison is carried out, so that the energy consumption suggestion for the target travel is output according to the energy consumption comparison result between the optimized travel and the target travel of the vehicle, the similar travel is searched through the road condition information, the obtained similar travel is the travel executed by the real vehicle, the feasibility of the energy consumption prediction is improved, the prediction accuracy can be effectively improved based on the energy consumption prediction of the road condition information and the driving information, and the method is applicable to different vehicles, and the universality is high.

Drawings

FIG. 1 is a flow chart of steps of a method for analyzing vehicle energy consumption according to an embodiment of the present invention;

FIG. 2 is a flow chart of steps of another method for analyzing vehicle energy consumption according to an embodiment of the present invention;

fig. 3 is a block diagram of a vehicle energy consumption analysis device according to an embodiment of the present invention.

Detailed Description

In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.

The travel is the distance travelled by the vehicle from power-on, running, ending running to power-off.

The vehicle driving information comprises vehicle signals generated by the vehicle in a certain journey, and the vehicle-mounted system can generate characteristic information corresponding to the current journey by acquiring the vehicle signals. The characteristic information may include at least road condition information and driving information, where the road condition information is used to represent energy consumption characteristic information corresponding to road conditions when the vehicle travels in the journey, for example, energy consumption information corresponding to road condition characteristics such as travel duration, travel distance, total travel duration, uphill time, uphill height, downhill time, etc. of different travel speeds; the driving information is information indicating energy consumption characteristics corresponding to a user operation when the vehicle is traveling on the trip, and is information indicating energy consumption corresponding to driving characteristics such as air conditioner use information, vehicle traveling mode use information, energy recovery mode use information, accelerator pedal use information, and brake pedal use information.

As an example, in a driving process of a certain section of travel, road conditions and user operations of the vehicle are all vehicle energy consumption with different degrees, and for the optimization of the vehicle energy consumption, the energy consumption caused by the road conditions cannot be avoided, so that the energy consumption optimization usually consists in the user operation, and through corresponding energy consumption suggestions, the energy consumption of the vehicle can be reduced in the driving process of the user, and the cruising time of the vehicle is improved. In this regard, a way is needed to diagnose and suggest vehicle energy consumption based on traffic information and driving information of the vehicle.

Therefore, one of the key points of the embodiment of the invention is that by acquiring the vehicle driving information of a certain section of travel of a vehicle, the vehicle driving information comprises target road condition information and target driving information, then obtaining corresponding similar travel through the target road condition information, screening the similar travel to obtain an optimized travel, then inputting the sample driving information of the optimized travel and the target road condition information of the target travel into an energy consumption estimation model to obtain the sample energy consumption of the optimized travel, then carrying out energy consumption comparison, and outputting an energy consumption suggestion for the section of travel according to the comparison result, thereby searching the similar travel through the road condition information, ensuring the possibility of energy consumption prediction for the travel searched by the real vehicle, and effectively improving the accuracy of the prediction based on the energy consumption prediction of the road condition information and the driving information.

Specifically, referring to fig. 1, a step flowchart of a method for analyzing vehicle energy consumption provided by an embodiment of the present invention may specifically include the following steps:

step 101, acquiring vehicle running information of a vehicle target journey, wherein the vehicle running information at least comprises target road condition information and vehicle energy consumption;

in practice, the vehicle driving information may be based on a CAN (Controller Area Network ) signal of the internet of vehicles, after the vehicle-mounted system may acquire a vehicle signal generated in the target journey, the vehicle-mounted system may upload the vehicle signal to the cloud, perform energy consumption prediction calculation by the cloud, reduce the system overhead of the vehicle, and then output energy consumption diagnosis and optimization advice after the cloud returns the energy consumption calculation result, and display the corresponding energy consumption diagnosis result and energy consumption advice through a vehicle-mounted central control screen of the vehicle.

In a specific implementation, the target journey may be a journey that the vehicle is powered up to powered down, or may be a journey that the vehicle is currently travelling, for example, after entering the vehicle, the user starts the vehicle through a vehicle starter and starts travelling, stops and closes the vehicle after travelling a certain distance, and the travelling journey may be taken as a complete journey; the current real-time changing distance can be used as a target journey after the user starts the vehicle and starts to drive, so that real-time energy consumption advice can be provided for the user in the driving process. Alternatively, the embodiment of the present invention is exemplified by taking the target travel as the history travel of the vehicle, and it is to be understood that the present invention is not limited thereto.

After determining a target journey requiring energy consumption diagnosis and energy consumption suggestion, acquiring vehicle driving information of the target journey, wherein the vehicle driving information at least comprises target road condition information and vehicle energy consumption, and the target road condition information is energy consumption characteristic information corresponding to road conditions in the target journey; the vehicle energy consumption is the total energy consumption of the vehicle in the target course.

102, obtaining at least one similar journey aiming at the target journey according to the target road condition information and a preset journey query model;

the travel query model can utilize an open source algorithm k-dimensional tree multidimensional data nearest neighbor search algorithm (KDTREE nearest neighbor search algorithm for short) to construct a model for similar travel search for a sample set, and can find at least one similar travel similar to a target travel according to target road condition information of the target travel, so that the similar travel is searched through the road condition information, the obtained similar travel is a travel executed by a real vehicle, and the energy consumption prediction executable performance is improved

Step 103, generating at least one optimized journey for the target journey according to at least one sample driving information of the similar journey and the target road condition information;

After at least one similar journey is obtained, feature optimization can be performed on the target journey, specifically, each similar journey has corresponding sample driving information, the sample driving information can be energy consumption feature information corresponding to user operation in the similar journey, and at least one optimized journey subjected to feature optimization for the target journey can be obtained by combining sample driving information corresponding to different similar journeys with target road condition information of the target journey, so that energy consumption conditions of the optimized journey can be predicted subsequently.

104, inputting the sample driving information corresponding to the optimized journey and the target road condition information into a preset energy consumption estimation model to generate sample energy consumption for the optimized journey;

the energy consumption prediction model is a model obtained by training based on road condition information and driving information corresponding to different strokes of different vehicles, so that sample driving information and target road condition information corresponding to the optimized strokes can be input into the energy consumption prediction model to obtain sample energy consumption of different optimized strokes, further energy consumption diagnosis and optimization suggestions can be made, the accuracy of prediction can be effectively improved based on the energy consumption prediction of the road condition information and the driving information, and the method is applicable to different vehicles and has high universality.

And step 105, comparing the sample energy consumption of at least one optimized journey with the vehicle energy consumption respectively, and outputting an energy consumption suggestion for the target journey according to the comparison result.

In a specific implementation, the sample energy consumption corresponding to each optimized journey is required to be compared with the vehicle energy consumption of the target journey, whether the sample energy consumption of the optimized journey is smaller than the vehicle energy consumption of the target journey is judged, and if the sample energy consumption is smaller than the vehicle energy consumption, an energy consumption suggestion for the target journey can be further output according to the comparison result of the sample energy consumption and the vehicle energy consumption; if the energy consumption of the sample is greater than or equal to the energy consumption of the vehicle, no further energy consumption advice is needed, so that energy consumption analysis can be effectively performed through comparison of energy consumption among different courses, and the specific energy consumption advice is output, so that a user can refer to the related energy consumption advice in the next driving process, and driving experience of the user is improved while the energy consumption of the vehicle can be reduced.

Referring to fig. 2, a step flow chart of another method for analyzing vehicle energy consumption according to an embodiment of the present invention may specifically include the following steps:

step 201, obtaining vehicle running information of a vehicle target journey, wherein the vehicle running information at least comprises target road condition information and vehicle energy consumption;

Step 202, obtaining at least one similar journey aiming at the target journey according to the target road condition information and a preset journey query model;

in specific implementation, the target road condition information of the target journey can be input into a preset journey query model to obtain at least one journey sample aiming at the target journey, wherein each journey sample is a journey with a certain similarity with the target journey; and then, comparing the position points between each travel sample and the target travel sample to obtain travel similarity between the travel samples and the target travel, for example, obtaining a plurality of geographic position points corresponding to the target travel and a plurality of geographic position points of the travel samples, and comparing longitude and latitude information of the geographic position points corresponding to the travel samples, so that the travel similarity between the travel samples and the target travel is calculated according to the similarity result of the longitude and latitude, thereby the travel sample with the travel similarity greater than or equal to a preset similarity threshold value is taken as a similar travel of the target travel, and the travel sample with the travel similarity less than the preset similarity threshold value is ignored, and the accuracy of the subsequent energy consumption prediction can be effectively improved through similarity detection of the similar travel, and the pertinence of energy consumption diagnosis and optimization suggestion is further improved. Alternatively, if the optimization run is 0, no energy consumption diagnosis and optimization advice is required.

In an alternative embodiment of the present invention, the travel query model may be generated by:

acquiring training sample data, wherein the training sample data comprises historical trips of different vehicles and historical driving information of the historical trips, and the historical driving information at least comprises historical road condition information; inputting the historical road condition information into a preset initial travel query model for iteration to generate a corresponding first predicted value; comparing the first predicted value with a preset first reference value, and performing reverse training on the initial travel query model according to the comparison result to generate a travel query model.

In a specific implementation, for each vehicle, at the end of each trip (each time power-up is the start of the trip and power-down is the end of the trip), the relevant vehicle signals generated in the trip may be counted, and corresponding energy consumption characteristics may be generated, which may specifically include at least the following characteristics:

1) Running duration of 0-30 yards: counting the vehicle speed per second, and adding (1/3600) to the counted time length in unit hour if the speed is lower than 30 codes;

2) Distance travelled by 0-30 yards: counting the speed of the vehicle per second, if the speed is lower than 30 codes, increasing the counting distance (speed/3600) and obtaining the unit kilometer;

3) 30-60 yards of driving duration: counting the vehicle speed per second, and adding (1/3600) to the counted time length in unit hour if the speed is lower than 60 codes;

4) 30-60 yards of travel distance: counting the speed of the vehicle per second, if the speed is lower than 60 codes, increasing the counting distance (speed/3600) and obtaining the unit kilometer;

5) 60-90 yards of travel duration: counting the vehicle speed per second, and adding (1/3600) to the counted time length in unit hour if the speed is lower than 90 codes;

6) 60-90 yards of travel: counting the speed of the vehicle per second, if the speed is lower than 90 codes, increasing the counting distance (speed/3600) and obtaining the unit kilometer;

7) 90-120 yards of travel duration: counting the vehicle speed per second, and adding (1/3600) the counted time length to unit hour if the speed is lower than 120 codes;

8) 90-120 yards travel distance: counting the speed of the vehicle per second, if the speed is lower than 120 codes, increasing the counting distance (speed/3600) and obtaining the unit kilometer;

9) 120 yards or more driving duration: counting the vehicle speed per second, and adding (1/3600) the counted time length to unit hour if the speed is lower than 120 codes;

10 120 yards or more travel distance: counting the speed of the vehicle per second, if the speed is lower than 120 codes, increasing the counting distance (speed/3600) and obtaining the unit kilometer;

11 0 speed duration: when the vehicle speed is 0, the statistical time length is added (1/3600) and the unit hour

12 Total length of travel: adding (1/3600) the statistical time length, and carrying out unit hour;

13 Uphill time): counting pitch angle signals (pitch) of the vehicle per second, and adding (1/3600) to the counted time length if the pitch is smaller than 0, wherein the counted time length is in unit hour;

14 Cumulative ramp height): counting pitch angle signals (pitch) of vehicles per second, if pitch is smaller than 0, adding (1/3600 x sin (abs (pitch))) to the counted time length, wherein the unit km sin is a trigonometric function, and abs obtains pitch absolute value;

15 Downhill time): counting pitch angle signals (pitch) of the vehicle per second, and adding (1/3600) to the counted time length if the pitch is larger than 0, wherein the counted time length is unit hour;

16 Cumulative ramp height): counting pitch angle signals (pitch) of vehicles per second, if pitch is greater than 0, adding (1/3600 x sin (abs (pitch))) to the counted time length, wherein the unit km sin is a trigonometric function, and abs obtains pitch absolute value;

17 Economic mode distance): counting the mode signals of the vehicle per second, and if the mode is an economic mode, increasing the counting distance (speed/3600) per kilometer;

18 Standard mode distance): counting the mode signals of the vehicle per second, and if the mode is a standard mode, increasing the counting distance (speed/3600) per kilometer;

19 Distance of motion pattern): counting the mode signals of the vehicle per second, and if the mode is a motion mode, increasing the counting distance (speed/3600) per kilometer;

20 Weak recovery distance): counting the vehicle energy recovery mode signal per second, and if the mode is weak recovery, increasing the counting distance (speed/3600) per kilometer;

21 Recovery distance in (c): counting the vehicle energy recovery mode signal per second, and if the mode is medium recovery, increasing the counting distance (speed/3600) per kilometer;

22 Strong recovery distance: counting the vehicle energy recovery mode signal per second, and if the mode is strong recovery, increasing the counting distance (speed/3600) per kilometer;

23 Pedal time: counting brake pedal signals of the vehicle per second, and adding (1/3600) to the counted time length if the brake pedal signals are larger than 0, wherein the counted time length is equal to unit hour;

24 Counting the brake pedal signals of the vehicle per second, and if the brake pedal signals are larger than 0, increasing the counted distance (speed/3600) per kilometer;

25 Counting the accelerator pedal signal of the vehicle per second, and adding (1/3600) to the counted time length if the accelerator pedal signal is greater than 0, wherein the counted time length is equal to unit hour;

26 Counting the accelerator pedal signals of the vehicle per second, and if the accelerator pedal signals are larger than 0, increasing the counted distance (speed/3600) per kilometer;

27 Cumulative accelerator pedal depth): counting the accelerator pedal signals of the vehicle per second, and if the accelerator pedal signals are larger than 0, increasing the accelerator pedal signal value to a statistic depth without units;

28 Deceleration distance): counting the speed signal of the vehicle per second, and if the speed signal is smaller than the speed signal of the previous second, increasing the counting distance (speed/3600) per kilometer;

29 Acceleration distance): counting the speed signal of the vehicle per second, and if the speed signal is greater than the speed signal of the previous second, increasing the counting distance (speed/3600) per kilometer;

30 Air conditioner on time period): counting the air conditioner starting signal of each second of vehicle, and adding (1/3600) the counting time length to unit hour if the air conditioner starting signal is larger than 0;

31 Average air conditioning load): counting air conditioning current and voltage signals of the vehicle per second, and dividing the accumulated current by voltage/36000000, namely, the unit kilowatt-hour, by the air conditioning starting time at the end of the stroke;

32 Total distance travelled): statistical distance increase (speed/3600), unit kilometers;

33 Total energy consumption): the current of the power battery per second is accumulated by voltage/36000000 and is measured in kilowatt-hours.

The energy consumption characteristics except the total driving distance and the average air conditioning load are divided by the total travel distance and multiplied by 100 to obtain hundred kilometers of average energy consumption; further, the energy consumption characteristic 33, divided by the total distance of travel times 100, yields a hundred kilometer average energy consumption. The energy consumption characteristics 1-16 can return road condition information, and the energy consumption characteristics 17-31 are called driving information.

Specifically, after the energy consumption characteristics (i.e., the historical driving information) are obtained, the historical driving information can be analyzed to obtain the travel marks, the historical road condition information and the like of different historical travel, then the travel marks and the historical road condition information are vectorized to generate a first training vector, the first training vector is input into a preset initial formation query model for iteration, a plurality of first loss functions of the initial travel query model after each iteration are calculated, and when the plurality of first loss functions of the initial travel query model after the iteration are minimized, the iteration is stopped to generate the travel query model.

In one example, assuming that the trip query model includes an input layer, a mapping layer, and an output layer, the output layer may include a plurality of output nodes, model training may be performed for each sample using 16-dimensional feature vectors (i.e., energy consumption features corresponding to 17-31 above) and trip identification. Meanwhile, for the parameter updating of the model activation function, the parameter can be updated in the target gradient direction based on the gradient descent strategy. Specifically, a learning rate can be preset, and the updating step length of the parameters in each iteration is controlled, so that a travel inquiry model is finally obtained. In addition, in practice, since the minimum value of the loss function is often difficult to reach, the model iteration can be controlled by setting the iteration times, and when the loss function reaches the expected value or remains unchanged basically, the model training can be regarded as ending.

Step 203, performing information combination by using the target road condition information and the sample driving information corresponding to the similar route respectively, so as to generate at least one optimized route aiming at the target route;

In an example, according to the obtained N similar routes with the similarity greater than or equal to the preset similarity threshold, sample driving information of each similar route is obtained respectively, and then the target road condition information of the target route is combined with the sample driving information of the N similar routes respectively to generate N complete samples, which can be used as an optimized route, and the method may include: target road condition information (x1...x16 and other 16 target road condition information) of the target journey, and similar journey 1 sample driving information (x17..x31) takes 31 dimension factors as an optimized journey 1; target road condition information of target journey (x1...x 16, etc. 16 target road condition information), similar journey 2 sample driving information (x17..x 31) total 31 dimension factors as optimized journey 2; target road condition information (x1...x 16 and other 16 target road condition information) of the target journey, 31 dimension factors are taken as the optimized journey 3 in total of similar journey 3 sample driving information (x 17..x 31), and the accuracy of follow-up energy consumption diagnosis and optimization suggestion is improved through fine adjustment and optimization of the energy consumption characteristics of the target journey.

Step 204, inputting the sample driving information corresponding to the optimized journey and the target road condition information into a preset energy consumption estimation model, and generating sample energy consumption for the optimized journey;

In an optional embodiment of the present invention, the historical driving information further includes historical driving information and a travel distance of the historical travel, and the energy consumption prediction model may be generated by:

dividing historical road condition information and historical driving information of historical journey in training sample data into a training sample set and a verification sample set by adopting journey distance; inputting the training sample set into a preset initial energy consumption estimation model for iteration to generate a corresponding second predicted value; comparing the second predicted value with a preset second reference value, and performing reverse training on the initial energy consumption estimated model according to the comparison result to generate a trained energy consumption estimated model; and performing cross verification on the trained energy consumption pre-estimation model according to the verification sample set to generate the energy consumption pre-estimation model.

In a specific implementation, the trip distance of the historical trip may be obtained first, and the training sample data may be divided into a plurality of trip sets according to the mileage of the trip distance, for example, the historical driving information of different vehicles within 180 days may be used as the training sample data, and then divided into 4 trip sets according to the trip distance, including: a travel set (1) with a travel distance of less than or equal to 5 km, a travel set (2) with a travel distance of more than 5 km and less than or equal to 25 km, a travel distance (3) with a travel distance of more than 25 km and less than or equal to 50 km, a travel set (4) with a travel distance of more than 50 km, etc., then the 4 travel sets can be randomly sampled, 70% of the sample data in the travel set is used as a training sample set, and 30% of the sample data is used as a verification sample set. Alternatively, the training sample set may be used as model training data, and the verification sample set may be used as model verification data after model training.

Specifically, the historical road condition information and the historical driving information of the training sample set can be vectorized to generate a second training vector, the second training vector is input into a preset initial energy consumption estimation model to iterate, a plurality of second loss functions of the initial travel query model after each iteration are calculated, and when the plurality of second loss functions of the initial energy consumption estimation model after each iteration are minimized, iteration is stopped to generate the first travel query model. After training is finished, the historical road condition information and the historical driving information of the verification sample set can be vectorized to generate verification feature vectors, the verification feature vectors are input into the first travel query model to carry out cross verification, a plurality of verification error values of the verified first travel query model are calculated, whether the plurality of verification error values meet a preset error threshold value or not is judged, and if yes, the first travel query model meeting the preset error threshold value is used as a travel query model.

In one example, after obtaining 4 travel sets, the following operations may be performed, respectively: randomly sampling 70% as a training sample set, 30% as a test sample set, taking the energy consumption characteristic X in the training sample set as a model input, taking the average energy consumption of hundred kilometers of travel as a sample regression target y (namely model output), and training a GBDT (Gradient Boosting Decision Tree, gradient lifting decision tree) regression model; 4 energy consumption pre-estimated models are obtained. And then, verifying the energy consumption estimation model according to the energy consumption characteristics in the verification sample set, so that the accuracy of model prediction is ensured, and the pertinence of energy consumption diagnosis and optimization suggestion is further improved.

It should be noted that the embodiments of the present invention include, but are not limited to, the foregoing examples, and it will be understood that those skilled in the art may also set the embodiments according to actual needs under the guidance of the present invention, and the present invention is not limited thereto.

Step 205, comparing the sample energy consumption of at least one optimized journey with the vehicle energy consumption, and outputting an energy consumption suggestion for the target journey according to the comparison result.

In the embodiment of the invention, the energy consumption coefficient aiming at the energy consumption of the vehicle can be obtained, then the energy consumption coefficient and the energy consumption of the vehicle are adopted to generate the energy consumption of the target vehicle, then the optimal journey with the sample energy consumption smaller than or equal to the energy consumption of the target vehicle is selected as the target optimal journey, the target vehicle driving information of the target optimal journey is obtained, and the energy consumption suggestion matched with the comparison result is output according to the comparison result between the vehicle driving information of the target journey and the target vehicle driving information of the target optimal journey.

In a specific implementation, the energy consumption coefficient may be a coefficient set for the vehicle energy consumption, which may be used as a coefficient for evaluating the energy consumption of the optimized journey, for example, the energy consumption coefficient may be set to 0.9, 0.8, 0.7, and the like, and accordingly, in the process of comparing the sample energy consumption with the vehicle energy consumption, that is, the sample energy consumption represented as the optimized journey is 10%, 20%, 30% and the like lower than the vehicle energy consumption of the target journey.

In one example, the energy consumption prediction model outputs hundred kilometers average energy consumption corresponding to the optimized travel, the vehicle energy consumption of the target travel is also hundred kilometers average energy consumption, and then the hundred kilometers average energy consumption corresponding to the optimized travel can be compared with the hundred kilometers average energy consumption of the target travel respectively, if the hundred kilometers average energy consumption of the optimized travel is less than or equal to 90% (or 80%) of the hundred kilometers average energy consumption of the target travel, the optimized travel meeting the energy consumption condition is taken as the target optimized travel, sample driving information of the target optimized travel is obtained, then the energy consumption average value of each energy consumption feature (the energy consumption features 17-31) in the sample driving information is counted respectively, and the energy consumption average value is compared with each energy consumption feature in the target driving information in the target travel respectively one by one to generate a comparison result aiming at the target driving information, for example, if the economic mode distance of the optimized travel is higher than the target travel or the motion mode distance of the optimized travel is lower than the target travel, the economic mode is considered in the proposal or the service time of the motion mode is reduced, so as to reduce the energy consumption; if the distance between the strong energy recovery mode of the optimized travel is higher than that of the target travel or the distance between the weak energy recovery mode of the optimized travel is lower than that of the target travel, outputting, and considering the strong energy recovery mode or reducing the service time of the weak energy recovery mode in the recommended travel so as to reduce the energy consumption; if the driving factors 23-29 of the optimized travel are lower than the target travel, outputting, and reducing unnecessary rapid acceleration and deceleration in the recommended travel, and keeping constant-speed stable travel so as to reduce energy consumption; if the using time of the air conditioner of the optimized journey is lower than that of the target journey or the average load of the air conditioner of the optimized journey is lower than that of the target journey, outputting, reasonably using the air conditioner in the recommended journey to reduce energy consumption and the like, and finally outputting the target road condition information, the target driving information, the sample driving information of the optimized journey with the lowest energy consumption and corresponding energy consumption advice through a vehicle-mounted central control screen of the vehicle, so that a user can refer to the energy consumption advice to drive in the next driving process, the driving energy consumption of the vehicle can be effectively reduced to a certain extent, the cruising ability of the vehicle is improved, and the driving experience of the user is ensured.

It should be noted that, for simplicity of description, the method embodiments are shown as a series of acts, but it should be understood by those skilled in the art that the embodiments are not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the embodiments. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred embodiments, and that the acts are not necessarily required by the embodiments of the invention.

Referring to fig. 3, a block diagram of a vehicle energy consumption analysis device according to an embodiment of the present invention is shown, which may specifically include the following modules:

the vehicle running information obtaining module 301 is configured to obtain vehicle running information of a target journey of a vehicle, where the vehicle running information at least includes target road condition information and vehicle energy consumption;

the similar journey determining module 302 is configured to obtain at least one similar journey for the target journey according to the target road condition information and a preset journey query model;

an optimized trip generation module 303, configured to generate at least one optimized trip for the target trip according to the sample driving information of at least one similar trip and the target road condition information;

the sample energy consumption determining module 304 is configured to input sample driving information corresponding to the optimized trip and the target road condition information into a preset energy consumption estimating model, and generate sample energy consumption for the optimized trip;

and the energy consumption suggestion output module 305 is configured to compare the sample energy consumption of at least one of the optimized routes with the vehicle energy consumption, and output an energy consumption suggestion for the target route according to the comparison result.

In an alternative embodiment of the present invention, the similar travel determination module 302 includes:

In an alternative embodiment of the present invention, the optimization travel generation module 303 is specifically configured to:

In an alternative embodiment of the present invention, the sample energy consumption determination module 304 includes:

In an optional embodiment of the present invention, the vehicle driving information further includes target driving information, and the energy consumption advice output submodule is specifically configured to:

In an alternative embodiment of the invention, the journey query model is generated by the following modules:

In an optional embodiment of the present invention, the historical driving information further includes a trip identifier of the historical trip, and the first predicted value generating module is specifically configured to:

the first model training module is specifically configured to:

In an optional embodiment of the present invention, the historical driving information further includes historical driving information and a trip distance of the historical trip, and the energy consumption prediction model is generated by the following modules:

In an optional embodiment of the invention, the second predicted value generating module is specifically configured to:

The second model training module is specifically configured to:

In an alternative embodiment of the present invention, the model verification module is specifically configured to:

For the device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments for relevant points.

The embodiment of the invention also provides a vehicle, which comprises:

one or more processors; and

a computer readable storage medium having instructions stored thereon, which when executed by the one or more processors, cause the vehicle to perform the method of the embodiments of the present invention.

Embodiments of the present invention also provide a computer-readable storage medium having instructions stored thereon, which when executed by one or more processors, cause the processors to perform the methods described in the embodiments of the present invention.

In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described by differences from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other.

It will be apparent to those skilled in the art that embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Moreover, embodiments of the invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, EEPROM, flash, eMMC, etc.) having computer-usable program code embodied therein.

Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the embodiments of the invention.

Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.

The above description of the method and the device for analyzing vehicle energy consumption provided by the invention applies specific examples to illustrate the principles and embodiments of the invention, and the above examples are only used to help understand the method and the core idea of the invention; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present invention, the present description should not be construed as limiting the present invention in view of the above.

Claims

1. A method for analyzing energy consumption of a vehicle, comprising:

2. The method of claim 1, wherein the obtaining at least one similar trip for the target trip according to the target traffic information and a preset trip query model comprises:

3. The method of claim 1, wherein the generating at least one optimized trip for the target trip based on the sample driving information for at least one of the similar trips and the target road condition information comprises:

4. The method of claim 1, wherein comparing the sample energy consumption of at least one of the similar trips with the vehicle energy consumption, respectively, and outputting an energy consumption recommendation for the target trip based on the comparison result, comprises:

5. The method according to claim 4, wherein the vehicle travel information further includes target driving information, the outputting of the energy consumption advice for the target trip based on a result of comparison between the vehicle travel information of the target trip and the target vehicle travel information of the target optimized trip includes:

6. The method of claim 1, wherein the travel query model is generated by:

7. The method of claim 6, wherein the historical driving information further includes a trip identifier of the historical trip, and the inputting the historical road condition information into a preset initial trip query model for iteration generates a corresponding first predicted value, including:

8. The method of claim 6, wherein the historical driving information further includes historical driving information and a trip distance of the historical trip, the energy consumption prediction model being generated by:

9. The method of claim 8, wherein the inputting the training sample set into a preset initial energy consumption prediction model for iteration to generate a corresponding second predicted value comprises:

10. The method of claim 9, wherein the cross-validating the trained energy consumption pre-estimation model according to the validation sample set to generate the energy consumption pre-estimation model comprises:

11. An analysis device for vehicle energy consumption, characterized by comprising:

12. A vehicle, characterized by comprising:

one or more processors; and

A computer readable storage medium having instructions stored thereon, which when executed by the one or more processors, cause the vehicle to perform the method of any of claims 1-10.

13. A computer-readable storage medium having instructions stored thereon, which when executed by one or more processors, cause the processors to perform the method of any of claims 1-10.