CN117575166B - Road traffic carbon emission prediction method and device based on deep learning - Google Patents
Road traffic carbon emission prediction method and device based on deep learning Download PDFInfo
- Publication number
- CN117575166B CN117575166B CN202311719331.6A CN202311719331A CN117575166B CN 117575166 B CN117575166 B CN 117575166B CN 202311719331 A CN202311719331 A CN 202311719331A CN 117575166 B CN117575166 B CN 117575166B
- Authority
- CN
- China
- Prior art keywords
- emission
- vehicles
- equivalent greenhouse
- greenhouse gas
- vehicle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 title claims abstract description 35
- 229910052799 carbon Inorganic materials 0.000 title claims abstract description 35
- 238000013135 deep learning Methods 0.000 title claims abstract description 17
- 239000005431 greenhouse gas Substances 0.000 claims abstract description 114
- 239000007789 gas Substances 0.000 claims abstract description 42
- 238000012544 monitoring process Methods 0.000 claims abstract description 40
- 238000012937 correction Methods 0.000 claims abstract description 31
- 238000012549 training Methods 0.000 claims abstract description 22
- 239000013598 vector Substances 0.000 claims description 70
- 238000001514 detection method Methods 0.000 claims description 42
- 230000008569 process Effects 0.000 claims description 12
- 238000012800 visualization Methods 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 3
- 230000002159 abnormal effect Effects 0.000 claims description 2
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 10
- 229910002092 carbon dioxide Inorganic materials 0.000 description 5
- 239000001569 carbon dioxide Substances 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- MWUXSHHQAYIFBG-UHFFFAOYSA-N nitrogen oxide Inorganic materials O=[N] MWUXSHHQAYIFBG-UHFFFAOYSA-N 0.000 description 5
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 description 4
- 229910002091 carbon monoxide Inorganic materials 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 239000000295 fuel oil Substances 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 2
- 239000013589 supplement Substances 0.000 description 2
- 238000002485 combustion reaction Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000000586 desensitisation Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000005183 environmental health Effects 0.000 description 1
- 239000003344 environmental pollutant Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 231100000719 pollutant Toxicity 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
- 239000012855 volatile organic compound Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Software Systems (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Operations Research (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a road traffic carbon emission prediction method and device based on deep learning, and relates to the field of traffic environment monitoring. The method comprises the steps of training a preliminary prediction model to be converged by taking contour features and running speed of a vehicle as input layers and taking the type of the vehicle and the emission rate of equivalent greenhouse gases as output layers; taking the detected emission concentration of each kind of tail gas as an input layer, taking the emission rate of the equivalent greenhouse gas of the vehicle as an output layer, and training the correction prediction model until convergence; setting monitoring points on the road; acquiring outline characteristics, running speed and emission concentration of tail gas of each type of passing vehicles at the monitoring points; and respectively inputting the profile characteristics, the running speed and the emission concentration of each kind of tail gas of the passing vehicles into the preliminary prediction model and the correction prediction model to obtain the equivalent greenhouse gas emission rate of the passing vehicles. The invention improves the accuracy of carbon emission prediction.
Description
Technical Field
The invention belongs to the technical field of traffic environment monitoring, and particularly relates to a road traffic carbon emission prediction method and device based on deep learning.
Background
Road traffic is one of the main sources of global carbon emissions, with significant environmental impact. In order to reduce carbon emissions, accurate prediction of carbon emissions from road traffic is required for effective management and control.
Traditional carbon emission prediction methods are mainly based on statistical models, such as linear regression, multiple linear regression, time series analysis, and the like. However, these methods have limited ability to handle complex, non-linear data relationships, and thus the prediction results may be subject to large errors.
Disclosure of Invention
The invention aims to provide a road traffic carbon emission prediction method and device based on deep learning, which are used for carrying out carbon emission detection on road traffic vehicles by obtaining a preliminary prediction model and a correction prediction model through historical data training, so that the accuracy of carbon emission prediction is improved.
In order to solve the technical problems, the invention is realized by the following technical scheme:
the invention provides a road traffic carbon emission prediction method based on deep learning, which comprises the following steps of,
Acquiring contour features of a plurality of types of vehicles;
Testing to obtain the detection emission concentration of a plurality of types of tail gases of different types of vehicles at different running speeds and the detection emission rate of equivalent greenhouse gases;
acquiring contour features of different types of vehicles;
Taking the contour features and the running speed of the vehicle as input layers, taking the types of the vehicle and the emission rate of equivalent greenhouse gases as output layers, and training the preliminary prediction model until convergence;
Taking the detected emission concentration of each kind of tail gas as an input layer, taking the emission rate of the equivalent greenhouse gas of the vehicle as an output layer, and training the correction prediction model until convergence;
Setting monitoring points on the road;
acquiring outline characteristics, running speed and emission concentration of tail gas of each type of passing vehicles at the monitoring points;
And respectively inputting the profile characteristics, the running speed and the emission concentration of each kind of tail gas of the passing vehicles into the preliminary prediction model and the correction prediction model to obtain the equivalent greenhouse gas emission rate of the passing vehicles.
The invention also discloses a road traffic carbon emission prediction method based on deep learning, which comprises the following steps,
Setting a plurality of monitoring points on a road;
continuously acquiring the equivalent greenhouse gas emission rate of the passing vehicles at each monitoring point;
And obtaining the equivalent greenhouse gas emission rate of the road passing vehicle according to the average value of the equivalent greenhouse gas emission rates of the passing vehicles at different moments at the monitoring points.
The invention also discloses a road traffic carbon emission prediction method based on deep learning, which comprises the following steps,
Acquiring a road space model;
Marking the positions of monitoring points in the road space model;
Acquiring the equivalent greenhouse gas emission rate of the passing vehicles at the monitoring points;
And displaying the equivalent greenhouse gas emission rate of the passing vehicles of the monitoring points in the road space model.
The invention also discloses a road traffic carbon emission prediction device based on deep learning, which comprises,
The model training module is used for acquiring contour features of a plurality of types of vehicles;
Testing to obtain the detection emission concentration of a plurality of types of tail gases of different types of vehicles at different running speeds and the emission rate of equivalent greenhouse gases;
acquiring contour features of different types of vehicles;
Taking the contour features and the running speed of the vehicle as input layers, taking the types of the vehicle and the emission rate of equivalent greenhouse gases as output layers, and training the preliminary prediction model until convergence;
Taking the detected emission concentration of each kind of tail gas as an input layer, taking the emission rate of the equivalent greenhouse gas of the vehicle as an output layer, and training the correction prediction model until convergence;
the global prediction module is used for setting monitoring points on the road;
The recognition prediction module is used for acquiring the contour characteristics, the running speed and the emission concentration of each kind of tail gas of the passing vehicles at preset monitoring points;
The contour features, the running speed and the emission concentration of each kind of tail gas of the passing vehicles are respectively input into the preliminary prediction model and the correction prediction model to obtain the equivalent greenhouse gas emission rate of the passing vehicles;
The global prediction module is also used for setting a plurality of monitoring points on the road;
Obtaining the equivalent greenhouse gas emission rate of the road passing vehicles according to the average value of the equivalent greenhouse gas emission rates of the passing vehicles at different moments at each monitoring point;
The visualization module is used for acquiring a road space model;
Marking the positions of monitoring points in the road space model;
acquiring the emission rate of equivalent greenhouse gases of passing vehicles;
And displaying the equivalent greenhouse gas emission rate of the passing vehicles of the monitoring points in the road space model.
According to the method, a preliminary prediction model and a correction prediction model for predicting the carbon emission rate are obtained through historical detection data training. In the prediction process, firstly, the emission rate of equivalent greenhouse gases of passing vehicles is obtained according to a preliminary prediction model, and verification is carried out on the emission rate. If the verification is passed, the equivalent greenhouse gas emission rate is obtained more accurately through correcting the prediction model if the verification is not passed. In this process, the accuracy of carbon emission prediction is improved by a dual means.
Of course, it is not necessary for any one product to practice the invention to achieve all of the advantages set forth above at the same time.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of functional modules and information flow of a deep learning-based road traffic carbon emission prediction device according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating steps performed by the model training module and the recognition prediction module according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating a global prediction module according to an embodiment of the present invention;
FIG. 4 is a flow chart illustrating steps of a visualization module according to an embodiment of the present invention;
FIG. 5 is a flowchart illustrating the step S8 according to an embodiment of the present invention;
FIG. 6 is a flowchart illustrating the step S81 according to an embodiment of the present invention;
FIG. 7 is a flowchart illustrating the step S812 of the present invention according to an embodiment;
FIG. 8 is a flowchart illustrating the steps of step S8122 according to an embodiment of the present invention;
FIG. 9 is a flowchart illustrating the step S84 according to an embodiment of the present invention;
fig. 10 is a flowchart illustrating a step of step S845 according to an embodiment of the invention.
In the drawings, the list of components represented by the various numbers is as follows:
1-model training module, 2-global prediction module, 3-recognition prediction module and 4-visualization module.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present application more apparent, the embodiments of the present application will be described in further detail with reference to the accompanying drawings.
It should be noted that the terms "first," "second," and the like herein are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the application described herein may be implemented in sequences other than those illustrated or otherwise described herein. The implementations described in the following exemplary examples do not represent all implementations consistent with the application. Rather, they are merely examples of apparatus and methods consistent with aspects of the application as detailed in the accompanying claims.
Carbon emissions refer to the amount of carbon dioxide (CO 2) gas released during energy consumption and production. It is one of the main greenhouse gases, and has an important influence on climate change. Road traffic is one of the important sources of carbon emissions. The combustion process of motor vehicles such as automobiles, trucks and motorcycles generates large amounts of carbon dioxide (CO 2) gas and is released into the atmosphere. This results in carbon emissions from road traffic.
In addition to carbon dioxide, automobile exhaust may contain other greenhouse gases and pollutants, such as carbon monoxide (CO), nitrogen oxides (NOx), and non-methane volatile organic compounds (NMVOCs). These substances also have a certain influence on the air quality and on the environmental health.
In order to predict the carbon emission of road traffic, please refer to fig. 1 to 4, the invention provides a road traffic carbon emission prediction device based on deep learning, which belongs to a device with multi-application module interaction, and specifically comprises a model training module 1, a global prediction module 2, an identification prediction module 3 and a visualization module 4. Of course, these divisions are not strictly according to the device configuration, but only the units that are implemented and have certain functions are divided, for example, the model training module 1 is used for training to obtain available preliminary prediction models and correction prediction models, and may be a series of test devices and calculation units and combinations in the practical process. The global prediction module 2 may be a calculation unit for global planning, the recognition prediction module 3 may be a combination of a speed sensor, an image acquisition camera, an exhaust gas detection sensor and a calculation unit, and the visualization module 4 may be a display screen.
Before each functional module is operated, the model training module 1 is required to train the model, and step S1 may be performed first to acquire contour features of a plurality of kinds of vehicles. Step S2 may then be performed to test for detected emission concentrations of multiple kinds of exhaust gases of different kinds of vehicles at different travel speeds, and for equivalent greenhouse gas emission rates. Step S3 may then be performed to obtain profile features for different kinds of vehicles. This is because different kinds of vehicles have different carbon emission rates, and the profile characteristics of the vehicle body are different. In general, a car has two to four doors and a closed body, the body is low and streamlined, has small front-rear suspension overhang, has smooth contour lines, the roof is in a smooth arc, and has moderate tire size and is in proportion to the body. SUVs (sport utility vehicles) have higher ground clearance and larger body dimensions, can accommodate different ground conditions, have relatively square contour lines, are longer and taller, generally have four doors and larger interior spaces, and have relatively larger tires with stronger off-road capability. Trucks have longer bodies and higher cargo boxes or cargo areas, longer heads, usually one or two cabins, larger tires, higher load carrying capacity, higher bodies, and larger chassis ground clearance. Vehicles with different carbon emission rates can be effectively distinguished by means of profile recognition. The license plate number of the pure electric energy automobile has obvious characteristics and can be identified in a mode of not identifying the outline.
In the model training process, step S4 may be performed to train the preliminary prediction model to convergence with the contour features and the running speed of the vehicle as input layers and the kind of the vehicle and the emission rate of the equivalent greenhouse gas as output layers. Next, step S5 may be performed to train the correction prediction model to converge with the detected emission concentration of each kind of exhaust gas as an input layer and the emission rate of the equivalent greenhouse gas of the vehicle as an output layer. Therefore, two models with different recognition accuracy are obtained, and the recognition prediction efficiency and accuracy are considered.
Before predicting the carbon emissions of the road, the global prediction module 2 is further required to perform step S6 to set a monitoring point on the road. And then the recognition prediction module 3 executes step S7 to acquire the contour characteristics, the running speed and the emission concentration of each kind of tail gas of the passing vehicles at preset monitoring points. Step S8 may be performed to input the profile features, the running speed, and the emission concentrations of the respective kinds of exhaust gases of the passing vehicle to the preliminary prediction model and the correction prediction model, respectively, to obtain the equivalent greenhouse gas emission rate of the passing vehicle.
Because the traffic states of the different positions of the road are different, a plurality of monitoring points can be set to realize more accurate prediction of the carbon emission of the road, and the global prediction module 2 is required to execute step S011 to set a plurality of monitoring points on the road. Next, steps S012 to S013 may be performed to receive and obtain the equivalent greenhouse gas emission rate of the road passing vehicle from the average of the equivalent greenhouse gas emission rates of the passing vehicles at different times at each monitoring point.
In order to intuitively demonstrate the state of the carbon emission of the road, the visualization module 4 may execute step S021 to obtain a road space model. Step S022 may then be performed to mark the location of the monitoring point in the road space model, and step S023 may then be performed to obtain the equivalent greenhouse gas emission rate of the passing vehicle. Finally, step S024 may be performed to display the equivalent greenhouse gas emission rate of the passing vehicle at the monitoring point in the road space model. Of course, the display information can be transmitted through the internet and displayed on various network terminals.
Referring to fig. 5, in the present solution, since a preliminary prediction model and a correction prediction model with different prediction accuracy are adopted, in order to consider the accuracy and the speed of prediction, the preliminary prediction model may be preferentially called, and if the output result is not satisfactory, the correction prediction model is called to correct the result. Specifically, in the implementation process of step S8, step S81 may be performed first to obtain the equivalent greenhouse gas emission ranges of the different types of vehicles at different driving speeds according to the detected emission rates of the equivalent greenhouse gases of the different types of vehicles at different driving speeds. Step S82 may be performed next to input the profile features and the running speed of the passing vehicle to the preliminary prediction model, resulting in the kind of passing vehicle and the equivalent greenhouse gas emission rate. Step S83 may be next performed to determine whether the equivalent greenhouse gas emission rate of the passing vehicle is in the equivalent greenhouse gas emission range at the corresponding travel speed. If yes, step S84 may be performed to output the emission rate of the equivalent greenhouse gas of the passing vehicle, and if not, step S85 may be performed to input the emission concentrations of the exhaust gases of the respective types of the passing vehicle to the correction prediction model, so as to obtain the emission rate of the equivalent greenhouse gas after the correction of the passing vehicle. And finally outputting the equivalent greenhouse gas emission rate of the passing vehicles. In the process, the accuracy and the efficiency of the identification prediction are considered.
To supplement the above-described implementation procedures of step S81 to step S85, source codes of part of the functional modules are provided, and a comparison explanation is made in the annotation section. In order to avoid data leakage involving trade secrets, a desensitization process is performed on portions of the data that do not affect implementation of the scheme, as follows.
The basic flow of the above codes is:
A session of Tensorflow is created and the trained preliminary prediction model and the corrected prediction model are read. And then, operating the preliminary prediction model to obtain a preliminary prediction result. If the preliminary prediction result is within a preset effective range, directly outputting the prediction result; and if the preliminary prediction result is not in the preset effective range, operating the correction prediction model to obtain a corrected prediction result, and outputting the corrected prediction result. And finally closing session.
Referring to fig. 6, since the data set for training the model generally includes a large amount of abnormal data, in order to avoid the insufficient output accuracy of the preliminary prediction model and the correction prediction model, it is necessary to resolve the accuracy of the output junction. Specifically, for each type of vehicle, step S81 described above may first be performed in a specific implementation to acquire the detected emission rates of the equivalent greenhouse gases of the vehicle at a plurality of traveling speeds a plurality of times. Step S812 may be performed next to obtain an emission rate range of the equivalent greenhouse gas of the vehicle at a plurality of traveling speeds from the detected emission rates of the equivalent greenhouse gas of the vehicle at a plurality of traveling speeds. Step S813 may be performed next to fit the equivalent greenhouse gas emission rate ranges of the vehicle at the plurality of driving speeds to obtain the equivalent greenhouse gas emission ranges of the vehicle at the different driving speeds, and step S814 may be performed finally to aggregate the data of the different types of vehicles to obtain the equivalent greenhouse gas emission ranges of the different types of vehicles at the different driving speeds. Of course, in specific practice, it may be set directly by an experienced automobile engineer.
Referring to fig. 7 to 8, the quality of the fuel oil of the vehicle is relatively consistent due to the domestic special system of the fuel oil. Meanwhile, the road condition of the domestic road is good. This allows the equivalent carbon dioxide in the exhaust to be reduced to that affected by the vehicle model and the speed of travel. In other words, the equivalent carbon dioxide emission speeds of vehicles of the same vehicle type in the same speed state converge, and the accuracy of the result output by the preliminary prediction model can be identified according to the principle. In order to obtain the range of the emission rates of the equivalent greenhouse gases of different types of vehicles at a plurality of driving speeds, step S812 may be executed first in a specific implementation process, where step S8121 forms a two-dimensional vector from each detected emission rate of the vehicle and the corresponding driving speed according to the detected emission rates of the equivalent greenhouse gases of the vehicle at the plurality of driving speeds, so as to obtain a plurality of detection feature vectors of the vehicle in the plurality of detection processes. Steps S81221 to S81226 in step S8122 may be performed next. That is, step S81221 may be performed first to select a plurality of detection feature vectors from the plurality of detection feature vectors as reference detection feature vectors. Step S81222 may be performed next to calculate a vector difference modulo length of the acquired other detection feature vector and the reference detection feature vector. Step S81223 may next be performed to combine the other detected feature vectors with the reference detected feature vector having the smallest vector difference modulo length into a vector set. Step S81224 may be performed next to calculate and acquire, as the updated reference detection feature vector, a detection feature vector having the smallest vector difference modulo length with the mean vector of all the detection feature vectors in each vector set. Step S81225 may then be performed to determine whether the reference detection feature vector changes before and after the update in the vector set. If so, steps S81223 to S81225 may be performed continuously to update the generated vector set and the reference detected feature vector, and if not, step S81226 may be performed continuously to set the vector set containing the largest number of detected feature vectors as the target vector set. Step S8123 may be performed next to obtain a plurality of detected emission rates corresponding to each travel speed from the two-dimensional vectors of the detected emission rates and the travel speeds corresponding to the detected feature vectors in the target vector set. Finally, step S8124 may be performed to take the range of the minimum value and the maximum value among the plurality of detected emission rates corresponding to each running speed as the range of the emission rates of the equivalent greenhouse gases of the vehicle at the plurality of running speeds.
To supplement the above-described implementation procedures of step S8121 to step S8124, source codes of part of the functional modules are provided, and a comparison explanation is made in the annotation section.
This procedure first obtains all the detected feature vectors and normalizes them. The feature vectors are then divided into sets by calculating the distances of all feature vectors from all other feature vectors. Next, the reference detected feature vector in each set, that is, the average of all feature vectors in the set, is calculated, and whether these new reference detected feature vectors are identical to the original reference detected feature vectors is compared. If there is any change, the process will be continued recursively, if otherwise the set containing the most feature vectors will be selected as the target set, and all feature vectors in this set will be returned, thereby indicating the emissions range of the vehicle at different speeds.
Referring to fig. 9, in order to make the output result of the correction prediction model more accurate, outliers in the input data may be removed. Specifically, in the specific implementation process of step S84, step S841 may be performed first to obtain the exhaust concentration ratio range of the plurality of types of exhaust gases of the different types of vehicles at the different traveling speeds according to the detected exhaust concentrations of the plurality of types of exhaust gases of the different types of vehicles at the different traveling speeds. Step S842 may then be performed to determine whether the ratio range of the detected exhaust concentration of each type of exhaust gas of the multiple-pass vehicle is within the ratio range of the exhaust concentrations of a plurality of types of exhaust gas of different types of vehicles at different travel speeds. If so, step S843 may be performed next, and if not, culling. Step S844 may be performed to input the detected emission concentrations of the exhaust gases of the respective kinds of the reserved plurality of groups of passing vehicles into the correction prediction model, respectively, so as to obtain a plurality of equivalent greenhouse gas emission rates of the passing vehicles. Finally, step S845 may be executed to obtain the corrected emission rate of the equivalent greenhouse gas of the passing vehicle according to the emission rates of the equivalent greenhouse gases of the passing vehicle.
However, in order to obtain the equivalent greenhouse gas emission rate more closely fitting the real state, referring to fig. 10, step S8451 may be performed to obtain the average running speed of the passing vehicle, which may be the active uploading of the data by the vehicle or the reading of the monitoring data of the traffic management department. Step S8452 may be next performed to acquire a running speed corresponding to the emission rate of each equivalent greenhouse gas of the passing vehicle. Finally, step S8453 may be performed to take the emission rate of the equivalent greenhouse gas corresponding to the running speed having the same or the smallest difference as the emission rate of the equivalent greenhouse gas after the correction of the passing vehicle. If the difference is too large, the running speed corresponding to the emission rate of each equivalent greenhouse gas can be fitted, so that the emission rate of the equivalent greenhouse gas under the average running speed of the vehicle can be obtained.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved.
It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by hardware, such as circuits or ASICs (Application SPECIFIC INTEGRATED circuits), which perform the corresponding functions or acts, or combinations of hardware and software, such as firmware and the like.
Although the invention is described herein in connection with various embodiments, other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the "a" or "an" does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.
The foregoing description of embodiments of the application has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the improvement of technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (9)
1. A road traffic carbon emission prediction method based on deep learning is characterized by comprising the following steps of,
Acquiring contour features of a plurality of types of vehicles;
Testing to obtain the detection emission concentration of a plurality of types of tail gases of different types of vehicles at different running speeds and the detection emission rate of equivalent greenhouse gases;
acquiring contour features of different types of vehicles;
Taking the contour features and the running speed of the vehicle as input layers, taking the types of the vehicle and the emission rate of equivalent greenhouse gases as output layers, and training the preliminary prediction model until convergence;
Taking the detected emission concentration of each kind of tail gas as an input layer, taking the emission rate of the equivalent greenhouse gas of the vehicle as an output layer, and training the correction prediction model until convergence;
Setting monitoring points on the road;
acquiring outline characteristics, running speed and emission concentration of tail gas of each type of passing vehicles at the monitoring points;
The contour features, the running speed and the emission concentration of each kind of tail gas of the passing vehicles are respectively input into the preliminary prediction model and the correction prediction model to obtain the equivalent greenhouse gas emission rate of the passing vehicles;
The step of inputting the profile characteristics, the running speed and the exhaust concentration of each kind of exhaust gas of the passing vehicles to the preliminary prediction model and the correction prediction model to obtain the equivalent greenhouse gas exhaust rate of the passing vehicles, respectively, comprises,
Obtaining the equivalent greenhouse gas emission ranges of different types of vehicles at different running speeds according to the detected emission rates of the equivalent greenhouse gases of the different types of vehicles at different running speeds;
inputting the contour features and the running speed of the passing vehicles into the preliminary prediction model to obtain the types of the passing vehicles and the equivalent greenhouse gas emission rate;
judging whether the emission rate of the equivalent greenhouse gases of the passing vehicles is in the equivalent greenhouse gas emission range under the corresponding running speed;
if yes, outputting the equivalent greenhouse gas emission rate of the passing vehicles;
if not, the emission concentration of each kind of tail gas of the passing vehicles is input into the correction prediction model, and the emission rate of the equivalent greenhouse gas after the passing vehicles are corrected is obtained;
and outputting the emission rate of the equivalent greenhouse gases of the passing vehicles.
2. The method according to claim 1, wherein the step of obtaining the equivalent greenhouse gas emission ranges of the different kinds of vehicles at the different traveling speeds based on the detected emission rates of the equivalent greenhouse gases of the different kinds of vehicles at the different traveling speeds includes,
For each type of vehicle,
The detected emission rates of the equivalent greenhouse gases of the vehicle at a plurality of traveling speeds are acquired a plurality of times,
Obtaining the range of the emission rate of the equivalent greenhouse gas of the vehicle at a plurality of running speeds according to the detected emission rate of the equivalent greenhouse gas of the vehicle at a plurality of running speeds,
Fitting the equivalent greenhouse gas emission rate ranges of the vehicle at a plurality of driving speeds to obtain equivalent greenhouse gas emission ranges of the vehicle at different driving speeds;
and summarizing to obtain the equivalent greenhouse gas emission ranges of different types of vehicles at different running speeds.
3. The method of claim 2, wherein the step of deriving the range of the emission rates of the equivalent greenhouse gases of the vehicle at the plurality of traveling speeds from the detected emission rates of the equivalent greenhouse gases of the vehicle at the plurality of traveling speeds comprises,
According to the detection emission rates of the multiple equivalent greenhouse gases of the vehicle at multiple running speeds, forming a two-dimensional vector by each detection emission rate of the vehicle and the corresponding running speed to obtain multiple detection feature vectors of the vehicle in multiple detection processes;
Removing an abnormal value from the plurality of detection feature vectors to obtain a target vector set;
obtaining a plurality of detection emission rates corresponding to each running speed according to the two-dimensional vectors of the detection emission rates and the running speeds corresponding to the detection feature vectors in the target vector set;
the range of the minimum value and the maximum value in the plurality of detected emission rates corresponding to each running speed is taken as the emission rate range of the equivalent greenhouse gas of the vehicle at the plurality of running speeds.
4. The method of claim 3, wherein the step of culling the set of target vectors from the plurality of detected feature vectors comprises,
Selecting a plurality of detection feature vectors from the plurality of detection feature vectors as reference detection feature vectors;
Calculating and obtaining vector difference modular lengths of other detection feature vectors and reference detection feature vectors;
Forming a vector set by the other detection feature vectors and the reference detection feature vector with the minimum vector difference module length;
Calculating to obtain a detection feature vector with the minimum vector difference modulus length with the mean value vector of all detection feature vectors in each vector set as an updated reference detection feature vector;
judging whether the reference detection feature vectors before and after updating in the vector set change or not;
If yes, continuously updating and generating the vector set and the reference detection feature vector;
If not, the vector set with the largest number of the detection feature vectors is taken as a target vector set.
5. The method according to claim 1, wherein the step of inputting the emission concentration of each kind of exhaust gas of the passing vehicle to the correction prediction model to obtain the corrected equivalent greenhouse gas emission rate of the passing vehicle comprises,
Obtaining the exhaust concentration proportion range of the tail gases of the different types of vehicles at different running speeds according to the detected exhaust concentrations of the tail gases of the different types of vehicles at different running speeds;
Judging whether the proportion range of the detected emission concentration of each kind of tail gas of the passing vehicle is in the proportion range of the emission concentration of a plurality of kinds of tail gas of different kinds of vehicles at different running speeds or not;
If yes, reserving;
If not, rejecting;
Respectively inputting the reserved detection emission concentrations of the tail gases of various types of the passing vehicles into the correction prediction model to obtain the emission rates of a plurality of equivalent greenhouse gases of the passing vehicles;
And obtaining the corrected equivalent greenhouse gas emission rate of the passing vehicle according to the emission rates of the plurality of equivalent greenhouse gases of the passing vehicle.
6. The method of claim 5, wherein the step of deriving the corrected equivalent greenhouse gas emission rate from the plurality of equivalent greenhouse gas emission rates of the passing vehicle comprises,
Acquiring the average running speed of the passing vehicles;
Acquiring a running speed corresponding to the emission rate of each equivalent greenhouse gas of the passing vehicles;
And taking the emission rate of the equivalent greenhouse gas corresponding to the running speed with the same average running speed or the minimum difference value as the emission rate of the equivalent greenhouse gas after the correction of the passing vehicle.
7. A road traffic carbon emission prediction method based on deep learning is characterized by comprising the following steps of,
Setting a plurality of monitoring points on a road;
Continuously acquiring the equivalent greenhouse gas emission rate of the passing vehicles at each monitoring point according to the deep learning-based highway traffic carbon emission prediction method of any one of claims 1 to 6;
And obtaining the equivalent greenhouse gas emission rate of the road passing vehicle according to the average value of the equivalent greenhouse gas emission rates of the passing vehicles at different moments at the monitoring points.
8. A road traffic carbon emission prediction method based on deep learning is characterized by comprising the following steps of,
Acquiring a road space model;
Marking the positions of monitoring points in the road space model;
obtaining the equivalent greenhouse gas emission rate of the passing vehicles at the monitoring points in the deep learning-based highway traffic carbon emission prediction method according to any one of claims 1 to 6;
And displaying the equivalent greenhouse gas emission rate of the passing vehicles of the monitoring points in the road space model.
9. A road traffic carbon emission prediction device based on deep learning is characterized by comprising,
The model training module is used for acquiring contour features of a plurality of types of vehicles;
Testing to obtain the detection emission concentration of a plurality of types of tail gases of different types of vehicles at different running speeds and the emission rate of equivalent greenhouse gases;
acquiring contour features of different types of vehicles;
Taking the contour features and the running speed of the vehicle as input layers, taking the types of the vehicle and the emission rate of equivalent greenhouse gases as output layers, and training the preliminary prediction model until convergence;
Taking the detected emission concentration of each kind of tail gas as an input layer, taking the emission rate of the equivalent greenhouse gas of the vehicle as an output layer, and training the correction prediction model until convergence;
the global prediction module is used for setting monitoring points on the road;
The recognition prediction module is used for acquiring the contour characteristics, the running speed and the emission concentration of each kind of tail gas of the passing vehicles at preset monitoring points;
The contour features, the running speed and the emission concentration of each kind of tail gas of the passing vehicles are respectively input into the preliminary prediction model and the correction prediction model to obtain the equivalent greenhouse gas emission rate of the passing vehicles;
The global prediction module is also used for setting a plurality of monitoring points on the road;
Obtaining the equivalent greenhouse gas emission rate of the road passing vehicles according to the average value of the equivalent greenhouse gas emission rates of the passing vehicles at different moments at each monitoring point;
The visualization module is used for acquiring a road space model;
Marking the positions of monitoring points in the road space model;
acquiring the emission rate of equivalent greenhouse gases of passing vehicles;
Displaying the equivalent greenhouse gas emission rate of the passing vehicles of the monitoring points in the road space model;
Wherein,
The step of inputting the profile characteristics, the running speed and the exhaust concentration of each kind of exhaust gas of the passing vehicles to the preliminary prediction model and the correction prediction model to obtain the equivalent greenhouse gas exhaust rate of the passing vehicles, respectively, comprises,
Obtaining the equivalent greenhouse gas emission ranges of different types of vehicles at different running speeds according to the detected emission rates of the equivalent greenhouse gases of the different types of vehicles at different running speeds;
inputting the contour features and the running speed of the passing vehicles into the preliminary prediction model to obtain the types of the passing vehicles and the equivalent greenhouse gas emission rate;
judging whether the emission rate of the equivalent greenhouse gases of the passing vehicles is in the equivalent greenhouse gas emission range under the corresponding running speed;
if yes, outputting the equivalent greenhouse gas emission rate of the passing vehicles;
if not, the emission concentration of each kind of tail gas of the passing vehicles is input into the correction prediction model, and the emission rate of the equivalent greenhouse gas after the passing vehicles are corrected is obtained;
and outputting the emission rate of the equivalent greenhouse gases of the passing vehicles.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311719331.6A CN117575166B (en) | 2023-12-14 | 2023-12-14 | Road traffic carbon emission prediction method and device based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311719331.6A CN117575166B (en) | 2023-12-14 | 2023-12-14 | Road traffic carbon emission prediction method and device based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117575166A CN117575166A (en) | 2024-02-20 |
| CN117575166B true CN117575166B (en) | 2024-05-28 |
Family
ID=89893916
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311719331.6A Active CN117575166B (en) | 2023-12-14 | 2023-12-14 | Road traffic carbon emission prediction method and device based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117575166B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106845371A (en) * | 2016-12-31 | 2017-06-13 | 中国科学技术大学 | A kind of city road network automotive emission remote sensing monitoring system |
| CN114820259A (en) * | 2022-03-28 | 2022-07-29 | 武汉大学 | Real-time calculation method for road vehicle tail gas carbon dioxide emission based on space-time deep learning model |
| CN115019513A (en) * | 2022-07-18 | 2022-09-06 | 桂林电子科技大学 | Traffic carbon emission monitoring system and monitoring method |
| CN115424448A (en) * | 2022-11-04 | 2022-12-02 | 八维通科技有限公司 | Traffic carbon emission evaluation method and system based on traffic travel data |
| CN115587526A (en) * | 2022-08-24 | 2023-01-10 | 襄阳达安汽车检测中心有限公司 | Neural network-based vehicle carbon dioxide real-time emission prediction method and system |
| CN116229607A (en) * | 2023-05-09 | 2023-06-06 | 深圳市城市交通规划设计研究中心股份有限公司 | Prediction method of running carbon emission of motor vehicle, electronic equipment and storage medium |
| CN116361711A (en) * | 2023-03-31 | 2023-06-30 | 襄阳达安汽车检测中心有限公司 | Method, device, equipment and storage medium for predicting real-time emission of carbon dioxide of automobile |
| CN116468152A (en) * | 2023-03-23 | 2023-07-21 | 东南大学 | Multi-situation regional prediction method for pollutant discharge amount of highway transportation |
| CN116611755A (en) * | 2023-04-25 | 2023-08-18 | 兰州交通大学 | Dangerous goods green heterogeneous vehicle path planning method based on uncertain risk |
| CN117060506A (en) * | 2023-10-12 | 2023-11-14 | 国网天津市电力公司培训中心 | Electric automobile and photovoltaic collaborative optimization method and device considering green electricity charging mode |
-
2023
- 2023-12-14 CN CN202311719331.6A patent/CN117575166B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106845371A (en) * | 2016-12-31 | 2017-06-13 | 中国科学技术大学 | A kind of city road network automotive emission remote sensing monitoring system |
| CN114820259A (en) * | 2022-03-28 | 2022-07-29 | 武汉大学 | Real-time calculation method for road vehicle tail gas carbon dioxide emission based on space-time deep learning model |
| CN115019513A (en) * | 2022-07-18 | 2022-09-06 | 桂林电子科技大学 | Traffic carbon emission monitoring system and monitoring method |
| CN115587526A (en) * | 2022-08-24 | 2023-01-10 | 襄阳达安汽车检测中心有限公司 | Neural network-based vehicle carbon dioxide real-time emission prediction method and system |
| CN115424448A (en) * | 2022-11-04 | 2022-12-02 | 八维通科技有限公司 | Traffic carbon emission evaluation method and system based on traffic travel data |
| CN116468152A (en) * | 2023-03-23 | 2023-07-21 | 东南大学 | Multi-situation regional prediction method for pollutant discharge amount of highway transportation |
| CN116361711A (en) * | 2023-03-31 | 2023-06-30 | 襄阳达安汽车检测中心有限公司 | Method, device, equipment and storage medium for predicting real-time emission of carbon dioxide of automobile |
| CN116611755A (en) * | 2023-04-25 | 2023-08-18 | 兰州交通大学 | Dangerous goods green heterogeneous vehicle path planning method based on uncertain risk |
| CN116229607A (en) * | 2023-05-09 | 2023-06-06 | 深圳市城市交通规划设计研究中心股份有限公司 | Prediction method of running carbon emission of motor vehicle, electronic equipment and storage medium |
| CN117060506A (en) * | 2023-10-12 | 2023-11-14 | 国网天津市电力公司培训中心 | Electric automobile and photovoltaic collaborative optimization method and device considering green electricity charging mode |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117575166A (en) | 2024-02-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107730425A (en) | Carbon emission amount computational methods, device and storage medium | |
| US8996290B2 (en) | Model-based efficiency scoring in automotive engineering | |
| CN106536315A (en) | System for assessing and/or optimising the operating behaviour of a vehicle | |
| Kaymaz et al. | Development of a driving cycle for Istanbul bus rapid transit based on real-world data using stratified sampling method | |
| CN112700201B (en) | Goods source recommending method, electronic equipment and storage medium | |
| CN114742413A (en) | Urban traffic carbon emission monitoring system and method | |
| CN106872001A (en) | A kind of motor-vehicle tail-gas detection method and system | |
| TW201339031A (en) | Method for calculating fuel consumption during driving and driving fuel consumption calculation system | |
| Kropiwnicki et al. | Test for assessing the energy efficiency of vehicles with internal combustion engines | |
| CN111783034B (en) | Emission road spectrum analysis method for vehicle-cloud cooperative computing | |
| CN117575166B (en) | Road traffic carbon emission prediction method and device based on deep learning | |
| CN112819229B (en) | Driving station optimization updating method and system based on distributed machine learning | |
| CN106055727A (en) | Fuel economy data analysis of vehicle infrastrucuture | |
| CN109145401B (en) | Method, system and terminal equipment for calculating emission list of motor vehicle | |
| CN111896264B (en) | Method and device for generating test working condition of range extender engine and electronic equipment | |
| DE102020003690A1 (en) | Method for simulating real driving scenarios of a vehicle using geographical information | |
| CN116048055A (en) | Vehicle fault detection method, device and storage medium | |
| Smirnov et al. | The application of transport telematics for the organization of an innovative system for the organization of the technical maintenance of vehicles | |
| WO2021149340A1 (en) | Abnormality detecting device, abnormality detecting method, and program | |
| US20220108569A1 (en) | Automated detection of vehicle data manipulation and mechanical failure | |
| Makarova et al. | Improving the reliability of autonomous vehicles in a branded service system using big data | |
| CN113642382A (en) | Heavy vehicle identification method based on multi-label target detection | |
| CN115527076A (en) | Construction method and system of identification model for abnormal driving behavior of commercial vehicle | |
| Sabde et al. | Scalable Machine Learning and Analytics of the Vehicle Data to derive Vehicle Health and Driving Characteristics | |
| Lobato et al. | Quantifying Repeatability of Real-World On-Road Driving Using Dynamic Time Warping |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |