CN113642241B - Road network fine particulate matter research method based on traffic running state - Google Patents
Road network fine particulate matter research method based on traffic running state Download PDFInfo
- Publication number
- CN113642241B CN113642241B CN202110942558.1A CN202110942558A CN113642241B CN 113642241 B CN113642241 B CN 113642241B CN 202110942558 A CN202110942558 A CN 202110942558A CN 113642241 B CN113642241 B CN 113642241B
- Authority
- CN
- China
- Prior art keywords
- road
- particulate matter
- time
- fine particulate
- traffic
- 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
- 239000013618 particulate matter Substances 0.000 title claims abstract description 68
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000011160 research Methods 0.000 title claims abstract description 18
- 238000012544 monitoring process Methods 0.000 claims abstract description 94
- 230000005540 biological transmission Effects 0.000 claims abstract description 35
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 18
- 239000010419 fine particle Substances 0.000 claims abstract description 14
- 230000008859 change Effects 0.000 claims abstract description 5
- 239000011159 matrix material Substances 0.000 claims description 24
- 238000004364 calculation method Methods 0.000 claims description 8
- 239000003344 environmental pollutant Substances 0.000 claims description 6
- 231100000719 pollutant Toxicity 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 4
- 238000007418 data mining Methods 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 claims description 2
- 238000012937 correction Methods 0.000 claims description 2
- 238000000605 extraction Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 claims description 2
- 238000012804 iterative process Methods 0.000 claims description 2
- 239000000203 mixture Substances 0.000 claims description 2
- 230000002457 bidirectional effect Effects 0.000 claims 1
- 230000000306 recurrent effect Effects 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 abstract 1
- 238000005259 measurement Methods 0.000 description 4
- 238000003915 air pollution Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 239000003570 air Substances 0.000 description 1
- 239000000809 air pollutant Substances 0.000 description 1
- 231100001243 air pollutant Toxicity 0.000 description 1
- 239000012080 ambient air Substances 0.000 description 1
- 239000000356 contaminant Substances 0.000 description 1
- 230000034994 death Effects 0.000 description 1
- 231100000517 death Toxicity 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 230000002028 premature Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 230000000241 respiratory effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a road network fine particulate matter research method based on traffic running state, which is characterized in that an urban fine particulate matter transmission recognition model is established based on a fine particulate matter region transmission recognition algorithm, so that information such as time, concentration and direction of regional fine particulate matter transmission is obtained, and monitoring interference of fine particulate matter region transmission on local traffic fine particulate matter emission is eliminated; then, extracting the space-time characteristics of road network traffic operation through urban road classification and road network traffic operation actual data, and establishing a road network fine particulate matter space-time emission model considering traffic operation states; the space-time emission model of the road network fine particles starts from the actual traffic running state of the road network and reflects the space-time emission characteristics of the road network fine particles generated by the traffic running of the road network. The invention is beneficial to researching the change condition of urban fine particles, provides a new view angle for purposefully implementing urban traffic guidance and planning urban traffic network, and makes it possible to take efficient traffic control from the urban traffic emission angle.
Description
Technical field:
the invention relates to the field of urban environment research, in particular to a road network fine particulate matter research method based on traffic running states.
The background technology is as follows:
since 2013, air pollution conditions such as haze suddenly intrude into the field of view of people, and knock the police clock for urban air pollution, thereby attracting attention to the air pollution problem. Among all air pollutants, PM2.5 is an important contributor to many diseases, especially in respiratory-related diseases, which have been demonstrated by many epidemic pathological studies over the last decade. The world health organization estimated that pollution of the urban environment caused about 370 tens of thousands of premature deaths in 2012. When a motor vehicle in a city runs, a large amount of exhaust gas is discharged, so that pollution of inhalable fine particles in the city air is increased, and wide attention is brought to people. According to the result released by the ecological environment bureau of Beijing city in 2018, the proportion of mobile sources in the PM2.5 sources in Beijing local is the largest and reaches 45%. Therefore, the influence of the emission of the traffic fine particles in the city on the quality of the ambient air is not negligible, and intensive research is necessary.
In the related research of urban traffic emission, the primary problem is to quantify the emission and pollution level generated by urban traffic operation, which is not only the premise of making scientific and reasonable emission reduction targets and implementing a practical and effective emission reduction scheme by a decision-making department, but also the main difficulty. The motor vehicle is moving as a source of pollution at all times, and monitoring of emissions generated by the motor vehicle is a great difficulty. In theory, the total emission of the city can be obtained by only monitoring and counting the emission of all the motor vehicles in the city, but in actual life, the whole running process of the motor vehicles cannot be monitored and counted in real time in a direct measurement mode due to the limitation of conditions such as technology, equipment and the like, and the motor vehicles in the city are huge in quantity, so that the emission of all the running vehicles in the city cannot be monitored due to the consideration of actual cost. Therefore, the total amount of vehicle emissions in a city must be reasonably estimated by a scientific and reasonable method. Moreover, the emission of the motor vehicle is not stable and unchanged, and is influenced by superposition of various factors such as the running state of the vehicle, the emission control technology of the vehicle, the type of the vehicle and the like, the running state of the traffic must be considered when the emission is arranged, but urban traffic is a complex system, the running state of the urban traffic changes constantly, and if the running state of the traffic changes can not be accurately mastered, the quantification of the emission of the motor vehicle in the running state of the urban traffic is not from talking.
At present, most of researches on traffic emission are based on macroscopic emission of the whole city, so that the urban interior is difficult to refine, and microscopic researches on road emission of partial road sections are difficult to grasp the relation between the traffic running state of the road network and the emission of fine particulate matters of the road network from the whole, so that the existing motor vehicle emission list is difficult to meet the requirement of fine emission management and control. Therefore, the reasonable mode is selected to quantify the influence of urban traffic exhaust emission on urban environment, which is a problem to be solved urgently.
The invention comprises the following steps:
aiming at the problems, the invention adopts a road network fine particulate matter research method based on traffic running state, and the method accurately analyzes the condition of influence of road traffic state and traffic tail gas emission on environment by removing the influence of pollutant regional transmission in other regions and applying various data including pollutant measurement data, taxi traffic driving data and weather data to a road network emission model.
The data sources used in the invention comprise published pollutant measurement data, road side autonomous installed pollutant measurement equipment data, vehicle data and weather data. Wherein the vehicle data includes vehicle number, vehicle instantaneous speed, longitude and latitude information, and steering angle data. The detector data includes contaminant concentration data at different times. The weather data includes regional wind speed, wind direction, temperature, humidity data on the same date as the detector and vehicle data.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
a road network fine particulate matter research method based on traffic running state comprises the following steps:
(1) Establishing an urban fine particulate matter transmission identification model based on a fine particulate matter region transmission identification algorithm, and identifying fine particulate matter region transmission from two dimensions of time and space;
(2) The method comprises the steps of performing type classification on roads in an urban road network according to the actual running state of a road vehicle, and obtaining urban road network classification which is more in line with the actual running state by adopting a self-adaptive density clustering algorithm based on vehicle running parameters;
(3) Using vehicle track data to establish a specific power emission model, and constructing a specific power distribution map-based fine particulate matter emission factor calculation method;
(4) The method for extracting the traffic running state of the road network driven by data is established, and the characteristics of the historical traffic running state are extracted and used as the input of a space-time emission model of the road network, so that the space-time emission of the road network considering the traffic running state is obtained.
In the step 1, the information such as the input monitoring points and the subsequent propagation directions of the regional input fine particulate matters is identified by establishing a monitoring zone threshold matrix, a fine particulate matter regional transmission database is constructed by combining historical data, and the effects of early discovery, real-time monitoring and the like of fine particulate matter regional transmission can be realized by the model through historical data mining and real-time data input.
In the step 2, the non-parameter clustering algorithm performs type fusion by initially generating numerous small classes and calculating the distance and the distance in the class, and finally performs clustering in a mode of calculating the density of the region without the need of initial class parameter setting.
In the step 3, the vehicle driving track data refers to a series of data of recording time series and vehicle GPS position. The method generally comprises the steps of vehicle numbering, recording time, longitude and latitude, speed, direction angle and the like, specific power distribution is established through track data of the vehicle running on an actual road, and compared with a traffic simulation transmission result and a driving simulator result, the motor vehicle running track is the vehicle actual road running data and has the best effect on reflecting an actual road state.
In the step 4, the model takes the road network traffic running state as one of the important inputs of the model, and because of the difficulty in acquiring real-time, high-precision and full-area road network traffic running data, the road network traffic running state database is constructed, the road network traffic running history data is subjected to characteristic learning by using a Bi-recurrent neural network (Bi-LSTM) in a data driving mode, the traffic running state of the road network is extracted by taking the traffic data at the current moment as the input, and the discharge condition of road network traffic pollutants changing along with time can be obtained according to the traffic running state and the road network discharge model.
By adopting the technical scheme, the invention has the following advantages:
(1) According to the actual vehicle running state, the road network roads are classified by using a self-adaptive density clustering algorithm based on the vehicle running state, and compared with the simple classification method based on the road grade in most of the existing researches, the novel classification method is more scientific and reasonable, and accurately reflects the difference of the road vehicle running states.
(2) And selecting a vehicle emission model based on the specific power principle and establishing specific power distribution based on vehicle track data, wherein the method brings the traffic running state on the road into the calculation of traffic emission, so that the result is more close to the emission of fine particles generated when the vehicle actually runs.
(3) The method is characterized in that urban road network space-time emission is calculated based on the road network traffic running state driven by data, the road network traffic running state is used as one of important inputs of the road network emission space-time model when the model is established, and the problem that the network running state is not considered enough and traffic running dynamic characteristics are ignored in the previous research is solved.
Description of the drawings:
FIG. 1 is a flow chart of a road network fine particulate matter research method based on traffic running state of the invention;
FIG. 2 is a fine particulate matter zone transport identification model framework of the present invention;
FIG. 3 is a matrix of fine particulate matter concentration thresholds according to the present invention.
The specific embodiment is as follows:
the following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments.
Example 1.
As shown in fig. 1, a road network fine particulate matter research method based on traffic running state comprises the following steps:
(1) Establishing an urban fine particulate matter transmission identification model based on a fine particulate matter region transmission identification algorithm, and identifying fine particulate matter region transmission from two dimensions of time and space;
by establishing monitoring bandsThe threshold matrix is used for identifying information such as input monitoring points of the regional input fine particulate matters and subsequent propagation directions. The fine particulate matter regional transmission database is constructed by combining historical data, and the model can realize the effects of early discovery, real-time monitoring and the like of fine particulate matter regional transmission through historical data mining and real-time data input. The model frame is shown in fig. 2. The information such as time and concentration of regional fine particulate matter input of a single monitoring point is obtained through a fine particulate matter concentration recognition algorithm of single monitoring point time data, a plurality of monitoring point data are associated, a monitoring zone fine particulate matter threshold matrix is established, and input monitoring points and drift directions of regional input particulate matters can be judged in space. The monitoring area fine particulate matter threshold matrix can also store threshold historical data of different time periods, and the monitoring area fine particulate matter threshold matrix can be judged in real time by combining with the real-time data access of all monitoring points on the identification area, and the monitoring area fine particulate matter threshold matrix can be updated in a self-iterative mode. The fine particulate matter concentration threshold matrix is shown in fig. 3. Through the monitoring point E on the outer and inner monitoring bands n And I n Reference concentration of fine particulate matter at month m time tAnd->Establishing a reference concentration vector of monitoring points of the regional fine particulate matter transmission identification region; by->And->And establishing an upper limit concentration vector of a monitoring point of the regional fine particulate matter transmission identification region. Even if the positions of the monitoring points on the inner layer monitoring and the outer layer monitoring at the same moment t are different, a certain difference exists between the reference concentration and the upper limit concentration of the monitoring points. Monitoring point I on inner layer monitoring band a For example, at month m time t monitoring point I a Is +.>The upper limit concentration is->The lower limit concentration is->At the same time, monitoring point I on inner layer monitoring band b Is +.>The upper limit concentration is->The lower limit concentration is->And the monitoring point E on the outer monitoring band c Is +.>The upper limit concentration is->The lower limit concentration is->Then at that moment, inner layer monitoring point I a And I b The upper limit of the concentration difference threshold (Upper limit ofconcentration difference threshold) between>For, see formula (1):
similarly, an inner layer monitoring point I a And I b Lower limit of concentration difference threshold (Lower limit of concentration difference threshold) betweenFor, see formula (2):
by usingMatrix values of the fine particulate matter upper limit threshold matrix between different monitoring points of the inner and outer monitoring bands of the identification area at the time t of month m are shown in a formula (3):
similarly, useAnd the matrix value of the fine particulate matter lower limit threshold matrix between different monitoring points of the inner and outer monitoring bands at the time t of month m is shown. When the transmission condition of the area with fine particles occurs, the peak concentration time, arrival time and departure time of each monitoring point can be obtained through a single monitoring point fine particle concentration identification algorithm. When the value of the change of the concentration of the fine particles at the monitoring point exceeds the monitoring zone threshold matrix +.>At the matrix value of month m time t, the monitoring point has the phenomenon of regional fine particulate matter input. Because the distance between each monitoring point and the transmission path is in different area transmission directions, the influence on the concentration of the fine particulate matters at different monitoring points is possibly inconsistent, and the transmission starting time of the different monitoring point areas is compared>And end time->And judging the monitoring point and the transmission direction of the primary area fine particulate matter which reach first.
(2) The method comprises the steps of performing type classification on roads in an urban road network according to the actual running state of a road vehicle, and obtaining urban road network classification which is more in line with the actual running state by adopting a self-adaptive density clustering algorithm based on vehicle running parameters;
the non-parameter clustering algorithm performs type fusion by initially generating a plurality of small classes and calculating the distance and the distance in the class, and finally, clustering in a mode of calculating the density of the region without the need of initial class parameter setting. The advantage of choosing a neighbor-propagation clustering algorithm (Affinity Propagation, AP) as a means to generate small classes is that the algorithm does not need to implement a specified number of clusters, i.e. all points may be the centers of the clusters. Using D (i, j) to represent any two data points x i And x j The distance between them can be expressed as: d (i, j) = ||x i -x j ||
Small class S i Is of the area density of (2)Can be expressed as:
the area density variance between small classes is taken as an objective function, and can be expressed as:
and continuously updating the value of the CRD until the CRD value is not reduced, and forming a stable clustering result, namely ending the iterative process. This merging is called density propagation, i.e. by merging small classes with similar area densities until several classes are available that cannot be merged.
(3) Using vehicle track data to establish a specific power emission model, and constructing a specific power distribution map-based fine particulate matter emission factor calculation method;
the upper and lower limits of the specific power section of the vehicle are respectively defined as-30 kW/t and 30kW/t, and the entire specific power section is divided into 60 sections. In D j A distribution value (0) indicating the jth specific power section in the specific power distribution of the vehicle<j<60). The horizontal axis of the road network specific power characteristic distribution is a vehicle specific power interval, and the vertical axis is the percentage D of different specific power intervals to the total road specific power j . And obtaining specific power distribution maps of different road types according to the road specific power calculation formula. The specific power distribution map of the road can reflect the running state of traffic flow carried by the road section. The coordinate axes at the bottom are the running speed and specific power interval of the vehicle respectively.
And after the specific power distribution patterns of the different road types are obtained, calculating the road emission factors corresponding to the different road types according to the specific power distribution patterns. The road emission factor is an important component of the research road network emission model and is used for calculating the road section emission of different road types. The road emission factor in this study was defined as: emission of fine particulate matter produced at a specific road type at a specific average speed running unit kilometer. The emission factor is calculated as follows: for a road type k to which a certain road section belongs, a length L (in kilometers) of the road, and PM generated by N vehicles at an average speed V (in kilometers per hour) in a period T (in seconds) 2.5 The discharge amounts are E respectively 1 ，E 2 …E N (in grams) and the emission factor (in grams/km) may be expressed as
Wherein the method comprises the steps ofThe emission factor of the vehicle when the vehicle is driving on road type k at average speed V, in g/km;
k is a set of road types (K 1 ,K 2 ,…,K n ) K epsilon K is a road of a certain type;
m is a set of vehicle types (M 1 ,m 2 ,…,m n ) M epsilon M is a type of vehicle of a certain type;
α m the proportion of the vehicle type m to the total traffic;
v is the average running speed of the vehicle, and the unit is km/h;
the unit is g/s for the instantaneous emission rate of the vehicle corresponding to the vehicle type m at the time t;
the unit is g/s for the vehicle emission rate corresponding to the jth specific power interval in the vehicle type m specific power distribution;
a distribution value of a j-th specific power interval in the specific power distribution of the vehicle type i;
delta is a constant and the value is 3600.
(4) Establishing a data-driven road network traffic running state extraction method, and obtaining road network space-time emission considering the traffic running state by extracting the characteristics of the historical traffic running state and taking the characteristics as the input of a road network space-time emission model;
by matching the trajectory data to each road segment in the road network, it is assumed that the set R of all road segments within the city contains (1, 2,3, …, R) in total R road segments. The 24 hours are divided into equal length time periods T, grouped as T (288 time periods a day if a single time period is 5 minutes in length).
The average speed of each road segment r over each time period t is calculated. Hypothesis roadSegment R is a segment of the road network set R,representing the number of vehicles passing through the road segment r during the time period t; s is(s) r Representing the length of the road segment r; Δt represents a set time interval, typically 5 minutes; />Representing the average speed of the vehicle i over the time Δt of the road segment r, the average speed +.>If a certain road segment r has no vehicle passing within Δt, replacing the average speed of the current time period with the average speed of the previous time period +.>And->Can be expressed as:
for various different types of roads contained in the urban road network, the road network emission factors based on the specific power distribution map are obtained based on the above, and the data driving is combined to obtain the basic data such as average speed, traffic flow, vehicle composition ratio and the like of each road section under different traffic running states. The urban road network is formed by combining a plurality of road sections, and the urban traffic operation is carried by different road sections in the urban road network, so that the urban traffic operation state can be overlapped by the operation state of each road section, and the fine particle space-time variation of the urban road network can also be formed by the fine particle space-time variation of each road section. Thus, establishing a road network space-time emission calculation taking into account road traffic operating conditions may be expressed as
NE t For urban road network emission (network emissions), representing the total amount of space-time emission of the urban road network at time t;
R K representing a complete set of road segments in road type K;
representing the length of a road r in a road type k, and the unit km;
the unit pcu/h is the traffic volume of the road r in the road type k in the traffic running state at the moment t;
the vehicle type m is the proportion of the road r in the road type k to the total road traffic;
the average speed of the road in the road type k is given by km/h, wherein the average speed of the road is equal to the average speed of the road in the road r in the traffic running state at the moment t;
considering the average driving speed of the vehicle for road r in road type k at time t +.>Is used in the emission factor of (1) in g/km.
Beta is a correction factor.
The above is merely a preferred embodiment of the present invention, the protection scope of the present invention is not limited to the above embodiment, it should be noted that several improvements and modifications are to be considered as the protection scope of the present invention without departing from the principle of the present invention.
Claims (2)
1. The road network fine particulate matter research method based on the traffic running state is characterized by comprising the following steps of:
step 1: establishing an urban fine particulate matter transmission identification model based on a fine particulate matter region transmission identification algorithm, and identifying fine particulate matter region transmission from two dimensions of time and space;
step 2: the method comprises the steps of performing type classification on roads in an urban road network according to the actual running state of a road vehicle, and obtaining urban road network classification which is more in line with the actual running state by adopting a self-adaptive density clustering algorithm based on vehicle running parameters;
step 3: using vehicle track data to establish a specific power emission model, and constructing a specific power distribution map-based fine particulate matter emission factor calculation method;
step 4: establishing a data-driven road network traffic running state extraction method, and obtaining road network space-time emission considering the traffic running state by extracting the characteristics of the historical traffic running state and taking the characteristics as the input of a road network space-time emission model;
in the step 1, an input monitoring point and subsequent propagation direction information of the regional input fine particulate matters are identified by establishing a monitoring zone threshold matrix, a fine particulate matter regional transmission database is constructed by combining historical data, and the early discovery and real-time monitoring effect on the fine particulate matter regional transmission is realized by a city fine particulate matter transmission identification model through historical data mining and real-time data input; the method comprises the following specific steps:
the time and concentration information of regional fine particulate matter input of a single monitoring point are obtained through a fine particulate matter concentration recognition algorithm of single monitoring point time data, a plurality of monitoring point data are associated, a monitoring zone fine particulate matter threshold matrix is established, and input monitoring points and drift directions of regional input particulate matters can be judged in space; the monitoring area fine particulate matter threshold matrix stores threshold historical data of different time periods, real-time judgment is carried out on area fine particulate matter input by combining the real-time data access of all monitoring points on the identification area, and the monitoring area fine particulate matter threshold matrix is self-iterated and updated;
through the monitoring points E on the outer and inner two layers of monitoring bands n And I n UsingIndicating the monitoring point E on the outer monitoring band n Reference concentration vector at month m time t, and->Indicating the monitoring point I on the inner layer monitoring band n A reference concentration vector at month m time t; wherein E is n And I n Respectively representing the nth point of the outer layer monitoring band and the nth point of the inner layer monitoring band; by->And->Establishing an upper limit concentration vector of a monitoring point of the regional fine particulate matter transmission identification region; even if the positions of the monitoring points are different in the inner layer monitoring and the outer layer monitoring at the same moment t, certain difference exists between the reference concentration and the upper limit concentration of the monitoring points; let the monitoring point on the inner layer monitoring band be I a Monitoring point I at time of month m a Is +.>The upper limit concentration isThe lower limit concentration is->At the same time, monitoring point I on inner layer monitoring band b Is +.>The upper limit concentration is->The lower limit concentration is->And the monitoring point E on the outer monitoring band c Is +.>The upper limit concentration isThe lower limit concentration is->Then at that moment, inner layer monitoring point I a And I b Upper limit of concentration difference threshold between +.>For, see formula (1):
similarly, an inner layer monitoring point I a And I b Lower limit of concentration difference threshold betweenFor, see formula (2):
by usingMatrix values of the fine particulate matter upper limit threshold matrix between different monitoring points of the inner and outer monitoring bands of the identification area at the time t of month m are shown in a formula (3):
similarly, useMatrix values of a fine particulate matter lower limit threshold matrix between different monitoring points of the inner and outer monitoring bands at a month m moment t are represented; when the transmission condition of the fine particulate matter area exists, the peak concentration moment, arrival and departure moment of each monitoring point can be obtained through a single monitoring point fine particulate matter concentration identification algorithm; when the value of the change of the concentration of the fine particles at the monitoring point exceeds the monitoring zone threshold matrix +.>Matrix values at the moment t of month m, wherein the monitoring point has the phenomenon of regional fine particulate matter input; because the distance between each monitoring point and the transmission path is in different area transmission directions, the influence on the concentration of the fine particulate matters at different monitoring points is possibly inconsistent, and the transmission starting time of the different monitoring point areas is compared>And end time->Judging the monitoring point and the transmission direction of the primary area fine particulate matter which reach at first;
in the step 2, the non-parameter clustering algorithm performs type fusion by initially generating a plurality of small classes and calculating the distance and the distance in the class, and finally clusters in a mode of calculating the density of the region without setting initial class parameters; the method comprises the following specific steps:
selecting a neighbor propagation clustering algorithm as a means for generating a small class, wherein D (i, j) is used for representing any two data points x i And x j The distance between them can be expressed as:
D(i,j)＝||x i -x j ||
small class S i Is of the area density of (2)Can be expressed as:
the area density variance between small classes is taken as an objective function, and can be expressed as:
continuously updating the CRD value until the CRD value is not reduced, forming a stable clustering result, namely ending the iterative process, wherein the merging mode is called density propagation, namely merging small classes with similar area densities until the small classes become a plurality of classes which cannot be merged;
in the step 3, the vehicle driving track data refers to a series of data of recording time sequence and vehicle GPS position, including vehicle number, recording time, longitude and latitude, speed and direction angle; establishing a specific power distribution map through track data of the vehicle running on an actual road;
after the specific power distribution patterns of the different road types are obtained, calculating road emission factors corresponding to the different road types according to the specific power distribution patterns, wherein the road emission factors are used for calculating road section emission of the different road types, and the road emission factors are defined as follows: generating subtle values in kilometers of travel units at a particular average speed over a particular road typeThe discharge amount of particulate matter; the calculation formula of the road emission factor is as follows: for a road type k to which a certain road section belongs, a road of length L, fine Particulate Matter (PM) is produced by N vehicles at an average speed V during a period T 2.5 ) The discharge amounts are E respectively 1 ，E 2 …E N The emission factor, which can be expressed as
Wherein the method comprises the steps ofThe emission factor of the vehicle when the vehicle is driving on road type k at average speed V, in g/km; the unit of the length L is kilometer, the unit of the period T is seconds, the unit of the average speed V is kilometer/hour, and the discharge amounts are E respectively 1 ，E 2 …E N In grams and the emission factor in grams/km;
k is a set of road types (K 1 ,K 2 ,…,K n ) K epsilon K is a road of a certain type;
m is a set of vehicle types (M 1 ,m 2 ,…,m n ) M epsilon M is a type of vehicle of a certain type;
α m the proportion of the vehicle type m to the total traffic;
v is the average running speed of the vehicle, and the unit is km/h;
the unit is g/s for the instantaneous emission rate of the vehicle corresponding to the vehicle type m at the time t;
the unit is g/s for the vehicle emission rate corresponding to the jth specific power interval in the vehicle type m specific power distribution;
a distribution value of a j-th specific power interval in the specific power distribution of the vehicle type i;
delta is a constant and the value is 3600.
2. The road network fine particulate matter research method based on traffic running state according to claim 1, characterized in that: in the step 4, the traffic running state of the road network is taken as one of important inputs of a space-time emission model of the road network, a traffic running state database is constructed, the database uses a bidirectional recurrent neural network to perform feature learning on traffic running history data of the road network in a data driving mode, and the traffic running state of the road network is extracted by taking the traffic data at the current moment as the input; according to the traffic running state and the road network emission model, the emission condition of road network traffic pollutants changing along with time can be obtained; the method comprises the following specific steps:
by matching the trajectory data to each road segment in the road network, let the set R of all road segments in the city contain (1, 2,3, …, R) totaling R road segments; dividing 24 hours into equal-length time periods T, and collecting the time periods as T;
calculating the average speed of each road segment R in each time period t, setting the road segment R as one road segment in the road network set R,representing the number of vehicles passing through the road segment r during the time period t; s is(s) r Representing the length of the road segment r; Δt represents a set time interval;representing the average speed of the vehicle i over the time Δt of the road segment r, the average speed +.>If a certain road section r does not pass through the vehicle within delta t, the vehicle is used for the last timeThe average speed of the segment replaces the average speed of the current time segmentAnd->Can be expressed as:
for various different types of roads contained in the urban road network, based on the road network emission factors based on the specific power distribution map, the data driving is combined to obtain basic data such as average speed, traffic flow, vehicle composition ratio and the like of each road section under different traffic running states; the urban road network is formed by combining a plurality of road sections, and urban traffic is carried by different road sections in the urban road network, so that the running state of the urban traffic can be overlapped by the running state of each road section, and the space-time change of fine particles of the urban road network can also be formed by the space-time change of the fine particles of each road section; thus, establishing a road network space-time emission calculation taking into account road traffic operating conditions may be expressed as
NE t For urban road network emission (network emissions), representing the total amount of space-time emission of the urban road network at time t;
R K representing a complete set of road segments in road type K;
representing the length of a road r in a road type k, and the unit km;
the unit pcu/h is the traffic volume of the road r in the road type k in the traffic running state at the moment t;
the vehicle type m is the proportion of the road r in the road type k to the total road traffic;
the average speed of the road in the road type k is given by km/h, wherein the average speed of the road is equal to the average speed of the road in the road r in the traffic running state at the moment t;
considering the average driving speed of the vehicle for road r in road type k at time t +.>Is arranged in units of g/km;
beta is a correction factor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110942558.1A CN113642241B (en) | 2021-08-17 | 2021-08-17 | Road network fine particulate matter research method based on traffic running state |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110942558.1A CN113642241B (en) | 2021-08-17 | 2021-08-17 | Road network fine particulate matter research method based on traffic running state |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113642241A CN113642241A (en) | 2021-11-12 |
| CN113642241B true CN113642241B (en) | 2023-10-31 |
Family
ID=78422391
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110942558.1A Active CN113642241B (en) | 2021-08-17 | 2021-08-17 | Road network fine particulate matter research method based on traffic running state |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113642241B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114509373B (en) * | 2022-04-20 | 2022-08-05 | 淄博众擎大数据科技合伙企业(有限合伙) | Automobile exhaust particulate matter detection method |
| CN114882614B (en) * | 2022-07-11 | 2022-10-11 | 杭州春来科技有限公司 | Motor vehicle carbon emission monitoring method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107563425A (en) * | 2017-08-24 | 2018-01-09 | 长安大学 | A kind of method for building up of the tunnel operation state sensor model based on random forest |
| CN108648452A (en) * | 2018-05-15 | 2018-10-12 | 高新兴科技集团股份有限公司 | A kind of road traffic state judgment method and traffic situation monitoring center |
| JP2018195332A (en) * | 2018-07-23 | 2018-12-06 | 三菱電機株式会社 | Mobile object travel management system |
| CN108985518A (en) * | 2018-08-01 | 2018-12-11 | 北京航空航天大学 | A kind of traffic congestion charging policy optimization method based on VISSIM emulation |
-
2021
- 2021-08-17 CN CN202110942558.1A patent/CN113642241B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107563425A (en) * | 2017-08-24 | 2018-01-09 | 长安大学 | A kind of method for building up of the tunnel operation state sensor model based on random forest |
| CN108648452A (en) * | 2018-05-15 | 2018-10-12 | 高新兴科技集团股份有限公司 | A kind of road traffic state judgment method and traffic situation monitoring center |
| JP2018195332A (en) * | 2018-07-23 | 2018-12-06 | 三菱電機株式会社 | Mobile object travel management system |
| CN108985518A (en) * | 2018-08-01 | 2018-12-11 | 北京航空航天大学 | A kind of traffic congestion charging policy optimization method based on VISSIM emulation |
Non-Patent Citations (2)
| Title |
|---|
| 城市交通排放高分辨率分析方法研究――北京实证;程颖;陈艳艳;刘莹;吴克寒;何巍楠;赵晋;;交通运输系统工程与信息(02);全文 * |
| 城市公共交通车辆配置研究;杨雨浓;任思佳;;物流技术(05);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113642241A (en) | 2021-11-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110209990B (en) | Single vehicle emission track calculation method based on vehicle identity detection data | |
| CN113642241B (en) | Road network fine particulate matter research method based on traffic running state | |
| CN102136190B (en) | Dispatching management system and method for event emergency response of urban bus passenger transport | |
| CN111696369B (en) | All-market road time-sharing and vehicle-division type traffic flow prediction method based on multi-source geographic space big data | |
| WO2017051411A1 (en) | Near real-time modeling of pollution dispersion | |
| WO2021115320A1 (en) | Traffic evaluation method and system | |
| Liu et al. | Intersection delay estimation from floating car data via principal curves: a case study on Beijing’s road network | |
| US20220335822A1 (en) | Method of determining the amount of pollutant emissions from a vehicle over a road network section | |
| CN105869402B (en) | Express highway section speed modification method based on polymorphic type floating car data | |
| CN111914217A (en) | Airport environment bearing capacity evaluation and prediction method | |
| CN114881356A (en) | Urban traffic carbon emission prediction method based on particle swarm optimization BP neural network optimization | |
| Shang et al. | Analyzing the effects of road type and rainy weather on fuel consumption and emissions: A mesoscopic model based on big traffic data | |
| CN112967493A (en) | Neural network-based prediction method for vehicle passing intersection travel time | |
| CN115291515A (en) | Automatic driving simulation test system and evaluation method based on digital twinning | |
| CN110738367A (en) | traffic exhaust emission prediction method based on deep residual error network | |
| CN112884014A (en) | Traffic speed short-time prediction method based on road section topological structure classification | |
| Passow et al. | Adapting traffic simulation for traffic management: A neural network approach | |
| CN112767686B (en) | Road network automobile emission estimation method based on multi-source data fusion | |
| Suleiman et al. | A comparative study of using Random Forests (RF), Extreme Learning Machine (ELM) and Deep Learning (DL) algorithms in modelling Roadside Particulate Matter (PM10 & PM2. 5) | |
| Kutlimuratov et al. | Impact of stops for bus delays on routes | |
| Golze et al. | Impact analysis of accidents on the traffic flow based on massive floating car data | |
| CN115270962A (en) | Atmospheric pollutant concentration calculation method based on urban traffic simulation | |
| CN114139984A (en) | Urban traffic accident risk prediction method based on flow and accident collaborative perception | |
| Hu et al. | City traffic forecasting using taxi GPS data: A coarse-grained cellular automata model | |
| Reveron et al. | A Framework for Collision Prediction Using Historical Accident Information and Real-time Sensor Data: A Case Study for the City of Ottawa |
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 |