CN109543874A - A kind of aerodrome air qualitative forecasting method that combination meteorological condition influences - Google Patents
A kind of aerodrome air qualitative forecasting method that combination meteorological condition influences Download PDFInfo
- Publication number
- CN109543874A CN109543874A CN201811187095.7A CN201811187095A CN109543874A CN 109543874 A CN109543874 A CN 109543874A CN 201811187095 A CN201811187095 A CN 201811187095A CN 109543874 A CN109543874 A CN 109543874A
- Authority
- CN
- China
- Prior art keywords
- aerodrome
- node
- airport
- quality
- air
- 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.)
- Granted
Links
- 238000013277 forecasting method Methods 0.000 title claims abstract description 14
- 238000000034 method Methods 0.000 claims abstract description 23
- 238000007637 random forest analysis Methods 0.000 claims abstract description 22
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 3
- 230000008878 coupling Effects 0.000 claims abstract description 3
- 238000010168 coupling process Methods 0.000 claims abstract description 3
- 238000005859 coupling reaction Methods 0.000 claims abstract description 3
- 239000003344 environmental pollutant Substances 0.000 claims description 29
- 231100000719 pollutant Toxicity 0.000 claims description 29
- 238000003066 decision tree Methods 0.000 claims description 17
- 238000012360 testing method Methods 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 10
- 238000009792 diffusion process Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 241000208340 Araliaceae Species 0.000 claims description 4
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims description 4
- 235000003140 Panax quinquefolius Nutrition 0.000 claims description 4
- 235000008434 ginseng Nutrition 0.000 claims description 4
- 239000012535 impurity Substances 0.000 claims description 4
- 238000002790 cross-validation Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 230000003455 independent Effects 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 2
- 238000012545 processing Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 claims 1
- 230000008859 change Effects 0.000 description 4
- 230000011218 segmentation Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000009472 formulation Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- SLVOVFVZZFUEAS-UHFFFAOYSA-N 2-[2-[2-[bis(carboxymethyl)amino]ethoxy]ethyl-(carboxymethyl)amino]acetic acid Chemical compound OC(=O)CN(CC(O)=O)CCOCCN(CC(O)=O)CC(O)=O SLVOVFVZZFUEAS-UHFFFAOYSA-N 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000004907 flux Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000007619 statistical method Methods 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/0004—Gaseous mixtures, e.g. polluted air
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/02—Instruments for indicating weather conditions by measuring two or more variables, e.g. humidity, pressure, temperature, cloud cover or wind speed
-
- 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)
- Tourism & Hospitality (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Environmental & Geological Engineering (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Pathology (AREA)
- Biochemistry (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Immunology (AREA)
- Quality & Reliability (AREA)
- Analytical Chemistry (AREA)
- Combustion & Propulsion (AREA)
- Food Science & Technology (AREA)
- Medicinal Chemistry (AREA)
- Atmospheric Sciences (AREA)
- Biodiversity & Conservation Biology (AREA)
- Ecology (AREA)
- Environmental Sciences (AREA)
- Educational Administration (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention discloses a kind of aerodrome air qualitative forecasting method that combination meteorological condition influences, and the prediction technique calculates concentration of emission and air quality index using LTO model and AERMOD models coupling, usage history meteorological data and flight operation data;Then random forest sorting algorithm is used to establish classifier to aerodrome air quality to using meteorological condition and the volume of traffic as feature.This method can significantly increase the predictive ability of air quality within the scope of airport, and provide data reference foundation for reasonable arrangement Airport Operation scheme.
Description
Technical field
It is that a kind of airport based on random forest classification method is empty the present invention relates to a kind of aerodrome air qualitative forecasting method
Makings amount Predicting Technique;More particularly to airport pollutant concentration appraisal procedure associated with aerodrome condition, aerodrome air
Method for evaluating quality.
Background technique
From reform and open up to the outside world it is later the overall national strength in China is quickly enhanced for decades, air-transport industry development it is fast
Speed, air traffic amount rapidly increase.However, thus bring environmental pollution the problems such as also increasingly aggravate, air quality is by people
More and more concerns, discharge influence to air quality to reduce aviation, how accurate evaluation weather condition is to airport sky
The influence of makings amount becomes the hot spot of aviation industry concern.
Air traffic amount and meteorological condition are to influence two key factors of aerodrome air quality, wherein air traffic amount master
The total release of aviation pollutant is influenced, meteorological condition mainly influences the diffusion of various emissions.
Currently to aerodrome air matter quantifier elimination primarily directed to pollutant during Airport Operation discharge to air matter
The influence of amount establishes airport pollutant emission inventory;Such technology lays particular emphasis on the sky that airport is faced from Airport Operation drain layer
Makings amount is analyzed and is studied, and seldom considers the influence that meteorological condition such as relative humidity spreads pollutant.Even its
The influence that aerodrome condition is considered in his research is also only limitted to influence of the analysis meteorological condition to Airport Operation, less combination
The prediction and monitoring capacity discharged to analyze the air quality on airport, to air quality are insufficient.It is empty due to lacking reliable airport
Gas quality prediction information causes inflexible in tactical level formulation airport operational plan.
Currently, both at home and abroad with meteorological condition and aerodrome traffic amount to the technological achievement in terms of aerodrome air prediction of quality compared with
Few, there are still certain field blank.Therefore, it needs to carry out the air quality on airport using a kind of efficient prediction technique
Science and Accurate Prediction, this enhances the prediction energy of aerodrome air quality for promoting air traffic control greenization developing goal
Power, auxiliary reasonable arrangement Airport Operation scheme etc. are of great significance.
Summary of the invention
It is a kind of based on random forest technical problem to be solved by the present invention lies in providing for the vacancy studied at present
Aerodrome air qualitative forecasting method can significantly increase the predictive ability of air quality within the scope of airport, and be reasonable arrangement airport
Operating scheme provides data reference foundation.
The present invention is using LTO model and AERMOD models coupling, using being based on history meteorological data and flight operation data,
Concentration of emission and air quality index are calculated, uses the random forest sorting algorithm to four kinds of main meteorological conditions and friendship thereafter
Flux is that main feature establishes classifier to aerodrome air quality.Realize that the technical solution of the method is as follows:
Step 1: the discharge of pollutant in analysis Airport Operation, combing influence all kinds of meteorologic factors of pollutant diffusion, meter
Aerodrome air performance figure is calculated, following sub-step is specifically included:
(1.1) discharge process of pollutant in Airport Operation is analyzed, and pollutant diffusion under the influence of meteorological condition
Process, analyzing influence aerodrome air quality critical meteorologic factor;
(1.2) based on standard LTO circulation and AERMOD atmospheric quality models, traffic data and pass using airport
Key meteorologic factor data count each type ratio, calculate the discharge amount of pollutant in the single LTO circulation of each type, in turn
Discharge of pollutant sources parameter in AERMOD is set accordingly, calculates the daily mean of concentration that airport discharges pollutants, and with this computer
Field air quality index.
(1.3) using the average value X of the intraday temperature in airport, wind speed, air pressure and relative humidity1、X2、X3、X4With
Aerodrome traffic amount X5Five variables are set element to construct aerodrome air influencing factors of quality initial set X={ X1,X2,X3,X4,
X5}。
Step 2: using random forest classification method, refines the corresponding aircraft airport air quality of different independents variable
Influence factor constructs aerodrome air influencing factors of quality Screening Samples collection, to the aerodrome air prediction of quality based on random forest
Model is trained, and specifically includes following sub-step:
(2.1) according to aerodrome air influencing factors of quality initial set X, training set S, test set T, intrinsic dimensionality F are given.Really
Determine parameter: using the quantity t of decision tree, the depth d of each tree, the feature quantity f that each node uses, termination condition:
Minimum sample number s on node, least information gain m on node
(2.2) from there is training set S (i) of the extraction size put back to as S in S, as the sample of root node, Cong Genjie
Point starts to train
(2.3) if reaching termination condition on present node, it is leaf node that present node, which is arranged, which is classification
Problem, then the prediction output of the leaf node is that one kind c (j), Probability p c that quantity is most in present node sample set
(j) ratio of Zhan Dangqian sample set then proceedes to train other nodes.If present node does not reach termination condition, from F
F dimensional feature is randomly selected without what is put back in dimensional feature.Using this f dimensional feature, find the best one-dimensional characteristic k of classifying quality and its
Threshold value th, sample of the sample kth dimensional feature less than th is divided into left sibling on present node, remaining is divided into right section
Point.Continue to train other nodes.
(2.4) (2.2) (2.3) are repeated until leaf node is all trained or be marked as to all nodes.
(2.5) (2.2) (2.3) (2.4) are repeated until all decision trees are all trained to.
Step 3: on the basis of being trained to random forest, using test set data to the machine based on random forest
Field Air Quality Forecast model is tested to achieve the purpose that cross validation, and following sub-step is specifically included:
(3.1) there is the stochastical sampling put back to from test set T, since the root node of present tree, according to present node
Threshold value th, judgement are to enter left sibling (<th) or enter right node (>=th), until arrival, some leaf node, and it is defeated
Predicted value out.
(3.2) (3.1) are repeated until all t trees all output predicted value.Output is prediction probability in all trees
That maximum class of summation, i.e., add up the Probability p of each c (j).
Step 4: classifier is carried out on the basis of step 3 to adjust ginseng optimization, improves predictablity rate to reach
Purpose.
(4.1) parameters are adjusted respectively, the variation tendency of predictablity rate when recording parameters adjustment.
(4.2) unified to arrange parameters, select the optimal maximum parameter combination of i.e. predictablity rate.
The invention has the benefit that the aerodrome air qualitative forecasting method proposed by the present invention based on random forest, it can
According to the meteorological condition and the volume of traffic on airport, any one day aerodrome air quality is accurately and quickly predicted, can also be
The formulation of tactics and pre- tactical level Airport Operation plan provide data reference foundation, efficiently accomplished to aerodrome air quality into
The realization of row prediction.
Detailed description of the invention
Fig. 1 is the aerodrome air qualitative forecasting method flow chart based on random forest;,
Fig. 2 is various type pollutant discharge amounts in single LTO circulation
Fig. 3 is the pollutant daily mean of concentration that AERMOD is calculated
Fig. 4 is to predict accurate situation of change when being adjusted to minimum blade size;
Fig. 5 be to single decision tree allow using characteristic be adjusted when predict accurate situation of change;
Fig. 6 is to predict accurate situation of change when being adjusted to decision tree number;
Fig. 7 is to predict accurate situation of change when being adjusted to decision tree depth.
Specific embodiment
In order to keep the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with 7 attached drawings and calculated examples,
The present invention is further described in detail, it should be understood that the core of example described herein for explaining only the invention
Principle, but be not intended to limit the present invention.
A certain Aviation Transportation Airport is chosen, prediction process is executed, calculated examples are as follows:
Step 1: the discharge of pollutant in analysis Airport Operation, combing influence all kinds of meteorologic factors of pollutant diffusion, meter
Calculate aerodrome air performance figure, sub-step is as follows: (1.1) analyze the discharge process of pollutant in Airport Operation, and in meteorology
Pollutant diffusion process under the influence of condition, analyzing influence aerodrome air quality critical meteorologic factor.Collect a certain air transportation
The history meteorological data (including METAR count off evidence and sounding data) and its operation data on airport.(1.2) to the data of collection into
Row processing, is screened and is cleaned to the history meteorological data on airport first, and the number that normal condition is not met in data set is removed
According to, and missing data is supplemented using interpolation method for missing data;It is for statistical analysis to the operation data on airport, filter out machine
The main type of field operation, and its engine type and engine number are inquired, and record number of run and the time of each type.
(1.3) it is recycled using standard LTO, according to aircraft fortune in each stage in the circulation of LTO specified in International Civil Aviation Organization (ICAO)
Aircraft engine emission factor specified in row time and engine parameter and BADA database is screened in conjunction in (1.2)
Type, calculate the discharge amount of pollutant in each type single LTO circulation, only calculate NO in this patent2,SO2, tri- kinds of CO
The discharge amount of pollutant.(1.4) the pollution sources row in AERMOD atmospheric quality models is set according to the calculated result in (1.3)
Parameter is put, wherein when aircraft is considered as non point source of pollution when scene slides, when aircraft takeoff is considered as line source.It is calculated with this
NO out2,SO2, the daily mean of concentration of tri- kinds of pollutants of CO, and then airport air quality index is calculated, calculated result is provided with China
Air quality grade 1-6 indicate.
Step 2: using random forest classification method, refines the corresponding aircraft airport air quality of different independents variable
Influence factor includes temperature, wind speed, air pressure, relative humidity and aerodrome traffic amount to construct the screening of aerodrome air influencing factors of quality
Sample set is trained these influence factors using the aerodrome air quality prediction model based on random forest, specifically includes
Following sub-step:
(2.1) feature, history meteorological data, Airport Operation data and the aerodrome air quality of calculating of finishing collecting are extracted
Index after temperature, wind speed, air pressure, relative humidity and aerodrome traffic amount are normalized, extracts mean temperature X1, it is average
Wind speed X2Average gas pressure X3, average relative humidity X4With aerodrome traffic amount X5Five variables are set element, construct aerodrome air matter
Measure influence factor initial set X={ X1,X2,X3,X4,X5And corresponding tally set Y.
(2.2) according to aerodrome air influencing factors of quality initial set X, training set S, test set T, intrinsic dimensionality F are given.Really
Determine parameter: using the quantity t of decision tree, the depth d of each tree, the feature quantity f that each node uses, termination condition:
Minimum sample number n on node.By major step reduced data, using aerodrome air performance figure as label, with temperature, wind
Speed, air pressure, relative humidity and aerodrome traffic amount are characterized, and provide original training set;Therein 80% is then chosen as training
Collect S, being used as test set T, intrinsic dimensionality F for remaining 20% is 5, and the quantity t of initial decision tree is 2, and each tree depth d is 3,
The characteristic f that each node uses is 2, and the minimum sample number s of node is 10.(2.3) there are the extraction size put back to and S mono- from S
The training set S (i) of sample is trained since root node as the sample of root node.
(2.4) if reaching termination condition on present node, it is leaf node that present node, which is arranged, which is classification
Problem, then the prediction output of the leaf node is that one kind c (j), Probability p c that quantity is most in present node sample set
(j) ratio of Zhan Dangqian sample set then proceedes to train other nodes.If present node does not reach termination condition, from F
F dimensional feature is randomly selected without what is put back in dimensional feature.Using this f dimensional feature, find the best one-dimensional characteristic k of classifying quality and its
Threshold value th, sample of the sample kth dimensional feature less than th is divided into left sibling on present node, remaining is divided into right section
Point then continues to train other nodes.Such as in the present invention, the decision tree that will be formed after segmentation, nothing is put from five kinds of features
The selection temperature and the volume of traffic returned, find the best one-dimensional characteristic of classifying quality and its threshold value finds best one-dimensional of classifying quality
Feature k and its threshold value th, sample of the sample kth dimensional feature less than th is divided into left sibling on present node, remaining is drawn
Right node is assigned to, then continues to train other nodes.Gini impurity level is calculated after segmentation
Gini=1- ∑ (P (i) * P (i))
In formula, P (i) is the ratio of the i-th class sample in data set on present node.As can be seen that category distribution is average,
Gini value is bigger, and classifying quality is poorer to need subseries again;Distribution is more uneven, and Gini value is smaller, illustrates that classifying quality is better.
That biggish node of Gini impurity level after segmentation is continued to divide, the Gini impurity level after calculating secondary splitting, not by Gini
That biggish node of purity continues to divide, until reaching parameter limitation.
(2.5) (2.3) (2.4) are repeated until leaf node is all trained or be marked as to all nodes.
(2.6) (2.3) (2.4) (2.5) are repeated until all decision trees are all trained to.At this time it can be concluded that training is accurate
Rate, training accuracy rate refer to that classifier predictablity rate is the ratio between the accurate sample size of prediction and total sample number, should
Value can be very good to assess the classification results of random forest, it can be seen that the value is bigger, and classification results are better;Conversely,
The value is smaller, and classification results are poorer.
Step 3: on the basis of being trained to random forest, using test set data to the machine based on random forest
Field Air Quality Forecast model is tested to achieve the purpose that cross validation, and following sub-step is specifically included:
(3.1) there is the stochastical sampling put back to from test set T, since the root node of present tree, according to present node
Threshold value th, judgement are to enter left sibling (<th) or enter right node (>=th), until arrival, some leaf node, and it is defeated
Predicted value out.
(3.2) (3.1) are repeated until all t trees all output predicted value.Output is prediction probability in all trees
That maximum class of summation, i.e., add up the Probability p of each c (j).
Step 4: classifier is carried out on the basis of step 3 to adjust ginseng optimization, improves predictablity rate to reach
Purpose.
(4.1) parameters are adjusted respectively, the variation tendency of predictablity rate when recording parameters adjustment.
Wherein respectively to minimum blade size, single decision tree allow using characteristic, decision tree number, decision tree depth adjusted
It is whole.
(4.2) unified to arrange parameters, select the optimal maximum parameter combination of i.e. predictablity rate.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
It for member, can also make several improvements without departing from the principle of the present invention, these improvement also should be regarded as of the invention
Protection scope.
Claims (8)
1. the aerodrome air qualitative forecasting method that a kind of combination meteorological condition influences, which is characterized in that the prediction technique uses
LTO model calculates concentration of emission and air matter with AERMOD models coupling, usage history meteorological data and flight operation data
Volume index;Then random forest sorting algorithm is used to establish to using meteorological condition and the volume of traffic as feature to aerodrome air quality
Classifier then predicts aerodrome air quality using the classifier that random forest is established.
2. the aerodrome air qualitative forecasting method that a kind of combination meteorological condition according to claim 1 influences, the method
The following steps are included:
Step 1: the discharge process of pollutant first in analysis Airport Operation, combing influence pollutant diffusion it is all kinds of it is meteorological because
Element, then calculates the per day pollutant concentration on airport using LTO model and AERMOD atmospheric quality models, and then calculates airport
Air quality index AQI;
Step 2: using random forest classification method, extracts the influence factor of the corresponding aerodrome air quality of different independents variable, structure
Aerodrome air influencing factors of quality Screening Samples collection is built, the aerodrome air quality prediction model based on random forest is instructed
Practice;
Step 3: empty to the airport based on random forest using test set data on the basis of being trained to random forest
Gas quality prediction model is tested to achieve the purpose that cross validation.
3. the aerodrome air qualitative forecasting method that a kind of combination meteorological condition according to claim 2 influences, feature exist
In the step 1 includes following sub-step:
Step (1.1) acquires pollutant emission data in Airport Operation, and pollutant under the influence of meteorological condition diffuses through
The data information of journey, analyzing influence aerodrome air quality critical meteorologic factor;
Step (1.2) is handled and is screened to the meteorological data of acquisition, carries out interpolation method processing to the data of scarce survey;It counts
The daily operation sortie of type and each type that airport is mainly run;
Step (1.3) is based on standard LTO circulation and AERMOD atmospheric quality models, traffic data and pass using airport
Key meteorologic factor data calculate the discharge amount of pollutant in the single LTO circulation of each type, and then it is every daily to calculate each type
The total amount of pollutant discharged in hour operation;
The discharge parameter of pollutant in AERMOD model is arranged according to the calculated result in (1.3) in step (1.4), calculates pollution
The daily mean of concentration of object, and then calculate the daily air quality index AQI in airport.
4. the aerodrome air qualitative forecasting method that a kind of combination meteorological condition according to claim 3 influences, feature exist
In the step 2 includes following sub-step:
(2.1) using the average value X of the intraday temperature in airport, wind speed, air pressure and relative humidity1、X2、X3、X4It is handed over airport
Flux X5Five variables are set element to construct aerodrome air influencing factors of quality initial set X={ X1,X2,X3,X4,X5};
(2.2) according to aerodrome air influencing factors of quality initial set X, training set S, test set T, intrinsic dimensionality F are given, determines ginseng
Number: the feature quantity f and termination condition for using the quantity t of decision tree, the depth d of each tree, each node to use:
Minimum sample number s on node, least information gain m on node;
(2.3) from there is training set S (i) of the extraction size put back to as S to open as the sample of root node from root node in S
Beginning is trained Random Forest model;
(2.4) reach termination condition on present node, then it is leaf node that present node, which is arranged, then the prediction of the leaf node is defeated
That one kind c (j) most for quantity in present node sample set out, Probability p are the ratio of c (j) Zhan Dangqian sample set, then
Continue to train other nodes;
If present node does not reach termination condition, f dimensional feature is randomly selected without what is put back to from F dimensional feature;Using institute
F dimensional feature is stated, classifying quality best one-dimensional characteristic k and its threshold value th are found, sample kth dimensional feature is less than th on present node
Sample be divided into left sibling, remaining is divided into right node;Continue to train other nodes;
(2.5) (2.3) (2.4) are repeated until leaf node is all trained or be marked as to all nodes;
(2.6) (2.3) (2.4) (2.5) are repeated until all decision trees are all trained to.
5. the aerodrome air qualitative forecasting method that a kind of combination meteorological condition according to claim 4 influences, feature exist
In the step 3 includes following sub-step:
(3.1) there is the stochastical sampling put back to from test set T, since the root node of present tree, according to the threshold value of present node
Th, judgement is into left sibling or to enter right node, until reaching, some leaf node, and export predicted value;
(3.2) (3.1) are repeated until all t trees all output predicted value;Output is prediction probability summation in all trees
That maximum class, i.e., add up the Probability p of each c (j).
6. the aerodrome air qualitative forecasting method that a kind of combination meteorological condition according to claim 5 influences, feature exist
In, the prediction technique further includes step 4, and the step 4 carries out classifier on the basis of step 3 to adjust ginseng optimization,
To achieve the purpose that improve predictablity rate.
7. the aerodrome air qualitative forecasting method that a kind of combination meteorological condition according to claim 6 influences, feature exist
In the step 4 includes sub-step:
(4.1) parameters are adjusted respectively, the variation tendency of predictablity rate when recording parameters adjustment.Wherein
Respectively to minimum blade size, single decision tree allow using characteristic, decision tree number, decision tree depth is adjusted;
(4.2) unified to arrange parameters, select the optimal maximum parameter combination of i.e. predictablity rate.
8. the aerodrome air prediction of quality side that a kind of combination meteorological condition according to any one of claims 4 to 7 influences
Method, which is characterized in that in the step (2.3), after sample kth dimensional feature lives through division on present node;Pass through calculating
Gini impurity level judges whether classifying quality meets the requirements:
Gini=1- ∑ (P (i) * P (i))
In formula, P (i) is the ratio of the i-th class sample in data set on present node, and category distribution is average, and Gini value is bigger, point
Class effect is poorer to need subseries again;Distribution is more uneven, and Gini value is smaller, illustrates that classifying quality is better.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811187095.7A CN109543874B (en) | 2018-10-12 | 2018-10-12 | Airport air quality prediction method combining meteorological condition influence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811187095.7A CN109543874B (en) | 2018-10-12 | 2018-10-12 | Airport air quality prediction method combining meteorological condition influence |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109543874A true CN109543874A (en) | 2019-03-29 |
CN109543874B CN109543874B (en) | 2021-06-22 |
Family
ID=65843881
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811187095.7A Active CN109543874B (en) | 2018-10-12 | 2018-10-12 | Airport air quality prediction method combining meteorological condition influence |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109543874B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110738354A (en) * | 2019-09-18 | 2020-01-31 | 北京建筑大学 | Method and device for predicting particulate matter concentration, storage medium and electronic equipment |
CN110751102A (en) * | 2019-10-22 | 2020-02-04 | 天津财经大学 | Kyojin Ji three-ground airport passenger flow correlation analysis method and device |
CN111023400A (en) * | 2019-12-30 | 2020-04-17 | 宁波奥克斯电气股份有限公司 | Air conditioner outdoor environment temperature prediction method and device and air conditioner |
CN111427101A (en) * | 2020-04-07 | 2020-07-17 | 江苏省气象科学研究所 | Thunderstorm strong wind grading early warning method, system, storage medium and computer program |
CN112783044A (en) * | 2020-12-31 | 2021-05-11 | 新奥数能科技有限公司 | Energy control system and energy control method |
CN113569440A (en) * | 2021-06-10 | 2021-10-29 | 上海工程技术大学 | Method for simulating emission and diffusion of airplane take-off and landing pollutants in airport area |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104834944A (en) * | 2015-05-26 | 2015-08-12 | 杭州尚青科技有限公司 | Cooperative training-based city region air quality estimation method |
CN105679103A (en) * | 2016-03-16 | 2016-06-15 | 南京航空航天大学 | Method for assessing air traffic volume accommodated in airport environment |
US20160299255A1 (en) * | 2014-09-12 | 2016-10-13 | The Climate Corporation | Weather forecasts through post-processing |
CN108053071A (en) * | 2017-12-21 | 2018-05-18 | 宇星科技发展(深圳)有限公司 | Regional air pollutant concentration Forecasting Methodology, terminal and readable storage medium storing program for executing |
-
2018
- 2018-10-12 CN CN201811187095.7A patent/CN109543874B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160299255A1 (en) * | 2014-09-12 | 2016-10-13 | The Climate Corporation | Weather forecasts through post-processing |
CN104834944A (en) * | 2015-05-26 | 2015-08-12 | 杭州尚青科技有限公司 | Cooperative training-based city region air quality estimation method |
CN105679103A (en) * | 2016-03-16 | 2016-06-15 | 南京航空航天大学 | Method for assessing air traffic volume accommodated in airport environment |
CN108053071A (en) * | 2017-12-21 | 2018-05-18 | 宇星科技发展(深圳)有限公司 | Regional air pollutant concentration Forecasting Methodology, terminal and readable storage medium storing program for executing |
Non-Patent Citations (1)
Title |
---|
雷明洁: "航空器排放对机场大气环境影响的研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110738354A (en) * | 2019-09-18 | 2020-01-31 | 北京建筑大学 | Method and device for predicting particulate matter concentration, storage medium and electronic equipment |
CN110738354B (en) * | 2019-09-18 | 2021-02-05 | 北京建筑大学 | Method and device for predicting particulate matter concentration, storage medium and electronic equipment |
CN110751102A (en) * | 2019-10-22 | 2020-02-04 | 天津财经大学 | Kyojin Ji three-ground airport passenger flow correlation analysis method and device |
CN110751102B (en) * | 2019-10-22 | 2023-12-22 | 天津财经大学 | Beijing Ji three-place airport passenger flow correlation analysis method and device |
CN111023400A (en) * | 2019-12-30 | 2020-04-17 | 宁波奥克斯电气股份有限公司 | Air conditioner outdoor environment temperature prediction method and device and air conditioner |
CN111427101A (en) * | 2020-04-07 | 2020-07-17 | 江苏省气象科学研究所 | Thunderstorm strong wind grading early warning method, system, storage medium and computer program |
CN111427101B (en) * | 2020-04-07 | 2022-04-26 | 南京气象科技创新研究院 | Thunderstorm strong wind grading early warning method, system and storage medium |
CN112783044A (en) * | 2020-12-31 | 2021-05-11 | 新奥数能科技有限公司 | Energy control system and energy control method |
CN112783044B (en) * | 2020-12-31 | 2022-06-03 | 新奥数能科技有限公司 | Energy control system and energy control method |
CN113569440A (en) * | 2021-06-10 | 2021-10-29 | 上海工程技术大学 | Method for simulating emission and diffusion of airplane take-off and landing pollutants in airport area |
Also Published As
Publication number | Publication date |
---|---|
CN109543874B (en) | 2021-06-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109543874A (en) | A kind of aerodrome air qualitative forecasting method that combination meteorological condition influences | |
CN111104981B (en) | Hydrological prediction precision evaluation method and system based on machine learning | |
CN108375808A (en) | Dense fog forecasting procedures of the NRIET based on machine learning | |
CN105702031B (en) | Road network key road segment recognition methods based on macroscopical parent map | |
CN105069537B (en) | A kind of construction method of combination type air Quality Prediction | |
CN105677791B (en) | For analyzing the method and system of the operation data of wind power generating set | |
CN110503245B (en) | Prediction method for large-area delay risk of airport flight | |
CN106600455A (en) | Electric charge sensitivity assessment method based on logistic regression | |
CN103984994B (en) | Method for predicting urban rail transit passenger flow peak duration | |
CN108364467A (en) | A kind of traffic information prediction technique based on modified decision Tree algorithms | |
CN110598726A (en) | Transmission tower bird damage risk prediction method based on random forest | |
CN110827169B (en) | Distributed power grid service monitoring method based on grading indexes | |
CN113887908A (en) | Bridge risk assessment method considering subjective and objective cross fusion weight | |
CN111160612A (en) | Off-site flight delay analysis and prediction method based on weather influence | |
CN112836720B (en) | Building operation and maintenance equipment abnormality diagnosis method, system and computer readable storage medium | |
CN115691120A (en) | Congestion identification method and system based on highway running water data | |
CN114596700A (en) | Real-time highway section flow estimation method based on portal data | |
CN112767684A (en) | Highway traffic jam detection method based on charging data | |
CN115953186A (en) | Network appointment demand pattern recognition and short-time demand prediction method | |
CN114898220B (en) | Intelligent production control method for structural member of overhead working truck | |
CN115526258A (en) | Power system transient stability evaluation method based on Spearman correlation coefficient feature extraction | |
CN116739376A (en) | Highway pavement preventive maintenance decision method based on data mining | |
CN110956316A (en) | Personnel level prediction model based on random forest | |
CN110458204A (en) | Vehicle supply chain method based on information gain and LightGBM model | |
CN110457374B (en) | Method for identifying typical rainstorm process in time period |
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 |