CN106408139B - Airport arrival rate prediction technique and device - Google Patents
Airport arrival rate prediction technique and device Download PDFInfo
- Publication number
- CN106408139B CN106408139B CN201611190122.7A CN201611190122A CN106408139B CN 106408139 B CN106408139 B CN 106408139B CN 201611190122 A CN201611190122 A CN 201611190122A CN 106408139 B CN106408139 B CN 106408139B
- Authority
- CN
- China
- Prior art keywords
- arrival rate
- data
- airport arrival
- airport
- practical
- 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 66
- 230000007613 environmental effect Effects 0.000 claims abstract description 70
- 238000012549 training Methods 0.000 claims abstract description 36
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 31
- 239000011159 matrix material Substances 0.000 claims description 21
- 238000007477 logistic regression Methods 0.000 claims description 10
- 238000010586 diagram Methods 0.000 description 18
- 230000006870 function Effects 0.000 description 13
- 230000008569 process Effects 0.000 description 11
- 241000288113 Gallirallus australis Species 0.000 description 10
- 230000000875 corresponding effect Effects 0.000 description 9
- 238000004458 analytical method Methods 0.000 description 8
- 238000012545 processing Methods 0.000 description 8
- 230000009467 reduction Effects 0.000 description 6
- 238000007418 data mining Methods 0.000 description 5
- 239000000284 extract Substances 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000004445 quantitative analysis Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 208000019901 Anxiety disease Diseases 0.000 description 1
- 241001269238 Data Species 0.000 description 1
- 206010012373 Depressed level of consciousness Diseases 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000036506 anxiety Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013524 data verification Methods 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000004907 flux Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 230000008450 motivation Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000010561 standard procedure Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
- 230000001755 vocal effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 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"
-
- G06Q50/40—
Abstract
A kind of airport arrival rate prediction technique provided in an embodiment of the present invention and device, belong to air traffic flow management technical field, which comprises obtain selection area environmental data within a preset period of time and practical airport arrival rate data corresponding with the preset time period;Environmental data is converted into environmental attribute data, environmental attribute data includes environment attribute training data, and practical airport arrival rate data include the first practical airport arrival rate;Based on the first practical airport arrival rate and preset classification thresholds rule, the first practical airport arrival rate classification data is obtained;Based on preset time period, environment attribute training data, the first practical airport arrival rate classification data and regression algorithm, airport arrival rate prediction model is established;Based on airport arrival rate prediction model, the first prediction airport arrival rate of selection area is calculated.This method predicts airport arrival rate by establishing airport arrival rate prediction model, can provide suggestion to air traffic policymaker.
Description
Technical field
The present invention relates to air traffic flow management technical fields, in particular to a kind of airport arrival rate prediction side
Method and device.
Background technique
With being continuously increased for aviation services, flight flow increasingly increases in limited civil aviaton airspace, and airspace resource is increasingly
Anxiety, this makes airspace system be easier the influence by weather.Atmospheric environment is fast changing, the influence that weather runs flight
Each stage of fly through, from the air route product of low visibility, surface wind, low clouds to Route reform stage when the landing of airport
Ice, torrent etc., are reduced which results in the capacity in the increase of flying distance, airport and airspace and flight is delayed, or even having can
Safety accident can be caused, reduce the operational efficiency of airspace system to a great extent, brought to airline and society larger
Economic loss.Currently, overwhelming majority air traffic control decision related with weather is all to be completed by manually, decision
Person can not directly acquire weather to influence caused by flight operation as a result, but needing to pass through by checking various Weather informations
Experience analyzes Weather information, and influences to eliminate on it, and this mode efficiency is lower, and with personal experience's difference
And change.
Summary of the invention
In view of this, the embodiment of the present invention is designed to provide a kind of airport arrival rate prediction technique and device, with solution
Certainly in the sky in traffic administration decision, policymaker can not directly acquire weather to influence caused by flight operation.
In a first aspect, the embodiment of the invention provides a kind of airport arrival rate prediction techniques, which comprises
Obtain the environmental data and practical machine corresponding with the preset time period of selection area within a preset period of time
Field arrival rate data;
Based on the environmental data and preset data attribute extracting rule, the environmental data is converted into environment attribute
Data, the environmental attribute data include environment attribute training data, and the practical airport arrival rate data include first practical
Airport arrival rate;
Based on the described first practical airport arrival rate and preset classification thresholds rule, the first practical airport arrival rate is obtained
Classification data;
Based on the preset time period, the environment attribute training data, the first practical airport arrival rate classification number
Accordingly and regression algorithm, airport arrival rate prediction model is established;
Based on the airport arrival rate prediction model, the first prediction airport arrival rate of the selection area is calculated.
Second aspect, the embodiment of the invention provides a kind of airport arrival rate prediction meanss, described device includes:
Obtain original data units, for obtain selection area environmental data within a preset period of time and with it is described pre-
If period corresponding practical airport arrival rate data;
Environmental attribute data unit is obtained, it, will for being based on the environmental data and preset data attribute extracting rule
The environmental data is converted to environmental attribute data, and the environmental attribute data includes environment attribute training data, the reality
Airport arrival rate data include the first practical airport arrival rate;
The first practical airport arrival rate taxon is obtained, for being based on the described first practical airport arrival rate and preset
Classification thresholds rule, obtains the first practical airport arrival rate classification data;
Airport arrival rate prediction model unit is established, for based on the preset time period, environment attribute training number
Airport arrival rate prediction model is established according to, the first practical airport arrival rate classification data and regression algorithm;
It calculates the first prediction airport arrival rate unit and calculates the choosing for being based on the airport arrival rate prediction model
Determine the first prediction airport arrival rate in region.
Other features and advantages of the present invention will be illustrated in subsequent specification, also, partly be become from specification
It is clear that by implementing understanding of the embodiment of the present invention.The objectives and other advantages of the invention can be by written theory
Specifically noted structure is achieved and obtained in bright book, claims and attached drawing.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 is a kind of structural block diagram that can be applied to the electronic equipment in the embodiment of the present invention;
Fig. 2 is the flow chart for the airport arrival rate prediction technique that first embodiment of the invention provides;
Fig. 3 is the environmental data signal of the airport arrival rate prediction technique that first embodiment of the invention provides obtained
Figure;
Fig. 4 be the METAR message (north America region) of airport arrival rate prediction technique that first embodiment of the invention provides and
Decode the schematic diagram of citing;
Fig. 5 is the airport arrival rate frequency distribution histogram that first embodiment of the invention provides;
Fig. 6 is the first airport arrival rate classification data schematic diagram that first embodiment of the invention provides;
Fig. 7 is the regression coefficient schematic diagram of airport arrival rate prediction model described in first embodiment of the invention;
The flow chart for the airport arrival rate prediction technique that Fig. 8 second embodiment of the invention provides;
Fig. 9 is the schematic diagram for the second confusion matrix that second embodiment of the invention provides;
Figure 10 is the flow chart for the airport arrival rate prediction technique that third embodiment of the invention provides;
Figure 11 is the first confusion matrix schematic diagram that third embodiment of the invention provides;
Figure 12 is the airport arrival rate prediction meanss that fourth embodiment of the invention provides;
Figure 13 is the airport arrival rate prediction meanss that fifth embodiment of the invention provides.
Specific embodiment
Below in conjunction with attached drawing in the embodiment of the present invention, technical solution in the embodiment of the present invention carries out clear, complete
Ground description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Usually exist
The component of the embodiment of the present invention described and illustrated in attached drawing can be arranged and be designed with a variety of different configurations herein.Cause
This, is not intended to limit claimed invention to the detailed description of the embodiment of the present invention provided in the accompanying drawings below
Range, but it is merely representative of selected embodiment of the invention.Based on the embodiment of the present invention, those skilled in the art are not doing
Every other embodiment obtained under the premise of creative work out, shall fall within the protection scope of the present invention.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need that it is further defined and explained in subsequent attached drawing.Meanwhile of the invention
In description, term " first ", " second " etc. are only used for distinguishing description, are not understood to indicate or imply relative importance.
Referring to Fig. 1, Fig. 1 shows a kind of structural block diagram of electronic equipment 100 that can be applied in the embodiment of the present application.
The electronic equipment 100 can be used as user terminal, computer or server.As shown in Figure 1, electronic equipment 100 may include depositing
Reservoir 110, storage control 111, processor 112 and airport arrival rate prediction meanss.
Memory 110, storage control 111 are directly or indirectly electrically connected between each element of processor 112, to realize
The transmission or interaction of data.For example, can realize electricity by one or more communication bus or signal bus between these elements
Connection.The airport arrival rate prediction technique respectively includes at least one can deposit in the form of software or firmware (firmware)
The software function module being stored in memory 110, for example, the airport arrival rate prediction meanss software function module that includes or
Computer program.
Memory 110 can store various software programs and module, such as airport arrival rate provided by the embodiments of the present application
Corresponding program instruction/the module of method and device.Processor 112 by operation storage software program in the memory 110 with
And module, thereby executing various function application and data processing, i.e. airport arrival rate prediction in realization the embodiment of the present application
Method.Memory 110 can include but is not limited to random access memory (Random Access Memory, RAM), read-only to deposit
Reservoir (Read Only Memory, ROM), programmable read only memory (Programmable Read-Only Memory,
PROM), erasable read-only memory (Erasable Programmable Read-Only Memory, EPROM), electric erasable
Read-only memory (Electric Erasable Programmable Read-Only Memory, EEPROM) etc..
Processor 112 can be a kind of IC chip, have signal handling capacity.Above-mentioned processor can be general
Processor, including central processing unit (Central Processing Unit, abbreviation CPU), network processing unit (Network
Processor, abbreviation NP) etc.;It can also be digital signal processor (DSP), specific integrated circuit (ASIC), ready-made programmable
Gate array (FPGA) either other programmable logic device, discrete gate or transistor logic, discrete hardware components.It can
To realize or execute disclosed each method, step and the logic diagram in the embodiment of the present application.General processor can be micro-
Processor or the processor are also possible to any conventional processor etc..
First embodiment
Referring to Fig. 2, the embodiment of the invention provides a kind of airport arrival rate prediction technique, the present embodiment describes needle
For the process flow of selection area environment within a preset period of time and traffic data, which comprises
Step S210: selection area environmental data within a preset period of time and corresponding with the preset time period is obtained
Practical airport arrival rate data;
Selection principle based on step S210, about selection area, it is preferable that selection is influenced more frequently regional by weather
Property or national hinge type airport.Its volume of traffic is big, is affected to surrounding area, is influenced more serious and frequency by weather
It is numerous, and have enough historical traffic datas and weather data for analysis and research.Domestic airport historical traffic data is difficult to
Obtain (certain airports announce daily flight in its official website and reach and leave the theatre in real time information, and data volume is very little).Secondly as weather
It is not also temporarily the primary factor for influencing air traffic in China, so determining that preliminary Foreign Airport progress case of choosing is ground herein
Study carefully.Philadelphia International Airport has main effect for the delay of the Eastern Seaboard, since adverse weather is more and hands over
Flux is larger, positioned at the middle position of busy east seashore air corridor, causes the delay within the scope of the whole America.
Preferably, the embodiment of the present invention selects Philadelphia International Airport (KPHL) as selection area, and preset time period is to work as
7 points to 22 points of night of morning of the ground time, 2011 to 2014 data, wherein the data conduct to 2013 Nian Sannian in 2011
Training data, data in 2014 are as test data.Airport arrival rate (Airport Arrival/Acceptance
Rates, AAR) be an Airport Operation situation index, indicate under certain condition, airport per hour can be with safe landing
Number of aircraft represents the number of aircraft that an airport can receive, can be used as the index of airfield handling capacity.It is arrived by airport
Aerodrome capacity can be speculated up to rate, be further associated with airport delay, in conjunction with the exploitation of decision support tool, machine
Field arrival rate classification prediction can be used to predict the status of implementation of delay and Ground-Holding program.
Step S220: it is based on the environmental data and preset data attribute extracting rule, the environmental data is converted
For environmental attribute data, the environmental attribute data includes environment attribute training data, the practical airport arrival rate data packet
Include the first practical airport arrival rate;
Based on step S220, the environmental data includes METAR message Weather information;It is described based on the environmental data and
The environmental data is converted to environmental attribute data by preset data attribute extracting rule, comprising:
Based on the METAR message Weather information and preset data attribute extracting rule, wind speed attribute data, energy are extracted
Degree of opinion attribute data, thunderstorm attribute data and the low high attribute data of cloud.
It chooses wind speed, visibility, whether have these high common weather phenomena of thunderstorm, cloud base, these attributes and airport are arrived
Up to rate correlation, and all environment attributes can be obtained from airport aviation routine weather report and weather forecast.The present invention is real
The Weather information for applying example is obtained from history METAR message, message accounting from meteorological resources website obtain (http: //
www.ogimet.com/metars.phtml.en).Referring to Fig. 3, Fig. 3 is that the airport that first embodiment of the invention provides reaches
The environmental data schematic diagram of rate prediction technique obtained.Referring to Fig. 4, Fig. 4 is the airport that first embodiment of the invention provides
The METAR message (north America region) of arrival rate prediction technique and the schematic diagram of decoding citing.
Original METAR message is handled, the data of Weather Elements and weather phenomenon needed for extracting, since METAR is reported
Text is text formatting, and data volume is larger, and the appearance form and position of verbal description and data have certain variability, institute
It being extracted with can use programming such as Perl realization according to the Weather information that hereafter rule carries out, being handled in order to facilitate follow-up data,
Generate the data file of comma separated value file format (Comma-Separated Values, CSV).
Determine the time for 7 points of morning local time to night 22 first, it should be noted that METAR message is according to the world
Universal time coordinated is formulated and submitted, then extracts this METAR message information.According to the sequencing of message marshalling, successively extracts wind speed, can see
Whether degree has thunderstorm to occur, cloud base height.
Wind direction and wind speed are extracted, it is dddffGfmfmKMH or MPS or KT that wind direction and wind velocity, which organizes into groups format, and wherein ddd is indicated
Observe preceding 10 minutes mean wind directions, ff indicates mean wind speed, and " G " is fitful wind indication code, and when no fitful wind omits, and " fmfm " is big
In the fitful wind of 5 meter per second of mean wind speed, when there is fitful wind, the embodiment of the present invention extracts " fmfm " and is used as wind speed.
Extract the thunderstorm information in the visibility value and weather phenomenon group in visibility group.When in weather phenomenon marshalling
When there is " TS " character, indicates that near airports have Thunderstorm Weather appearance, then " 1 " is denoted as to thunderstorm information, indicates thunderstorm, otherwise
For " 0 ".
Cloud group is formulated and submitted using the format of NsNsNshshshs (CC), this group can formulate and submit several groups, by the height of cloud base
It sequentially formulates and submits from low to high, wherein " NsNsNs " is cloud amount, simple language " FEW " (1/8~2/8 cloud amount, partly cloudy), " SCT " (3/ need to be used
8~4/8 cloud amount dredge cloud), " BKN " (5/8~7/8, cloud amount is cloudy), " OVC " (8/8, the cloudy day)." hshshs " is that cloud base is high,
It is code multiplied by 30m or 100ft (north America region)." CC " is cloud form, is only compiled to cumulonimbus (Cb) and cumulus congestus (TCu)
Report.
Observation to cloud amount: cloud amount refers to the number of obnubilation lid sky view.When ground observation, whole skies are in hemisphere
Shape.Sky is divided into eight equal portions by civil aviaton's agencies dictate, wherein the number being hidden by clouds is exactly cloud amount, such as 6/8 sky quilt
Cloud layer covers, and cloud amount is exactly 6.Flight influence on system operation is analyzed according to cloud, only extracting BKN and OVC respective heights herein is this
The hour height of cloud base (extracts lower value by appearance sequence).Particularly, when occurring to cumulonimbus (Cb) when formulating and submitting, such as in message
When fruit cumulonimbus cloud amount is SCT, cloud base height is subject to the code height after SCT, and gives up BKN and OVC.In cumulonimbus with
And cumulonimbus area forbids flying, so especially being handled when there is cumulonimbus cloud form its cloud base height.More than not occurring
Cloud amount and when cloud form, then it is assumed that the high factor in cloud base ignores flight influence on system operation, and the height of cloud base is denoted as one and is much larger than
The high value in normal cloud base does not influence flight operation to show, is denoted as 99900.
Step S230: based on the described first practical airport arrival rate and preset classification thresholds rule, it is practical to obtain first
Airport arrival rate classification data;
Based on step S230, the described first practical airport arrival rate data are based on, the described first practical airport is obtained and reaches
The frequency distribution histogram of rate data;Based on the frequency distribution histogram and the preset classification thresholds rule, institute is obtained
State the first airport arrival rate classification data.Sliding-model control is carried out to first airport arrival rate, it is high, normal, basic to choose three values
It is substituted.It should be noted that being possible to meeting under a few cases, there are two METAR within one hour, or lack some hour
METAR data should be noted that when corresponding to airport arrival rate and need to be corrected by every month.If lacked within certain day certain hour
Data delete this day whole day data in order to not influence this independent variable of hour.
Airport arrival rate (AAR) is categorized into different levels firstly the need of based on certain threshold value, is defined horizontal
Threshold value selection is considered based on the following:
1) airport AAR is distributed;
2) the AAR level that may cause airport blocking or delay is distinguished;
3) regression variable, that is, environmental variance predictability.
As the starting point of airport arrival rate horizontal classification, pass through airport arrival rate within histogram analysis several years first
Distribution and its frequency of occurrences, referring to Fig. 5, Fig. 5 is the airport arrival rate frequency disribution histogram that first embodiment of the invention provides
Figure, based on the considerations of histogram and to demand and traffic jam possibility, numerical value airport arrival rate value is divided into three by selection
It is horizontal.The embodiment of the present invention does not develop the systematic procedure of selection airport arrival rate classification, can be further in following research
Develop this process.The embodiment of the present invention is distributed the practical airport arrival rate in historical data using frequency distribution histogram
Observation, and as the basis for determining classification thresholds.Preferably, Fig. 6 is please referred to, Fig. 6 is that first embodiment of the invention provides
One airport arrival rate classification data schematic diagram.In conjunction with the analysis influenced on traffic circulation, airport arrival rate is divided into basic, normal, high
Three class, the range of receiving of airport arrival rate are less than or equal to 35 and correspond to low-level classification, greater than 35 and be less than or equal to
50 be middle horizontal classification, is greater than 50 for high level classification, correspondence is assigned a value of 1,2,3.
In addition, it is necessary to which explanation, the traffic data of the embodiment of the present invention are obtained from U.S. ASPM (Aviation
System Performance Metrics) database, it is selected according to system prompt, the history on available specific airport
Traffic condition counts per hour, since data open to the public are the contact traffic condition statistics between 77 main airports,
Rather than the whole volume of traffic of the airport at that time.
Step S240: it is reached based on the preset time period, the environment attribute training data, the first practical airport
Rate classification data and regression algorithm establish airport arrival rate prediction model;
Based on step S240, instructed using multinomial logistic regression algorithm, with the preset time period, the environment attribute
Practicing data is regressor, using the described first practical airport arrival rate classification data as sort capacity, establishes airport arrival rate prediction mould
Type.
Logistical regression algorithm can predict interested classified variable according to multiple independent attributes, and independent attribute can be with
For numerical quantities or sort capacity.It is as follows using several motivations of multinomial logistic regression analytic approach: firstly, this technology is fine
Be suitable for rough estimate airport arrival rate classification target, when especially data available is limited.Predict that AAR is horizontal or classifies
Inherent nature be particularly likely that some multiple numerical attributes mixing (such as visibility and wind speed) and categorical attribute (such as
Whether thunderstorm is had), and multinomial logistic regression algorithm allows to use while classifying type and numeric type regressor.Logistic regression
The stochastic model of airport arrival rate classification can be quickly generated, and provides regression parameter.In addition, multivariate logistic regression model
Theoretical premise want much more relaxed with respect to techniques of discriminant analysis, and the not stringent vacation about distribution pattern, covariance matrix etc.
It is fixed.
Next multinomial logistic regression analytic approach is briefly introduced, and discusses and airport is arrived using regression technique
It is modeled up to rate classification.Model assumes initially that the amount Y (herein referring to airport arrival rate) that will be modeled has k different classification
(be referred to as "horizontal" later, and be denoted as 1, K, k).Dependent on m regressor, (finger ring of embodiment of the present invention border belongs to these level probabilities
Property), it is denoted as x1,x2,...,xm.When logistical regression specifies given attribute value, the probability of each classification of Y.It is specific next
It says, if X=[x1 x2…xm], probability of the Y in horizontal j ≠ k is expressed as the function of X, shown in following formula one:
Wherein, cjIt is a constant, Bj=[b1 b2…bm]TIt is the coefficient column vector for the horizontal j that will be determined.Y is in level
Probability in k can be byIt obtains.If certain attribute is classification type, for example work as xiIt assume that as m difference
One of value, i.e.,The then binary variable that algorithm will be assumed with mThis attribute is replaced,
These binary variables can be taken as 0 or 1.In this way, algorithm is just that each hypothesis variable in above-mentioned formula one is assigned with one
A logic coefficient.
Constant parameter cjCommonly referred to as intercept parameter, as attribute x1,x2,...,xmWhen being 0, it can be obtained by intercept parameter
Take Pj(X).In this case, the probability in horizontal j are as follows:
Next, enabling Xi=[xi,1…xi,m] it is i-th in n historical record, the historical data of Y corresponds to Yi.It is multinomial
Logistical regression algorithm is estimated to obtain each j=1 ..., k-1 by minimizing negatively correlated multinomial log-likelihood function L
Logic coefficient vector BjAnd intercept parameter cj:
Preferably, logistical regression study in generally use gradient descent method and quasi-Newton method to objective function L into
Row optimizes.
Logic coefficient captures the relative importance of each attribute in prediction Y, for some given horizontal Y, height system
Number indicates the high correlation between attribute and level.In brief, a specific level, such as the horizontal p ∈ in Y
{ 1 ..., k }, if for p level, attribute xjThere is positive (or negative) coefficient, then xjThe increase (or reduction) of value will increase Y
Belong to the probability of p level.The size of coefficient is indicated with xjIn unit incrementss, it will cause observed Y p level in
Probability increase the determinant of (or reduce).
Multinomial logistic regression analysis method is a kind of in the relatively broad technology used of multiple fields, and various software can lead to
Cross historical data building model.The software category that can carry out data mining at present is more, Orange, RapidMiner,
JHepWork, KNIME, Weka etc. are the free data mining softwares of open source.Selection uses Weka herein, in addition to multinomial logic
Regression algorithm, which provides the method for data processing and filtering simultaneously, while there are also model measurement options.The full name of Weka
It is Waikato intellectual analysis environment, is a based on the machine learning increased income under JAVA environment and data mining software, source
Code can be downloaded in its official website.As a disclosed data mining workbench, Weka, which has gathered, a large amount of can undertake number
It pre-processes, classifies according to the machine learning algorithm of mining task, including to data, return, cluster, correlation rule and new
Interactive interface on visualization, while the environment can also compare and assess the performance of different learning algorithms, be now
One of most complete Data Mining Tools.
The private data format of Weka is ARFF (Attribute-Relation File Format), and this document is
ASCII text file describes the example list of shared one group of attribute structure, is made of independent and unordered example, is Weka table
Show the standard method of data set.Weka provides the support to csv file, and this format is supported by a lot of other softwares
, processing of the embodiment of the present invention generates the data file of CSV format, and is converted into ARFF format by Weka.In Weka
In, output target to be predicted is referred to as Class attribute, i.e. " class " of classification task, and the attribute of the last one statement is referred to as
Class attribute, in classification task, it is the target variable of default (embodiment of the present invention is airport arrival rate).
With the variation of hour in one day, airport arrival rate can show some changing patteries, in order to obtain in one day by
Airport arrival rate changes over time caused by the operating standard (such as the maintenance of plan runway or configuration variation) on airport
Mode, the regressor that the increase local time classifies as airport arrival rate.
The environment attribute training data in the embodiment of the present invention includes preset time period, wind speed attribute data, can see
Attribute data, thunderstorm attribute data and the low high attribute data of cloud are spent as regressor, first time airport arrival rate classification data
As sort capacity, Multinomial Logistic Regression model is established.
Preset time period refers to the hour instruction of predicted time, (classifying type, 7 points of morning arrive at 22 points at night, totally 16 class);Thunder
Sudden and violent attribute data refers in predicted time in termination environment whether there is or not thunderstorm appearance, (categorical attribute, respectively 0,1, totally 2 class);Wind speed
Attribute data refers to the wind speed in predicted time, (Numeric Attributes);Visibility attribute data refers to the visibility in predicted time
(Numeric Attributes);The low high attribute data of cloud refers to that the cloud base in predicted time is high (Numeric Attributes).
Application Weka software of the embodiment of the present invention carries out the multinomial of airport arrival rate classification and patrols this meaning regression modeling, passes through three
The training data in year obtains model parameter.The multinomial logistic regression algorithm generated in Weka produces recurrence in regression formula
Coefficient, i.e. vector B in above-mentioned formula onejAnd intercept parameter cj.Model result, which gives, corresponds to three given water in algorithm
The coefficient of low and middle horizontal generation in flat, by it can be concluded that total probability, so not providing high-level coefficient of correspondence.It please join
It is the regression coefficient schematic diagram of airport arrival rate prediction model described in first embodiment of the invention according to Fig. 7, Fig. 7.Every a line table
Show each regressor, every one kind represents sort capacity, numerical value indicate each sort capacity to the regression coefficient value of each regressor, i.e.,
Dependence.Regression coefficient obtains each airport arrival rate level to a variety of attribute possibilities according to lazyness.For specific airport
The positive coefficient of arrival rate level illustrates that the increase of attribute value is conducive to the appearance of this airport arrival rate level.It is similar, one
It is horizontal that a negative coefficient illustrates that the reduction of attribute value is conducive to this airport arrival rate.As shown in Figure 7, HOUR indicates time, WS
Indicate that wind speed, VIS indicate visibility, TS indicates that storm, CEILING indicate that cloud base is high;Class presentation class, 1 expression airport are arrived
Indicate horizontal in airport arrival rate (AAR) up to rate (AAR) low-level, 2.The regression coefficient 0.0383 and 0.0108 of wind speed, wind speed
Increase result in low AAR situation and be easier to occur, the reduction of visibility equally also results in low AAR event and occurs.Coefficient
Size indicates correlative factor, by correlative factor, the variation of one unit of some attribute value increase (or reduction) some
The trend of specific airport arrival rate level.Wherein, positive number indicates to increase, and negative number representation is reduced.For example, by the comparison of coefficient,
It can be concluded that the reduction (- 0.2103) of visibility can make low airport arrival rate event compared to the increase (0.0383) of wind speed
It becomes more serious, be easier to occur.In this way, by logistical regression model, to specific airport different weather situation to traffic
The influence of situation has further quantitative analysis.
In addition, since the data of acquisition are the contact traffic condition statistics between main airports, rather than the airport was at that time
Whole volume of traffic, when weather conditions change, the amplitude that this part volume of traffic is affected and changes is smaller, leads
Cause weather smaller compared to hour coefficient of discharge.
Although it should be noted that being to be directed to some specific classification, logistical regression can be to any classification
It uses.In addition, improve the classification of airport arrival rate if necessary, as described above, recurrence the result is that it can be used repeatedly.
Step S250: being based on the airport arrival rate prediction model, and the first prediction airport for calculating the selection area is arrived
Up to rate.
Based on the airport arrival rate prediction model after acquisition regression coefficient, the environment attribute training data is brought into
The airport arrival rate prediction model after obtaining regression coefficient, calculates and the first prediction airport for obtaining the selection area is arrived
Up to rate.
The embodiment of the invention provides a kind of airport arrival rate prediction techniques, pass through the environmental data and practical machine to acquisition
Field arrival rate data establish airport arrival rate prediction model with algorithm with regress analysis method, obtain model parameter as training data,
Prediction airport arrival rate value is obtained, environmental data has tentatively been done into quantitative analysis to the influence of airport arrival rate, can predict future
Environment influences airport arrival rate, provides suggestion to the policymaker in traffic administration decision in the sky.
Second embodiment
Referring to Fig. 8, the embodiment of the invention provides a kind of airport arrival rate prediction technique, the present embodiment describes needle
For the process flow of selection area environment within a preset period of time and traffic data, which comprises
Step S310: selection area environmental data within a preset period of time and corresponding with the preset time period is obtained
Practical airport arrival rate data;
Step S320: it is based on the environmental data and preset data attribute extracting rule, the environmental data is converted
For environmental attribute data, the environmental attribute data includes environment attribute training data, the practical airport arrival rate data packet
Include the first practical airport arrival rate;
Step S330: based on the described first practical airport arrival rate and preset classification thresholds rule, it is practical to obtain first
Airport arrival rate classification data;
Step S340: it is reached based on the preset time period, the environment attribute training data, the first practical airport
Rate classification data and regression algorithm establish airport arrival rate prediction model;
Step S350: being based on the airport arrival rate prediction model, and the first prediction airport for calculating the selection area is arrived
Up to rate.
Step S360: it is based on the airport arrival rate prediction model and the environment attribute prediction data, verifies the machine
The performance of field arrival rate prediction model.
Based on step S360, the practical airport arrival rate further includes the second practical airport arrival rate, is based on described second
Practical airport arrival rate and the preset classification thresholds rule, obtain the second practical airport arrival rate classification data;The base
In the airport arrival rate prediction model and the environment attribute prediction data, the property of the airport arrival rate prediction model is verified
It can, comprising:
Based on the preset time period, the environment attribute prediction data, the second practical airport arrival rate classification number
Accordingly and airport arrival rate is predicted in the airport arrival rate prediction model, acquisition second;
Reach rate based on the second prediction airport arrival rate and the described second practical airport, obtains the second confusion matrix;
Based on second confusion matrix, the second classification performance index is calculated, to verify the airport arrival rate prediction
The performance of regression model, wherein the second classification performance index includes the second overall classification accuracy.
The embodiment of the present invention uses the performance of Logic Regression Models of data verification in 2014.For 2014 annual data collection
In each record, regression model predicts the level of airport arrival rate based on attribute value, and by these prediction results and reality
The airport arrival rate level on border compares.
The embodiment of the present invention is indicated by performance of the confusion matrix to model, wherein each column represent prediction class
Not, the classification results of test data are indicated, the sum of each column indicates the number for being predicted as the data of the category;Every a line generation
The table true belonging kinds of data, the data count of every a line indicate the number of the data instance of the category.Confusion matrix pair
Linea angulata value is bigger, and presentation class accuracy is higher.
Referring to Fig. 9, Fig. 9 is the schematic diagram of the second confusion matrix provided in second embodiment of the invention, every a line table
Show the second practical airport arrival rate (3 class), every a kind of the second prediction airport arrival rate (3 class) of expression, such as in the first row, records
The low airport arrival rate record of 1409+144+8=1561 in 2014, model correctly confirmed 1409 records, while wrong
144 medium levels records of erroneous judgement, 8 high level records, this kind of accuracy rate areDiagonal line position
The classification set shows that the second practical airport arrival rate is consistent with the second prediction airport arrival rate classification, there is 1409+681+240=
2330,3 class sums are 1409+681+240+144+8+87+136+111=2816, and the second overall classification accuracy is 2330/
2816=82.7%, this result illustrate that the performance of model is standard compliant // small in statistics and data mining document
Mark.The confusion matrix of regressor indicates airport arrival rate level and has obtained effective differentiation by returning, especially low airport
Arrival rate level can be effectively obtained prediction.
It is appreciated that the most important difference of airport arrival rate prediction technique that the present embodiment and first embodiment provide exists
In arrival rate prediction technique in airport provided in this embodiment further includes based on the airport arrival rate prediction model and the environment
Attribute forecast data verify the airport arrival rate prediction model performance.In other data processing principles and first embodiment
Principle it is similar, similarity can be repeated no more in the present embodiment with reference to the content in first embodiment.
The embodiment of the invention provides a kind of airport arrival rate prediction techniques, pass through the environmental data and practical machine to acquisition
Field arrival rate data establish airport arrival rate prediction model with algorithm with regress analysis method, obtain model parameter as training data,
And the airport arrival rate model performance is demonstrated, it can predict that FUTURE ENVIRONMENT influences airport arrival rate, to traffic pipe in the sky
The policymaker managed in decision provides suggestion.
3rd embodiment
Referring to Fig. 10, Figure 10 is the flow chart for the airport arrival rate prediction technique that third embodiment of the invention provides, this
Embodiment describes the process flow of the environment being directed to selection area within a preset period of time and traffic data, the method
Include:
Step S310: selection area environmental data within a preset period of time and corresponding with the preset time period is obtained
Practical airport arrival rate data;
Step S320: it is based on the environmental data and preset data attribute extracting rule, the environmental data is converted
For environmental attribute data, the environmental attribute data includes environment attribute training data, the practical airport arrival rate data packet
Include the first practical airport arrival rate;
Step S330: based on the described first practical airport arrival rate and preset classification thresholds rule, it is practical to obtain first
Airport arrival rate classification data;
Step S340: it is reached based on the preset time period, the environment attribute training data, the first practical airport
Rate classification data and regression algorithm establish airport arrival rate prediction model;
Step S350: being based on the airport arrival rate prediction model, and the first prediction airport for calculating the selection area is arrived
Up to rate.
Step S360: it is based on the airport arrival rate prediction model and the environment attribute prediction data, verifies the machine
The performance of field arrival rate prediction model;
Based on step S360, the practical airport arrival rate further includes the second practical airport arrival rate, is based on described second
Practical airport arrival rate and the preset classification thresholds rule, obtain the second practical airport arrival rate classification data;The base
In the airport arrival rate prediction model and the environment attribute prediction data, the property of the airport arrival rate prediction model is verified
It can, comprising:
Based on the preset time period, the environment attribute prediction data, the second practical airport arrival rate classification number
Accordingly and airport arrival rate is predicted in the airport arrival rate prediction model, acquisition second;
Reach rate based on the second prediction airport arrival rate and the described second practical airport, obtains the second confusion matrix;
Based on second confusion matrix, the second classification performance index is calculated, to verify the airport arrival rate prediction
Regression model performance, wherein the second classification performance index includes the second overall classification accuracy.
Step S370: the airport arrival rate prediction model, the environment attribute prediction data grade and the environment are based on
Attribute forecast data obtain the stability of the airport arrival rate prediction model.
Reach rate based on the described first practical airport arrival rate and first prediction airport, obtains the first confusion matrix;
Based on first confusion matrix, the first classification performance index is calculated, wherein the first classification performance index packet
Include the first overall classification accuracy;
Based on first overall classification accuracy and second overall classification accuracy, described first overall point is obtained
The calculating gap of class accuracy rate and second overall classification accuracy, if the calculating gap is greater than preset difference value, table
Show that the airport arrival rate prediction model is unstable, constantly modify the regression coefficient of the regression algorithm, obtains modified time
Reduction method, then based on the modified regression algorithm, the preset time period, the environment attribute training data, described the
One practical airport arrival rate classification data, establishes airport arrival rate prediction model, until the calculating gap is no more than described pre-
If difference, to obtain the stability of the airport arrival rate model.
Figure 11 is please referred to, Figure 11 is the first confusion matrix schematic diagram that third embodiment of the invention provides, i.e., 2011 arrive
Environmental attribute data in 2013 and the first practical airport arrival rate, based on the airport arrival rate for having obtained regression coefficient
Model obtains the first prediction airport arrival rate.Diagonal line indicates that the first practical airport arrival rate and the first prediction airport reach
Rate is classified correctly simultaneously, and 4028+2000+902=6930 totally has recorded 4028+317+12+207+2000+175+3+308+
902=7952, the first overall classification accuracy are 6930/7952=87.1%.Referring to Fig. 9, similarly, obtaining second overall point
Class accuracy rate is 82.7%.
The stability of one model is also a factor in need of consideration, and stability refers to the pre- of training sample and test sample
Survey the gap between accuracy rate.Accuracy rate difference is smaller, and model is more stable.More stable model, with the prediction on new data
Effect is better, can more generate and similarly classify with training sample set and prediction effect.Incorporated by reference to referring to Fig.1 1 and Fig. 9, can obtain
It is 87.1%-82.7%=4.4% to the first overall classification accuracy and the accurate gap of the second general classification, it is poor compares calculating
Away from the size with preset difference value, it is preferable that preset difference value 5%, 4.4% < 5% belongs to acceptable range, it was demonstrated that the machine
Field arrival rate prediction model has good stability.
It is appreciated that the most important difference of airport arrival rate prediction technique that the present embodiment and first embodiment provide exists
In arrival rate prediction technique in airport provided in this embodiment further includes based on the airport arrival rate prediction model and the environment
Attribute forecast data verify the airport arrival rate prediction model performance;Based on the airport arrival rate prediction model, the ring
Border attribute forecast data level and the environment attribute prediction data, obtain the stability of the airport arrival rate prediction model.Its
His data processing principle is similar to the principle in first embodiment, and similarity can refer to the content in first embodiment,
It is repeated no more in the present embodiment.
The embodiment of the invention provides a kind of airport arrival rate prediction techniques, pass through the environmental data and practical machine to acquisition
Field arrival rate data establish airport arrival rate prediction model with algorithm with regress analysis method, obtain model parameter as training data,
And the airport arrival rate model performance is demonstrated, and obtain the good stability of airport arrival rate model, it can predict
FUTURE ENVIRONMENT influences airport arrival rate, provides suggestion to the policymaker in traffic administration decision in the sky.
Fourth embodiment
Figure 12 is please referred to, fourth embodiment of the invention provides a kind of airport arrival rate prediction meanss 400, described device
400 include:
Obtain original data units 410, for obtain selection area environmental data within a preset period of time and with institute
State the corresponding practical airport arrival rate data of preset time period;
Environmental attribute data unit 420 is obtained, for being based on the environmental data and preset data attribute extracting rule,
The environmental data is converted into environmental attribute data, the environmental attribute data includes environment attribute training data, the reality
Border airport arrival rate data include the first practical airport arrival rate;
The first practical airport arrival rate taxon 430 is obtained, for based on the described first practical airport arrival rate and in advance
If classification thresholds rule, obtain the first practical airport arrival rate classification data;
Airport arrival rate prediction model unit 440 is established, for based on the preset time period, environment attribute training
Data, the first practical airport arrival rate classification data and regression algorithm, establish airport arrival rate prediction model;
The first prediction airport arrival rate unit 450 is calculated, for being based on the airport arrival rate prediction model, described in calculating
First prediction airport arrival rate of selection area.
It should be noted that each unit in the present embodiment can be and be realized by ageng, the above each unit equally may be used
To be realized by hardware such as IC chip.
5th embodiment
Figure 13 is please referred to, fifth embodiment of the invention provides a kind of airport arrival rate prediction meanss 500, described device
500 include:
Obtain original data units 510, in obtain selection area environmental data within a preset period of time and with it is described
The corresponding practical airport arrival rate data of preset time period;
Environmental attribute data unit 520 is obtained, it, will in being based on the environmental data and preset data attribute extracting rule
The environmental data is converted to environmental attribute data, and the environmental attribute data includes environment attribute training data, the reality
Airport arrival rate data include the first practical airport arrival rate;
Obtain the first practical airport arrival rate taxon 530, in based on the described first practical airport arrival rate and preset
Classification thresholds rule, obtain the first practical airport arrival rate classification data;
Airport arrival rate prediction model unit 540 is established, in based on the preset time period, environment attribute training number
According to, the first practical airport arrival rate classification data and regression algorithm, airport arrival rate prediction model is established;
The first prediction airport arrival rate unit 550 is calculated, for being based on the airport arrival rate prediction model, described in calculating
First prediction airport arrival rate of selection area;
The airport arrival rate prediction model unit 560 is verified, further includes environment attribute for the environmental attribute data
Prediction data is based on the airport arrival rate prediction model and the environment attribute prediction data, verifies the airport arrival rate
The performance of prediction model.
The airport arrival rate prediction model stability unit 570 is obtained, for predicting mould based on the airport arrival rate
Type, the environment attribute training data and the environment attribute prediction data, to obtain the airport arrival rate model
Stability.
In several embodiments provided herein, it should be understood that disclosed device and method can also pass through it
Its mode is realized.The apparatus embodiments described above are merely exemplary, for example, the flow chart and block diagram in attached drawing are aobvious
The device of multiple embodiments according to the present invention, architectural framework in the cards, the function of method and computer program product are shown
It can and operate.In this regard, each box in flowchart or block diagram can represent one of a module, section or code
Point, a part of the module, section or code includes one or more for implementing the specified logical function executable
Instruction.It should also be noted that function marked in the box can also be attached to be different from some implementations as replacement
The sequence marked in figure occurs.For example, two continuous boxes can actually be basically executed in parallel, they sometimes may be used
To execute in the opposite order, this depends on the function involved.It is also noted that each of block diagram and or flow chart
The combination of box in box and block diagram and or flow chart can be based on the defined function of execution or the dedicated of movement
The system of hardware is realized, or can be realized using a combination of dedicated hardware and computer instructions.
In addition, each functional module in each embodiment of the present invention can integrate one independent portion of formation together
Point, it is also possible to modules individualism, an independent part can also be integrated to form with two or more modules.
It, can be with if the function is realized and when sold or used as an independent product in the form of software function module
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic or disk.It needs
Illustrate, herein, relational terms such as first and second and the like be used merely to by an entity or operation with
Another entity or operation distinguish, and without necessarily requiring or implying between these entities or operation, there are any this realities
The relationship or sequence on border.Moreover, the terms "include", "comprise" or its any other variant are intended to the packet of nonexcludability
Contain, so that the process, method, article or equipment for including a series of elements not only includes those elements, but also including
Other elements that are not explicitly listed, or further include for elements inherent to such a process, method, article, or device.
In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including the element
Process, method, article or equipment in there is also other identical elements.
The foregoing is only a preferred embodiment of the present invention, is not intended to restrict the invention, for the skill of this field
For art personnel, the invention may be variously modified and varied.All within the spirits and principles of the present invention, made any to repair
Change, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.It should also be noted that similar label and letter exist
Similar terms are indicated in following attached drawing, therefore, once being defined in a certain Xiang Yi attached drawing, are then not required in subsequent attached drawing
It is further defined and explained.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.
Claims (7)
1. a kind of airport arrival rate prediction technique, which is characterized in that the described method includes:
It obtains selection area environmental data within a preset period of time and practical airport corresponding with the preset time period is arrived
Up to rate data;
Based on the environmental data and preset data attribute extracting rule, the environmental data is converted into environment attribute number
According to the environmental attribute data includes environment attribute training data, and the practical airport arrival rate data include the first practical machine
Field arrival rate;
Based on the described first practical airport arrival rate and preset classification thresholds rule, the first practical airport arrival rate classification is obtained
Data;
Based on the preset time period, the environment attribute training data, the first practical airport arrival rate classification data with
And regression algorithm, establish airport arrival rate prediction model;
Based on the airport arrival rate prediction model, the first prediction airport arrival rate of the selection area is calculated;
Wherein, the environmental attribute data further includes environment attribute prediction data, described based on the preset time period, described
Environment attribute training data, the first practical airport arrival rate classification data and regression algorithm, it is pre- to establish airport arrival rate
It surveys after model step, the method also includes: number is predicted based on the airport arrival rate prediction model and the environment attribute
According to verifying the performance of the airport arrival rate prediction model;
The practical airport arrival rate further includes the second practical airport arrival rate, is based on the described second practical airport arrival rate and institute
Preset classification thresholds rule is stated, the second practical airport arrival rate classification data is obtained;It is described pre- based on the airport arrival rate
Model and the environment attribute prediction data are surveyed, the performance of the airport arrival rate prediction model is verified, comprising:
Based on the preset time period, the environment attribute prediction data, the second practical airport arrival rate classification data with
And the airport arrival rate prediction model, obtain the second prediction airport arrival rate;
Reach rate based on the second prediction airport arrival rate and the described second practical airport, obtains the second confusion matrix;
Based on second confusion matrix, the second classification performance index is calculated, to verify the airport arrival rate prediction model
Performance, wherein the second classification performance index include the second overall classification accuracy.
2. the method as described in claim 1, which is characterized in that the environmental data includes METAR message Weather information;It is described
Based on the environmental data and preset data attribute extracting rule, the environmental data is converted into environmental attribute data, is wrapped
It includes:
Based on the METAR message Weather information and preset data attribute extracting rule, wind speed attribute data, visibility are extracted
Attribute data, thunderstorm attribute data and the low high attribute data of cloud.
3. method according to claim 2, which is characterized in that based on the preset time period, environment attribute training number
According to, the first practical airport arrival rate classification data and regression algorithm, airport arrival rate prediction model is established, comprising:
Using multinomial logistic regression algorithm, using the preset time period, the environment attribute training data as regressor, with
The first practical airport arrival rate classification data is sort capacity, establishes airport arrival rate prediction model.
4. the method as described in claim 1, which is characterized in that the method also includes:
Reach rate based on the described first practical airport arrival rate and first prediction airport, obtains the first confusion matrix;
Based on first confusion matrix, the first classification performance index is calculated, wherein the first classification performance index includes the
One overall classification accuracy;
Based on first overall classification accuracy and second overall classification accuracy, it is quasi- to obtain first general classification
The calculating gap of true rate and second overall classification accuracy, if the calculating gap is greater than preset difference value, then it represents that institute
It is unstable to state airport arrival rate prediction model, constantly modifies the regression coefficient of the regression algorithm, obtains modified recurrence and calculates
Method, then in fact based on the modified regression algorithm, the preset time period, the environment attribute training data, described first
Border airport arrival rate classification data establishes airport arrival rate prediction model, until the calculating gap is not more than the default difference
Value, to obtain the stability of the airport arrival rate model.
5. the method as described in claim 1, which is characterized in that described based on the described first practical airport arrival rate data and pre-
If classification thresholds rule, obtain the first airport arrival rate classification data, comprising:
Based on the described first practical airport arrival rate data, the frequency disribution for obtaining the first practical airport arrival rate data is straight
Fang Tu;Based on the frequency distribution histogram and the preset classification thresholds rule, first airport arrival rate point is obtained
Class data.
6. a kind of airport arrival rate prediction meanss, which is characterized in that described device includes:
Obtain original data units, for obtain selection area environmental data within a preset period of time and with it is described default when
Between the corresponding practical airport arrival rate data of section;
Environmental attribute data unit is obtained, it, will be described for being based on the environmental data and preset data attribute extracting rule
Environmental data is converted to environmental attribute data, and the environmental attribute data includes environment attribute training data, the practical airport
Arrival rate data include the first practical airport arrival rate;
The first practical airport arrival rate taxon is obtained, for being based on the described first practical airport arrival rate and preset classification
Threshold rule obtains the first practical airport arrival rate classification data;
Airport arrival rate prediction model unit is established, for being based on the preset time period, the environment attribute training data, institute
The first practical airport arrival rate classification data and regression algorithm are stated, airport arrival rate prediction model is established;
It calculates the first prediction airport arrival rate unit and calculates the selected area for being based on the airport arrival rate prediction model
The first prediction airport arrival rate in domain;
Wherein, the environmental attribute data further includes environment attribute prediction data, described device further include:
The airport arrival rate prediction model unit is verified, for being based on the airport arrival rate prediction model and the environment category
Property prediction data, verifies the performance of the airport arrival rate prediction model;
The practical airport arrival rate further includes the second practical airport arrival rate, described device further include:
Second practical airport arrival rate classification data obtaining unit, for based on the described second practical airport arrival rate and described pre-
If classification thresholds rule, obtain the second practical airport arrival rate classification data;
The verifying airport arrival rate prediction model unit includes:
Second prediction airport arrival rate obtaining unit, for being based on the preset time period, the environment attribute prediction data, institute
The second practical airport arrival rate classification data and the airport arrival rate prediction model are stated, the second prediction airport is obtained and reaches
Rate;
Second confusion matrix obtaining unit, for being reached based on the second prediction airport arrival rate with the described second practical airport
Rate obtains the second confusion matrix;
Second classification performance indicator calculating unit, for calculating the second classification performance index based on second confusion matrix, from
And verify the performance of the airport arrival rate prediction model, wherein the second classification performance index includes the second general classification
Accuracy rate.
7. device as claimed in claim 6, which is characterized in that described device further include:
The airport arrival rate prediction model stability unit is obtained, for based on the airport arrival rate prediction model, described
Environment attribute training data and the environment attribute prediction data, to obtain the stability of the airport arrival rate model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611190122.7A CN106408139B (en) | 2016-12-20 | 2016-12-20 | Airport arrival rate prediction technique and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611190122.7A CN106408139B (en) | 2016-12-20 | 2016-12-20 | Airport arrival rate prediction technique and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106408139A CN106408139A (en) | 2017-02-15 |
CN106408139B true CN106408139B (en) | 2019-08-06 |
Family
ID=58087676
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611190122.7A Active CN106408139B (en) | 2016-12-20 | 2016-12-20 | Airport arrival rate prediction technique and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106408139B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108152866B (en) * | 2017-11-06 | 2020-07-07 | 南京航空航天大学 | Aviation weather forecast quality evaluation method based on weather influence flight quantity |
CN107844602B (en) * | 2017-11-24 | 2021-03-16 | 重庆邮电大学 | Prediction method based on spatio-temporal attribute association rule |
CN109409587A (en) * | 2018-10-09 | 2019-03-01 | 南京航空航天大学 | A kind of airport excavated based on weather data is into traffic flow forecasting method of leaving the theatre |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104156594A (en) * | 2014-08-11 | 2014-11-19 | 中国民航大学 | Dynamic flight station-crossing time estimation method based on Bayes network |
WO2015134425A1 (en) * | 2014-03-03 | 2015-09-11 | Inrix Inc. | Assessing environmental impact of vehicle transit |
CN105528647A (en) * | 2015-11-25 | 2016-04-27 | 南京航空航天大学 | Airport traffic demand possibility prediction method based on big data analysis |
CN105844346A (en) * | 2016-03-17 | 2016-08-10 | 福州大学 | Flight delay prediction method based on ARIMA model |
-
2016
- 2016-12-20 CN CN201611190122.7A patent/CN106408139B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015134425A1 (en) * | 2014-03-03 | 2015-09-11 | Inrix Inc. | Assessing environmental impact of vehicle transit |
CN104156594A (en) * | 2014-08-11 | 2014-11-19 | 中国民航大学 | Dynamic flight station-crossing time estimation method based on Bayes network |
CN105528647A (en) * | 2015-11-25 | 2016-04-27 | 南京航空航天大学 | Airport traffic demand possibility prediction method based on big data analysis |
CN105844346A (en) * | 2016-03-17 | 2016-08-10 | 福州大学 | Flight delay prediction method based on ARIMA model |
Non-Patent Citations (1)
Title |
---|
空域容量评估与预测技术研究;赵征;《中国博士学位论文全文数据库》;20160731;第C031-35页 |
Also Published As
Publication number | Publication date |
---|---|
CN106408139A (en) | 2017-02-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11501039B2 (en) | Optimizing aircraft flows at airports using data driven predicted capabilities | |
CN109409587A (en) | A kind of airport excavated based on weather data is into traffic flow forecasting method of leaving the theatre | |
Cui et al. | Wind power ramp event forecasting using a stochastic scenario generation method | |
Conde Rocha Murca et al. | Trajectory clustering and classification for characterization of air traffic flows | |
CN110503245B (en) | Prediction method for large-area delay risk of airport flight | |
Palutikof et al. | A review of methods to calculate extreme wind speeds | |
CN103530704A (en) | Predicating system and method for air dynamic traffic volume in terminal airspace | |
CN106408139B (en) | Airport arrival rate prediction technique and device | |
CN106651088B (en) | Flight quality monitoring method based on temporal GIS | |
Xue et al. | Using stochastic, dynamic weather-impact models in strategic traffic flow management | |
Provan et al. | A probabilistic airport capacity model for improved ground delay program planning | |
Geetha et al. | Time-series modelling and forecasting: Modelling of rainfall prediction using ARIMA model | |
Nakamura et al. | HITS: Hurricane intensity and track simulator with North Atlantic Ocean applications for risk assessment | |
CN106779222B (en) | Airport ground waiting time prediction method and device | |
CN106710316A (en) | Method and device of airspace capacity determination based on bad weather condition | |
CN112232535A (en) | Flight departure average delay prediction method based on supervised learning | |
CN111667108A (en) | Distribution network first-aid repair time length prediction method | |
CN110796315B (en) | Departure flight delay prediction method based on aging information and deep learning | |
Klüver et al. | A self-enforcing neural network as decision support system for air traffic control based on probabilistic weather forecasts | |
Angalakudati et al. | Improving emergency storm planning using machine learning | |
Schultz et al. | Classification of weather impacts on airport operations | |
CN111427101B (en) | Thunderstorm strong wind grading early warning method, system and storage medium | |
Dhal et al. | An operations-structured model for strategic prediction of airport arrival rate and departure rate futures | |
Du et al. | Decision support for aircraft taxi time based on deep metric learning | |
McGovern et al. | Enhanced spatiotemporal relational probability trees and forests |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |