CN110095777A - Fuzzy logic method meteorology particle identification method based on shuffling technology - Google Patents
Fuzzy logic method meteorology particle identification method based on shuffling technology Download PDFInfo
- Publication number
- CN110095777A CN110095777A CN201910326685.1A CN201910326685A CN110095777A CN 110095777 A CN110095777 A CN 110095777A CN 201910326685 A CN201910326685 A CN 201910326685A CN 110095777 A CN110095777 A CN 110095777A
- Authority
- CN
- China
- Prior art keywords
- data
- particle
- meteorology
- fuzzy logic
- shuffling technology
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/95—Radar or analogous systems specially adapted for specific applications for meteorological use
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Electromagnetism (AREA)
- Devices For Executing Special Programs (AREA)
Abstract
The present invention discloses a kind of fuzzy logic method meteorology particle identification method based on shuffling technology, comprising the following steps: is (1) blurred；(2) rule induction；(3) integrate；(4) move back fuzzy.The parameter matrix and obtained sounding data that recognition methods of the invention utilizes radar detection to arrive, by the transformation rule as described in membership function, final inverting obtains the matrix of particle phase.Recognition methods process of the invention is succinctly understandable, improves recognition efficiency using Matlab and Java shuffling technology, friendly interactive interface provides better usage experience for user.
Description
Technical field:
The present invention relates to weather radar field more particularly to a kind of meteorological grains based on Java Yu Matlab shuffling technology
Sub- recognition methods.
Background technique:
The phase of water-setting object is developed in cloud leads in weather modification, weather phenomenon forecast, aircraft icing aviation safety etc.
Domain all has larger impact, but the shortage of detection means and data inversion precision deficiency are to restrict the development of correlative study
Main cause.Mainly there are laser radar, microwave radiometer, ceilometer and radio to the detection means of cloud both at home and abroad
Sounding etc. can effectively obtain the macro microphysics parameter information in cloud by these modes.Weather radar is outstanding to cloud particle
It is that the detection echo strength of cirrus ice crystals is weaker, can not utilize the echo information accurately inverting cloud particle of weather radar
The macro microphysics parameter of son, and millimetre-wave radar is to the cumulus under non-Precipitation Clouds such as shallow-layer cloud, altostratus, altocumulus, fine day
And cirrus has good detectivity.The millimetre-wave radar having gradually developed these years is in addition to being capable of continuous observation cloud
Horizontal vertical structure change can also obtain micro-physical feature in cloud, as cloud particle size, concentration, drop-size distribution distribution, ice with
The content etc. of liquid water, this provides support for the parameter of inverting cloud particle, improves worker in meteorology and identifies water-setting object grain in cloud
The accuracy of son, has great importance for the research of weather and weather.
Currently, in cloud water-setting object particle types identification with classification main algorithm there are four types of.It is respectively: (1) refreshing
Through network algorithm；(2) decision Tree algorithms；(3) statistical decision algorithm and (4) fuzzy logic algorithm, these algorithms can distinguish water
The type of condensate, wherein there is decision tree method the mode that speed is fast, precision is high, generates to be easily understood and be easy conversion ingredient
The advantages that rule-like, but due to using greedy algorithm in the construction process of decision tree, thus cause decision tree scale mistake
Greatly, the disadvantages of regular length generated is too long.Statistic decision method can not only disclose the knot of problem with the formal intuition of graph theory
Structure, and can be rearranged according to structure of the probability theory principle to problem, it is a system complicated joint probability distribution PROBLEM DECOMPOSITION
Relatively simple module is arranged, greatly reduces the difficulty and complexity solved the problems, such as, but it is that it is apparent scarce that uncertainty is excessive
Point.Neural network have certain learning functionality, when learning sample is larger, the feature difference of different mode sample compared with
The problem of hour, convergence rate can be relatively slow, locally will appear smaller value.And what Zedah was proposed is used to solve fuzzy problem
Fuzzy logic algorithm ambiguous object can accurately be described, be handled, fuzzy logic algorithm has certain extendibility and simultaneous
Capacitive, algorithmic rule are relatively simple compared to other algorithms, are appropriate for cloud particle identification work, meanwhile, it is patrolled with fuzzy
The repeated work of particle identification system can be reduced by collecting algorithm, so that whole system is succinctly efficient.
Simply by compare range, the methods of table look-up to water-setting object particle in cloud carry out identification be it is unworkable, it is former
Because packet is lived: the case where being (1) overlapped between the obtained each variable of radar detection, supplemental characteristic there is variable.(2) to grain
During subclassification, the value range of the particle of different phase is not to uniquely determine.
Summary of the invention:
To solve the above problems, the present invention provides a kind of fuzzy logic method gas based on Java Yu Matlab shuffling technology
As particle identification method, its technical solution is as follows:
A kind of fuzzy logic method meteorology particle identification method based on shuffling technology, comprising the following steps:
Step 1: blurring reads reflectivity factor, doppler velocity, spectrum in millimetre-wave radar data by Matlab
Temperature data in wide data and sonde data, then using the data of reading as four input parameters, and establish be subordinate to respectively
Input parameter is converted Fuzzy dimension by function；
Step 2: rule induction passes through Matlab by the foundation of membership function and the blurring of four input parameters
To in millimetre-wave radar data and sonde data each put input parameter corresponding to snow, ice, mix phase, liquid water,
The threshold values of the phase of this 6 kinds of particle types of drizzle, rain carries out regular deduction；
Step 3: it is integrated, intensity S is contributed to each particle phase obtained in rule deductionjData use and take maximum
The method of value carries out value；
Step 4: move back it is fuzzy, finally obtain each point particle phase type and phase contribution intensity SjThe index of data
Value, determines that the point corresponds to the type of any particle phase according to index value.
Preferably, in step 1, the membership function is shown below:
Preferably, in step 2, regular deduction is carried out as follows:
Wherein: PijIndicate i-th of parameter to the degree of membership of jth kind particle types；AiFor weight coefficient, i-th ginseng is indicated
The several pairs of weights and ratio for judging processing result, 0≤Ai≤1.
Preferably, in step 2, it is preferable that weight coefficient Ai=1.
Preferably, further include following steps:
Step 5: import data interface using Java creation and show interface, method in 1~step 4 of invocation step, to
Family imports data and is handled.
Preferably, in step 5, method in the 1~step 4 of invocation step specifically: write step first with Matlab
Multiple M files of rapid 1~step 4 method call Deployment Tool that the multiple M files write are packaged into one
JAR packet, then JAR packet is introduced into established Java project.
Preferably, in step 5, data interface is for allowing user to import millimetre-wave radar, sonde data file；It shows
Interface is for showing meteorological particle identification effect picture.
Preferably, in step 1, millimetre-wave radar detection data is nc data file, and sonde detection data is dat number
According to file.
The present invention has the following beneficial effects: compared with the prior art
The present invention utilizes fuzzy logic algorithm pair by pre-processing to millimetre-wave radar data and sonde data
Meteorological particle classifying simultaneously marks, and has the following beneficial effects:
(1) present invention incorporates the succinct quick computing capability of the interactive interface of Java close friend and Matlab, be conducive to mention
High service efficiency and calculating speed carry out correlative study for meteorological research worker and provide conveniently.
(2) present invention needs effectively to handle all kinds of radar datas and explicitly divide, and selection is patrolled with fuzzy
It collects algorithm radar data is handled and classified, the repeated work of particle identification system is reduced, so that whole system is succinctly high
Effect
(3) present invention directly finds corresponding particle types using calculated index value and exports as a result, convenient for scholar
It calculates and uses when study, research.
(4) present invention can show particle types by image, and effect is intuitively and conveniently.
Detailed description of the invention:
Fig. 1 is control flow chart in the embodiment of the present invention；
Fig. 2 is fuzzy logic method identification process schematic diagram in the embodiment of the present invention；
Fig. 3 is operational flow diagram during the present invention is implemented；
Specific embodiment:
Embodiment one:
The present embodiment uses a kind of fuzzy logic method meteorology particle identification side based on shuffling technology provided by the invention
Method, as shown in Figure 1, comprising the following steps:
Step 1: blurring reads reflectivity factor, doppler velocity, spectrum in millimetre-wave radar data by Matlab
Temperature data in wide data and sonde data, then using the data of reading as four input parameters, and establish be subordinate to respectively
Input parameter is converted Fuzzy dimension by function；Millimetre-wave radar detection data is nc data file；Sonde detection data is
Dat data file.Wherein, membership function is shown below:
Step 2: rule induction passes through Matlab by the foundation of membership function and the blurring of four input parameters
To in millimetre-wave radar data each put input parameter corresponding to snow, ice, mix phase, liquid water, drizzle, rain this 6
The threshold values (24 values) of the phase of kind particle types carries out regular deduction；
Step 3: it is integrated, intensity S is contributed to each particle phase obtained in rule deductionjData use and take maximum
The method of value carries out value；Wherein, rule is inferred and is shown below:
Wherein: PijIndicate i-th of parameter to the degree of membership of jth kind particle types.AiFor weight coefficient, i-th of ginseng is indicated
The several pairs of weights and ratio for judging processing result, 0≤Ai≤1.The present embodiment, Ai=1.
Step 4: move back it is fuzzy, finally obtain each point particle phase type and phase contribution intensity SjThe index of data
Value, determines that the point corresponds to the type of any particle phase according to index value.
Step 5: importing data interface using Java creation and show interface, recycle and called in Matlab
Multiple M files of method in step 1~step 4 are packaged into a JAR packet by DeploymentTool together, then JAR packet is drawn
Enter into the Java project established to call directly method in step 1~step 4 (class and method in JAR packet), to
Family imports data and is handled.Wherein, data interface is shown for allowing user to import millimetre-wave radar, sonde data file
Interface is for showing meteorological particle identification effect picture.
Embodiment two:
The present embodiment uses a kind of fuzzy logic method meteorology particle identification side based on shuffling technology provided by the invention
Method, comprising the following steps:
1) the related M file of fuzzy logic algorithm is write in Matlab, reads in radar data, carries out data identification, with
It exports image and is packaged into a JAR packet, i.e. a .JAR file as recognition result, and by this .M file；
2) Java engineering project is created, a main window class MainFrame, the building method in the forms class are edited
In forms attribute itself, initial window inter-related components, such as label (JLabel), text box are respectively set
(JTextField), button (JButton) etc. creates child window, for showing meteorological particle identification effect picture；
3) JavaBuilderAPI provided by JAR packet that Matlab is obtained and official website is building up to project translation and compiling environment
Under copy in project under lib catalogue；
4) the Java project finished writing is packaged, generates the executable file under Windows environment, i.e. .exe file.
Wherein, the principle of fuzzy logic algorithm identification cloud particle phase is mainly the ginseng for utilizing radar detection to arrive in step 1
Matrix number and obtained sounding data, by the transformation rule as described in membership function, final inverting obtains particle phase
The matrix of state.Because temperature is to identify a very crucial parameter of meteorological particle phase, be conducive to promote the correct of identification
Rate, so the temperature data of sonde measurement is also by as input variable.Using membership function system to reflectivity factor, how general
It strangles this four average speed, spectrum width, temperature input parameters to be blurred, be referred in the specific different membership functions of construction
The threshold value that Shupe is provided.For 6 kinds of meteoric water condensate particle types, four kinds of parameters establish membership function, and output variable is
The particle types (Q) of water-setting object, after this genealogical classification, the output type definition of 6 kinds of rainfall particle phases is as shown in table 1:
The output type of 1 fuzzy logic system of table
| Precipitation particles type (Hydrometeortype) | It exports result (OutputQ) |
| It avenges (Snow) | 1 |
| Ice (Ice) | 2 |
| It mixes phase (Mixed) | 3 |
| Liquid water (Liquid) | 4 |
| Drizzle (Drizzle) | 5 |
| Rain (Rain) | 6 |
Known by table 1, fuzzy logic system is for snow, ice, mixing phase, liquid water, drizzle, rain this six kinds of precipitation grains
The identifying processing result of subtype is respectively 1,2,3,4,5,6.
Fuzzy logic algorithm in step 1 as shown in Fig. 2, its detailed process is as follows:
1.1) using the reflectivity factor of millimeter wave cloud radar, Doppler's average speed, spectrum width, temperature as input variable,
Fuzzy dimension is converted by 4 input parameters, the process of conversion then relies primarily on membership function realization；The present embodiment, which uses, to be subordinate to
Citation form of the wherein asymmetric trapezoidal function of function as membership function, concrete form are as follows:
Definition T is degree of membership, and x is parameter, and the value range of parameter variable is different, and x has different degrees of membership.Determining T
The coefficient X of function1、X2、X3、X4When provided with reference to Shupe threshold value --- Shupe summarized millimetre-wave radar in 2007
Four kinds of parametric datas correspond to the threshold value of water-setting object particle in different type cloud, these four parametric datas include reflectivity factor
(Zh), Doppler's average speed (VD), spectrum width (WD) and temperature (T) (present invention in these parameters can manually be adjusted
It is whole), as shown in the table:
Water-setting object member function coefficient in 2 cloud of table
(in table, P function is a kind of expression form of T function)
As seen from the above table, our available 4 parameters, 4 parameters can each corresponding meteorological particle phase,
It amounts to and needs 24 membership functions.
It 1.2), can be to 4 parameters of each point by the foundation of membership function and the blurring of four parametric variables
24 values of corresponding 6 kinds of particle types carry out regular deduction, are as a result exactly four parameters of this point to jth in 6 class particles
The contribution intensity S of class particlej.Inferred using the rule of following particle phase type, and provides Ai=1:
Wherein: PijIndicate i-th of parameter to the degree of membership of jth kind particle types.AiFor weight coefficient, i-th of ginseng is indicated
The several pairs of weights and ratio for judging processing result.
1.3) to S obtained in rule deductionjData are using the method being maximized, i.e., maximum integrated, selection is maximum
Sj, i.e. Q=maxSj；
1.4) index value for finally, obtaining each particle phase type shown in table 1, can obtain according to index value
The point corresponds to the type of any particle phase.
In technical solution mentioned above, key is the shuffling technology of Matlab and Java.
Matlab is by the publication of mathworks company, the U.S. mainly in face of scientific algorithm, visualization and interactive journey
The high-tech of sequence design calculates environment.It calculates numerical analysis, matrix, science data visualize and Nonlinear Dynamic system
Many powers such as the modeling and simulation of system are integrated in a wieldy windowing environment, are set for scientific research, engineering
Meter and the numerous scientific domains that must carry out Effective Numerical calculating provide a kind of comprehensive solution, and in very great Cheng
The edit pattern that traditional noninteractive program design language (such as C, Fortran) is got rid of on degree, represents current international scientific
The advanced level of software for calculation.
Java is that the world today is most important, is also one of most popular computer language.Java is initially by Sun public affairs
Department's publication, becomes more powerful instead over time.Java is an Object-Oriented Programming Language, not only absorbs C
++ the various advantages of language have also abandoned the concepts such as elusive more succession, pointers in C++, therefore Java language has function
It can powerful and easy to use two features.Representative of the Java language as static Object-Oriented Programming Language, admirably realizes
Object-oriented theory allows programmer to carry out complicated programming with the graceful mode of thinking.Java has simplicity, towards right
As, distribution, robustness, safety, platform it is independent with portability, multithreading, dynamic the features such as.Java can write
Multipad, web application, distributed system and embedded system application program etc..
JDK (Java Development Kit) is the Software Development Kit of Java language, is mainly used for movement and sets
Java application standby, on embedded device.JDK is the core of entire Java exploitation, it contains the running environment of Java
(JVM+Java system class libraries) and Java tool.Nowadays JDK has been updated to 10 versions.
Since Java has portability, so that may operate in different hardware environment using the program of written in Java
In operating system, such as Windows, Linux, Android, IOS etc..
It includes following two method that this .M file, which is packaged into a JAR packet, in step 1:
Matlab program is compiled into JAR packet by method one:
Call Deployment Tool M file, i.e. Matlab program can be broken into JAR packet using in Matlab, it will
JAR packet is introduced into Java project the class and method that can be called directly in JAR packet.
Matlab program is compiled as C/C++ dynamic base DLL by method two；Matlab program is compiled into DLL dynamic chain
After connecing library, need to call in java applet by JNI (JavaNative Interface) technology.This method is more multiple
It is miscellaneous, need to write Java Native class and C/C++ code, but it is more advantageous in operational efficiency.
Either any implementation method all be unable to do without MCR (Matlab Complier Runtime) environment, and MCR is one
Group standard DLL, Essential Environment needed for operation program is provided.It to run on other computers, must just install
MCR.Java version: Java (TM) SE Runtime Environment (build 1.8.0_91-b15)；Develop environment:
Matlab R2015b, EclipseOxygen.3Release (4.7.3)；MCR version: Matlab Runtime 9.0；System
Environment: Windows 10x64
The method that the present embodiment step 1 uses comprises the following steps: inputting in the command window of Matlab first
Deploytool opens compilation tool case, selects Java Package, clicks after the .M file compiled required for importing
Package button waits completion to be compiled, obtains JAR packet.After .M file is compiled into JAR packet, JAR packet is imported into project
Under classpath environment --- by the dependence JAR packet Javabuilder.JAR of Matlab and obtained JAR packet replication to item
In mesh under lib catalogue, and add build path.
Software utilizes Java Swing frame, using MVC three-tier design mode.MVC design mode is model (Model)-
View (View)-controller (Controller) design pattern abbreviation, is a kind of a kind of modular design mould of software design
Formula.MVC design mode is a kind of software design pattern that Xerox PARC is invented in the 1980s, wide
The use of general ground is in field of software development, and wherein model layer includes data model and business model, data mould in the method for the present invention
Type exists in the form of nc file, and is obtained from user interface, and the processing of data and ways of presentation are business model；Depending on
Figure layer is absorbed in the data exhibiting of user and close friend's interaction, and user data input, software function navigation and help are such as obtained
Deng；Controller is present between model layer and view layer, and there are three tasks for controller main, obtains the input of user, calls
Data or business model complete corresponding business, and the result returned to business model is packed, and then update view layer, do
It is synchronous to the data between data Layer and view layer.
The technology for implementing MVC design mode is Java Swing, and Swing is one for developing Java application journey
The development kit of sequence user interface.The view interface of the software is by file input frame, parameter input frame and a series of button groups
At.All controls in view all use event model, user by mouse, keyboard light terminal generate event in event source it
In, the generation of event is monitored in event source by monitor (Listener), calls event handling function finishing service logic.
Java gets the inputted related data of user by event sniffer, creates model layer instance objects, calls its related side
Method.Calculation processing is carried out by MCR environment, generates associated picture, and be output to the place mesh for the data file that user is inputted
Record.
Embodiment three:
The present embodiment uses a kind of fuzzy logic method meteorology particle identification side based on shuffling technology provided by the invention
Method identifies meteorological particle, comprising the following steps:
Correlated variables in step 1. millimetre-wave radar data file is stored in the form of array, is read by Matlab
The correlated variables of millimetre-wave radar, is followed successively by that " time (time) ", " rang (height) ", " Zh (reflectivity factor) ", " v is (more
General Le speed) ", " width (spectrum width) "；Read sonde temperature data.Wherein, millimetre-wave radar " Zh (reflectivity factor) ",
" v (doppler velocity) ", " width (spectrum width) " and sonde temperature " T (temperature) " are that four inputs of fuzzy logic algorithm are joined
Number；" time (time) ", " rang (height) " are for establishing coordinate system, output identification image.Matlab is built-in for reading thunder
Up to data nc file and the function of sounding temperature dat file, user is facilitated to read related data
Millimetre-wave radar all data is all stored in nc format in the form of array, and sounding temperature data is also with array
Form be stored in dat format.Method need to read variable " time (time) " in millimetre-wave radar data file,
" range (height) " is two identical one-dimension arrays；The variable " Zh (reflectivity factor) " of reading millimetre-wave radar,
" v (doppler velocity) ", " width (spectrum width) ", three variables are the identical two-dimensional array of three sizes, this two-dimensional array
It is characterized in that, one of dimension size of variable is identical as variable range, that is, representing the variable can be with matched, side
What is used in method uses Matlab to have the function for easily reading our above-mentioned required variables in millimetre-wave radar,
And variable is input in Matlab in the form of array；The data characteristics that sonde detects are that parameters are stored in
In 114 × 11 array, second is classified as " height (height) ", and third is classified as " T (temperature) ", and linear correlation.
Step 2. is input variable using reflectivity factor, doppler velocity, spectrum width, environment temperature, it is known that blurring
Concrete form as shown by the following formula, therefore write first four for blurring function, respectively correspond reflectivity factor, more
General Le speed, spectrum width, temperature；
The every meteorological data read is blurred by step 3. using fuzzification function, is made using following formula
Inferred for rule of inference, and provides Ai=1；The data that rule of inference obtains are maximized, are finally obtained such as 1 institute of table
The index value for each particle phase type shown can obtain the class that the point corresponds to any particle phase according to index value
Type, wherein rule of inference formula are as follows:
Deploytool is inputted in the command window of Matlab and opens compilation tool case, is selected Java Package, is led
Package button is clicked after the .M file compiled required for entering, is waited completion to be compiled, is obtained JAR packet.
Step 4. creates a Java project, wraps view for realizing interface function, wherein have DateInputFrame,
3 classes of DisplayPicture and MainFrame.A main window class MainFrame is edited, such is inherited from Java
JFrame window class.Forms attribute itself, correlation inside initial window are respectively set in the building method in the forms class
Component, such as label (JLabel), text box (JTextField), button (JButton).Code is write to be arranged outside each component
Pattern, such as window size, window title, each button size are seen, interface is finally obtained.
Know figure button event monitoring call service class run function, be passed to ncFilePathTextField and
The value of the Text of DatFilePathTextField calls the code for connecting the part Matlab.And it is passed to
The merging character string of ncFilePathTextField and " particle types .png " obtain the picture that Matlab code section generates
File path for new window show picture, class DisplayPicture be used for show Matlab generate image results.
Step 5. creates a Java project, for creating child window.Structure in the affiliated JFrame forms class of the window
Itself attribute that the window is respectively set in method is made, each component appearance style is arranged in initial window inter-related components,
Add panel container class JPanel.A class ShowImage Panel is edited for showing image, is inherited from Java
JPanel panel container class.These components are the event source of generation event, add monitor to these components, write at event
Function is managed, --- meteorological particle phase diagram.
JavaBuilderAPI provided by the JAR packet that Matlab is obtained and official website is building up to project compiling by step 6.
It is copied under environment in project under lib catalogue.A Java project is created, corresponding calling JAR packet is write in grassroot project
Code, write corresponding class call method --- using run function call Matlab code, parameter is in run function
NcPath, datPath, xDatpath indicate data file path, and write the customized exception class in Java, enable program
Enough in the case where user's operation correctly opens nc file and dat file, Matlab algorithm is called to handle data.
Step 7. is packaged the Java project finished writing using exe4j tool, generates holding under Windows environment
Style of writing part, i.e. .exe file, specific steps are as follows:
Step 7.1. creates a file as exe4j output directory, needs to break into the text of exe for storing us
Part, and following catalogue: bin file folder is established in this document folder, for storing program classes file；Lib file is used
In the item that stock item relies on；Jre file, for storing jdk；
Step 7.2. opens exe4j software, and the step according to software is operated, selected when selecting works category
Select Regular mode mode.The main classes Launcher for adding lib packet in software according to step, bin file folder, running, adds
File copy in need to pressing from both sides the bin file in eclipse when bin file being added to press from both sides is pressed from both sides to the bin file in built output directory
In；
Step 7.3. configures jre and client mode, complete until being packaged according to software prompt；
As shown in figure 3, the present invention is after opening program, by user software interface input corresponding radar data file and
Path where the data file of sonde detection, click recognition button, if the file presence such as file format of user's input,
The mistakes such as file address, then interface can provide friendly guidance, if the file of user's input does not have mistake, may wait for identifying
It completes, and data processing is carried out to data file and is analyzed, result is shown with pictorial manner.Mistake of the present invention for user
Operation and incorrect operation, software makes friendly interface guidance will be in image and Matlab while image display
.fig file be output under the place catalogue of data file together, facilitate user to the Reusability of image.
Claims (8)
1. a kind of fuzzy logic method meteorology particle identification method based on shuffling technology, it is characterised in that: the following steps are included:
Step 1: blurring reads reflectivity factor, doppler velocity, spectrum width data in millimetre-wave radar data by Matlab
And temperature data in sonde data, then using the data of reading as four input parameters, and membership function is established respectively, it will
Input parameter is converted into Fuzzy dimension；
Step 2: rule induction, by the foundation of membership function and the blurring of four input parameters, by Matlab to milli
Snow corresponding to the input parameter that each in metre wave radar data and sonde data is put, ice, mixing phase, liquid water, mao mao
The threshold values of the phase of this 6 kinds of particle types of rain, rain carries out regular deduction；
Step 3: it is integrated, intensity S is contributed to each particle phase obtained in rule deductionjData are using the side being maximized
Method carries out value；
Step 4: move back it is fuzzy, finally obtain each point particle phase type and phase contribution intensity SjThe index value of data, according to
Index value determines that the point corresponds to the type of any particle phase.
2. the fuzzy logic method meteorology particle identification method according to claim 1 based on shuffling technology, in step 1, institute
Membership function is stated to be shown below:
3. the fuzzy logic method meteorology particle identification method according to claim 2 based on shuffling technology, in step 2, press
Following formula carries out regular deduction:
Wherein: PijIndicate i-th of parameter to the degree of membership of jth kind particle types；AiFor weight coefficient, i-th of parameter pair is indicated
Judge the weight and ratio of processing result, 0≤Ai≤1。
It is excellent in step 2 4. the fuzzy logic method meteorology particle identification method according to claim 3 based on shuffling technology
Selection of land, weight coefficient Ai=1.
5. the fuzzy logic method meteorology particle identification method according to claim 1 to 4 based on shuffling technology, is also wrapped
Include following steps:
Step 5: importing data interface using Java creation and show interface, method in 1~step 4 of invocation step leads user
Enter data to be handled.
6. the fuzzy logic method meteorology particle identification method according to claim 5 based on shuffling technology, in step 5, institute
State method in 1~step 4 of invocation step specifically: the multiple M files of step 1~step 4 method are write first with Matlab,
It calls the Deployment Tool the multiple M files write are packaged into a JAR packet, then JAR packet is introduced into and is established
In Java project.
7. the fuzzy logic method meteorology particle identification method according to claim 6 based on shuffling technology, in step 5, number
According to interface for allowing user to import millimetre-wave radar, sonde data file；Show interface for showing meteorological particle identification effect
Fruit figure.
8. the fuzzy logic method meteorology particle identification method according to claim 1 based on shuffling technology, in step 1, milli
Metre wave radar detection data is nc data file, and sonde detection data is dat data file.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910326685.1A CN110095777A (en) | 2019-04-23 | 2019-04-23 | Fuzzy logic method meteorology particle identification method based on shuffling technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910326685.1A CN110095777A (en) | 2019-04-23 | 2019-04-23 | Fuzzy logic method meteorology particle identification method based on shuffling technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110095777A true CN110095777A (en) | 2019-08-06 |
Family
ID=67445537
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910326685.1A Pending CN110095777A (en) | 2019-04-23 | 2019-04-23 | Fuzzy logic method meteorology particle identification method based on shuffling technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110095777A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114970308A (en) * | 2021-12-30 | 2022-08-30 | 成都流体动力创新中心 | Aircraft icing prediction method and system and computer program product |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160252614A1 (en) * | 2013-07-11 | 2016-09-01 | Furuno Electric Co., Ltd. | Meteorological radar system and method and device for processing meteorological information |
| CN107238826A (en) * | 2017-06-09 | 2017-10-10 | 杨波 | The method being distributed using Dual-Polarized Doppler Weather Radar echo inverting thunderstorm charge |
| CN107607954A (en) * | 2017-09-11 | 2018-01-19 | 中国民航大学 | A kind of FNN rainfall particle phase identification methods based on T S models |
| CN108680920A (en) * | 2018-04-28 | 2018-10-19 | 深圳市雅码科技有限公司 | A kind of hazard weather identification early warning system and method based on dual polarization radar |
-
2019
- 2019-04-23 CN CN201910326685.1A patent/CN110095777A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160252614A1 (en) * | 2013-07-11 | 2016-09-01 | Furuno Electric Co., Ltd. | Meteorological radar system and method and device for processing meteorological information |
| CN107238826A (en) * | 2017-06-09 | 2017-10-10 | 杨波 | The method being distributed using Dual-Polarized Doppler Weather Radar echo inverting thunderstorm charge |
| CN107607954A (en) * | 2017-09-11 | 2018-01-19 | 中国民航大学 | A kind of FNN rainfall particle phase identification methods based on T S models |
| CN108680920A (en) * | 2018-04-28 | 2018-10-19 | 深圳市雅码科技有限公司 | A kind of hazard weather identification early warning system and method based on dual polarization radar |
Non-Patent Citations (3)
| Title |
|---|
| 王薇 等: "Java和Matlab混合编程及其应用", 《长春大学学报》 * |
| 范盼 等: "基于模糊逻辑法的粒子识别研究", 《河南师范大学学报(自然科学版)》 * |
| 赵相伟 等: "《MATLAB与测量数据处理》", 31 March 2014, 中国矿业大学出版社 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114970308A (en) * | 2021-12-30 | 2022-08-30 | 成都流体动力创新中心 | Aircraft icing prediction method and system and computer program product |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bromley et al. | Applications of near-term photonic quantum computers: software and algorithms | |
| Romero-Shaw et al. | Bayesian inference for compact binary coalescences with bilby: validation and application to the first LIGO–Virgo gravitational-wave transient catalogue | |
| Wikstrà et al. | The Heureka forestry decision support system: an overview | |
| Nachman | A guide for deploying Deep Learning in LHC searches: How to achieve optimality and account for uncertainty | |
| Filippi et al. | JDEVS: an implementation of a DEVS based formal framework for environmental modelling | |
| Tardioli et al. | A methodology for calibration of building energy models at district scale using clustering and surrogate techniques | |
| Beucler et al. | Machine learning for clouds and climate | |
| Richiardi et al. | JAS-mine: A new platform for microsimulation and agent-based modelling | |
| Ciomek et al. | Polyrun: A Java library for sampling from the bounded convex polytopes | |
| Radosevic et al. | Solar radiation modeling with KNIME and Solar Analyst: Increasing environmental model reproducibility using scientific workflows | |
| Monteiro et al. | sympl (v. 0.4. 0) and climt (v. 0.15. 3)–towards a flexible framework for building model hierarchies in Python | |
| Zhu | Implementation of support-vector machine algorithm to develop a model for electronic commerce energy regulatory system | |
| Romor et al. | ATHENA: advanced techniques for high dimensional parameter spaces to enhance numerical analysis | |
| Dhawale et al. | A levy flight based strategy to improve the exploitation capability of arithmetic optimization algorithm for engineering global optimization problems | |
| Vitiello et al. | Visual analytics: A sensemaking framework for systems thinking in systems engineering | |
| CN110095777A (en) | Fuzzy logic method meteorology particle identification method based on shuffling technology | |
| Wang et al. | Deep learning-based gas-phase chemical kinetics kernel emulator: Application in a global air quality simulation case | |
| Cheng et al. | A framework to calibrate ecosystem demography models within Earth system models using parallel surrogate global optimization | |
| CN110058208A (en) | Threshold method subcooled water area recognizing method based on shuffling technology | |
| Vijayakamal et al. | A Novel Approach for WEKA & Study On Data Mining Tools | |
| Zhao et al. | Online generic diagnostic reservoir operation tools | |
| Dueben et al. | Technical memo | |
| CN111695989A (en) | Modeling method and platform of wind-control credit model | |
| Davidović et al. | Implementation of the WRF-ARW prognostic model on the Grid | |
| Hilton et al. | The spark framework |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190806 |
|
| WD01 | Invention patent application deemed withdrawn after publication |