CN109946762A - A kind of method and system based on probability distribution Short-term Forecast precipitation - Google Patents
A kind of method and system based on probability distribution Short-term Forecast precipitation Download PDFInfo
- Publication number
- CN109946762A CN109946762A CN201910166319.4A CN201910166319A CN109946762A CN 109946762 A CN109946762 A CN 109946762A CN 201910166319 A CN201910166319 A CN 201910166319A CN 109946762 A CN109946762 A CN 109946762A
- Authority
- CN
- China
- Prior art keywords
- precipitation
- data
- probability
- rainfall
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000001556 precipitation Methods 0.000 title claims abstract description 150
- 238000000034 method Methods 0.000 title claims abstract description 45
- 230000005540 biological transmission Effects 0.000 claims abstract description 11
- 238000013480 data collection Methods 0.000 claims abstract description 11
- 230000008569 process Effects 0.000 claims description 14
- 238000000354 decomposition reaction Methods 0.000 claims description 10
- 108090000623 proteins and genes Proteins 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 5
- 238000013461 design Methods 0.000 claims description 3
- 238000013139 quantization Methods 0.000 claims description 3
- 241001365789 Oenanthe crocata Species 0.000 claims 1
- 230000004069 differentiation Effects 0.000 claims 1
- 239000011159 matrix material Substances 0.000 abstract description 12
- 238000011160 research Methods 0.000 abstract description 5
- 230000002123 temporal effect Effects 0.000 abstract description 5
- 230000000694 effects Effects 0.000 description 8
- 238000012544 monitoring process Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 241000475481 Nebula Species 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000004304 visual acuity Effects 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to Precipitation forecast technical fields, a kind of method and system based on probability distribution Short-term Forecast precipitation are disclosed, the system based on probability distribution Short-term Forecast precipitation includes: cloud atlas acquisition module, meteorological data collection module, data transmission module, main control module, precipitation probability generation module, precipitation estimation block, Characteristics of Precipitation determining module, data disaply moudle.The present invention can play the advantage of multi-source data by precipitation estimation block, obtain the rainfall product data of the higher more confidence of precision；Simultaneously, determine that mould utilizes the rainfall rainfall pattern eigenmatrix and temporal characteristics matrix of precipitation station by Characteristics of Precipitation, the rainfall of play rainfall and the situation of movement of storm centre from the time and spatially are described, to utilize math matrix, the quantitative spatial-temporal distribution characteristic for featuring rainfall, the subsequent math matrix based on quantitative description play rainfall, studies the feature of rainfall, to improve the accuracy of research characteristics of rainfall and comprehensive.
Description
Technical field
The invention belongs to Precipitation forecast technical field more particularly to a kind of methods based on probability distribution Short-term Forecast precipitation
And system.
Background technique
Weather forecast (survey) or weather forecast (survey) are using modern science and technology to the following a certain place earth atmosphere
State predicted.Weather forecast is mainly using collection a large amount of data (gas epidemic disaster, wind direction and wind speed, air pressure etc.
Deng), then using determining that following air changes to the understanding of Atmospheric processes (meteorology) at present.The Central Meteorological Observatory, China defends
Nebula figure is exactly absorbed from meteorological satellites such as " wind and cloud No.1s ".It is analyzed using satellite cloud picture photo, it is pre- weather to be improved
The accuracy rate of report.Weather forecast is generally divided into three kinds with regard to the length of timeliness: short-range weather forecast (2~3 days), medium-range forecast
(4~9 days), long-range weather forecasting (10~15 days or more).However, existing Short-term Forecast precipitation is using a kind of data of exclusive use
Rainfall estimation is carried out, Yi Yinqi estimation result generates deviation with the error of observation data, to be unfavorable for as geological disaster
Monitoring, the work such as flood-control and drought relief provide accurate data and support；Meanwhile in practical rainfall, region where precipitation station with
The characteristics of rainfall in other regions in city may respectively have difference, that is to say, the practical rainfall in the city and the above method are true
Fixed characteristics of rainfall is not consistent, and the accuracy so as to cause above-mentioned characteristics of rainfall research method is poor.
In conclusion problem of the existing technology is:
(1) existing Short-term Forecast precipitation carries out rainfall estimation, Yi Yinqi estimation result using a kind of data are used alone
Deviation is generated with the error of observation data, to be unfavorable for monitoring for geological disaster, the work such as flood-control and drought relief provide accurate
Data are supported；Meanwhile in practical rainfall, the characteristics of rainfall in other regions in region and city where precipitation station may be each
There is difference, that is to say, the practical rainfall in the city is not consistent with the characteristics of rainfall that the above method determines, so as to cause upper
The accuracy for stating characteristics of rainfall research method is poor.
(2) it is denoised in the prior art using traditional algorithm, noise cannot be completely inhibited, causing can not be preferably
The characteristic spikes point for cashing original signal is preserved, denoising effect is reduced.
(3) it in the prior art using traditional algorithm in the image processing process of earth cloud atlas, cannot effectively avoid
Except detailed information such as edge and angle points, the clarity of image detail is reduced, it cannot reliable accurate data.
(4) probability forecast is carried out using traditional algorithm in the prior art, the effect of fitting cannot be effectively improved, and not
Can be obviously improved numerical model in the forecast skill singly stood, reduce the accuracy rate of forecast.
Summary of the invention
In view of the problems of the existing technology, the present invention provides a kind of methods based on probability distribution Short-term Forecast precipitation
And system.
The invention is realized in this way a method of it is described based on probability point based on probability distribution Short-term Forecast precipitation
The method of cloth Short-term Forecast precipitation includes:
The first step acquires earth cloud atlas data, weather temperature, humidity, wind direction, wind speed, barometric information information；
The various data of acquisition are transmitted to monitoring center by second step；
Third step is handled and is stored to the various data of acquisition, according to the various data of acquisition, judges precipitation
Probability and precipitation determine the feature of rainfall, obtain a result；It is pre- using precipitation probability that probability forecast is carried out according to the collected data
Sorting algorithm is reported, detailed process is as follows:
Step 1 calculates prior probability；The precipitation sample of weather station corresponding time is divided into nothing and there are two classes, their elder generation
Test probability f1And f2It is respectively as follows:
N in formula1, n2It is respectively no rain and rainy sample size；
Step 2, design conditions probability, by 22 predictor (X=[x in sample1, x2, x3..., x21, x22]), it uses
Following formula is standardized, formula are as follows:
In formula, g=1,2.；μK, g, σK, gIt is k-th of predictor average value and even variance in g class sample respectively；
It is assumed that each predictor Normal Distribution, calculates conditional probability r of 22 predictors in precipitation with and without inK, 1
And rK, 2；
Step 3, sample classification, with building 30 individual gene position strings, are " 1 " base in each individual in more sample set
Because being maintained at 4 to 6, by genes of individuals bit string, asThe predictor of Bayes classifier is sentenced according to following formula foundation
Other equation group:
The f that above-mentioned steps are obtained1, f2, rK, gIt substitutes into respectively
Obtain discriminant value q1And q2, compare the size of the two, and then differentiate whether this sample has precipitation, sample set
Middle whole sample classification finishes；
4th step is believed using the cloud atlas of display display acquisition, meteorology, precipitation probability, precipitation, Characteristics of Precipitation data
Breath.
Further, the method based on probability distribution Short-term Forecast precipitation acquires earth cloud atlas number by meteorological satellite
According to handling the image information of acquisition, using the algorithm for image enhancement based on LIP model, detailed process is as follows:
Normalizing is carried out to gamma function and mends the simplified algorithm of transformation, gamma function normalizing mends transform definition, as follows:
Wherein the codomain of f is defined on [0, M] section in function, for 8bit image, M=256, simplified algorithm are as follows:
WhereinWithIt is original respectively and treated gamma function, α and β are real numbers, in addition
Then willAsIt substitutes intoAnd it asksInverse operationF (i, j) is obtained, as enhanced image.
Further, the method data based on probability distribution Short-term Forecast precipitation are transmitted through wireless base station and receive acquisition
Data Concurrent give, using wavelet threshold denoising algorithm, specifically includes the following steps:
Step 1, the wavelet decomposition of signal: including choosing a kind of wavelet basis function of determination and determining optimal small wavelength-division
Number of plies N is solved, N layers of wavelet decomposition is carried out to containing hot-tempered signal f (t), obtains one group of wavelet coefficient ωJ, k；
Step 2, the threshold value quantizing of wavelet decomposition high frequency coefficient: according to the quantization criterion of uniform threshold, to first to N
Each layer of high frequency coefficient of layer, determines a threshold value appropriate, by ωJ, kThreshold value quantizing processing is carried out, estimation small echo is obtained
CoefficientAnd makeIt is small as far as possible；
Step 3, wavelet reconstruction: by threshold value quantizing handle later the 1st to n-th layer high-frequency wavelet coefficient ωJ, kWith N
The low-frequency wavelet coefficients of layer carry out wavelet inverse transformation, obtain estimation signal and areSignal after as denoising.
Another object of the present invention is to provide a kind of methods based on probability distribution Short-term Forecast precipitation described in realize
Based on the system of probability distribution Short-term Forecast precipitation, the system based on probability distribution Short-term Forecast precipitation includes:
Cloud atlas acquisition module, connect with main control module, for acquiring earth cloud atlas data by meteorological satellite；
Meteorological data collection module, connect with main control module, for by Weather station apparatus acquisition weather temperature, humidity,
The data informations such as wind direction, wind speed, air pressure；
Data transmission module is connect with main control module, and the Data Concurrent for receiving acquisition by wireless base station is sent to master
Control module；
Main control module generates mould with cloud atlas acquisition module, meteorological data collection module, data transmission module, precipitation probability
Block, precipitation estimation block, Characteristics of Precipitation determining module, data disaply moudle connection, for receiving acquisition by single-chip microcontroller
Cloud atlas, meteorological data, and control modules normal work；
Precipitation probability generation module, connect with main control module, more for being generated meteorological data by probability distribution program
A rate probability, wherein each of the multiple rate probability rate probability indicates that precipitation will be in multiple precipitation rates
Corresponding precipitation rate a possibility that occurring in this region；
Precipitation estimation block, connect with main control module, for passing through estimation program according to meteorological data Rainfall Estimation amount
Data；
Characteristics of Precipitation determining module, connect with main control module, drops for being determined by signature analysis program according to precipitation
Water feature；
Data disaply moudle is connect with main control module, general for cloud atlas, meteorology, the precipitation by display display acquisition
Rate, precipitation, Characteristics of Precipitation data information.
Another object of the present invention is to provide a kind of methods based on probability distribution Short-term Forecast precipitation described in application
Precipitation forecast system.
Advantages of the present invention and good effect are as follows: the present invention can play the excellent of multi-source data by precipitation estimation block
Gesture obtains the rainfall product data of the higher more confidence of precision；The result of multiple data sources convergence analysis of the present invention reduce due to
The inaccurate bring Regional Rainfall amount of certain single class rainfall data is estimated uncertain；For reinforced region high-precision disaster alarm,
Evade flooding risk or Storm flood of small basins estimation provide the input of more structurally sound data and it is richer more refine build
Modulus evidence；Meanwhile determining that mould utilizes the rainfall rainfall pattern eigenmatrix and temporal characteristics matrix of precipitation station by Characteristics of Precipitation,
The rainfall of play rainfall and the situation of movement of storm centre from the time and spatially are described, thus using math matrix, it is fixed
The spatial-temporal distribution characteristic for featuring rainfall of amount, the subsequent math matrix based on quantitative description play rainfall, studies rainfall
Feature, to improve the research accuracy of characteristics of rainfall and comprehensive.
Data transmission module is sent to the process of main control module by the Data Concurrent that wireless base station receives acquisition in the present invention
In, signal is denoised using wavelet threshold denoising algorithm, noise can be completely inhibited, can preferably preserve realization
The originally characteristic spikes point of signal, obtains good denoising effect.
Cloud atlas acquisition module is acquired in earth cloud atlas data procedures by meteorological satellite in the present invention, using based on LIP mould
The algorithm for image enhancement of type handles the image information of acquisition, can effectively avoid the removal details such as edge and angle point letter
Breath, enhances the clarity of image detail, obtains reliable accurate data, improves the forecast accuracy of precipitation.
During main control module carries out probability forecast according to the collected data in the present invention, using Precipitation Probability Forecast point
Class algorithm carry out probability forecast, the effect of fitting can be effectively improved, and can be obviously improved numerical model in the forecast singly stood
Skill improves the accuracy rate of forecast.
Detailed description of the invention
Fig. 1 is the method flow diagram provided in an embodiment of the present invention based on probability distribution Short-term Forecast precipitation.
Fig. 2 is the system structure diagram provided in an embodiment of the present invention based on probability distribution Short-term Forecast precipitation；
In figure: 1, cloud atlas acquisition module；2, meteorological data collection module；3, data transmission module；4, main control module；5, it drops
Water probability generation module；6, precipitation estimation block；7, Characteristics of Precipitation determining module；8, data disaply moudle.
Specific embodiment
In order to further understand the content, features and effects of the present invention, the following examples are hereby given, and cooperate attached drawing
Detailed description are as follows.
Structure of the invention is explained in detail with reference to the accompanying drawing.
As shown in Figure 1, the method provided by the invention based on probability distribution Short-term Forecast precipitation the following steps are included:
S101: firstly, the data informations such as acquisition earth cloud atlas data, weather temperature, humidity, wind direction, wind speed, air pressure；
S102: the various data of above-mentioned acquisition are transmitted to monitoring center；
S103: being handled and stored to the various data of acquisition, according to the various data of acquisition, judges the general of precipitation
Rate and precipitation determine the feature of rainfall, obtain a result；
S104: the cloud atlas of display display acquisition, meteorology, precipitation probability, precipitation, Characteristics of Precipitation data information are utilized.
As shown in Fig. 2, the system provided in an embodiment of the present invention based on probability distribution Short-term Forecast precipitation includes: that cloud atlas is adopted
Collection module 1, meteorological data collection module 2, data transmission module 3, main control module 4, precipitation probability generation module 5, precipitation are estimated
Calculate module 6, Characteristics of Precipitation determining module 7, data disaply moudle 8.
Cloud atlas acquisition module 1 is connect with main control module 4, for acquiring earth cloud atlas data by meteorological satellite；
Meteorological data collection module 2 is connect with main control module 4, for passing through Weather station apparatus acquisition weather temperature, wet
The data informations such as degree, wind direction, wind speed, air pressure；
Data transmission module 3 is connect with main control module 4, and the Data Concurrent for receiving acquisition by wireless base station is sent to
Main control module；
Main control module 4, it is raw with cloud atlas acquisition module 1, meteorological data collection module 2, data transmission module 3, precipitation probability
It is connected at module 5, precipitation estimation block 6, Characteristics of Precipitation determining module 7, data disaply moudle 8, for being connect by single-chip microcontroller
Cloud atlas, the meteorological data of acquisition are received, and controls modules normal work；
Precipitation probability generation module 5 is connect with main control module 4, for being generated meteorological data by probability distribution program
Multiple rate probability, wherein each of the multiple rate probability rate probability indicates that precipitation will be with multiple precipitation rates
In corresponding precipitation rate a possibility that occurring in this region；
Precipitation estimation block 6 is connect with main control module 4, for passing through estimation program according to meteorological data Rainfall Estimation
Measure data；
Characteristics of Precipitation determining module 7 is connect with main control module 4, for being determined by signature analysis program according to precipitation
Characteristics of Precipitation；
Data disaply moudle 8 is connect with main control module 4, for the cloud atlas by display display acquisition, meteorology, precipitation
Probability, precipitation, Characteristics of Precipitation data information.
The cloud atlas acquisition module 1 is acquired in earth cloud atlas data procedures by meteorological satellite, needs the image to acquisition
Information is handled, and in order to avoid removing the detailed information such as edge and angle point, is enhanced the clarity of image detail, is obtained reliable essence
True data improve the forecast accuracy of precipitation, and using the algorithm for image enhancement based on LIP model, detailed process is as follows:
Normalizing is carried out to gamma function and mends the simplified algorithm of transformation, gamma function normalizing mends transform definition, as follows:
Wherein in function the codomain of f be defined on [0, M) section, for 8bit image, M=256, simplified algorithm are as follows:
WhereinWithIt is original respectively and treated gamma function, α and β are real numbers, in addition
Then willAsIt substitutes intoAnd it asksInverse operationF (i, j) is obtained, as enhanced image.
During the data transmission module 3 is sent to main control module by the Data Concurrent that wireless base station receives acquisition,
In order to completely inhibit noise, the characteristic spikes point for cashing original signal can be preferably preserved, obtains good go
It makes an uproar effect, using wavelet threshold denoising algorithm, specifically includes the following steps:
Step 1, the wavelet decomposition of signal: including choosing a kind of wavelet basis function of determination and determining optimal small wavelength-division
Number of plies N is solved, N layers of wavelet decomposition is carried out to containing hot-tempered signal f (t), obtains one group of wavelet coefficient ωJ, k；
Step 2, the threshold value quantizing of wavelet decomposition high frequency coefficient: according to the quantization criterion of uniform threshold, to first to N
Each layer of high frequency coefficient of layer, determines a threshold value appropriate, by ωJ, kThreshold value quantizing processing is carried out, estimation small echo is obtained
CoefficientAnd makeIt is small as far as possible；
Step 3, wavelet reconstruction: by threshold value quantizing handle later the 1st to n-th layer high-frequency wavelet coefficient ωJ, kWith N
The low-frequency wavelet coefficients of layer carry out wavelet inverse transformation, obtain estimation signal and areSignal after as denoising.
During the main control module 4 carries out probability forecast according to the collected data, in order to improve the effect of fitting, and
And can be obviously improved numerical model in the forecast skill singly stood, improve the accuracy rate of forecast, classified using Precipitation Probability Forecast
Algorithm, detailed process is as follows:
Step 1 calculates prior probability；The precipitation sample of weather station corresponding time is divided into nothing and there are two classes, their elder generation
Test probability f1And f2It is respectively as follows:
N in formula1, n2It is respectively no rain and rainy sample size；
Step 2, design conditions probability, by 22 predictor (X=[x in sample1, x2, x3..., x21, x22]), it adopts
It is standardized with following formula, formula are as follows:
In formula, g=1,2；μK, g, σK, gIt is k-th of predictor average value and even variance in g class sample respectively；
It is assumed that each predictor Normal Distribution, calculates conditional probability r of 22 predictors in precipitation with and without inK, 1
And rK, 2；
Step 3, sample classification, with building 30 individual gene position strings, are " 1 " base in each individual in more sample set
Because being maintained at 4 to 6, by these genes of individuals bit strings, asThe predictor of Bayes classifier is built according to following formula
Vertical discriminant equation group:
The f that above-mentioned steps are obtained1, f2, rK, gIt substitutes into respectively
Obtain discriminant value q1And q2, compare the size of the two, and then differentiate whether this sample has precipitation, sample set
Middle whole sample classification finishes.
6 evaluation method of precipitation estimation block provided by the invention is as follows:
(1) earth station in multi-source satellite rainfall data and survey region is obtained by meteorological satellite and observes rainfall data；
(2) multi-source satellite rainfall data are pre-processed, comprising: to multi-source satellite rainfall data carry out uniform format,
Region is cut and scale matching, forms survey region multi-source satellite rainfall product data body；
(3) earth station's observation rainfall data in survey region are mutually tied with survey region multi-source satellite rainfall product data body
It closes, constitutes survey region multi-source data collection；
(4) the survey region multi-source data collection is used, using earth station's observation rainfall data as constraint, is embodied different
The satellite rainfall data in source establish the research based on Dynamic Bayesian theory to the weighing factor of survey region rainfall
Bayes's precipitation predicting model in region；Wherein, Bayes's precipitation predicting model is that weight will be as the more of prior probability
Source satellite rainfall data are converted into the process of earth station's observation rainfall data as posterior probability；
(5) target dynamic training sample extracts: the Bayes's precipitation predicting model established for step (4) utilizes ground
As training sample, the Bayes's precipitation predicting model established to step (4) is trained, is trained observation rainfall data of standing
Bayes's precipitation predicting model afterwards；
(6) Bayes's precipitation predicting model after seeking the training that step (5) obtains using maximum entropy method it is non-linear
Optimal solution, and then determine the optimal weight and unascertained information in each satellite data source；Wherein, optimal weight and uncertainty
Change with spatial position, the variation of time, dynamic is presented；
(7) optimal weight and uncertainty based on satellite data source generate and apply Multi-source Information Fusion in survey region
The estimation result of rainfall.
Multi-source satellite rainfall data provided by the invention refer to the rainfall data from different platform difference satellite type；
After obtaining original multi-source satellite rainfall data, after excluding outlier, then uniform format, region cutting and scale matching are carried out
Processing.
In step (2) provided by the invention, the detailed process of the uniform format are as follows: by the lattice of multi-source satellite rainfall data
Formula is unified for binary format；Wherein, include coordinate range and data source in header file, include corresponding position in file body
Rainfall product data matrix；
The detailed process that the region is cut are as follows: cut out from multi-source satellite rainfall data and belong to survey region range
Multi-source satellite rainfall data, the data space ranges cut out are rectangle, and coordinate range is determined by the extreme value of survey region coordinate
Fixed, that is, the region cut out is the minimum rectangle comprising survey region；
The matched detailed process of scale are as follows: scale matching process includes that the moment is unified and space scale is unified；Due to
The monitoring moment of various satellite rainfall data and spatial resolving power are distinct, and therefore, the multi-source of survey region range is defended
Star rainfall data are converted into be unified constantly, the consistent data of space lattice；
It is cut by uniform format, region and scale matches, convert uniform format, time for multi-source satellite rainfall data
Interval and the consistent data volume in spatial position, the multi-source that each moment rainfall in survey region internal standard grid is consequently formed are defended
Star rainfall product data body.
Characteristics of Precipitation determining module 7 provided by the invention determines that method is as follows:
1) the rainfall rainfall pattern for obtaining each precipitation station in multiple precipitation stations in city to be studied by Weather station apparatus is special
Matrix is levied, rainfall pattern index of the rainfall rainfall pattern eigenmatrix by each precipitation station in multiple default duration of raining forms；
2) the temporal characteristics matrix of the multiple precipitation station is obtained, the temporal characteristics matrix is maximum default by rainfall
The relative time composition that maximum rainfall occurs in duration of raining, the relative time of the appearance are the maximum drop of each precipitation station
The difference for the time that maximum rainfall occurs earliest in the time of occurrence of rainfall and all precipitation stations；
3) it according to the sample rainfall rainfall pattern eigenmatrix of the multiple sample precipitation station, is calculated by the first default cluster
Method determines the sample rain types of each sample precipitation station；
4) special to the rainfall rainfall pattern of each precipitation station according to the sample rain types of each sample precipitation station
It levies matrix and carries out fuzzy diagnosis, obtain rain types belonging to each precipitation station；
5) it is counted the multiple according to the temporal characteristics matrix of the multiple precipitation station by the second default clustering algorithm
The relative time that maximum rainfall of the precipitation station after cluster occurs；
6) relative time and the multiple rainfall that the maximum rainfall according to the multiple precipitation station after cluster occurs
Rain types belonging to standing, determine the rain types in the city to be studied.
Sample rainfall rainfall pattern eigenmatrix provided by the invention according to the multiple sample precipitation station, it is pre- by first
If clustering algorithm, the sample rain types of each sample precipitation station are determined, comprising:
According to the sample rainfall rainfall pattern eigenmatrix of the multiple sample precipitation station, by the first default clustering algorithm,
Determine the rainfall rainfall pattern feature clustering center of multiple sample precipitation stations；
According to the rainfall rainfall pattern feature clustering center of the multiple sample precipitation station, the multiple sample precipitation station is pressed
Classify according to rainfall rainfall pattern feature clustering center, takes multiple sample precipitation stations in every kind of classification in such purpose rainfall
Envelope under rainfall pattern feature clustering center, according to taking the multiclass sample precipitation station rainfall rainfall pattern feature clustering center after envelope,
Determine the sample rain types of each sample precipitation station.
The above is only the preferred embodiments of the present invention, and is not intended to limit the present invention in any form,
Any simple modification made to the above embodiment according to the technical essence of the invention, equivalent variations and modification, belong to
In the range of technical solution of the present invention.
Claims (5)
1. a kind of method based on probability distribution Short-term Forecast precipitation, which is characterized in that described to be based on probability distribution Short-term Forecast
The method of precipitation includes:
The first step acquires earth cloud atlas data, weather temperature, humidity, wind direction, wind speed, barometric information information；
The various data of acquisition are transmitted to monitoring center by second step；
Third step is handled and is stored to the various data of acquisition, according to the various data of acquisition, judges the probability of precipitation
And precipitation, it determines the feature of rainfall, obtains a result；Probability forecast is carried out according to the collected data using Precipitation Probability Forecast point
Class algorithm, detailed process is as follows:
Step 1 calculates prior probability；The precipitation sample of weather station corresponding time is divided into nothing and there are two classes, their priori is general
Rate f1And f2It is respectively as follows:
N in formula1, n2It is respectively no rain and rainy sample size；
Step 2, design conditions probability, by 22 predictor X=[x in sample1, x2, x3..., x21, x22]), using following formula into
Row standardization, formula are as follows:
In formula, g=1,2；μk·g, σk·gIt is k-th of predictor average value and even variance in g class sample respectively；
It is assumed that each predictor Normal Distribution, calculates conditional probability r of 22 predictors in precipitation with and without ink·1With
rk·2；
Step 3, sample classification, with building 30 individual gene position strings, are protected in more sample set in each individual for " 1 " gene
It holds at 4 to 6, by genes of individuals bit string, asThe predictor of Bayes classifier establishes differentiation side according to following formula
Journey group:
The f that above-mentioned steps are obtained1, f2, rk·gIt substitutes into respectively
Obtain discriminant value q1And q2, the size both compared, and then differentiate whether this sample has precipitation, it is complete in sample set
Portion's sample classification finishes；
4th step utilizes the cloud atlas of display display acquisition, meteorology, precipitation probability, precipitation, Characteristics of Precipitation data information.
2. the method as described in claim 1 based on probability distribution Short-term Forecast precipitation, which is characterized in that described to be based on probability
The method for being distributed Short-term Forecast precipitation acquires earth cloud atlas data by meteorological satellite, handles the image information of acquisition,
Using the algorithm for image enhancement based on LIP model, detailed process is as follows:
Normalizing is carried out to gamma function and mends the simplified algorithm of transformation, gamma function normalizing mends transform definition, as follows:
Wherein in function the codomain of f be defined on [0, M) section, for 8bit image, M=256, simplified algorithm are as follows:
WhereinWithIt is original respectively and treated gamma function, α and β are real numbers, in addition
Then willAsIt substitutes intoAnd it asksInverse operationF (i, j) is obtained, as enhanced image.
3. the method as described in claim 1 based on probability distribution Short-term Forecast precipitation, which is characterized in that described to be based on probability
The Data Concurrent that the method data of distribution Short-term Forecast precipitation are transmitted through wireless base station reception acquisition is sent, and is gone using wavelet threshold
It makes an uproar algorithm, specifically includes the following steps:
Step 1, the wavelet decomposition of signal: including choosing a kind of wavelet basis function of determination and determining optimal wavelet decomposition layer
Number N carries out N layers of wavelet decomposition to containing hot-tempered signal f (t), obtains one group of wavelet coefficient ωJ, k；
The threshold value quantizing of wavelet decomposition high frequency coefficient: step 2 according to the quantization criterion of uniform threshold, arrives n-th layer to first
Each layer of high frequency coefficient determines a threshold value appropriate, by ωJ, kThreshold value quantizing processing is carried out, estimation wavelet coefficient is obtainedAnd makeIt is small as far as possible；
Step 3, wavelet reconstruction: by threshold value quantizing handle later the 1st to n-th layer high-frequency wavelet coefficient ωJ, kWith n-th layer
Low-frequency wavelet coefficients carry out wavelet inverse transformation, obtain estimation signal and areSignal after as denoising.
4. a kind of method realized based on probability distribution Short-term Forecast precipitation described in claim 1 is pre- in short-term based on probability distribution
The system for reporting precipitation, which is characterized in that the system based on probability distribution Short-term Forecast precipitation includes:
Cloud atlas acquisition module, connect with main control module, for acquiring earth cloud atlas data by meteorological satellite；
Meteorological data collection module, connect with main control module, for by Weather station apparatus acquisition weather temperature, humidity, wind direction,
The data informations such as wind speed, air pressure；
Data transmission module is connect with main control module, and the Data Concurrent for receiving acquisition by wireless base station is sent to master control mould
Block；
Main control module, with cloud atlas acquisition module, meteorological data collection module, data transmission module, precipitation probability generation module, drop
Instream flow estimate module, Characteristics of Precipitation determining module, data disaply moudle connection, for by single-chip microcontroller receive acquire cloud atlas,
Meteorological data, and control modules normal work；
Precipitation probability generation module, connect with main control module, for meteorological data to be generated multiple speed by probability distribution program
Rate probability, wherein each of the multiple rate probability rate probability indicates that precipitation will be with the phase in multiple precipitation rates
A possibility that precipitation rate answered occurs in this region；
Precipitation estimation block, connect with main control module, for passing through estimation program according to meteorological data Rainfall Estimation amount data；
Characteristics of Precipitation determining module, connect with main control module, for determining precipitation spy according to precipitation by signature analysis program
Sign；
Data disaply moudle is connect with main control module, for the cloud atlas by display display acquisition, meteorology, precipitation probability, drop
Water, Characteristics of Precipitation data information.
5. a kind of precipitation using the method based on probability distribution Short-term Forecast precipitation described in claims 1 to 3 any one is pre-
Reporting system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910166319.4A CN109946762B (en) | 2019-03-06 | 2019-03-06 | Method and system for short-time rainfall forecast based on probability distribution |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910166319.4A CN109946762B (en) | 2019-03-06 | 2019-03-06 | Method and system for short-time rainfall forecast based on probability distribution |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109946762A true CN109946762A (en) | 2019-06-28 |
| CN109946762B CN109946762B (en) | 2021-05-18 |
Family
ID=67008500
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910166319.4A Expired - Fee Related CN109946762B (en) | 2019-03-06 | 2019-03-06 | Method and system for short-time rainfall forecast based on probability distribution |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109946762B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110826810A (en) * | 2019-11-13 | 2020-02-21 | 吉林农业大学 | Regional rainfall prediction method combining spatial reasoning and machine learning |
| CN110865425A (en) * | 2019-11-28 | 2020-03-06 | 中国有色金属长沙勘察设计研究院有限公司 | Rain gauge gross error detection method based on prior information |
| CN111159640A (en) * | 2019-11-20 | 2020-05-15 | 北京玖天气象科技有限公司 | Small rain emptying method, system, electronic equipment and storage medium suitable for grid forecast |
| CN112800634A (en) * | 2021-04-07 | 2021-05-14 | 水利部交通运输部国家能源局南京水利科学研究院 | Rainfall estimation method and system coupling dry-wet state identification and multi-source information fusion |
| CN113033957A (en) * | 2021-02-26 | 2021-06-25 | 兰州中心气象台(兰州干旱生态环境监测预测中心) | Multi-mode rainfall forecast and real-time dynamic inspection and evaluation system |
| CN113050195A (en) * | 2021-02-07 | 2021-06-29 | 国家气象中心(中央气象台) | Hourly resolution precipitation process identification method |
| CN113095590A (en) * | 2021-04-29 | 2021-07-09 | 中国人民解放军国防科技大学 | High spatial-temporal resolution reconstruction analysis and short-term prediction method for microwave horizontal rainfall field |
| CN114677059A (en) * | 2022-05-26 | 2022-06-28 | 水利部交通运输部国家能源局南京水利科学研究院 | Method and system for comprehensively evaluating precision of inversion precipitation product by integrating time-space indexes |
| CN117033935A (en) * | 2023-08-03 | 2023-11-10 | 北京市市政工程设计研究总院有限公司广东分院 | Prediction method of rainfall characteristic under statistics and monitoring based on Bayesian fusion |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036330A (en) * | 2014-05-22 | 2014-09-10 | 南京信息工程大学 | Rainfall classification prediction method based on MapReduce |
| CN106991278A (en) * | 2017-03-21 | 2017-07-28 | 武汉大学 | It is a kind of to gather precipitation forecast and the coupling process of real-time flood probability forecast |
-
2019
- 2019-03-06 CN CN201910166319.4A patent/CN109946762B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036330A (en) * | 2014-05-22 | 2014-09-10 | 南京信息工程大学 | Rainfall classification prediction method based on MapReduce |
| CN106991278A (en) * | 2017-03-21 | 2017-07-28 | 武汉大学 | It is a kind of to gather precipitation forecast and the coupling process of real-time flood probability forecast |
Non-Patent Citations (1)
| Title |
|---|
| 韩焱红: "《基于贝叶斯理论的集合降水概率预报方法研究》", 《气象》 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110826810A (en) * | 2019-11-13 | 2020-02-21 | 吉林农业大学 | Regional rainfall prediction method combining spatial reasoning and machine learning |
| CN110826810B (en) * | 2019-11-13 | 2022-07-15 | 吉林农业大学 | Regional rainfall prediction method combining spatial reasoning and machine learning |
| CN111159640A (en) * | 2019-11-20 | 2020-05-15 | 北京玖天气象科技有限公司 | Small rain emptying method, system, electronic equipment and storage medium suitable for grid forecast |
| CN110865425B (en) * | 2019-11-28 | 2021-10-26 | 中国有色金属长沙勘察设计研究院有限公司 | Rain gauge gross error detection method based on prior information |
| CN110865425A (en) * | 2019-11-28 | 2020-03-06 | 中国有色金属长沙勘察设计研究院有限公司 | Rain gauge gross error detection method based on prior information |
| CN113050195A (en) * | 2021-02-07 | 2021-06-29 | 国家气象中心(中央气象台) | Hourly resolution precipitation process identification method |
| CN113050195B (en) * | 2021-02-07 | 2022-07-15 | 国家气象中心(中央气象台) | Hourly resolution precipitation process identification method |
| CN113033957B (en) * | 2021-02-26 | 2023-10-27 | 兰州中心气象台(兰州干旱生态环境监测预测中心) | Multi-mode rainfall forecast and real-time dynamic inspection and evaluation system |
| CN113033957A (en) * | 2021-02-26 | 2021-06-25 | 兰州中心气象台(兰州干旱生态环境监测预测中心) | Multi-mode rainfall forecast and real-time dynamic inspection and evaluation system |
| CN112800634A (en) * | 2021-04-07 | 2021-05-14 | 水利部交通运输部国家能源局南京水利科学研究院 | Rainfall estimation method and system coupling dry-wet state identification and multi-source information fusion |
| CN112800634B (en) * | 2021-04-07 | 2021-06-25 | 水利部交通运输部国家能源局南京水利科学研究院 | Rainfall estimation method and system coupling dry-wet state identification and multi-source information fusion |
| CN113095590A (en) * | 2021-04-29 | 2021-07-09 | 中国人民解放军国防科技大学 | High spatial-temporal resolution reconstruction analysis and short-term prediction method for microwave horizontal rainfall field |
| CN113095590B (en) * | 2021-04-29 | 2022-04-29 | 中国人民解放军国防科技大学 | High spatial-temporal resolution reconstruction analysis and short-term prediction method for microwave horizontal rainfall field |
| CN114677059A (en) * | 2022-05-26 | 2022-06-28 | 水利部交通运输部国家能源局南京水利科学研究院 | Method and system for comprehensively evaluating precision of inversion precipitation product by integrating time-space indexes |
| CN114677059B (en) * | 2022-05-26 | 2022-08-23 | 水利部交通运输部国家能源局南京水利科学研究院 | Method and system for comprehensively evaluating precision of inversion precipitation product by integrating time-space indexes |
| CN117033935A (en) * | 2023-08-03 | 2023-11-10 | 北京市市政工程设计研究总院有限公司广东分院 | Prediction method of rainfall characteristic under statistics and monitoring based on Bayesian fusion |
| CN117033935B (en) * | 2023-08-03 | 2024-02-23 | 北京市市政工程设计研究总院有限公司 | Prediction method of rainfall characteristic under statistics and monitoring based on Bayesian fusion |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109946762B (en) | 2021-05-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109946762A (en) | A kind of method and system based on probability distribution Short-term Forecast precipitation | |
| Hong et al. | Precipitation estimation from remotely sensed imagery using an artificial neural network cloud classification system | |
| WO2022016884A1 (en) | Method for extracting sea surface wind speed on basis of k-means clustering algorithm | |
| CN106127725B (en) | A kind of millimetre-wave radar cloud atlas dividing method based on multiresolution CNN | |
| CN110363327A (en) | Short based on ConvLSTM and 3D-CNN faces Prediction of Precipitation method | |
| CN111666656A (en) | Rainfall estimation method and rainfall monitoring system based on microwave rainfall attenuation | |
| Kohail et al. | Implementation of data mining techniques for meteorological data analysis | |
| CN113936142A (en) | Rainfall approach forecasting method and device based on deep learning | |
| Zhou et al. | Individual tree parameters estimation for plantation forests based on UAV oblique photography | |
| CN111210483B (en) | Simulated satellite cloud picture generation method based on generation of countermeasure network and numerical mode product | |
| CN111401602A (en) | Assimilation method for satellite and ground rainfall measurement values based on neural network | |
| Gagne et al. | Day-ahead hail prediction integrating machine learning with storm-scale numerical weather models | |
| CN105069295A (en) | Assimilation method for satellite and ground rainfall measured values based on Kalman filtering | |
| CN105974495A (en) | Method for pre-judging future average cloud amount of target area by using classification fitting method | |
| CN115357847A (en) | Day scale star-ground precipitation fusion method based on error decomposition | |
| CN110516552B (en) | Multi-polarization radar image classification method and system based on time sequence curve | |
| CN116609858A (en) | Tropical cyclone atmospheric waveguide prediction method and system based on interpretability XGBoost model | |
| McCollum et al. | Microwave rainfall estimation over coasts | |
| Guo et al. | Correction of sea surface wind speed based on SAR rainfall grade classification using convolutional neural network | |
| CN114254692A (en) | Multiscale thunderstorm intelligent classification and identification method based on multisource lightning data | |
| Miller et al. | A preliminary assessment of using spatiotemporal lightning patterns for a binary classification of thunderstorm mode | |
| CN112347926A (en) | High-resolution image urban village detection method based on building form distribution | |
| CN110826526A (en) | Method for cloud detection radar to identify clouds | |
| CN115421220A (en) | Multi-factor local precipitation indication method and system based on deep learning | |
| CN105893744A (en) | Tibet Plateau snow water equivalent estimation method and system based on passive microwave remote sensing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP02 | Change in the address of a patent holder |
Address after: 401520 No.1 holiday Avenue, Hechuan District, Chongqing Patentee after: COLLEGE OF MOBILE TELECOMMUNICATIONS CHONGQING University OF POSTS AND TELECOMMUNICATIONS Address before: 401520 No.1 holiday Avenue, Hechuan District, Shapingba District, Chongqing Patentee before: COLLEGE OF MOBILE TELECOMMUNICATIONS CHONGQING University OF POSTS AND TELECOMMUNICATIONS |
|
| CP02 | Change in the address of a patent holder | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210518 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |