CN109614744A - A kind of precipitation quantity measuring method and system based on big data - Google Patents
A kind of precipitation quantity measuring method and system based on big data Download PDFInfo
- Publication number
- CN109614744A CN109614744A CN201811617914.7A CN201811617914A CN109614744A CN 109614744 A CN109614744 A CN 109614744A CN 201811617914 A CN201811617914 A CN 201811617914A CN 109614744 A CN109614744 A CN 109614744A
- Authority
- CN
- China
- Prior art keywords
- space
- detection
- model
- optimization
- data
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/14—Rainfall or precipitation gauges
-
- 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
Abstract
The present invention provides a kind of precipitation quantity measuring method and system based on big data, the control method include: acquisition detection data, and establish space-time detection model according to detection data;Judge that the detection parameters in space-time detection model with the presence or absence of singular value, and when determining in detection parameters there are when singular value, are corrected detection parameters;It is data-optimized to the progress of space-time detection model according to the default principle of optimality, to obtain space and time optimization model;Model costing bio disturbance is carried out for space and time optimization model, to obtain loss function, and when determining loss function less than loss threshold value, space and time optimization model is exported, to obtain testing result.The present invention is by establishing the design of space-time detection model, so that will test data is converted into concrete model parameter, facilitate subsequent processing and calculating to data, by carrying out data-optimized design to space-time detection model, influence of the disturbing factor to space-time detection model is prevented, to improve the accuracy of testing result.
Description
Technical field
The present invention relates to precipitation detection technique fields, detect in particular to a kind of precipitation based on big data
Method and system.
Background technique
An important factor for spatial and temporal distributions of precipitation are the natural calamities such as initiation flood, landslide, mud-rock flow extremely,
China is more than every year hundred billion yuan by the direct economic loss that flood causes, and average population suffered from disaster is more than 1.2 hundred million people.Although rain
A variety of Rainfall estimation means such as meter, weather radar and meteorological satellite have been widely applied, but precipitation is in the weight such as city, mountain area
Point region is there are extremely complex change in time and space, and rainfall gauge website is unevenly distributed, even if relatively intensive in websites such as cities
Area is still difficult to monitor the fine change in time and space of precipitation;Weather radar can only measure part precipitation under the conditions of the high elevation angle
Body or cloud body influence under the conditions of the low elevation angle vulnerable to background return, thus the measurement effect in city and mountain area is limited;Rain is surveyed to defend
Star can only measure from top to bottom cloud top or penetrate cloud top, with drop to adjacent ground surface practical precipitation between there are bigger difference,
It is difficult to according to echo inverting and detects accurate rainfall distribution.
Method used by the existing detection for precipitation is the precipitation detection based on radar, and due to existing
In the detection mode use process of precipitation based on radar, detection precision it is lower, and can effectively for precipitation into
Row detection.
Summary of the invention
Based on this, the purpose of the embodiment of the present invention is to solve in the prior art, low for the detection precision of rainfall
The problem of.
In a first aspect, the present invention provides a kind of precipitation quantity measuring method based on big data, which comprises
Detection data is obtained, and space-time detection model is established according to the detection data;
Judge that the detection parameters in the space-time detection model whether there is singular value, and works as and determine the detection parameters
In there are when the singular value, the detection parameters are corrected;
It is data-optimized to space-time detection model progress according to the local default principle of optimality, to obtain space and time optimization mould
Type;
Model costing bio disturbance is carried out for the space and time optimization model, to obtain loss function, and works as and determines the damage
When losing function less than loss threshold value, the space and time optimization model is exported, to obtain testing result.
Further, described that space-time detection model is established according to the detection data in preferred embodiments of the present invention
The step of include:
The radar reflectivity factor combination stored in the detection data is obtained, and using MSE as optimization aim, to establish drop
Hydrospace relies on mapping relations;
The correlation of the Precipitation Process stored in the detection data is obtained, and using the correlation of the Precipitation Process as mesh
Mark, to establish Precipitation Time Series mapping relations.
Further, the detection parameters in preferred embodiments of the present invention, in the judgement space-time detection model
Include: with the presence or absence of the step of singular value
The parameter difference between the adjacent detection parameters is calculated separately, and judges whether the parameter difference is greater than difference
Threshold value;
If so, determining that the corresponding detection parameters of the parameter difference are singular value.
Further, in preferred embodiments of the present invention, described the step of being corrected to the detection parameters, includes:
Delete the detection parameters, and obtain the parameter between the detection parameters adjacent parameter and;
The average value of the parameter sum is calculated, to obtain replacement values, and the replacement values are carried out the detection parameters
Replacement.
Further, described that the space-time is examined according to the local default principle of optimality in preferred embodiments of the present invention
Survey model carries out data-optimized step and includes:
Variance calculating is carried out to the space-time detection model, to obtain variance yields, and the variance yields is prestored with local
The optimization table of storage is matched, to obtain the first optimization fluctuation;
Fluctuation optimization is carried out to the space-time detection model according to the first optimization fluctuation;
Calculating is filtered to the space-time detection model, to obtain filter value, and by the filter value and the optimization
Table is matched, to obtain the second optimization fluctuation;
Fluctuation optimization is carried out to the space-time detection model according to the second optimization fluctuation;
It is filtered variance to the space-time detection model to calculate, to obtain filter error variance value, and by the filter error variance
Value is matched with the optimization table, to obtain third optimization fluctuation;
Fluctuation optimization is carried out to the space-time detection model according to third optimization fluctuation.
Further, described to carry out model loss for the space and time optimization model in preferred embodiments of the present invention
Calculation formula used by calculating are as follows:
The above-mentioned precipitation quantity measuring method based on big data, by establishing the design of the space-time detection model, so that will
The detection data is converted into concrete model parameter, and then effectively facilitates subsequent processing and calculating to data, by sentencing
It whether there is the design of the singular value in the detection parameters of breaking, so as to be effectively and timely corrected for abnormal data,
To improve the accuracy of testing result, by carrying out data-optimized design to the space-time detection model, and then it effectively prevent
Influence of the disturbing factor to the space-time detection model, further improves the precision of the testing result, by institute
It states space and time optimization model and carries out model costing bio disturbance, to determine the accuracy of the space and time optimization model, and it is described when determining
When space and time optimization model is correct, result is exported.
Second aspect, the present invention provides a kind of precipitation amount detection systems based on big data, comprising:
Modeling module establishes space-time detection model for obtaining detection data, and according to the detection data;
Judgment module, for judging that the detection parameters in the space-time detection model whether there is singular value, and when judgement
Into the detection parameters there are when the singular value, the detection parameters are corrected;
Optimization module, it is data-optimized for being carried out according to the local default principle of optimality to the space-time detection model, with
To space-time Optimized model;
Computing module to obtain loss function, and is worked as carrying out model costing bio disturbance for the space and time optimization model
When determining the loss function less than loss threshold value, the space and time optimization model is exported, to obtain testing result.
Further, in preferred embodiments of the present invention, the modeling module includes:
First modeling unit combines for obtaining the radar reflectivity factor stored in the detection data, and is with MSE
Optimization aim relies on mapping relations to establish precipitation space;
Second modeling unit, for obtaining the correlation of the Precipitation Process stored in the detection data, and with the drop
The correlation of water process is target, to establish Precipitation Time Series mapping relations.
Further, in preferred embodiments of the present invention, the judgment module includes:
First computing unit, for calculating separately the parameter difference between the adjacent detection parameters;
Judging unit, for judging whether the parameter difference is greater than difference threshold;If so, determining the parameter difference
The corresponding detection parameters are singular value.
Further, in preferred embodiments of the present invention, the judgment module further include:
Acquiring unit, for deleting the detection parameters, and obtain the parameter between the detection parameters adjacent parameter and;
Second computing unit, for calculating the average value of the parameter sum, to obtain replacement values, and by the replacement values pair
The detection parameters are replaced.
The above-mentioned precipitation amount detection systems based on big data establish the space-time detection model by the modeling module
Design, so as to convert concrete model parameter for the detection data, and then effectively facilitate the subsequent processing to data and
It calculates, the design that whether there is the singular value in the detection parameters is judged by the judgment module, so that effectively in time
Be corrected for abnormal data, to improve the accuracy of testing result, the space-time is detected by the optimization module
Model carries out data-optimized design, and then effectively prevents influence of the disturbing factor to the space-time detection model, further
The precision for improving the testing result carries out model loss meter to the space and time optimization model by the computing module
Calculate, to determine the accuracy of the space and time optimization model, and when determine the space and time optimization model it is correct when, export result.
Detailed description of the invention
It, below will be to use required in embodiment in order to illustrate more clearly of the technical solution of embodiment of the present invention
Attached drawing be briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not to be seen as
It is the restriction to range, it for those of ordinary skill in the art, without creative efforts, can be with root
Other relevant attached drawings are obtained according to these attached drawings.
Fig. 1 is the flow chart for the precipitation quantity measuring method based on big data that first embodiment of the invention provides;
Fig. 2 is the flow chart for the precipitation quantity measuring method based on big data that second embodiment of the invention provides;
Fig. 3 is the flow chart of the specific implementation step of step S61 in Fig. 2;
Fig. 4 is the structural schematic diagram for the precipitation amount detection systems based on big data that third embodiment of the invention provides:
Specific embodiment
For the ease of more fully understanding the present invention, the present invention is carried out further below in conjunction with related embodiment attached drawing
It explains.The embodiment of the present invention is given in attached drawing, but the present invention is not limited in above-mentioned preferred embodiment.On the contrary, providing
The purpose of these embodiments be in order to make disclosure of the invention face more sufficiently.
Expression or logic and/or step described otherwise above herein in flow charts, for example, being considered use
In the order list for the executable instruction for realizing logic function, may be embodied in any computer-readable medium, for
Instruction execution system, device or equipment (such as computer based system, including the system of processor or other can be held from instruction
The instruction fetch of row system, device or equipment and the system executed instruction) it uses, or combine these instruction execution systems, device or set
It is standby and use.For the purpose of this specification, " computer-readable medium ", which can be, any may include, stores, communicates, propagates or pass
Defeated program is for instruction execution system, device or equipment or the dress used in conjunction with these instruction execution systems, device or equipment
It sets.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms are not
Centainly refer to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be any
One or more embodiment or examples in can be combined in any suitable manner.
Referring to Fig. 1, the flow chart of the precipitation quantity measuring method based on big data provided for first embodiment of the invention,
Including step S10 to S60.
Step S10 obtains detection data, and establishes space-time detection model according to the detection data;
Wherein, the monitoring acquisition methods of the detection data can monitor for surface weather station's monitoring, satellite monitoring or radar,
Satellite monitoring is mainly detected by way of satellite cloud picture, and what is mainly reflected is cloud top information, and radar monitoring can be
Airspace surface sweeping is completed in several minutes.Spatial resolution is about 1 kilometer, time can produce millions of groups of spaces when each radar is each
Monitoring data, and therefore radar reflection has direct correlativity with precipitation, therefore, by using radar monitoring in the step
Mode is to carry out the acquisition of the detection data, it is preferred that by establishing the design of the space-time detection model in the step, with
Make to convert concrete model parameter for the detection data, and then effectively facilitates subsequent processing and calculating to data;
Step S20 judges the detection parameters in the space-time detection model with the presence or absence of singular value;
Wherein, by the design for judging to whether there is the singular value in the detection parameters, so that effectively and timely needle
Abnormal data is corrected, to improve the accuracy of testing result;
When step S20 is determined in the detection parameters there are when the singular value, step S30 is executed;
Step S30 is corrected the detection parameters;
Wherein, bearing calibration employed in the step can delete for parameter replacement, parameter fluctuation modification or parameter
Mode is corrected, and is then effectively prevented and is detected mistake as caused by wrong parameter, and described be based on is further improved
The accuracy of the precipitation quantity measuring method of big data;
Step S40, it is data-optimized to space-time detection model progress according to the local default principle of optimality, to obtain space-time
Optimized model;
Wherein, multiple optimal conditions are stored in the default principle of optimality, which can be according to the need of user
It asks and is independently configured, which can be data variance optimization, data fluctuations optimization etc., by detecting to the space-time
Model carries out data-optimized design, and then effectively prevents influence of the disturbing factor to the space-time detection model, further
Improve the precision of the testing result;
Step S50 carries out model costing bio disturbance for the space and time optimization model, to obtain loss function;
Wherein, by carrying out model costing bio disturbance to the space and time optimization model, to determine the space and time optimization model
Accuracy;
Step S60 exports the space and time optimization model when determining the loss function less than loss threshold value, with
To testing result;
Wherein, the corresponding precipitation information in each region is stored in the testing result, user can be based on the precipitation
Information is measured to carry out the prediction of precipitation;
In the present embodiment, by establishing the design of the space-time detection model, so as to convert tool for the detection data
Body Model parameter, and then effectively facilitate subsequent processing and calculating to data, by judge in the detection parameters whether
There are the designs of the singular value, so as to effectively and timely be corrected for abnormal data, to improve the accurate of testing result
Property, by carrying out data-optimized design to the space-time detection model, and then disturbing factor is effectively prevented to the space-time
The influence of detection model further improves the precision of the testing result, by carrying out mould to the space and time optimization model
Type costing bio disturbance, to determine the accuracy of the space and time optimization model, and when determine the space and time optimization model it is correct when, it is defeated
Result out.
Referring to Fig. 2, the flow chart of the precipitation quantity measuring method based on big data provided for second embodiment of the invention,
The method includes the steps S11 to S81.
Step S11 obtains detection data, obtains the radar reflectivity factor stored in the detection data and combines, and with
MSE is optimization aim, relies on mapping relations to establish precipitation space;
Wherein, influence of the space correlation system to testing result in order to prevent in the present embodiment, the step is by introducing space
Difference approach relies on mapping pass to establish the precipitation space, and then effectively reduces space correlation system to the shadow of testing result
It rings, improves the precision of the precipitation quantity measuring method based on big data;
Step S21 obtains the correlation of the Precipitation Process stored in the detection data, and with the phase of the Precipitation Process
Guan Xingwei target obtains space-time detection model to establish Precipitation Time Series mapping relations;
Wherein, in the present embodiment in order to prevent the time when ordered pair testing result influence, the step by introducing condition with
The mode on airport relies on mapping to establish the precipitation space and closes, so when effectively reducing the time ordered pair testing result shadow
It rings, improves the precision of the precipitation quantity measuring method based on big data;
Step S31 calculates separately the parameter difference in the space-time detection model between adjacent detection parameters, and judges institute
State whether parameter difference is greater than difference threshold;
Wherein, in the step by way of calculating the parameter difference, to judge the accurate of the corresponding detection parameters
Property, and then the data accuracy of the detection parameters in the effective space-time detection model is determined, detection is improved
The accuracy of method, it is preferred that the parameter difference and the difference threshold can be carried out in the step by the way of comparator
Size between value calculates, and difference threshold described in the present embodiment can be independently configured according to the demand of user, in turn
Effectively meet the multifarious demand of user data;
When step S31 determines the parameter difference greater than the difference threshold, step S41 is executed;
Step S41 determines that the corresponding detection parameters of the parameter difference are singular value, deletes the detection parameters,
And obtain the parameter between the detection parameters adjacent parameter and;
Wherein, it uses the mode deleted to prevent from detecting mistake caused by wrong parameter in the step, improves detection side
The accuracy of method;
Step S51 calculates the average value of the parameter sum, to obtain replacement values, and by the replacement values to the detection
Parameter is replaced;
Wherein, the mode supplementary data replaced in the step by using average value, so that data are complete, and then improves
The integrality of data, improves user experience;
Step S61, it is data-optimized to space-time detection model progress according to the local default principle of optimality, to obtain space-time
Optimized model;
Wherein, by carrying out data-optimized design to the space-time detection model, and then disturbing factor is effectively prevented
Influence to the space-time detection model, further improves the precision of the testing result;
Referring to Fig. 3, for the specific implementation step flow chart of step S61 in Fig. 2:
Step S610 carries out variance calculating to the space-time detection model, to obtain variance yields, and by the variance yields with
Locally pre-stored optimization table is matched, to obtain the first optimization fluctuation;
Step S611 carries out fluctuation optimization to the space-time detection model according to the first optimization fluctuation;
Step S612 is filtered calculating to the space-time detection model, to obtain filter value, and by the filter value with
The optimization table is matched, to obtain the second optimization fluctuation;
Step S613 carries out fluctuation optimization to the space-time detection model according to the second optimization fluctuation;
Step S614 is filtered variance to the space-time detection model and calculates, to obtain filter error variance value, and will be described
Filter error variance value is matched with the optimization table, to obtain third optimization fluctuation;
Step S615 carries out fluctuation optimization to the space-time detection model according to third optimization fluctuation;
Please continue to refer to Fig. 2, step S71, model costing bio disturbance is carried out for the space and time optimization model, to be lost
Function;
Wherein, by carrying out model costing bio disturbance to the space and time optimization model, to determine the space and time optimization model
Accuracy;
Step S81 exports the space and time optimization model when determining the loss function less than loss threshold value, with
To testing result;
Wherein, the corresponding precipitation information in each region is stored in the testing result, user can be based on the precipitation
Information is measured to carry out the prediction of precipitation;
Preferably, described to carry out meter used by model costing bio disturbance for the space and time optimization model in the present embodiment
Calculate formula are as follows:
In the present embodiment, by establishing the design of the space-time detection model, so as to convert tool for the detection data
Body Model parameter, and then effectively facilitate subsequent processing and calculating to data, by judge in the detection parameters whether
There are the designs of the singular value, so as to effectively and timely be corrected for abnormal data, to improve the accurate of testing result
Property, by carrying out data-optimized design to the space-time detection model, and then disturbing factor is effectively prevented to the space-time
The influence of detection model further improves the precision of the testing result, by carrying out mould to the space and time optimization model
Type costing bio disturbance, to determine the accuracy of the space and time optimization model, and when determine the space and time optimization model it is correct when, it is defeated
Result out.
Referring to Fig. 4, the structure of the precipitation amount detection systems 100 based on big data provided for third embodiment of the invention
Schematic diagram, comprising:
Modeling module 10 establishes space-time detection model for obtaining detection data, and according to the detection data, wherein
The monitoring acquisition methods of the detection data can be main for surface weather station's monitoring, satellite monitoring or radar monitoring, satellite monitoring
It is detected by way of satellite cloud picture, what is mainly reflected is cloud top information, and radar monitoring can be completed in several minutes
Airspace surface sweeping.Spatial resolution is about 1 kilometer, time can produce millions of groups of space monitoring data when each radar is each, and
Therefore radar reflection has direct correlativity with precipitation, therefore, by using the mode of radar monitoring to carry out in the module
The acquisition of the detection data, it is preferred that by establishing the design of the space-time detection model in the module, so that by the inspection
Measured data is converted into concrete model parameter, and then effectively facilitates subsequent processing and calculating to data.
Judgment module 20 for judging that the detection parameters in the space-time detection model whether there is singular value, and is worked as and is sentenced
Break into the detection parameters there are when the singular value, the detection parameters are corrected, wherein by judging the inspection
The design that whether there is the singular value is surveyed in parameter, so as to effectively and timely be corrected for abnormal data, to improve inspection
Survey the accuracy of result, it is preferred that bearing calibration employed in the module can be parameter replacement, parameter fluctuation modification or ginseng
The mode that number is deleted is corrected, and is then effectively prevented and is detected mistake as caused by wrong parameter, further improves
The accuracy of the precipitation quantity measuring method based on big data.
Optimization module 30, it is data-optimized for being carried out according to the local default principle of optimality to the space-time detection model, with
Obtain space and time optimization model, wherein be stored with multiple optimal conditions in the default principle of optimality, which can basis
The demand of user is independently configured, which can be data variance optimization, data fluctuations optimization etc., by described
Space-time detection model carries out data-optimized design, and then effectively prevents disturbing factor to the shadow of the space-time detection model
It rings, further improves the precision of the testing result.
Computing module 40, for carrying out model costing bio disturbance for the space and time optimization model, to obtain loss function, and
When determining the loss function less than loss threshold value, the space and time optimization model is exported, to obtain testing result, wherein
By carrying out model costing bio disturbance to the space and time optimization model, to determine the accuracy of the space and time optimization model, the inspection
It surveys in result and is stored with the corresponding precipitation information in each region, user can be based on the precipitation information to carry out precipitation
Prediction.
Preferably, meter used by model costing bio disturbance is carried out for the space and time optimization model in the computing module 40
Calculate formula are as follows:
Specifically, in the present embodiment, the modeling module 10 includes:
First modeling unit 11, for obtaining the radar reflectivity factor stored in the detection data combination, and with MSE
For optimization aim, rely on mapping relations to establish precipitation space, wherein in the present embodiment in order to prevent space correlation system to detection
As a result influence, the unit establish the precipitation space by introducing space interpolation method and rely on mapping pass, and then effective
Influence of the space correlation system to testing result is reduced, the accurate of the precipitation quantity measuring method based on big data is improved
Degree.
Second modeling unit 12, for obtaining the correlation of the Precipitation Process stored in the detection data, and with described
The correlation of Precipitation Process is target, to establish Precipitation Time Series mapping relations, wherein time in order to prevent in the present embodiment
When ordered pair testing result influence, which establishes the precipitation space by way of introducing condition random field and relies on mapping
The influence of ordered pair testing result when closing, and then effectively reducing the time improves the precipitation detection based on big data
The precision of method.
Further, the judgment module 20 includes:
First computing unit 21, for calculating separately the parameter difference between the adjacent detection parameters;
Judging unit 22, for judging whether the parameter difference is greater than difference threshold;If so, determining the parameter difference
Being worth the corresponding detection parameters is singular value, wherein in the unit by way of calculating the parameter difference, with judgement pair
Answer the accuracy of the detection parameters, and then the data accuracy of the detection parameters in the effective space-time detection model
Determined, improve the accuracy of detection method, it is preferred that the ginseng can be carried out in the unit by the way of comparator
Size between number difference and the difference threshold calculates, and difference threshold described in the present embodiment can be according to the demand of user
It is independently configured, and then effectively meets the multifarious demand of user data.
Preferably, the judgment module 20 further include:
Acquiring unit 23 for deleting the detection parameters, and obtains the parameter between the detection parameters adjacent parameter
With, wherein it uses the mode deleted to prevent from detecting mistake caused by wrong parameter in the unit, improves the standard of detection method
True property.
Second computing unit 24, for calculating the average value of the parameter sum, to obtain replacement values, and by the replacement values
The detection parameters are replaced, wherein the mode supplementary data replaced in the unit by using average value, so that data
Completely, and then the integralities of data is improved, improves user experience.
Specifically, the optimization module 30 includes:
First optimization unit 31, for carrying out variance calculating to the space-time detection model, to obtain variance yields, and by institute
It states variance yields and is matched with the optimization table being locally pre-stored, to obtain the first optimization fluctuation, fluctuated according to first optimization
Fluctuation optimization is carried out to the space-time detection model;
Second optimization unit 32, for being filtered calculating to the space-time detection model, to obtain filter value, and by institute
Filter value is stated to be matched with the optimization table, with obtain the second optimization fluctuation, according to it is described second optimization fluctuation to it is described when
Empty detection model carries out fluctuation optimization;
Third optimizes unit 33, calculates for being filtered variance to the space-time detection model, to obtain filter error variance
Value, and the filter error variance value is matched with the optimization table, to obtain third optimization fluctuation, optimized according to the third
Fluctuation carries out fluctuation optimization to the space-time detection model.
The above-mentioned precipitation amount detection systems 100 based on big data are established the space-time by the modeling module 10 and are detected
The design of model so as to convert concrete model parameter for the detection data, and then effectively facilitates subsequent to data
Processing and calculating judge the design that whether there is the singular value in the detection parameters by the judgment module 20, so that
It is effectively and timely corrected for abnormal data, to improve the accuracy of testing result, passes through 30 pairs of institutes of the optimization module
It states space-time detection model and carries out data-optimized design, and then effectively prevent disturbing factor to the shadow of the space-time detection model
It rings, further improves the precision of the testing result, the space and time optimization model is carried out by the computing module 40
Model costing bio disturbance, to determine the accuracy of the space and time optimization model, and when determine the space and time optimization model it is correct when,
Export result.
The present embodiment additionally provides a kind of mobile terminal, including storage equipment and processor, the storage equipment are used for
Computer program is stored, the processor runs the computer program so that the mobile terminal execution is above-mentioned based on big number
According to precipitation quantity measuring method.
The present embodiment additionally provides a kind of storage medium, is stored thereon with computer journey used in above-mentioned mobile terminal
Sequence, the program when being executed, include the following steps:
Detection data is obtained, and space-time detection model is established according to the detection data;
Judge that the detection parameters in the space-time detection model whether there is singular value, and works as and determine the detection parameters
In there are when the singular value, the detection parameters are corrected;
It is data-optimized to space-time detection model progress according to the local default principle of optimality, to obtain space and time optimization mould
Type;
Model costing bio disturbance is carried out for the space and time optimization model, to obtain loss function, and works as and determines the damage
When losing function less than loss threshold value, the space and time optimization model is exported, to obtain testing result.The storage medium, such as:
ROM/RAM, magnetic disk, CD etc..
It is apparent to those skilled in the art that for convenience and simplicity of description, only with above-mentioned each function
The division progress of unit, module can according to need and for example, in practical application by above-mentioned function distribution by different function
Energy unit or module are completed, i.e., the internal structure of storage device is divided into different functional unit or module, more than completing
The all or part of function of description.Each functional unit in embodiment, module can integrate in one processing unit,
It can be each unit to physically exist alone, can also be integrated in one unit with two or more units, it is above-mentioned integrated
Unit both can take the form of hardware realization, can also realize in the form of software functional units.In addition, each function list
Member, the specific name of module are also only for convenience of distinguishing each other, the protection scope being not intended to limit this application.
It will be understood by those skilled in the art that composed structure shown in Fig. 4 is not constituted to of the invention based on big number
According to precipitation amount detection systems restriction, may include than illustrating more or fewer components, perhaps combine certain components or
Different component layouts, and the precipitation quantity measuring method based on big data in Fig. 1-3 also uses shown in Fig. 4 more or more
Few component combines certain components or different component layouts perhaps to realize.The so-called unit of the present invention, module etc. are
Refer to that one kind can simultaneously function have reached performed by the processor (not shown) in the precipitation amount detection systems based on big data
At the series of computation machine program of specific function, the storage of the precipitation amount detection systems based on big data can be stored in
In equipment (not shown).
Above embodiment described technical principles of the invention, and the description is merely to explain the principles of the invention, and
It cannot be construed to the limitation of the scope of the present invention in any way.Based on the explanation herein, those skilled in the art is not required to
Other specific embodiments of the invention can be associated by paying creative labor, these modes fall within of the invention
In protection scope.
Claims (10)
1. a kind of precipitation quantity measuring method based on big data, which is characterized in that the described method includes:
Detection data is obtained, and space-time detection model is established according to the detection data;
Judge that the detection parameters in the space-time detection model are deposited in the detection parameters with the presence or absence of singular value, and when determining
In the singular value, the detection parameters are corrected;
It is data-optimized to space-time detection model progress according to the local default principle of optimality, to obtain space and time optimization model;
Model costing bio disturbance is carried out for the space and time optimization model, to obtain loss function, and works as and determines the loss letter
When number is less than loss threshold value, the space and time optimization model is exported, to obtain testing result.
2. the precipitation quantity measuring method according to claim 1 based on big data, which is characterized in that described according to the inspection
Measured data establishes the step of space-time detection model and includes:
The radar reflectivity factor combination stored in the detection data is obtained, and using MSE as optimization aim, to establish precipitation sky
Between rely on mapping relations;
The correlation of the Precipitation Process stored in the detection data is obtained, and using the correlation of the Precipitation Process as target,
To establish Precipitation Time Series mapping relations.
3. the precipitation quantity measuring method according to claim 1 based on big data, which is characterized in that it is described judgement it is described when
Detection parameters in empty detection model whether there is the step of singular value and include:
The parameter difference between the adjacent detection parameters is calculated separately, and judges whether the parameter difference is greater than difference threshold
Value;
If so, determining that the corresponding detection parameters of the parameter difference are singular value.
4. the precipitation quantity measuring method according to claim 1 based on big data, which is characterized in that described to the detection
The step of parameter is corrected include:
Delete the detection parameters, and obtain the parameter between the detection parameters adjacent parameter and;
The average value of the parameter sum is calculated, to obtain replacement values, and the replacement values are replaced the detection parameters.
5. the precipitation quantity measuring method according to claim 1 based on big data, which is characterized in that described according to local pre-
If the principle of optimality carries out data-optimized step to the space-time detection model
Variance calculating is carried out to the space-time detection model, to obtain variance yields, and by the variance yields with it is locally pre-stored
Optimization table is matched, to obtain the first optimization fluctuation;
Fluctuation optimization is carried out to the space-time detection model according to the first optimization fluctuation;
Calculating is filtered to the space-time detection model, to obtain filter value, and by the filter value and the optimization table into
Row matching, to obtain the second optimization fluctuation;
Fluctuation optimization is carried out to the space-time detection model according to the second optimization fluctuation;
Variance is filtered to the space-time detection model to calculate, to obtain filter error variance value, and by the filter error variance value with
The optimization table is matched, to obtain third optimization fluctuation;
Fluctuation optimization is carried out to the space-time detection model according to third optimization fluctuation.
6. the precipitation quantity measuring method according to claim 1 based on big data, which is characterized in that it is described for it is described when
Empty Optimized model carries out calculation formula used by model costing bio disturbance are as follows:
。
7. a kind of precipitation amount detection systems based on big data characterized by comprising
Modeling module establishes space-time detection model for obtaining detection data, and according to the detection data;
Judgment module for judging that the detection parameters in the space-time detection model whether there is singular value, and works as and determines institute
It states in detection parameters there are when the singular value, the detection parameters is corrected;
Optimization module, it is data-optimized for being carried out according to the local default principle of optimality to the space-time detection model, when obtaining
Empty Optimized model;
Computing module, for carrying out model costing bio disturbance for the space and time optimization model, to obtain loss function, and when judgement
When being less than loss threshold value to the loss function, the space and time optimization model is exported, to obtain testing result.
8. the precipitation amount detection systems according to claim 7 based on big data, which is characterized in that the modeling module packet
It includes:
First modeling unit is optimization for obtaining the radar reflectivity factor stored in the detection data combination, and with MSE
Target relies on mapping relations to establish precipitation space;
Second modeling unit, for obtaining the correlation of the Precipitation Process stored in the detection data, and with the precipitation mistake
The correlation of journey is target, to establish Precipitation Time Series mapping relations.
9. the precipitation amount detection systems according to claim 7 based on big data, which is characterized in that the judgment module packet
It includes:
First computing unit, for calculating separately the parameter difference between the adjacent detection parameters;
Judging unit, for judging whether the parameter difference is greater than difference threshold;If so, determining that the parameter difference is corresponding
The detection parameters be singular value.
10. the precipitation amount detection systems according to claim 9 based on big data, which is characterized in that the judgment module
Further include:
Acquiring unit, for deleting the detection parameters, and obtain the parameter between the detection parameters adjacent parameter and;
Second computing unit, for calculating the average value of the parameter sum, to obtain replacement values, and by the replacement values to described
Detection parameters are replaced.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811617914.7A CN109614744B (en) | 2018-12-28 | 2018-12-28 | Big data-based precipitation detection method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811617914.7A CN109614744B (en) | 2018-12-28 | 2018-12-28 | Big data-based precipitation detection method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109614744A true CN109614744A (en) | 2019-04-12 |
CN109614744B CN109614744B (en) | 2022-11-22 |
Family
ID=66011595
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811617914.7A Active CN109614744B (en) | 2018-12-28 | 2018-12-28 | Big data-based precipitation detection method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109614744B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111123410A (en) * | 2019-12-26 | 2020-05-08 | 国网北京市电力公司 | Precipitation monitoring system and method, storage medium and processor |
Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101647215A (en) * | 2007-03-29 | 2010-02-10 | 索尼株式会社 | Method of and apparatus for analyzing noise in a signal processing system |
CN103792524A (en) * | 2014-03-06 | 2014-05-14 | 兰州大学 | Radar quantitative precipitation estimation method based on cloud classification |
CN103948393A (en) * | 2014-05-21 | 2014-07-30 | 华侨大学 | Near-infrared noninvasive detection method and device for contents of blood components |
CN104398271A (en) * | 2014-11-14 | 2015-03-11 | 西安交通大学 | Method using three-dimensional mechanics and tissue specific imaging of blood vessels and plaques for detection |
CN104656163A (en) * | 2015-02-04 | 2015-05-27 | 中国人民解放军理工大学 | Rainfall distribution and dynamic measurement method based on big-data mobile communication network |
CN104751580A (en) * | 2015-04-08 | 2015-07-01 | 武汉理工光科股份有限公司 | Distributed optical fiber sensing signal mode identifying method and system |
CN104820222A (en) * | 2014-12-25 | 2015-08-05 | 邓勇 | Rainfall radar real-time dynamic attenuation correction and precipitation calculation method |
CN105425319A (en) * | 2015-09-16 | 2016-03-23 | 河海大学 | Rainfall satellite rainstorm assimilation method based on ground measuring data correction |
CN106127242A (en) * | 2016-06-21 | 2016-11-16 | 河海大学 | Year of based on integrated study Extreme Precipitation prognoses system and Forecasting Methodology thereof |
US20170220708A1 (en) * | 2016-01-29 | 2017-08-03 | Fujitsu Limited | Operating a solar power generating system |
CN107423190A (en) * | 2017-04-19 | 2017-12-01 | 国家电网公司 | A kind of daily record data points to recognition methods and device extremely |
CN107703564A (en) * | 2017-10-13 | 2018-02-16 | 中国科学院深圳先进技术研究院 | A kind of precipitation predicting method, system and electronic equipment |
CN108009632A (en) * | 2017-12-14 | 2018-05-08 | 清华大学 | Confrontation type space-time big data Forecasting Methodology |
-
2018
- 2018-12-28 CN CN201811617914.7A patent/CN109614744B/en active Active
Patent Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101647215A (en) * | 2007-03-29 | 2010-02-10 | 索尼株式会社 | Method of and apparatus for analyzing noise in a signal processing system |
CN103792524A (en) * | 2014-03-06 | 2014-05-14 | 兰州大学 | Radar quantitative precipitation estimation method based on cloud classification |
CN103948393A (en) * | 2014-05-21 | 2014-07-30 | 华侨大学 | Near-infrared noninvasive detection method and device for contents of blood components |
CN104398271A (en) * | 2014-11-14 | 2015-03-11 | 西安交通大学 | Method using three-dimensional mechanics and tissue specific imaging of blood vessels and plaques for detection |
CN104820222A (en) * | 2014-12-25 | 2015-08-05 | 邓勇 | Rainfall radar real-time dynamic attenuation correction and precipitation calculation method |
CN104656163A (en) * | 2015-02-04 | 2015-05-27 | 中国人民解放军理工大学 | Rainfall distribution and dynamic measurement method based on big-data mobile communication network |
CN104751580A (en) * | 2015-04-08 | 2015-07-01 | 武汉理工光科股份有限公司 | Distributed optical fiber sensing signal mode identifying method and system |
CN105425319A (en) * | 2015-09-16 | 2016-03-23 | 河海大学 | Rainfall satellite rainstorm assimilation method based on ground measuring data correction |
US20170220708A1 (en) * | 2016-01-29 | 2017-08-03 | Fujitsu Limited | Operating a solar power generating system |
CN106127242A (en) * | 2016-06-21 | 2016-11-16 | 河海大学 | Year of based on integrated study Extreme Precipitation prognoses system and Forecasting Methodology thereof |
CN107423190A (en) * | 2017-04-19 | 2017-12-01 | 国家电网公司 | A kind of daily record data points to recognition methods and device extremely |
CN107703564A (en) * | 2017-10-13 | 2018-02-16 | 中国科学院深圳先进技术研究院 | A kind of precipitation predicting method, system and electronic equipment |
CN108009632A (en) * | 2017-12-14 | 2018-05-08 | 清华大学 | Confrontation type space-time big data Forecasting Methodology |
Non-Patent Citations (1)
Title |
---|
李九龙等: ""基于"3σ法则"的显著误差检测"", 《计算机与现代化》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111123410A (en) * | 2019-12-26 | 2020-05-08 | 国网北京市电力公司 | Precipitation monitoring system and method, storage medium and processor |
Also Published As
Publication number | Publication date |
---|---|
CN109614744B (en) | 2022-11-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110174044B (en) | Bridge longitudinal displacement deformation monitoring method based on PSI technology | |
US20120188119A9 (en) | Identification and Analysis of Persistent Scatterers In Series of SAR Images | |
CN106871776A (en) | A kind of real-time deformation monitoring system based on GNSS | |
CN109358350A (en) | A kind of three frequency Cycle Slips Detection of Beidou and device | |
CN109991601A (en) | A kind of house methods of risk assessment based on PS-InSAR technology | |
Binjammaz et al. | GPS integrity monitoring for an intelligent transport system | |
CN109493569A (en) | Come down method for early warning, device, computer equipment and storage medium | |
CN110398753A (en) | GNSS survey station coordinate time sequence periodicity detection method and system | |
CN109540095A (en) | Roadbed settlement monitoring method based on satellite navigation and least square | |
CN105158778B (en) | Multisystem combined implementation carrier phase difference fault satellites elimination method and its system | |
CN114236645B (en) | Large-scale rainfall monitoring abnormal site screening method | |
CN106291542A (en) | A kind of tunnel three-D imaging method | |
CN108153979A (en) | Deformation information extraction method based on InSAR, terminal and storage medium | |
CN115792974A (en) | GNSS deformation monitoring result quality evaluation method | |
CN109614744A (en) | A kind of precipitation quantity measuring method and system based on big data | |
Bie et al. | Online method to impute missing loop detector data for urban freeway traffic control | |
CN116719073B (en) | GNSS (Global navigation satellite System) solution domain-oriented coarse difference detection and rejection method | |
CN115980317B (en) | Foundation GNSS-R data soil moisture estimation method based on corrected phase | |
CN110749307A (en) | Power transmission line displacement settlement determination method and system based on Beidou positioning | |
CN111650570A (en) | Three-dimensional atmospheric correction method and system for ground-based interference radar | |
CN107179064B (en) | A kind of determination method of the confidence level of wheelset profile on-line detecting system measured value | |
CN115201864A (en) | Method, device, storage medium and equipment for detecting clock difference jump of satellite | |
CN108168440A (en) | Ultra-high-tension power transmission line arc sag early-warning detection method based on radar satellite image | |
CN115166874A (en) | Meteorological drought index SPI construction method based on machine learning | |
JP4868471B2 (en) | Real-time earthquake risk prediction method |
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 | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20200918 Address after: Hangzhou City, Zhejiang province 310018 Xiasha Higher Education Park No. 2 Street No. 928 Applicant after: ZHEJIANG SCI-TECH University Address before: 330013 Jiangxi Province, Nanchang Qingshan Lake East Shuanggang Street No. 808 South Campus of East China Jiaotong University Applicant before: East China Jiaotong University |
|
GR01 | Patent grant | ||
GR01 | Patent grant |