CN108427041A - Lightning Warning method, system, electronic equipment and storage medium - Google Patents
Lightning Warning method, system, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN108427041A CN108427041A CN201810208869.3A CN201810208869A CN108427041A CN 108427041 A CN108427041 A CN 108427041A CN 201810208869 A CN201810208869 A CN 201810208869A CN 108427041 A CN108427041 A CN 108427041A
- Authority
- CN
- China
- Prior art keywords
- lightning
- data
- thunder
- target area
- data information
- 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
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R29/00—Arrangements for measuring or indicating electric quantities not covered by groups G01R19/00 - G01R27/00
- G01R29/08—Measuring electromagnetic field characteristics
- G01R29/0807—Measuring electromagnetic field characteristics characterised by the application
- G01R29/0814—Field measurements related to measuring influence on or from apparatus, components or humans, e.g. in ESD, EMI, EMC, EMP testing, measuring radiation leakage; detecting presence of micro- or radiowave emitters; dosimetry; testing shielding; measurements related to lightning
- G01R29/0842—Measurements related to lightning, e.g. measuring electric disturbances, warning systems
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/10—Devices for predicting weather conditions
Landscapes
- Environmental & Geological Engineering (AREA)
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Atmospheric Sciences (AREA)
- Biodiversity & Conservation Biology (AREA)
- Ecology (AREA)
- Environmental Sciences (AREA)
- Electromagnetism (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A kind of Lightning Warning method of present invention offer, system, electronic equipment and storage medium, Lightning Warning method include：Obtain the atmospheric electric field intensity data information and lighting location data information of target area；The temporal characteristics data that target area is determined according to atmospheric electric field intensity data information and lighting location data information determine the space characteristics data of target area according to lighting location data information；And according to temporal characteristics data, space characteristics data and the preset length with convolutional layer memory network LSTM models in short-term, determine the thunder and lightning prediction result of target area, and Lightning Warning is carried out according to the thunder and lightning prediction result.The present invention can carry out Lightning Warning from time and two, space data dimension, utilize the atmospheric electric field intensity data information and lighting location data information, obtain the time series of atmospheric electric field intensity and some spatial data informations of lighting location, the two is combined and carries out thunder and lightning prediction, the reliability and accuracy of Lightning Warning can be effectively improved.
Description
Technical field
The present invention relates to Lightning Warning technical fields, and in particular to a kind of Lightning Warning method, system, electronic equipment and deposits
Storage media.
Background technology
Thunder and lightning is the weather phenomenon for betiding a kind of Transient Currents in air, high voltage, strong electromagnetic radiation, its electricity
The physical effects such as pressure, high temperature can instantaneously generate great destructive power, and in today's society production activity, lightning still threatens
Forest, inflammable chemicals, and bring about great losses to the lives and properties of the mankind, to aviation, space flight communicates, electric power, oil depot,
There are very great influence in many departments of the national defence such as building and national economy.Since the life cycle of thunderstorm formation is short, range
Small feature so that accurately thunderstorm early warning is more difficult, this is also that meteorological department of China and global every country are faced
Problem and challenge.
Currently, the equipment that China carries out used in weather forecasting and thunder and lightning prediction is roughly divided into following classification：It is mostly general
Strangle weather radar, Lighting position machine, atmospheric electric field detector.We carry out weather forecasting and thunder and lightning predicts required data nothing more than
Radar map, Lighting Position Data and some region of atmospheric electric field intensity data utilize these various data, research
Personnel use different methods and carry out weather forecasting and thunder and lightning prediction, but the practical data detected to these real-time clocks are special
Sign, especially the application study work of thunderstorm feature, physical characteristic of atmospheric electric field etc. and these data in lightning warning aspect
Still be weak, between data potential rule and the excavation work of correlation it is yet fewer.These research work are main
Concentrate on several aspects：Establish physical equation progress numerical prediction, predicted using the method that radar map is extrapolated,
It establishes machine learning model to be predicted, such as decision-tree model, Logic Regression Models, supporting vector machine model.
But due to thunder and lightning when occurring possessed periodicity, space-time randomness, instantaneity and different regions it
Between the factors such as existing lightning Characteristics otherness, both increase difficulty of the people to the knowledge of regularity of lightening activity.Thunder and lightning occurs
When and the variation of atmospheric electric field have certain relationship, at the time of thunder and lightning occurs, electric potential gradient need to reach air breakdown potential in air
Potential gradient.When thunder cloud is close to atmospheric electric field detector, electric field value generally will appear the fast feature for becoming shake, this is that cloud layer is frequently put
Electrical phenomena.However ground electric field might not have a notable fluctuation after lightning occurs, and electric field small change but also not necessarily generation
Occur without lightning in table area, but can be effectively reduced using the networking of more atmospheric electric field detectors observation and not observe lightning
The case where.Previous achievement in research mostly emphasis carries out thunder and lightning prediction by electric field strength data by has ignored lightning information
Space characteristics so that space characteristics are but difficult to capture, and substantially reduce the dimension of feature and the reliability of Lightning Warning, accurate in this way
Property.
Invention content
For the problems of the prior art, a kind of Lightning Warning method of present invention offer, system, electronic equipment and storage are situated between
Matter can effectively improve the reliability and accuracy of Lightning Warning.
In order to solve the above technical problems, the present invention provides following technical scheme：
In a first aspect, the present invention provides a kind of Lightning Warning method, the Lightning Warning method includes：
Obtain the atmospheric electric field intensity data information and lighting location data information of target area；
The time of the target area is determined according to the atmospheric electric field intensity data information and lighting location data information
Characteristic, and determine according to the lighting location data information space characteristics data of the target area；
And according to the temporal characteristics data, space characteristics data and the preset long short-term memory with convolutional layer
Network LSTM models determine the thunder and lightning prediction result of the target area, and carry out Lightning Warning according to the thunder and lightning prediction result.
Further, described according to the temporal characteristics data, space characteristics data and the preset length with convolutional layer
Short-term memory network LSTM models determine the thunder and lightning prediction result of the target area, and are carried out according to the thunder and lightning prediction result
Lightning Warning, including：
The temporal characteristics data and space characteristics data are formed into data set, and by the data set according to default ratio
Graduation is training set and test set；
Data training is carried out to the preset shot and long term memory network LSTM models with convolutional layer using the training set,
Obtain the target weight parameter for showing thunder and lightning prediction result；
And recruitment evaluation is carried out to the target weight parameter according to the training set, if the knot of the recruitment evaluation
Fruit meets preset requirement, then carries out Lightning Warning according to the target weight parameter.
Further, the atmospheric electric field intensity data information and lighting location data information for obtaining target area, packet
It includes：
The atmospheric electric field intensity data information that atmospheric electric field detector in the target area acquires the target area is set,
And the change curve of atmospheric electric field intensity is drawn according to the atmospheric electric field intensity data information；
And the lighting location data letter that the Lighting position machine in the target area acquires the target area is set
Breath, wherein the lighting location data information includes：The location information and thunder and lightning of temporal information, thunder and lightning generation that thunder and lightning occurs
Apart from the elevation information on ground when generation.
Further, described that the mesh is determined according to the atmospheric electric field intensity data information and lighting location data information
The temporal characteristics data in region are marked, including：
According to the change curve of the atmospheric electric field intensity, obtained respectively for indicating thunderstorm cloud cluster close to or away from described
The curve polarity smooth features data A1 of atmospheric electric field detector, for indicating that the curve of the severe degree of thunderstorm cloud cluster energy variation is trembled
Move characteristic A2, the curve steepness characteristic A3 for indicating thunderstorm cloud cluster fuel deposition rate and for indicating thunder cloud
The curve peak characteristic A4 of group's energy size；
And the position of the temporal information, thunder and lightning generation occurred according to the thunder and lightning in the lighting location data information is believed
Elevation information when breath and thunder and lightning occur apart from ground, obtains respectively using the Lighting position machine as the center of circle and is with pre-determined distance
The thunder and lightning cloud of thunder and lightning total quantity characteristic A5, the target area in the range of radius dodge the ratio characteristic data dodged with ground
A6, and, the closing speed A7 of the thunderstorm cloud cluster；
Wherein, the curve polarity smooth features data A1, curve jitter feature data A2, curve steepness characteristic
A3, curve peak characteristic A4, thunder and lightning total quantity characteristic A5, thunder and lightning cloud dodge with ground dodge ratio characteristic data A6 and
The closing speed A7 of thunderstorm cloud cluster collectively constitutes the temporal characteristics data of the target area.
Further, the space characteristics data that the target area is determined according to the lighting location data information,
Including：
It is recommended that the network of a N*N, and each cell in the network represents 1 square kilometre；
And thunder and lightning whereabouts time numerical value is marked in corresponding each unit lattice according to the lighting location data information, it obtains
To N*N matrixes, which is the space characteristics data of the target area；
Wherein, N is the positive integer more than 1.
Further, described that the temporal characteristics data and space characteristics data are formed into data set, and by the data
Integrate according to default ratio graduation as training set and test set, including：
The temporal characteristics data are formed into a row vector a1；
Space characteristics data input convolutional layer is subjected to process of convolution, the space characteristics data after process of convolution are turned
It is changed to one-dimensional row vector a2；
Combine the row vector a1 and the one-dimensional row vector a2, the data set a3；
And by the data set a3 according to default ratio graduation be training set and test set.
Further, it is described using the training set to the preset shot and long term memory network LSTM models with convolutional layer
Data training is carried out, the target weight parameter for showing thunder and lightning prediction result is obtained, including：
Data training is carried out to the preset shot and long term memory network LSTM models with convolutional layer using the training set,
And the corresponding loss function of shot and long term memory network LSTM models with convolutional layer is to intersect entropy function；
And the intersection entropy function is iterated using gradient descent algorithm, until obtaining for showing that thunder and lightning is pre-
Survey the target weight parameter of result.
Second aspect, the present invention provide a kind of Lightning Warning system, and the Lightning Warning system includes：
Data acquisition module, the atmospheric electric field intensity data information for obtaining target area and lighting location data letter
Breath；
Characteristic extracting module, for determining institute according to the atmospheric electric field intensity data information and lighting location data information
The temporal characteristics data of target area are stated, and determine the space characteristics of the target area according to the lighting location data information
Data；
Lightning Warning module, for according to the temporal characteristics data, space characteristics data and it is preset carry convolutional layer
Length memory network LSTM models in short-term, determine the thunder and lightning prediction result of the target area, and according to the thunder and lightning prediction result
Carry out Lightning Warning.
The third aspect, the present invention provides a kind of electronic equipment, including memory, processor and storage are on a memory and can
The computer program run on a processor, the processor realize the step of the Lightning Warning method when executing described program
Suddenly.
Fourth aspect, the present invention provide a kind of computer readable storage medium, are stored thereon with computer program, the calculating
Machine program realizes the step of Lightning Warning method when being executed by processor.
As shown from the above technical solution, Lightning Warning method provided by the invention, by the atmospheric electricity for obtaining target area
Field intensity data information and lighting location data information；Believed according to the atmospheric electric field intensity data information and lighting location data
Breath determines the temporal characteristics data of the target area, and determines the target area according to the lighting location data information
Space characteristics data；And in short-term according to the temporal characteristics data, space characteristics data and the preset length with convolutional layer
Memory network LSTM models determine the thunder and lightning prediction result of the target area, and carry out thunder and lightning according to the thunder and lightning prediction result
Early warning can carry out Lightning Warning from two data dimensions of time and space, using the atmospheric electric field intensity data information and
Lighting location data information obtains the time series of atmospheric electric field intensity and some spatial data informations of lighting location, two
Person, which combines, carries out thunder and lightning prediction, can effectively improve the reliability and accuracy of Lightning Warning.
Description of the drawings
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is the present invention
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.
Fig. 1 is the flow diagram of the Lightning Warning method in the embodiment of the present invention one；
Fig. 2 be the present invention Lightning Warning method in step 100 flow diagram；
Fig. 3 be the present invention Lightning Warning method in step 200 flow diagram；
Fig. 4 be the present invention Lightning Warning method in step 300 flow diagram；
Fig. 5 is the flow diagram of the Lightning Warning method in the application example of the present invention；
Fig. 6 is the schematic diagram that the thunderstorm cloud cluster in the application example of the present invention is being moved to electric field instrument；
Fig. 7 is that the thunderstorm cloud cluster in the application example of the present invention is leaving the schematic diagram of electric field instrument；
Fig. 8 be the present invention application example in thunderstorm cloud cluster not close to electric field instrument when electric field strength change over time
Curve synoptic diagram；
Fig. 9 be thunderstorm cloud cluster in the application example of the present invention close to electric field instrument when the song that changes over time of electric field strength
Line schematic diagram；
Figure 10 is the atmospheric electric field shake schematic diagram in the application example of the present invention；
Figure 11 is the atmospheric electric field steepness schematic diagram in the application example of the present invention；
Figure 12 is the LSTM loop body structural schematic diagrams in the application example of the present invention；
Figure 13 is the structural schematic diagram of the Lightning Warning system in the embodiment of the present invention two；
Figure 14 is the structural schematic diagram of the electronic equipment in the embodiment of the present invention three.
Specific implementation mode
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical solution in the embodiment of the present invention carries out clear, complete description, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
The every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
The embodiment of the present invention one provides a kind of specific implementation mode of Lightning Warning method, and referring to Fig. 1, the thunder and lightning is pre-
Alarm method specifically includes following content：
Step 100：Obtain the atmospheric electric field intensity data information and lighting location data information of target area.
In step 100, the Lightning Warning system obtains the atmospheric electric field intensity data information and thunder and lightning of target area
Location data information.It is understood that the Lightning Warning system obtains the monitoring device being arranged in the target area
The atmospheric electric field intensity data information and lighting location data information of the target area sent.
It is understood that effectively improve the accuracy of data acquisition, the monitoring in the target area is set
Equipment can have multiple, and the target area can be completely covered in the monitoring range summation of multiple monitoring devices.
Step 200：The target area is determined according to the atmospheric electric field intensity data information and lighting location data information
The temporal characteristics data in domain, and determine according to the lighting location data information space characteristics data of the target area.
In step 200, the Lightning Warning system is according to the atmospheric electric field intensity data information and lighting location number
It is believed that breath determines the temporal characteristics data of the target area, and the target area is determined according to the lighting location data information
The space characteristics data in domain.It is understood that the Lightning Warning system can be first according to the atmospheric electric field intensity number
According to n temporal characteristics data of acquisition of information, when determining latter m of the target area further according to the lighting location data information
Between characteristic, then n temporal characteristics data and m temporal characteristics data be combined, obtain whole temporal characteristics numbers
According to.
Step 300：Remembered in short-term according to the temporal characteristics data, space characteristics data and the preset length with convolutional layer
Recall network LSTM models, determines the thunder and lightning prediction result of the target area, and pre- according to thunder and lightning prediction result progress thunder and lightning
It is alert.
In step 300, the Lightning Warning system is according to the temporal characteristics data, space characteristics data and preset
Length with convolutional layer memory network LSTM models in short-term, determine the thunder and lightning prediction result of the target area, and according to the thunder
Electric prediction result carries out Lightning Warning.It is understood that the Lightning Warning system is by the temporal characteristics data and space
Input of the characteristic as the preset length with convolutional layer memory network LSTM models in short-term so that described preset
Memory network LSTM models are computed output thunder and lightning prediction result to length with convolutional layer in short-term, then predict to tie according to the thunder and lightning
Fruit carries out Lightning Warning.
Further it will be understood that the preset long short-term memory with convolutional layer (Convention layer)
Network LSTM (Long Short-Term Memory) model can be simultaneously to the temporal characteristics data and space characteristics data
Data processing is carried out, i.e. the preset length with convolutional layer includes in short-term that can carry out space in memory network LSTM models
The structure of characteristic processing, wherein LSTM (Long Short-Term Memory) is a kind of time recurrent neural network, is fitted
Together in being spaced and postpone relatively long critical event in processing and predicted time sequence, and convolutional layer is that processing space data are special
A kind of structure of sign.
As can be seen from the above description, the Lightning Warning method that the embodiment of the present invention provides, from time and two, space data
Dimension carries out Lightning Warning and obtains atmospheric electric field using the atmospheric electric field intensity data information and lighting location data information
The time series of intensity and some spatial data informations of lighting location combine the two and carry out thunder and lightning prediction, Neng Gouyou
Effect improves the reliability and accuracy of Lightning Warning.
In a specific embodiment, the present invention also provides the specific implementations of step 100 in the Lightning Warning method
Mode, referring to Fig. 2, the step 100 specifically includes following content：
Step 101：The atmospheric electric field intensity that atmospheric electric field detector in the target area acquires the target area is set
Data information, and according to the change curve of atmospheric electric field intensity data information drafting atmospheric electric field intensity.
Step 102：The lighting location data that Lighting position machine in the target area acquires the target area are set
Information, wherein the lighting location data information includes：The location information and thunder of temporal information, thunder and lightning generation that thunder and lightning occurs
Elevation information when electricity occurs apart from ground.
In step 101 and 102, the Lightning Warning system is believed using atmospheric electric field detector acquisition atmospheric electric field intensity data
Breath can utilize drawing tool to depict the change curve of atmospheric electric field intensity, understand atmospheric electric field intensity accurate and visually
Change conditions, and, include mainly thunder using the lighting location data information occurred within the scope of Lighting position machine pickup area
The attribute informations such as the height when position (latitude and longitude information) of the generation of time, thunder and lightning that electricity occurs, thunder and lightning occur apart from ground.
As can be seen from the above description, the Lightning Warning method that the embodiment of the present invention provides, can accurately obtain pneumoelectric field strength
Degrees of data information and lighting location data information provide accurate data basis for subsequent Lightning Warning.
In a specific embodiment, the present invention also provides the specific implementations of step 200 in the Lightning Warning method
Mode, referring to Fig. 3, the step 200 specifically includes following content：
Step 201：According to the change curve of the atmospheric electric field intensity, obtain respectively for indicate thunderstorm cloud cluster it is close or
Curve polarity smooth features data A1 far from the atmospheric electric field detector, the severe degree for indicating thunderstorm cloud cluster energy variation
Curve jitter feature data A2, the curve steepness characteristic A3 for indicating thunderstorm cloud cluster fuel deposition rate and be used for table
Show the curve peak characteristic A4 of thunderstorm cloud cluster energy size.
Step 202：The position of the temporal information, thunder and lightning generation that are occurred according to the thunder and lightning in the lighting location data information
Elevation information when information and thunder and lightning occur apart from ground, obtains respectively using the Lighting position machine as the center of circle and with pre-determined distance
Thunder and lightning cloud for thunder and lightning total quantity characteristic A5, the target area in the range of radius dodges the ratio characteristic number dodged with ground
According to A6, and, the closing speed A7 of the thunderstorm cloud cluster.
It is understood that the curve polarity smooth features data A1, curve jitter feature data A2, curve steepness are special
It levies data A3, curve peak characteristic A4, thunder and lightning total quantity characteristic A5, thunder and lightning cloud and dodges the ratio characteristic number dodged with ground
The temporal characteristics data of the target area are collectively constituted according to the closing speed A7 of A6 and thunderstorm cloud cluster.
Step 203：It is recommended that the network of a N*N, and each cell in the network represents 1 square of public affairs
In.
Step 204：Thunder and lightning whereabouts time numerical value is marked in corresponding each unit lattice according to the lighting location data information,
N*N matrixes are obtained, which is the space characteristics data of the target area.
It is understood that N is the positive integer more than 1.
As can be seen from the above description, the Lightning Warning method that the embodiment of the present invention provides, can accurately obtain temporal characteristics
Data and space characteristics data provide accurate data basis for subsequent Lightning Warning.
In a specific embodiment, the present invention also provides the specific implementations of step 300 in the Lightning Warning method
Mode, referring to Fig. 4, the step 300 specifically includes following content：
Step 301：The temporal characteristics data and space characteristics data are formed into data set, and by the data set according to
Default ratio graduation is training set and test set.
In the step 301, the temporal characteristics data are formed a row vector by the Lightning Warning system first
a1；Then space characteristics data input convolutional layer is subjected to process of convolution, the space characteristics data after process of convolution is turned
It is changed to one-dimensional row vector a2；Combine the row vector a1 and the one-dimensional row vector a2, the data set a3；And it will be described
Data set a3 is training set and test set according to default ratio graduation.
Step 302：The preset shot and long term memory network LSTM models with convolutional layer are carried out using the training set
Data are trained, and the target weight parameter for showing thunder and lightning prediction result is obtained.
In step 302, using the training set to the preset shot and long term memory network LSTM models with convolutional layer
Data training is carried out, and the corresponding loss function of shot and long term memory network LSTM models with convolutional layer is cross entropy letter
Number；And the intersection entropy function is iterated using gradient descent algorithm, until obtaining for showing thunder and lightning prediction result
Target weight parameter.The shot and long term memory network LSTM models with convolutional layer are a bases based on common RNN
Plinth model LSTM (Long Short Term Memory) is improved, since LSTM has the cycle containing loop structure
Body, it, which has the information characteristics of front, retains memory function, therefore it can be carried out according to the temporal aspect of front well
Prediction.But containing spatial data information in our data, if direct construction LSTM networks, we may lose space number
According to space characteristics.Then, we add convolutional coding structure in LSTM neural networks, we have just obtained our needs at this time
Model carry the LSTM deep neural networks of convolutional coding structure, our spatial information in this way can be extracted well
Come.At this point, the LSTM deep neural networks with convolutional coding structure that we obtain are provided simultaneously with memory function and spatial information is special
Abstraction function is levied, the precision and accuracy of Lightning Warning model are further increased.Lightning Warning system is special by the time
It levies data and space characteristics data forms data set, and be training set and test according to default ratio graduation by the data set
Collection；Data training is carried out to the preset shot and long term memory network LSTM models with convolutional layer using the training set, is obtained
Target weight parameter for showing thunder and lightning prediction result；And the target weight parameter is carried out according to the training set
Recruitment evaluation carries out Lightning Warning if the result of the recruitment evaluation meets preset requirement according to the target weight parameter.
Wherein, shown in the following formula of the shot and long term memory network LSTM models with convolutional layer one：
FC-LSTM
Above-mentioned formula is the calculation formula of a state cell of LSTM, and in this formula, σ is sigmoid functions, it
For state cell input activation primitive,It is state cell output activation primitive, W**It is weight ginseng
Number, xtIt is the input data of t moment state cell, b*It is offset parameter, ctIt is the state value of t moment state cell.htWhen being t
Carve the output valve of state cell.
Step 303：And recruitment evaluation is carried out to the target weight parameter according to the training set, if the effect
The result of assessment meets preset requirement, then carries out Lightning Warning according to the target weight parameter.
In step 303, the LSTM deep neural networks with convolutional coding structure that the Lightning Warning uses to data into
Row training, we will form a row vector by the feature (the one before feature in feature extraction) of extraction in every six minutes, the 9th
The feature that a matrix character obtained after process of convolution carries out conversion and becomes one-dimensional row vector, then with the row vector of front into
Data set is divided into two parts by the row vector of row splicing one bigger of composition by a certain percentage first after handling data set well：
Training set and test set, then training set is brought into the LSTM deep neural network models with convolutional coding structure of structure into
Row training study, wherein loss function are to intersect entropy function, are iterated using gradient descent algorithm until obtaining best power
Weight parameter.
As can be seen from the above description, the embodiment of the present invention provides Lightning Warning method, obtain with convolutional coding structure
LSTM deep neural networks are provided simultaneously with memory function and spatial information feature extraction functions, further increase Lightning Warning
The precision and accuracy of model.
For further instruction this programme, the present invention also provides a kind of concrete application example of Lightning Warning method, referring to
Fig. 5, the Lightning Warning method specifically include following content：
The Lightning Warning method is mainly using the LSTM deep neural networks with convolutional coding structure to atmospheric electricity field strength
Degrees of data and lighting location data are handled, are predicted.On the one hand, since the LSTM deep neural networks with convolutional coding structure have
There are long Memorability, the correlated characteristic before thunder and lightning can be occurred well in short-term to remain, better foundation is provided for prediction.
On the other hand, since the LSTM deep neural networks with convolutional coding structure have convolutional coding structure, the space in data can be tied
Structure feature extraction comes out, and further increases the precision and accuracy rate of prediction.
The process flow of the Lightning Warning method includes the following steps：
Step 1：Data acquire
The electric field region of the atmospheric electric field detector equipment detection used in this application example is limited in scope, so this application example
For some areas Lightning Warning, here is the data source of this application example.
(1) it utilizes atmospheric electric field detector to acquire atmospheric electric field intensity data information, air can be depicted using drawing tool
The change curve of electric field strength understands the change conditions of atmospheric electric field intensity accurate and visually.
(2) the lighting location data information occurred within the scope of Lighting position machine pickup area is utilized, includes mainly thunder and lightning
The attribute informations such as the height when time of generation, the position (latitude and longitude information) of the generation of thunder and lightning, thunder and lightning occur apart from ground.
Step 2：Feature extraction
The extraction of data characteristics is necessary, and is also important.In this application example, according to the periodicity of thunder and lightning, wink
It is as follows that the features such as Shi Xing, mobility, has carried out targetedly feature extraction, main feature to the data source of acquisition：
Atmospheric electric field intensity data information correlated characteristic：
(1) as shown in Fig. 6 to 9, it is characterized as that smooth polarity upset, this feature show thunderstorm group close to electric field in Fig. 6 to 9
Instrument or the process for leaving electric field instrument.Under normal conditions, the close initial mark of thundercloud is that field strength is born by just switching to, meaning
Taste thundercloud and is moved toward overhead.Watch window at least 40 minutes.Sliding window (width：10 minutes, step-length 1 minute) in whether have
Polarity is smoothly overturn, including from positive to negative and from negative to positive, is had, and is set to 1, is otherwise 0.
(2) as shown in Figure 10, it is characterized as that degree of jitter, this feature represent the violent journey of thunderstorm group energy variation in Figure 10
Degree generally means that nearby there is lightning stroke.The definition shaken greatly：Short time (be defined as 1 minute) electric field intensity inside high change dramatically with
It is rapid afterwards to restore.The value of this variable is the amplitude peak of all big shakes in watch window.
(3) as shown in figure 11, it is characterized as that steepness, this feature description thunderstorm roll into a ball fuel deposition rate, that is, electricity in Figure 11
Field intensity value Ramp Rate：A how many minute electric-field strengths increase Mkv/m.The feelings that electric field absolute value does not thunder when big but very flat
When condition, the time of being climbed is longer.
(4) this is characterized as electric field strength maximum value, this feature description size of thundercloud energy.
Lighting Position Data correlated characteristic information：
(5) short distance thunder and lightning total quantity
(6) the thunder and lightning cloud of guard plot dodges and ground dodges ratio：Because thunderstorm group generates the rule for having ground after first cloud, cloud than with
Thunderstorm group's life cycle is related, which represents thunderstorm group maturity feature.
(8) the thunderstorm closing speed of guard plot：Speed of the nearest position in distance protection area of thunderstorm group close to guard plot.
(9) a 50*50 network is built, each lattice in grid represents 1 square kilometre, if there is several thunders
It is several that electricity, which falls and just marks the value of the grid within a grid, such as：There are 5 thunders and lightnings to fall in a certain grid, just marks the grid
Value is 5.We just obtain the matrix of a 50*50 in this way.
Step 3：Build model
The model that the present invention uses is a basic model LSTM (the Long Short Term based on common RNN
Memory it) improves.Since LSTM has the loop body containing loop structure, referring to Figure 12, its information for front
Feature, which has, retains memory function, therefore it can be predicted according to the temporal aspect of front well.But our data
In contain spatial data information, if direct construction LSTM networks, we may lose the space characteristics of spatial data.Then,
We add convolutional coding structure in LSTM neural networks, we have just obtained the model of our needs with convolutional coding structure at this time
LSTM deep neural networks, our spatial information in this way can be extracted well.At this point, the band that we obtain
There are the LSTM deep neural networks of convolutional coding structure to be provided simultaneously with memory function and spatial information feature extraction functions, further carries
The high precision and accuracy of Lightning Warning model.
Step 4：Training pattern
In this application example, the LSTM deep neural networks with convolutional coding structure of use are trained data, will
A row vector is formed by the feature (the one before feature in feature extraction) of extraction in every six minutes, the 9th matrix character is carried out
The feature obtained after process of convolution carries out conversion and becomes one-dimensional row vector, and splicing composition one is then carried out with the row vector of front
The row vector of bigger.According to such regulation, we are the 2016 and 2017 electric-field strength number of degrees in Beijing world Capital Airport
Processing and feature extraction are carried out according to lightning data, then labeling.After handling data set well, we first press data set
Certain proportion (7：3) be divided into two parts training set and test set, then training set be brought into we structure carry convolution knot
Study is trained in the LSTM deep neural network models of structure, wherein loss function is to intersect entropy function, is declined using gradient
Algorithm is iterated until obtaining best weight parameter.
Step 5：Assessment models
For trained model, it is necessary to the quality for assessing a drag, in the Lightning Warning model of the present invention,
Seek to assess the effect of weight parameter, we only need the data input model test set, check the loss of model
Functional value and accuracy.
Therefore, Lightning Warning model using the present invention so that we are simultaneously the information and Spatial Dimension of time dimension
Information all apply in prediction well, for we subsequent Lightning Warning experiment in provide more wide thinking and
Theoretical foundation.
Memorability and space of this application example by the LSTM deep neural networks with convolutional coding structure to time series
The feature extraction functions of information so that we can be comprehensive the time of data source and spatial signature information when carrying out thunder and lightning prediction
Altogether.
As can be seen from the above description, compared with Lightning Warning model before, the Lightning Warning model of this application example is simultaneously
Time series feature with memory front thunder and lightning, and also convolutional coding structure extracts spatial signature information.So of the invention
Lightning Warning model atmospheric electric field intensity and the thunder and lightning attribute information that has occurred and that can be utilized to promote the precision of Lightning Warning,
It eliminates and changes over time the various interference that the surrounding enviroment variation brought and ageing equipment etc. are brought, realization is more acurrate timely to dodge
Electric early warning.
The embodiment of the present invention two provides a kind of Lightning Warning that can realize full content in the Lightning Warning method
The specific implementation mode of system, referring to Figure 13, the Lightning Warning system specifically includes following content：
Data acquisition module 10, the atmospheric electric field intensity data information for obtaining target area and lighting location data letter
Breath.
Characteristic extracting module 20, for being determined according to the atmospheric electric field intensity data information and lighting location data information
The temporal characteristics data of the target area, and determine that the space of the target area is special according to the lighting location data information
Levy data.
Lightning Warning module 30, for according to the temporal characteristics data, space characteristics data and it is preset carry convolution
The length memory network LSTM models in short-term of layer determine the thunder and lightning prediction result of the target area, and predict to tie according to the thunder and lightning
Fruit carries out Lightning Warning.
The embodiment of Lightning Warning system provided by the invention specifically can be used for executing the thunder and lightning in above-described embodiment one
The process flow of the embodiment of method for early warning, details are not described herein for function, is referred to retouching in detail for above method embodiment
It states.
As can be seen from the above description, the Lightning Warning system that the embodiment of the present invention provides, from time and two, space data
Dimension carries out Lightning Warning and obtains atmospheric electric field using the atmospheric electric field intensity data information and lighting location data information
The time series of intensity and some spatial data informations of lighting location combine the two and carry out thunder and lightning prediction, Neng Gouyou
Effect improves the reliability and accuracy of Lightning Warning.
The embodiment of the present invention three, which provides, can realize Overall Steps in the Lightning Warning method in above-described embodiment one
The specific implementation mode of a kind of electronic equipment, referring to Figure 14, the electronic equipment specifically includes following content：
Processor (processor) 601, memory (memory) 602, communication interface (Communications
Interface) 603 and bus 604；
Wherein, the processor 601, memory 602, communication interface 603 complete mutual lead to by the bus 604
Letter；The communication interface 603 is transmitted for realizing the information between the relevant devices such as each monitoring station and central server；
The processor 601 is used to call the computer program in the memory 602, the processor to execute the meter
The Overall Steps in above-described embodiment one are realized when calculation machine program, for example, reality when the processor executes the computer program
Existing following step：
Step 100：Obtain the atmospheric electric field intensity data information and lighting location data information of target area.
Step 200：The target area is determined according to the atmospheric electric field intensity data information and lighting location data information
The temporal characteristics data in domain, and determine according to the lighting location data information space characteristics data of the target area.
Step 300：Remembered in short-term according to the temporal characteristics data, space characteristics data and the preset length with convolutional layer
Recall network LSTM models, determines the thunder and lightning prediction result of the target area, and pre- according to thunder and lightning prediction result progress thunder and lightning
It is alert.
As can be seen from the above description, the electronic equipment that the embodiment of the present invention provides, from time and two, space data dimension
It carries out Lightning Warning and obtains atmospheric electric field intensity using the atmospheric electric field intensity data information and lighting location data information
Time series and lighting location some spatial data informations, the two combine carry out thunder and lightning prediction, can effectively carry
The reliability and accuracy of high Lightning Warning.
The embodiment of the present invention four provides one of Overall Steps in the Lightning Warning method that can realize above-described embodiment one
Computer readable storage medium is planted, is stored with computer program on the computer readable storage medium, the computer program quilt
Processor realizes the Overall Steps of above-described embodiment one when executing, for example, reality when the processor executes the computer program
Existing following step：
Step 100：Obtain the atmospheric electric field intensity data information and lighting location data information of target area.
Step 200：The target area is determined according to the atmospheric electric field intensity data information and lighting location data information
The temporal characteristics data in domain, and determine according to the lighting location data information space characteristics data of the target area.
Step 300：Remembered in short-term according to the temporal characteristics data, space characteristics data and the preset length with convolutional layer
Recall network LSTM models, determines the thunder and lightning prediction result of the target area, and pre- according to thunder and lightning prediction result progress thunder and lightning
It is alert.
As can be seen from the above description, the computer readable storage medium that the embodiment of the present invention provides, from time and space two
A data dimension carries out Lightning Warning, using the atmospheric electric field intensity data information and lighting location data information, obtains big
The time series of pneumoelectric field intensity and some spatial data informations of lighting location combine the two and carry out thunder and lightning prediction,
The reliability and accuracy of Lightning Warning can be effectively improved.
In the description of the present invention, it should be noted that the orientation or positional relationship of the instructions such as term "upper", "lower" is base
It in orientation or positional relationship shown in the drawings, is merely for convenience of description of the present invention and simplification of the description, rather than indicates or imply
Signified device or element must have a particular orientation, with specific azimuth configuration and operation, therefore should not be understood as to this
The limitation of invention.Unless otherwise clearly defined and limited, term " installation ", " connected ", " connection " shall be understood in a broad sense, example
Such as, it may be fixed connection or may be dismantle connection, or integral connection；It can be mechanical connection, can also be to be electrically connected
It connects；It can be directly connected, can also can be indirectly connected through an intermediary the connection inside two elements.For this
For the those of ordinary skill in field, the specific meanings of the above terms in the present invention can be understood according to specific conditions.
It should also be noted that, herein, relational terms such as first and second and the like are used merely to one
Entity or operation are distinguished with another entity or operation, without necessarily requiring or implying between these entities or operation
There are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant are intended to contain
Lid non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also include other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.
Above example is only used to illustrate the technical scheme of the present invention, rather than its limitations；Although with reference to the foregoing embodiments
Invention is explained in detail, it will be understood by those of ordinary skill in the art that：It still can be to aforementioned each implementation
Technical solution recorded in example is modified or equivalent replacement of some of the technical features；And these are changed or replace
It changes, the spirit and scope for various embodiments of the present invention technical solution that it does not separate the essence of the corresponding technical solution.
Claims (10)
1. a kind of Lightning Warning method, which is characterized in that the Lightning Warning method includes：
Obtain the atmospheric electric field intensity data information and lighting location data information of target area；
The temporal characteristics of the target area are determined according to the atmospheric electric field intensity data information and lighting location data information
Data, and determine according to the lighting location data information space characteristics data of the target area；
And according to the temporal characteristics data, space characteristics data and the preset length with convolutional layer memory network in short-term
LSTM models determine the thunder and lightning prediction result of the target area, and carry out Lightning Warning according to the thunder and lightning prediction result.
2. Lightning Warning method according to claim 1, which is characterized in that described according to the temporal characteristics data, sky
Between characteristic and the preset length with convolutional layer memory network LSTM models in short-term, determine that the thunder and lightning of the target area is pre-
Survey as a result, and according to the thunder and lightning prediction result carry out Lightning Warning, including：
The temporal characteristics data and space characteristics data are formed into data set, and by the data set according to default ratio graduation
For training set and test set；
Data training is carried out to the preset shot and long term memory network LSTM models with convolutional layer using the training set, is obtained
Target weight parameter for showing thunder and lightning prediction result；
And recruitment evaluation is carried out to the target weight parameter according to the training set, if the result symbol of the recruitment evaluation
Preset requirement is closed, then Lightning Warning is carried out according to the target weight parameter.
3. Lightning Warning method according to claim 1, which is characterized in that the atmospheric electricity field strength for obtaining target area
Degrees of data information and lighting location data information, including：
The atmospheric electric field intensity data information that atmospheric electric field detector in the target area acquires the target area, and root are set
The change curve of atmospheric electric field intensity is drawn according to the atmospheric electric field intensity data information；
And the lighting location data information that the Lighting position machine in the target area acquires the target area is set,
In, the lighting location data information includes：When the location information and thunder and lightning of the temporal information, thunder and lightning generation that thunder and lightning occurs occur
Elevation information apart from ground.
4. Lightning Warning method according to claim 3, which is characterized in that described according to the atmospheric electric field intensity data
Information and lighting location data information determine the temporal characteristics data of the target area, including：
According to the change curve of the atmospheric electric field intensity, obtained respectively for indicating thunderstorm cloud cluster close to or away from the air
The curve polarity smooth features data A1 of electric field instrument, the curve shake of the severe degree for indicating thunderstorm cloud cluster energy variation are special
Levy data A2, the curve steepness characteristic A3 for indicating thunderstorm cloud cluster fuel deposition rate and for indicating thunderstorm cloud cluster energy
Measure the curve peak characteristic A4 of size；
And according in the lighting location data information thunder and lightning occur temporal information, thunder and lightning occur location information and
Elevation information apart from ground when thunder and lightning occurs, it is the center of circle and using pre-determined distance as radius to be obtained respectively using the Lighting position machine
In the range of thunder and lightning total quantity characteristic A5, the target area thunder and lightning cloud dodge with ground dodge ratio characteristic data A6,
And the closing speed A7 of the thunderstorm cloud cluster；
Wherein, the curve polarity smooth features data A1, curve jitter feature data A2, curve steepness characteristic A3, song
Line peak characteristic A4, thunder and lightning total quantity characteristic A5, thunder and lightning cloud dodge the ratio characteristic data A6 and thunderstorm dodged with ground
The closing speed A7 of cloud cluster collectively constitutes the temporal characteristics data of the target area.
5. Lightning Warning method according to claim 3, which is characterized in that described according to the lighting location data information
Determine the space characteristics data of the target area, including：
It is recommended that the network of a N*N, and each cell in the network represents 1 square kilometre；
And thunder and lightning whereabouts time numerical value is marked in corresponding each unit lattice according to the lighting location data information, obtain N*N
Matrix, the N*N matrixes are the space characteristics data of the target area；
Wherein, N is the positive integer more than 1.
6. Lightning Warning method according to claim 2, which is characterized in that described by the temporal characteristics data and space
Characteristic composition data collection, and by the data set according to default ratio graduation be training set and test set, including：
The temporal characteristics data are formed into a row vector a1；
Space characteristics data input convolutional layer is subjected to process of convolution, the space characteristics data after process of convolution are converted to
One-dimensional row vector a2；
Combine the row vector a1 and the one-dimensional row vector a2, the data set a3；
And by the data set a3 according to default ratio graduation be training set and test set.
7. Lightning Warning method according to claim 2, which is characterized in that the application training set is to preset band
There are the shot and long term memory network LSTM models of convolutional layer to carry out data training, obtains the target for showing thunder and lightning prediction result and weigh
Weight parameter, including：
Data training, and institute are carried out to the preset shot and long term memory network LSTM models with convolutional layer using the training set
It is to intersect entropy function to state the corresponding loss function of shot and long term memory network LSTM models with convolutional layer；
And the intersection entropy function is iterated using gradient descent algorithm, until obtaining for showing thunder and lightning prediction knot
The target weight parameter of fruit.
8. a kind of Lightning Warning system, which is characterized in that the Lightning Warning system includes：
Data acquisition module, the atmospheric electric field intensity data information for obtaining target area and lighting location data information；
Characteristic extracting module, for determining the mesh according to the atmospheric electric field intensity data information and lighting location data information
The temporal characteristics data in region are marked, and determine the space characteristics number of the target area according to the lighting location data information
According to；
Lightning Warning module, for according to the temporal characteristics data, space characteristics data and the preset length with convolutional layer
Short-term memory network LSTM models determine the thunder and lightning prediction result of the target area, and are carried out according to the thunder and lightning prediction result
Lightning Warning.
9. a kind of electronic equipment, including memory, processor and storage are on a memory and the calculating that can run on a processor
Machine program, which is characterized in that the processor realizes that the thunder and lightning as described in any one of claim 1 to 7 is pre- when executing described program
The step of alarm method.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program quilt
It is realized when processor executes as described in any one of claim 1 to 7 the step of Lightning Warning method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810208869.3A CN108427041B (en) | 2018-03-14 | 2018-03-14 | Lightning early warning method, system, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810208869.3A CN108427041B (en) | 2018-03-14 | 2018-03-14 | Lightning early warning method, system, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108427041A true CN108427041A (en) | 2018-08-21 |
| CN108427041B CN108427041B (en) | 2020-03-17 |
Family
ID=63158480
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810208869.3A Active CN108427041B (en) | 2018-03-14 | 2018-03-14 | Lightning early warning method, system, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108427041B (en) |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109444990A (en) * | 2019-01-13 | 2019-03-08 | 卜俊伟 | A kind of Lightning Warning method based on weather satellite data |
| CN109599852A (en) * | 2018-12-05 | 2019-04-09 | 北京雷布斯雷电安全科技有限公司 | Lightning-protection system |
| CN109961192A (en) * | 2019-04-03 | 2019-07-02 | 南京中科九章信息技术有限公司 | Object event prediction technique and device |
| CN110618474A (en) * | 2019-10-24 | 2019-12-27 | 广东省气象公共安全技术支持中心 | Lightning monitoring and early warning method and system based on multi-source data |
| CN110806509A (en) * | 2019-11-29 | 2020-02-18 | 广州供电局有限公司 | Lightning activity spatial feature detection method and device |
| CN111175852A (en) * | 2019-12-27 | 2020-05-19 | 中国电子科技集团公司第十四研究所 | Airport fog forecast early warning method based on long-time memory algorithm |
| CN111242374A (en) * | 2020-01-10 | 2020-06-05 | 上海眼控科技股份有限公司 | Lightning prediction method, device, computer equipment and computer readable storage medium |
| CN112017090A (en) * | 2020-08-14 | 2020-12-01 | 广东电网有限责任公司广州供电局 | Lightning activity intensity characterization method and device, computer equipment and storage medium |
| CN112085731A (en) * | 2020-09-18 | 2020-12-15 | 深圳市易图资讯股份有限公司 | Security early warning method, device and equipment based on satellite map and storage medium |
| CN112329346A (en) * | 2020-11-06 | 2021-02-05 | 国网四川省电力公司泸州供电公司 | Analysis and optimization method for lightning ground flashover data of power transmission line |
| CN112668790A (en) * | 2020-12-30 | 2021-04-16 | 南京信息工程大学 | Thunder and lightning prediction method based on space-time sequence clustering algorithm and LSTM neural network |
| WO2021077729A1 (en) * | 2019-10-23 | 2021-04-29 | 国网电力科学研究院武汉南瑞有限责任公司 | Lightning prediction method |
| CN113219259A (en) * | 2021-04-26 | 2021-08-06 | 吉林省气候中心 | Lightning early warning method, device, equipment and storage medium |
| CN113295935A (en) * | 2021-06-01 | 2021-08-24 | 兰州资源环境职业技术学院 | Lightning stroke risk assessment method based on high-precision lightning positioning technology |
| CN113408803A (en) * | 2021-06-24 | 2021-09-17 | 国网浙江省电力有限公司双创中心 | Thunder and lightning prediction method, device, equipment and computer readable storage medium |
| CN113406726A (en) * | 2021-06-01 | 2021-09-17 | 中国石油大学(北京) | Oil and gas station lightning accident early warning method, device, equipment and storage medium |
| CN113533835A (en) * | 2020-04-20 | 2021-10-22 | 中国石油化工股份有限公司 | Thunder early warning system and method based on thunder and electric field detection technology |
| CN113533834A (en) * | 2020-04-20 | 2021-10-22 | 中国石油化工股份有限公司 | Lightning early warning system and method based on lightning positioning and electric field detection technology |
| CN113986999A (en) * | 2021-09-13 | 2022-01-28 | 广东电网有限责任公司广州供电局 | Thunder and lightning early warning method, early warning device, electronic equipment and computer storage medium |
| CN113985145A (en) * | 2021-09-13 | 2022-01-28 | 广东电网有限责任公司广州供电局 | Thunder and lightning early warning method, early warning device and computer readable storage medium |
| CN114252706A (en) * | 2021-12-15 | 2022-03-29 | 华中科技大学 | Lightning early warning method and system |
| CN114994801A (en) * | 2022-08-05 | 2022-09-02 | 中国气象局公共气象服务中心(国家预警信息发布中心) | Lightning monitoring and early warning method and device |
| CN115456248A (en) * | 2022-08-15 | 2022-12-09 | 国网电力科学研究院武汉南瑞有限责任公司 | Thunderstorm prediction model construction method based on convolutional neural network |
| CN115859222A (en) * | 2022-12-28 | 2023-03-28 | 四川省气象灾害防御技术中心 | Lightning automatic prediction method based on multi-gradient feature fusion network |
| CN115993488A (en) * | 2023-03-24 | 2023-04-21 | 天津安力信通讯科技有限公司 | Intelligent monitoring method and system for electromagnetic environment |
| CN116910491A (en) * | 2023-09-11 | 2023-10-20 | 四川弘和数智集团有限公司 | Lightning monitoring and early warning system and method, electronic equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3737463B2 (en) * | 2002-08-06 | 2006-01-18 | 財団法人日本気象協会 | Lightning strike prediction method |
| JP4173701B2 (en) * | 2002-07-31 | 2008-10-29 | 関西電力株式会社 | Winter lightning lightning forecasting method and forecasting device |
| CN102156739A (en) * | 2011-04-15 | 2011-08-17 | 南京信息工程大学 | GIS (Geographic Information System) platform processing method for mass lightning data |
| CN105004932A (en) * | 2015-07-17 | 2015-10-28 | 云南电力试验研究院(集团)有限公司 | Thunder and lightning early warning data correction method based on real-time thunder and lighting positioning data correlation analysis |
| CN105279565A (en) * | 2014-05-27 | 2016-01-27 | 北京中科九章软件有限公司 | Lightning early warning method and lightning early warning system |
| CN106353605A (en) * | 2016-11-02 | 2017-01-25 | 安徽锦坤电子科技有限公司 | Pre-warning system used for lightening |
-
2018
- 2018-03-14 CN CN201810208869.3A patent/CN108427041B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4173701B2 (en) * | 2002-07-31 | 2008-10-29 | 関西電力株式会社 | Winter lightning lightning forecasting method and forecasting device |
| JP3737463B2 (en) * | 2002-08-06 | 2006-01-18 | 財団法人日本気象協会 | Lightning strike prediction method |
| CN102156739A (en) * | 2011-04-15 | 2011-08-17 | 南京信息工程大学 | GIS (Geographic Information System) platform processing method for mass lightning data |
| CN105279565A (en) * | 2014-05-27 | 2016-01-27 | 北京中科九章软件有限公司 | Lightning early warning method and lightning early warning system |
| CN105004932A (en) * | 2015-07-17 | 2015-10-28 | 云南电力试验研究院(集团)有限公司 | Thunder and lightning early warning data correction method based on real-time thunder and lighting positioning data correlation analysis |
| CN106353605A (en) * | 2016-11-02 | 2017-01-25 | 安徽锦坤电子科技有限公司 | Pre-warning system used for lightening |
Cited By (35)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109599852A (en) * | 2018-12-05 | 2019-04-09 | 北京雷布斯雷电安全科技有限公司 | Lightning-protection system |
| CN109444990A (en) * | 2019-01-13 | 2019-03-08 | 卜俊伟 | A kind of Lightning Warning method based on weather satellite data |
| CN109961192A (en) * | 2019-04-03 | 2019-07-02 | 南京中科九章信息技术有限公司 | Object event prediction technique and device |
| CN109961192B (en) * | 2019-04-03 | 2021-11-26 | 南京中科九章信息技术有限公司 | Target event prediction method and device |
| WO2021077729A1 (en) * | 2019-10-23 | 2021-04-29 | 国网电力科学研究院武汉南瑞有限责任公司 | Lightning prediction method |
| CN110618474A (en) * | 2019-10-24 | 2019-12-27 | 广东省气象公共安全技术支持中心 | Lightning monitoring and early warning method and system based on multi-source data |
| CN110806509B (en) * | 2019-11-29 | 2021-12-17 | 广东电网有限责任公司广州供电局 | Lightning activity spatial feature detection method and device |
| CN110806509A (en) * | 2019-11-29 | 2020-02-18 | 广州供电局有限公司 | Lightning activity spatial feature detection method and device |
| CN111175852A (en) * | 2019-12-27 | 2020-05-19 | 中国电子科技集团公司第十四研究所 | Airport fog forecast early warning method based on long-time memory algorithm |
| CN111242374A (en) * | 2020-01-10 | 2020-06-05 | 上海眼控科技股份有限公司 | Lightning prediction method, device, computer equipment and computer readable storage medium |
| CN113533835A (en) * | 2020-04-20 | 2021-10-22 | 中国石油化工股份有限公司 | Thunder early warning system and method based on thunder and electric field detection technology |
| CN113533834A (en) * | 2020-04-20 | 2021-10-22 | 中国石油化工股份有限公司 | Lightning early warning system and method based on lightning positioning and electric field detection technology |
| CN112017090A (en) * | 2020-08-14 | 2020-12-01 | 广东电网有限责任公司广州供电局 | Lightning activity intensity characterization method and device, computer equipment and storage medium |
| CN112085731A (en) * | 2020-09-18 | 2020-12-15 | 深圳市易图资讯股份有限公司 | Security early warning method, device and equipment based on satellite map and storage medium |
| CN112329346B (en) * | 2020-11-06 | 2022-06-14 | 国网四川省电力公司泸州供电公司 | Analysis and optimization method for lightning ground flashover data of power transmission line |
| CN112329346A (en) * | 2020-11-06 | 2021-02-05 | 国网四川省电力公司泸州供电公司 | Analysis and optimization method for lightning ground flashover data of power transmission line |
| CN112668790A (en) * | 2020-12-30 | 2021-04-16 | 南京信息工程大学 | Thunder and lightning prediction method based on space-time sequence clustering algorithm and LSTM neural network |
| CN112668790B (en) * | 2020-12-30 | 2023-07-25 | 南京信息工程大学 | Lightning prediction method based on space-time sequence clustering algorithm and LSTM neural network |
| CN113219259A (en) * | 2021-04-26 | 2021-08-06 | 吉林省气候中心 | Lightning early warning method, device, equipment and storage medium |
| CN113295935A (en) * | 2021-06-01 | 2021-08-24 | 兰州资源环境职业技术学院 | Lightning stroke risk assessment method based on high-precision lightning positioning technology |
| CN113406726B (en) * | 2021-06-01 | 2022-09-27 | 中国石油大学(北京) | Oil and gas station lightning accident early warning method, device, equipment and storage medium |
| CN113406726A (en) * | 2021-06-01 | 2021-09-17 | 中国石油大学(北京) | Oil and gas station lightning accident early warning method, device, equipment and storage medium |
| CN113408803A (en) * | 2021-06-24 | 2021-09-17 | 国网浙江省电力有限公司双创中心 | Thunder and lightning prediction method, device, equipment and computer readable storage medium |
| CN113986999A (en) * | 2021-09-13 | 2022-01-28 | 广东电网有限责任公司广州供电局 | Thunder and lightning early warning method, early warning device, electronic equipment and computer storage medium |
| CN113985145A (en) * | 2021-09-13 | 2022-01-28 | 广东电网有限责任公司广州供电局 | Thunder and lightning early warning method, early warning device and computer readable storage medium |
| CN113986999B (en) * | 2021-09-13 | 2024-09-06 | 广东电网有限责任公司广州供电局 | Lightning early warning method, early warning device, electronic equipment and computer storage medium |
| CN114252706B (en) * | 2021-12-15 | 2023-03-14 | 华中科技大学 | Lightning early warning method and system |
| CN114252706A (en) * | 2021-12-15 | 2022-03-29 | 华中科技大学 | Lightning early warning method and system |
| CN114994801A (en) * | 2022-08-05 | 2022-09-02 | 中国气象局公共气象服务中心(国家预警信息发布中心) | Lightning monitoring and early warning method and device |
| CN114994801B (en) * | 2022-08-05 | 2022-10-25 | 中国气象局公共气象服务中心(国家预警信息发布中心) | Lightning monitoring and early warning method and device |
| CN115456248A (en) * | 2022-08-15 | 2022-12-09 | 国网电力科学研究院武汉南瑞有限责任公司 | Thunderstorm prediction model construction method based on convolutional neural network |
| CN115859222A (en) * | 2022-12-28 | 2023-03-28 | 四川省气象灾害防御技术中心 | Lightning automatic prediction method based on multi-gradient feature fusion network |
| CN115993488A (en) * | 2023-03-24 | 2023-04-21 | 天津安力信通讯科技有限公司 | Intelligent monitoring method and system for electromagnetic environment |
| CN116910491A (en) * | 2023-09-11 | 2023-10-20 | 四川弘和数智集团有限公司 | Lightning monitoring and early warning system and method, electronic equipment and storage medium |
| CN116910491B (en) * | 2023-09-11 | 2024-01-23 | 四川弘和数智集团有限公司 | Lightning monitoring and early warning system and method, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108427041B (en) | 2020-03-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108427041A (en) | Lightning Warning method, system, electronic equipment and storage medium | |
| Zhou et al. | Forecasting different types of convective weather: A deep learning approach | |
| CN108957595A (en) | A kind of lightning forecasting method and system | |
| CN113327022B (en) | Lightning protection safety risk management system and method | |
| Saxena et al. | A review study of weather forecasting using artificial neural network approach | |
| CN108304536A (en) | A kind of geographical environmental simulation of the geographical environmental element of coupling and predicting platform | |
| Gao et al. | Heuristic failure prediction model of transmission line under natural disasters | |
| Li et al. | Multivariable time series prediction for the icing process on overhead power transmission line | |
| Zhao et al. | Extracting and classifying typhoon disaster information based on volunteered geographic information from Chinese Sina microblog | |
| Vega-Oliveros et al. | From spatio-temporal data to chronological networks: An application to wildfire analysis | |
| Yao et al. | A new condition-monitoring method based on multi-variable correlation learning network for wind turbine fault detection | |
| Baudhanwala et al. | Machine learning approaches for improving precipitation forecasting in the Ambica River basin of Navsari District, Gujarat | |
| Šaur | Evaluation of the accuracy of numerical weather prediction models | |
| CN109410527B (en) | Space weather disaster monitoring and early warning method, system, storage medium and server | |
| Alves et al. | Lightning Warning Prediction with Multi-source Data | |
| Khan et al. | A factual flash flood evaluation using SVM and K-NN | |
| CN112162336A (en) | Visibility prediction method and device based on two-dimensional meteorological element field | |
| Mpanga et al. | Spatial and temporal variation of Zambia lightning for designing lightning protection of infrastructure | |
| Wu et al. | Forecast of thunderstorm cloud trend based on monitoring data of thunder mobile positioning system | |
| Goymann et al. | Flood Prediction through Artificial Neural Networks | |
| CN114545098B (en) | Thunderstorm forecasting method and lightning positioning method | |
| Singh et al. | FlashBench: A lightning nowcasting framework based on the hybrid deep learning and physics-based dynamical models | |
| CN117434624B (en) | Strong convection weather identification and development prejudgment method based on semantic segmentation | |
| Sangpradid et al. | Estimates of PM2. 5 Concentration Based on Aerosol Optical Thickness Data Using Ensemble Learning with Support Vector Machine and Decision Tree | |
| Luo et al. | A Multiscale Attention Network for the Classification of Lightning Safety Risk Warnings |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230113 Address after: Room 401, Floor 4, Xingfa Building, No. 45, Zhongguancun Street, Haidian District, Beijing, 100089 Patentee after: BEIJING JOZZON CAS SOFTWARE CO.,LTD. Address before: Room 501, building 2, Shuanglong science and Technology Industrial Park, No. 2, Shuanglong street, Qinhuai District, Nanjing, Jiangsu 210006 Patentee before: NANJING ZHONGKE JIUZHANG INFORMATION TECHNOLOGY Co.,Ltd. |