JP3150302B2 - Lightning prediction device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a lightning prediction device for predicting the occurrence of lightning (hereinafter referred to as lightning) using observation values such as data on thunderclouds obtained from weather sensors.

[0002]

2. Description of the Related Art FIG. 30 shows, for example, "ground lightning detector"
(Koichi Hirashima, Go Mio, Hisashi Ichiyama, Tsutomu Sankono, IEICE Technical Report, SA
It is a block diagram which shows the conventional lightning prediction apparatus shown by NE97-87). The lightning prediction device shown in FIG. 30 includes a weather sensor 1 that observes and outputs a lightning situation, a radar echo analyzer 2 that outputs echo intensity data and a three-dimensional echo height distribution with the lightning situation as an input, An image memory 3 for storing intensity data for N times, an echo intensity image display 4 for inputting and displaying the N times of echo intensity data, and a source for inputting the three-dimensional echo intensity distribution and calculating and outputting a lightning determination element. A lightning judgment element calculator 5, a lightning risk calculator 6 that calculates and outputs a lightning risk with the lightning judgment element as an input, a lightning risk calculator 7 that displays an image of the lightning risk, A weather condition data file 8 for storing lightning judgment elements, a graph display 9 for inputting the lightning judgment element data stored in the weather condition data file 8 and outputting data for the past N times, And a lightning predictor 10 to output a lightning judgment result data of the echo intensity and the lightning risk and the past N times as inputs.

Next, the operation will be described. Weather sensor 1
Are, for example, a weather radar for observing the echo of a thundercloud, a static direction finder for observing the position of the lightning where the lightning occurred, or an electric field meter for measuring the electric field strength. Observes and outputs lightning conditions such as echo, lightning location, and electric field strength.

[0004] The radar echo analyzer 2 receives the above lightning situation as input, performs smoothing processing of echo data and the like, and then performs
It performs a process of extracting radar echoes at a constant altitude called Constant Altitude PPI (Constant Altitude PPI, CAPPI) and a process of converting polar coordinate data of distance and azimuth obtained by radar echoes into rectangular coordinates, and divides one section into one. Converted into a three-dimensional echo height distribution represented by the echo intensity value in each section divided into 200 × 200 sections and 18 sections in the altitude direction on a horizontal plane equivalent to the ground, with a cube having a side of 1 km, Output to the element calculator 5. Further, from the three-dimensional echo intensity distribution data, the maximum value of the echo heights of 18 sections in the altitude direction is calculated for each of 200 × 200 sections on the horizontal plane, and the two-dimensional echo intensity distribution data is calculated. The distribution data is output to the image memory 3 as echo intensity data.

[0005] An image memory 3 receives and stores echo intensity data from the radar echo analyzer 2. This image memory 3 stores the echo intensity data from the newest to N
It can record the number of sheets. The echo intensity image display 4 receives the echo intensity data stored by the image memory 3 and displays the data on the screen.

The lightning determining element calculator 5 receives the three-dimensional echo intensity distribution data as input, calculates and outputs a lightning determining element of a thundercloud corresponding to each of the 200 × 200 plane sections. Here, the lightning judgment element is data that serves as a guideline for judging lightning, such as the result of judging the thickness of the thunder cloud in the height direction, the development or weakness of the echo intensity, or the number of polarity changes of each electric field meter. It is.

The lightning risk calculator 6 receives the lightning determination element as input, calculates lightning risk data, and outputs the data. Here, the lightning risk is obtained by comprehensively quantifying the possibility of lightning based on the above-mentioned lightning determination factor. For example, if the thickness of the thundercloud in the height direction is high, or if the echo intensity is developed, or if the number of polarities is large, the risk of lightning is high. These lightning determination elements are calculated for each 200 × 200 plane section. These lightning risk levels are collectively output as lightning risk data.

The lightning risk image display 7 inputs and displays the lightning risk data. The current weather data file 8 receives the above-mentioned lightning judgment element and lightning risk data,
It is stored as two-dimensional data defined for each 200 × 200 plane section. The weather status data file 8 always stores at least N past data of the two-dimensional database calculated by the weather sensor 1.

The graph display 9 receives the above-mentioned lightning judgment element data as input and displays the data of the past N times on a screen in time series. The lightning prediction judging unit 10 judges lightning based on the output from the echo intensity image display 4, the output from the lightning risk image display 7, and the output from the graph display 9. And outputs the lightning judgment result.

[0010]

As described above, in the conventional lightning prediction device, the purpose is to determine the lightning based on the current situation. Could not do. In addition, there is a problem that data is insufficient to perform empirical prediction.

SUMMARY OF THE INVENTION The present invention has been made to solve such a problem, and a past weather case similar to the current weather situation is searched for and used for prediction, thereby achieving higher accuracy than before. An object is to obtain a lightning prediction device capable of predicting lightning.

[0012]

A lightning prediction device according to the present invention includes a weather sensor for observing and outputting a lightning situation, echo intensity data and three-dimensional echo intensity based on the lightning situation from the weather sensor. A radar echo analyzer that outputs distribution data, an image memory that stores echo intensity data from the radar echo analyzer, an echo intensity image display that inputs and displays the echo intensity data stored by the image memory, A lightning judgment element calculator for calculating and outputting a lightning judgment element based on the three-dimensional echo height distribution data from the radar echo analyzer, and a lightning judgment based on the lightning judgment element from the lightning judgment element calculator A lightning risk calculator that calculates and outputs risk data, and a lightning risk image display that inputs and displays lightning risk data from the lightning risk calculator And, input the current weather data file storing the lightning judgment element from the lightning judgment element calculator and the lightning judgment element data stored in the weather condition data file and output the past lightning judgment element data In the lightning prediction device for predicting and determining lightning, the output of the echo intensity image display, the output of the lightning risk image display, and stored in the weather status data file are provided. A past case search unit that outputs a past weather condition similar to the current lightning judgment element based on the lightning judgment element that is being output, the output of the echo intensity image display, and the lightning risk image display And a past case considering lightning prediction judging unit that outputs a lightning prediction result in consideration of a past case based on the past weather condition from the past case search unit and the output of the graph display unit. , Serial past case search unit
Is the output of the echo intensity image display or the
Perform a search on the image based on the output of the
Image / position selector to select the specified position and specified time
And the specified position and specified time from the image / position selector
As input, search target for position and search for time
A search condition determiner that determines a search target as a search condition;
Meteorological cases that store past case data as elephant case data
Data file and search conditions from the above search condition determination unit
And the lightning strikes stored in the weather status data file
Judgment element data and stored in the above weather case data file
The past case data entered as input
Targets for calculating similarity from past case data
The composed data is extracted, the as an off-similarity score calculation target data
Similarity calculation target data selection to be output together with the lightning judgment element
Selector and similarity calculation from the above similarity calculation data selector
The similarity between the target data and the
Similarity calculation to calculate and output for each similarity calculation target data
Device and calculation target data with similarity from the similarity calculator
Is processed by similarity of data to create past case search results
Search result creator and output past weather image data
Meteorological image data file to be stored and search result creation
Past case search results from the
Based on past case search results based on weather images
Images are compared with past weather conditions similar to the current lightning judgment factor.
And a search result display for displaying and outputting .

[0013]

Further, a weather sensor for observing and outputting a lightning situation.
And echo based on the lightning situation from the weather sensor
Outputs intensity data and three-dimensional echo intensity distribution data
Radar echo analyzer and the radar echo analyzer
Image memory for storing echo intensity data from the
Enter the echo intensity data stored by the image memory.
Echo intensity image display for force display and the above radar eco
-Based on 3D echo height distribution data from the analyzer
A lightning judgment element calculator for calculating and outputting a lightning judgment element by
Based on the lightning judgment element from the above lightning judgment element calculator
A lightning risk calculator that calculates and outputs lightning risk data;
Enter the lightning risk data from the lightning risk calculator
The lightning risk image display to be displayed and the calculation
Meteorological situation data file that stores the lightning
And the lightning strikes stored in the above weather data file
Input judgment element data and output past lightning judgment element data
Lightning to predict and judge lightning
In the prediction device, the output of the echo intensity image display
And the output of the lightning risk image display and the weather conditions
Based on the lightning judgment element stored in the data file
To report past weather conditions similar to the current
Past case search unit and the echo intensity image display
Output, the output of the lightning risk image display, and the past
Past weather conditions from the case search section and the above graph display
Based on the output of
And a past case considering lightning prediction judging unit to be output, wherein the past case search unit should perform a search on an image based on the output of the echo intensity image display or the output of the lightning risk image display. Image to select specified position and specified time
A position selector, a designated position and designated time from the image / position selector as input, a search condition determiner for determining a search target for the position and a search target for the time as search conditions, and similarity target data not extracted And a single search condition extractor that extracts a single search condition from the remaining condition, a search condition from the search condition determiner, and a search condition stored in the weather condition data file. Sequential searcher that executes the sequential processing of the past case search based on the detected lightning determination element data and the single search condition from the single search condition extractor and outputs the search result and the remaining condition And a search result based on the weather image data file storing the past weather image data, the search result from the search result creator and the weather image data of the weather image data file A search result display device for displaying and outputting a corresponding image as a past weather condition similar to the current lightning determination element, and the sequential processing search device stores the past case data as weather case data. A case data file, a single search condition from the single search condition extractor, a lightning determination element data stored in the weather status data file, and a past lightning determination data stored in the weather case data file. Similarity calculation which takes case data as input, extracts data for which similarity should be calculated from the past case data based on the single search condition, and outputs the similarity calculation target data together with the lightning determination element The similarity between the target data selector and the similarity calculation target data from the similarity calculation data selector and the lightning determination element data is calculated for each similarity calculation target. A similarity calculator that calculates and outputs past case data with similarity and a search condition for each data, a search result file that stores search results output to the search result display, and the similarity calculator. A sequential search result creator that outputs a new search result created by modifying a search result based on past case data with similarity and the search result from the search result file, and a search from the search condition determiner It is determined whether or not the remaining condition remains based on the condition, the single search condition from the single search condition extractor, and the search result from the sequential search result creator. Are provided with a search termination determiner that outputs the remaining condition to the single search condition extractor, and outputs the search result to the search result display when there is no remaining condition.

In addition, a meteorological sensor for observing and outputting a lightning situation.
And echo based on the lightning situation from the weather sensor
Outputs intensity data and three-dimensional echo intensity distribution data
Radar echo analyzer and the radar echo analyzer
Image memory for storing echo intensity data from the
Enter the echo intensity data stored by the image memory.
Echo intensity image display for force display and the above radar eco
-Based on 3D echo height distribution data from the analyzer
A lightning judgment element calculator for calculating and outputting a lightning judgment element by
Based on the lightning judgment element from the above lightning judgment element calculator
A lightning risk calculator that calculates and outputs lightning risk data;
Enter the lightning risk data from the lightning risk calculator
The lightning risk image display to be displayed and the calculation
Meteorological situation data file that stores the lightning
And the lightning strikes stored in the above weather data file
Input judgment element data and output past lightning judgment element data
Lightning to predict and judge lightning
In the prediction device, the output of the echo intensity image display
And the output of the lightning risk image display and the weather conditions
Based on the lightning judgment element stored in the data file
To report past weather conditions similar to the current
Past case search unit and the echo intensity image display
Output, the output of the lightning risk image display, and the past
Past weather conditions from the case search section and the above graph display
Based on the output of
And a past case considering lightning prediction judging unit to be output, wherein the past case search unit should perform a search on an image based on the output of the echo intensity image display or the output of the lightning risk image display. Image to select specified position and specified time
A position selector, a search condition determiner that receives a designated position and a designated time from the image / position selector as input, and determines a search target for the position and a search target for the time as search conditions, and past case data as weather case data The weather case data file storing the search conditions from the search condition determiner, the lightning judgment element data stored in the weather status data file, and the past case data stored in the weather case data file A parallel processing searcher that processes and outputs search results for each search condition in parallel based on each search condition; a search result integrator that integrates and outputs search results for each search condition from the parallel processing searcher; A weather image data file storing image data, a search result from the search result integrator, and a weather image data of the weather image data file. And a search result display for displaying and outputting an image corresponding to the search result as a past weather situation similar to the current lightning determination element based on the search result, and the parallel processing searcher determines the search condition. The search condition from the section, the lightning judgment element data stored in the weather status data file, and the past case data stored in the weather case data file are input. A similarity calculation target data selector that extracts data for which similarity is to be calculated from the data and outputs the similarity calculation target data together with the lightning determination element; and a similarity calculation from the similarity calculation data selector. A similarity calculator for calculating and outputting the similarity between the target data and the lightning determination element data for each similarity calculation target data, and a class from the similarity calculator. A plurality of search result creators that process data to be calculated with similarity of data and generate and output past case search results are arranged, and the search results for each search condition are processed in parallel and output. It is.

Further, the search condition deciding device is configured to execute
A designated coordinate search condition determiner that inputs a designated position and a designated time from a position selector and outputs a specific position corresponding to the designated position as one of search conditions, wherein the similarity calculation target data selector is The search condition from the designated coordinate search condition determination unit, the lightning judgment element data stored in the weather status data file, and the past case data stored in the weather case data file are input, and the search is performed. According to a condition, data for which similarity is to be calculated is extracted from past case data and output as similarity calculation target data together with the lightning determination element.

Further, the search condition deciding device is adapted to execute
A search condition candidate determiner that inputs a specified position and a specified time from the position selector and outputs search condition candidates, a specified and specified time from the image / position selector, and a search from the search condition candidate determiner A maximum feature search condition determiner that inputs a condition candidate and lightning risk data from the lightning risk image display and echo intensity data from the echo intensity image display and outputs a search condition, The similarity calculation target data selector is stored in the search condition from the maximum feature search condition determiner, the lightning determination element data stored in the weather status data file, and the weather case data file. Using past case data as input, extract data for which similarity should be calculated from the past case data based on the above search conditions, and use the above lightning as the similarity calculation target data. It is characterized in that the output with a constant element.

In addition, a meteorological sensor for observing and outputting a lightning situation.
And echo based on the lightning situation from the weather sensor
Outputs intensity data and three-dimensional echo intensity distribution data
Radar echo analyzer and the radar echo analyzer
Image memory for storing echo intensity data from the
Enter the echo intensity data stored by the image memory.
Echo intensity image display for force display and the above radar eco
-Based on 3D echo height distribution data from the analyzer
A lightning judgment element calculator for calculating and outputting a lightning judgment element by
Based on the lightning judgment element from the above lightning judgment element calculator
A lightning risk calculator that calculates and outputs lightning risk data;
Enter the lightning risk data from the lightning risk calculator
The lightning risk image display to be displayed and the calculation
Meteorological situation data file that stores the lightning
And the lightning strikes stored in the above weather data file
Input judgment element data and output past lightning judgment element data
Lightning to predict and judge lightning
In the prediction device, the output of the echo intensity image display
And the output of the lightning risk image display and the weather conditions
Based on the lightning judgment element stored in the data file
To report past weather conditions similar to the current
Past case search unit and the echo intensity image display
Output, the output of the lightning risk image display, and the past
Past weather conditions from the case search section and the above graph display
Based on the output of
And a past case considering lightning prediction judging unit to be output, wherein the past case search unit should perform a search on an image based on the output of the echo intensity image display or the output of the lightning risk image display. Image to select specified position and specified time
A position selector, a search condition determiner that receives a designated position and a designated time from the image / position selector as input, and determines a search target for the position and a search target for the time as search conditions, and past case data as weather case data , The search condition from the search condition determiner, the lightning judgment element data stored in the weather status data file, and the past case data stored in the weather case data file A similarity calculation target data selector for extracting data to be subjected to the similarity calculation from the past case data according to the search condition, and outputting the similarity calculation target data together with the lightning determination element. A search condition that inputs a search condition from the search condition determiner, weights the search condition, and outputs a search condition weight. A weighter, the similarity calculation target data from the similarity calculation target data selector and the search condition weight from the search condition weighter are input, and the similarity calculation target data and the lightning determination element data are input. A weighted similarity calculator that calculates and outputs weighted similarity for each element of each similarity calculation target data, and processes the similarity calculation target data from the weighted similarity calculator by data similarity A search result creator for creating and outputting past case search results, a weather image data file for storing past weather image data, a past case search result from the search result creator and a weather image of the weather image data file And a search result display for displaying and outputting an image corresponding to the past case search result as a past weather condition similar to the current lightning determination element based on the Is shall.

The similarity calculation target data selector receives the search condition and the lightning determination element data as input, extracts past case data at all positions corresponding to the search condition, and calculates similarity. It is characterized by being a similarity calculation target all data extractor output as target data.

The search condition from the search condition determiner and the designated time from the image / position selector are input.
The apparatus further includes an extraction target time determiner that determines and outputs an extraction target time, wherein the similarity calculation target data selector includes a search condition from the search condition determiner and a source stored in the weather status data file. The lightning judgment element data, the past case data stored in the weather case data file, and the extraction target time from the extraction target time determiner are input, and all the data having the extraction target time are similar from the past case data. A similarity calculation target data selector for extracting the degree as a target to be calculated and outputting the similarity calculation target data together with the lightning determination element, and the similarity calculator calculates the similarity based on the time. The similarity between the similarity calculation target data based on the time from the calculation data selector and the lightning determination element data is calculated for each similarity calculation target data. It is characterized in that force.

The similarity calculation target data selector receives the search condition and the lightning determination element data as inputs and manually calculates the similarity while considering the search condition and the lightning determination element data. It is characterized by being a similarity calculation target data compulsory designator for selecting and outputting target data.

Further, the apparatus further comprises a maximum similarity calculator which receives a search result from the search result creator as input, and outputs a maximum value of similarity of the search case in the search result as a search case similarity. The result display is a search result display that uses the similarity of the search case as an input and displays the search result.

Further, the apparatus further comprises a minimum similarity calculator which receives a search result from the search result creator as input, and outputs a minimum value of similarity of the search case in the search result as a similarity of the search case. The result display device is characterized in that it is a search result display device that uses the minimum similarity to display the search result by inputting the similarity of the search case.

Also, a search condition weight file for storing search condition weights, a search result from the search result generator and a search condition weight stored in the search condition weight file are input, and the search condition weight is used. A search condition weight consideration similarity calculator for calculating and outputting the similarity of each search case, and wherein the search result display is configured to input the similarity of the search case as input and to consider search condition weights. Is displayed on the display screen.

Further, the past case data in advance which position further comprises a location index file to store the position index of whether the stored anywhere weather case data file, the similarity calculation target data selector, said A similarity calculation target data selector using a position index that inputs a position index file, the search condition, and the lightning determination element, and outputs similarity calculation target data.

[0026] In addition, pre-past case data any time, anywhere in further comprising a time index file to store the time index of whether it is stored, the similarity calculation target data selector of weather case data file, the above A similarity calculation target data selector using a time index for inputting a time index file, the search condition and the lightning determination element, and outputting similarity calculation target data.

[0027]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiment 1 FIG. 1 is a block diagram showing a lightning prediction device according to Embodiment 1 of the present invention. In FIG. 1, the same portions as those of the conventional example shown in FIG. 30 are denoted by the same reference numerals, and description thereof will be omitted. As a new code, reference numeral 11 denotes the output of the echo intensity image display 4, the output of the lightning risk degree image display 7, and the weather status data file 8.
The past case search unit 12 receives the output of the above as an input and outputs a past weather condition similar to the current lightning determination element, and outputs the output of the echo intensity image display 4 and the output of the lightning risk image display 7 And the output of the graph display 9 and the past case search unit 11
This is a past case-based lightning prediction judging unit that receives the output from the system as an input, and outputs a lightning prediction result in consideration of similar lightning situations that occurred in the past.

Next, the operation according to the embodiment of the present invention will be described. The weather sensor 1 observes and outputs a lightning situation. The radar echo analysis device 2 receives the above lightning situation as input, and outputs echo intensity data and three-dimensional echo height distribution. The image memory 3 stores the echo intensity data in N
Store for batches. The echo intensity image display 4 inputs and displays the echo intensity data for N times. The lightning determination element calculator 5 receives the three-dimensional echo intensity distribution and outputs a lightning determination element. The lightning risk calculator 6 receives the lightning determination element and outputs the lightning risk. The lightning risk image display 7 displays the lightning risk image. The weather status data file 8 stores the lightning determination element. The graph display 9 receives the lightning determination element data stored in the weather status data file as input, and outputs the past N
Outputs the lightning judgment element data for each time. These are known as described in the conventional example. The past case search unit 11 outputs the output of the echo intensity image display 4, the output of the lightning risk image display 7, and the weather status data file 8.
And outputs past weather conditions similar to the current lightning judgment element.

Here, the operation of the past case search section 11 will be described in more detail. FIG. 2 shows the past case search unit 1
1 is a block diagram showing a configuration of a past case search unit 11. The past case search unit 11 includes an image / position selector 13, a search condition determiner 14,
It comprises a similarity calculation target data selector 15, a weather case data file 16, a similarity calculator 17, a search result creator 18, a search result display 19, and a weather image data file 20.

The image / position selector 13 receives the output of the echo intensity image display 4 or the output of the lightning risk image display 7 and outputs a designated position and a designated time. The search condition determiner 14 inputs the specified position and the specified time, determines a search condition, and outputs the result. The similarity calculation target data selector 15 receives the above search conditions, the lightning determination element data stored in the weather case data file 8 and the past case data stored in the weather case data file 16 as inputs, and performs similarity calculation. The degree calculation target data and the above-mentioned lightning judgment element data are output. The similarity calculator 17 receives the similarity calculation target data and the lightning determination element data as inputs, calculates the similarity between the similarity calculation target data and the lightning determination element data, and outputs the similarity-added past case data. Is output. The search result creator 18 receives the past case data with similarity as input, creates and outputs a past case search result. The search result display 19 receives the past case search result and past weather image data stored in the weather image data file 20 and inputs an image corresponding to the past case search result as a current lightning determination element. Display and output as similar past weather conditions.

Next, the operation of the past case search section 11 will be described. The image / position selector 13 receives the output of the echo intensity image 4 or the output of the lightning risk image display 7 as an input and allows an operator or the like to output the echo intensity image or the lightning risk image. Select a search position on the image with reference to the output. Such a selection is made by, for example, clicking a corresponding position with a mouse provided in a computer or the like.

FIG. 3 shows how the image position selector 13 selects one of the images. As shown in FIG. 3, the selection result specifies the time (year / month / day / hour / minute) of the lightning risk image or the echo intensity image with the coordinates of the selected position as the specified position. Output as time. For example, in the case of FIG. 3, the image is an image at 15:10 on December 15, 1997, and the coordinates of the selected position are (value of X coordinate, value of Y coordinate) = (50, 150)
Therefore, 15:10 on December 15, 1997 is output as the specified time, and (50, 150) is output as the specified position.

Although an example of one image has been described here for simplicity, it is also possible to set a specified time and a specified position for a plurality of different images. In that case, the designated position and the designated time are output to the separate images in the same manner as described above. In this case, a plurality of designated positions and designated times are output, and designated positions and designated times corresponding to a maximum of N images can be set.

The search condition determiner 14 receives the specified position and the specified time as input, determines which data on the image or which data is to be searched, and
The search target for the position and the search target for the time are determined. The search target for the position is determined using the specified position, and the search target for the time is determined using the specified time. For example, in the case of FIG. 3, since the designated time is 15:10 on December 15, 1997, for example, if one month before and after this is to be searched for the time,
All times from November 15, 1997 to January 15, 1998 are output as search targets for time. Since the designated position is (50, 150), for example, X
In the case where the coordinates and the Y coordinate are to be searched for 10 positions before and after the coordinates, (40, 140) to (60, 16)
All the coordinates included between 0) are output as search targets for the position. The search target for these positions and the search target for time are output as search conditions.
If a plurality of designated positions and designated times are designated, the above operation is executed for all of the designated positions and designated times, and a plurality of search conditions are output.

The similarity calculation target data selector 15 receives the search condition, the lightning judgment element data stored in the weather status data file 8, and the past case data stored in the weather case data file 16. Is input from the past case data according to the search condition, and the data for which the similarity should be calculated is extracted and output as the similarity calculation target data together with the lightning determination element. Here, past case data is a record of past lightning judgment factors.
This is a collection of stored data.
It is assumed that it is composed of a position, a past lightning judgment element, and the like.

FIG. 4 shows an example of past case data. As described above, in the past case data, each row has one piece of data and each column has the content of specific data. For example, the data of the first row has a time of October 1, 1995, Is (50, 150), and the lightning determination element is composed of the thickness of the cloud and the number of polarities, which means that the values are 3 km and 2 times, respectively. For example, from the data of FIG.
When the similarity calculation target data is extracted based on the search condition, the seventh, eighth, twelfth, fourteenth, and eighteenth rows that match the search condition are obtained. For example, as shown in FIG. Is output together with the lightning determination element data as similarity calculation target data. If a plurality of search conditions are input, the above operation is performed for all of the plurality of search conditions, and a plurality of pairs of similarity calculation target data and lightning determination element data are output.

The similarity calculator 17 calculates the similarity between the similarity calculation target data and the lightning determination element data for each similarity calculation target data. The method will be described below. First, the similarity is calculated for each type of data included in the lightning determination element. For example, FIG.
In the case of, the lightning determination element calculates the similarity to the cloud thickness and the similarity to the number of polarity changes, respectively, based on the cloud thickness and the number of polarity changes. As the calculation formula, for example, a formula such as Expression 1 is used.

[0038]

(Equation 1)

As described above, after the similarity is obtained for each type of element, the similarity for each type is integrated by an equation such as Expression 2 to finally calculate the similarity to be used.

[0040]

(Equation 2)

Such total similarity calculation is performed for each data stored in the similarity calculation target data, and data obtained by adding the total similarity to the similarity calculation target data is used as similarity calculation target data. Output. FIG. 6 shows an example of past case data with similarity. In FIG. 6, for example, the data in the first row corresponds to the data in the first row in FIG. 5, and is obtained by adding the similarity 0.57 with the lightning determination element to the data in FIG.

The search result creator 18 processes each row of the data to be calculated with similarity according to the similarity of the data to create a search result. Figure 7 shows an example of search results
Shown in As shown in FIG. 7, the search result includes a search case in which a row indicates a similar case extracted as a result of the search, and a column in which data represents each case. In addition, as shown in the first row of FIG. 7, the data representing each search case is composed of, for example, a time, a search condition, a similarity, and a lightning determination element. Taking search example C (1) as an example, the data is the data at the position (40, 150) at 22:50 on December 27, 1995, and the search condition is 199.
At 15:00 on December 16, 2007, it means that the similarity with the search condition of the search case is 0.95.
When there are a plurality of search conditions, there may be a plurality of columns corresponding to the plurality of search conditions, as in search case C (2).

Next, a detailed operation of the search result generator 18 will be described. First, it is determined whether or not data at the same time is included in the similarity-attached calculation target data created corresponding to each search condition. If the data at the same time is included, only the similarity-attached calculation target data having the maximum similarity is left, and the subsequent data is deleted. After executing the processing, a search result is created for each row (hereinafter, referred to as current data) of the calculation data with similarity according to the flowchart of FIG. 8 as follows.

First, in step ST1, I = 1, and in step ST2, it is determined whether or not a search result already exists. If there is, the process proceeds to step ST3. If not, step ST4
And the process ends with the current data as the search case C (I).
If there is a search result, it is determined in step ST3 whether the current data can be time-series data with the search case C (I) data. This is performed as follows. The time of case C (I) is tC (I), the designated time is tC (I) cond, the time of the current data is t0, and the search condition (specified time) is t0cond. At this time, the following values are calculated. t1 = tC (I) -t0 t2 = tC (I) cond-t0cond t1-t2 ≦ Δt0 (where Δt0 is given in advance), it is determined that the data is time-series data, and in step ST5, Add the current data to Case C (I),
finish. Otherwise, it is determined that it is not a time series, the process proceeds to step ST6, I = I + 1, and step ST7
In step ST4, if I> N, the current data is added as a new search case C (I), and the process ends. If I> N, the process returns to step ST3.

Although only one search condition has been described here, the same applies to a plurality of search conditions. In this case, the above processing may be performed on all the calculation data with similarity corresponding to a plurality of search conditions.

The search result display 19 processes the search result so that it can be easily displayed, and displays it as a search result list. As a display method, when a case similar to the lightning determination element occurs, the similarity, and when a plurality of search conditions are used, the presence or absence of data corresponding to each condition is displayed. Further, it is determined whether or not lightning has occurred at the time of each search case, and the presence or absence of lightning is also displayed.

FIG. 9 shows an example of the search list created in this way. In FIG. 9, rows correspond to respective search cases, and columns correspond to specific contents. For example, the case of the first line is a case of 22:50 on December 27, 1997, which means that lightning has occurred and the similarity with the original search condition is 95%. Here, the similarity with the original search condition shown in the table is calculated using, for example, the average value of the similarity stored in the data corresponding to the same search case in the search result list. Also,
Regarding whether or not lightning has occurred, if any one of the data corresponding to the same case includes data indicating that lightning has occurred, it is determined that lightning has occurred. Further, in the search result display 19, when an arbitrary search case displayed in the search result list is selected, a weather image such as a time when the search case occurred, or a lightning risk or an echo intensity before and after the time. Data is input from the weather image data file 20 and displayed together.

The past case consideration lightning prediction / judgment unit 12 outputs the output of the echo intensity image display 4, the output of the lightning risk image display 7, the output of the graph display 9, and the past case search unit 11. And outputs the result of lightning prediction, taking into account similar lightning situations that occurred in the past. FIG. 10 shows an example of lightning prediction. In FIG. 10, the upper part is an image representing the current weather condition, and the lower part is an image representing the state of the past retrieval case C (I). Here, it is assumed that a lightning strike is detected 10 minutes later in a search case C (I) similar to the current situation. At this time, even after 10 minutes in the current situation, it can be predicted that there is a high possibility that lightning will occur. For example, “large possibility of lightning after 10 minutes” is output as a lightning prediction result.

As described above, according to the lightning prediction device according to the first embodiment of the present invention, not only the current weather condition but also past lightning cases similar to the current weather condition are searched, and prediction is performed. Since it can be used as a reference, more accurate lightning prediction can be performed.

Embodiment 2 FIG. 11 shows a past case search unit 11 in a lightning prediction device according to Embodiment 2 of the present invention.
FIG. 3 is a block diagram showing the configuration of FIG. In the second embodiment, the configuration other than the past case search unit 11 is the same as that of FIG. As illustrated in FIG. 11, the past case search unit 11 according to the second embodiment receives a remaining condition that is a search condition for which similarity target data has not been extracted, and extracts a single search condition from the remaining condition. One search condition extractor 21, a single search condition from the single search condition extractor 21, lightning determination element data stored in the weather status data file 8, A sequential processing searcher 22 for executing sequential processing of past case searches based on search conditions and outputting search results and remaining conditions.

Here, the sequential processing search unit 22 includes a weather case data file 16 for storing past case data as weather case data, a single search condition from the single search condition extractor 21, The lightning judgment element data stored in the current situation data file 8 and the past case data stored in the weather case data file 16 are input, and the similarity is determined from the past case data by the single search condition. A similarity calculation target data selector 15 for extracting data to be calculated and outputting the same as the similarity calculation target data together with the lightning determination element, and similarity calculation target data from the similarity calculation data selector 15. A similarity calculator that calculates similarity between lightning judgment element data for each similarity calculation target data and outputs past case data with similarity and search conditions 7, a search result file 24 storing search results output to the search result display 19, and a search result based on the past case data with similarity from the similarity calculator 17 and the search results from the search result file 24. A sequential search result creator 23 that outputs a new search result created by modifying the above, a search condition from the search condition creator 23,
It is determined whether or not the remaining condition remains based on the search result from the single search condition extractor 21. If the remaining condition remains, the remaining condition is output to the single search condition extractor 21. If it is not, a search end determiner 25 that outputs the search result to the search result display 19 is provided.

Next, the operation according to the second embodiment of the present invention will be described. The single search condition extractor 21 inputs the remaining conditions, and when there are a plurality of remaining conditions, selects one of them and outputs it as a single search condition. Here, the remaining condition is a search condition from which the similarity target data has not been extracted, and the search condition input from the search condition determiner becomes the remaining condition for the first time, and is input from the search end determiner 25 for the second and subsequent times. Take advantage of the remaining conditions that are used. Next, the similarity calculation target data selector 15 in the sequential processing searcher 22
Receives the single search condition, the past case data, and the lightning determination element data as input, and, based on the single search condition, sets data for which similarity is to be calculated from the past case data. It is extracted and output as similarity calculation target data together with the lightning determination element data. That is, while the similarity calculation target data selector according to the first embodiment outputs similarity calculation target data and lightning determination element data for all search conditions, the similarity calculation unit 21 Calculation target data selector 1
Reference numeral 5 outputs only similarity calculation data and lightning determination element data corresponding to one search condition.

The similarity calculator 17 receives the similarity calculation target data and the lightning determination element data as input, and outputs past case data with similarity and the search condition. This is similar to the similarity calculator 17. The sequential search result creator 23 receives the past case data with similarity and the search result input to the search result file 24, and corrects the search result.
Output together with the search condition. Here, the search result is
This is the same as that described in the first embodiment. First, using the above-mentioned past case data with similarity, each row of the past case data with similarity is processed according to the similarity of the data, and a search result for a single condition is created. The method of making the same is the same as that of the search result creator 18 in the first embodiment. Next, a search result for a single condition created in this way and a search result file 24
To create a new search result by integrating with the search results existing in.

FIG. 12 shows an example of creating a new search result. As shown in FIG. 12, when a search result based on a single search condition is added to search cases C (1) and C (2) stored in a search file, the search result based on the single search condition is: Since it is understood that the search result is added to the search case C (2), a new search result as shown in FIG. 12 is obtained. The new search result created in this way is output to the search end determiner 25.

The search end determiner 25 inputs the search result and the remaining condition, and determines whether to end the search. That is, from the remaining conditions, the one corresponding to the single condition and used for calculating the similarity calculation target data for the single condition is deleted. As a result, when all the remaining conditions are gone, it is considered that the search is completed, and the search result is output to the search result display 19. Otherwise, if the remaining condition still remains, the search condition is stored in a single search condition extractor 21 and the search result is stored in the search result file 2.
4 is output. After that, using the above residual conditions,
The same is repeated from a single search condition extractor 21.

As described above, according to the lightning prediction apparatus according to the second embodiment of the present invention, it is possible to smoothly perform the past case search processing by the computer.

Embodiment 3 FIG. 13 shows a past case search unit 1 in the lightning prediction device according to the embodiment of the present invention.
1 is a block diagram showing a configuration of FIG. In the configuration of the past case search unit 11 shown in FIG. 13, the first embodiment shown in FIG.
The same parts as those described above are denoted by the same reference numerals and description thereof will be omitted. A new code 26 is based on the search condition from the search condition determiner 14, the lightning judgment element data stored in the weather status data file 8, and the past case data stored in the weather case data file 16. Is a parallel processing searcher that processes and outputs search results for each search condition in parallel, and 27 is a search result integrator that integrates and outputs search results for each search condition from the parallel processing searcher 26. .

Here, the parallel processing search unit 26 stores the search conditions from the search condition determiner 14, the lightning judgment element data stored in the weather status data file 8, and the weather case data file 16. The similarity calculation target, which inputs past case data that has been input, extracts data for which similarity should be calculated from the past case data based on search conditions, and outputs the similarity calculation target data together with the lightning judgment element A data selector 15A and a similarity calculator for calculating and outputting the similarity between the similarity calculation target data and the lightning determination element data from the similarity calculation data selector 15A for each similarity calculation target data. 17A and a search result creator 18 that processes the calculation data with similarity from the similarity calculator 17A based on the similarity of the data to create and output a past case search result. DOO Results for each search condition by arranging a plurality outputs parallel processing to the.

Next, the operation according to the third embodiment of the present invention will be described. In the parallel processing searcher 26, for a single search condition, a single similarity calculation target data selector 1
5A, a single similarity calculator 17A, and a single search result creator 18A, and the similarity calculation target data selector 15 according to the first embodiment only for a single condition.
The same operation as the similarity calculator 17 and the search result creator 18 is performed, and a search result for each search condition is output.

The search result integrator 27 integrates the search results for each of the above conditions, combines them into one search result, and outputs the search result. FIG. 14 shows an example of integration. In FIG. 14, in the parallel processing searcher, a search result (1) based on a single search condition is converted to a search result (3) based on a single search condition.
Suppose you have obtained At this time, it is determined whether or not the above search results correspond to the same search case, and if the search results are combined, a search result after integration as shown in FIG. 14 is obtained, for example.

As described above, according to the lightning prediction apparatus according to the third embodiment of the present invention, the search process can be performed for each condition, so that the past case search process can be executed quickly.
It is possible to predict lightning.

Embodiment 4 FIG. 15 is a past case search unit 1 in the lightning prediction device according to the embodiment of the present invention.
1 is a block diagram showing a configuration of FIG. In the past case search unit 11 shown in FIG. 15, the same parts as those in the first embodiment shown in FIG. 2 are denoted by the same reference numerals, and description thereof will be omitted. Reference numeral 28 denotes a designated coordinate search condition determiner which inputs a designated position and a designated time from the image / position selector 13 and outputs a specific position corresponding to the designated position as one of search conditions. .

Next, an operation according to the fourth embodiment of the present invention will be described. The designated coordinate search condition determination unit 28
The designated position and designated time in the position selector 13 are input, and only the coordinate data for the designated position is output as a search target for the position. Other operations are the same as those of the search condition determiner 14. For example, in the case of FIG. 3 in the first embodiment, since the designated position is (50, 150), only the data at the position (50, 150) is output. That is, the output of the designated coordinate search condition determiner 28 is
As shown in FIG.

As described above, according to the lightning prediction device according to the fourth embodiment of the present invention, it is possible to carefully search and search past lightning cases closer to the current weather conditions and perform lightning prediction. It becomes possible.

Embodiment 5 FIG. 17 is a past case search unit 1 in the lightning prediction device according to the embodiment of the present invention.
1 is a block diagram showing a configuration of FIG. In the past case search unit 11 shown in FIG. 17, the same parts as those in the first embodiment shown in FIG. 2 are denoted by the same reference numerals, and description thereof will be omitted. As a new code, 29 is a search condition candidate determiner, 30 is a maximum feature search condition determiner, which replaces the search condition determiner 14 shown in FIG. , The designated position and designated time from the image / position selector 13 are input, and a search condition candidate is output. The search condition candidates from the condition candidate determiner 29, the lightning risk data from the lightning risk image display 7 and the echo intensity data from the echo intensity image display 4 are input, and the search conditions are output.

Next, an operation according to the fifth embodiment of the present invention will be described. The search condition candidate determiner 29 receives the designated position and the designated time in the image / position designator 13 as inputs, and determines, for example, up to which coordinates around the designated position to check the lightning risk or the echo intensity, The determined value is used as a search condition position candidate. The determination of the search target with respect to the time is performed in the same manner as the search condition determiner 14. A search target for these search condition position candidates and time is output as search condition candidates.

The maximum feature search condition determiner 30 receives the above search condition candidate and lightning risk data or echo intensity data and inputs the lightning risk among the search condition position candidates in the search condition. Alternatively, the search target for the position is determined in consideration of the coordinates having the highest echo intensity. FIG.
Shows an example of the search target determination. In FIG. 18, it is assumed that the input designated position is (50, 150). At this time, for example, in the search condition candidate determiner 29, five coordinates before and after this, that is, from (45,145) to (55,15)
It is assumed that up to 5) are determined as search condition position candidates. If (53, 146) has the highest lightning risk among the search condition position candidates, for example, (53, 14)
The 10 coordinates before and after 6), that is, all coordinates between (43, 136) and (63, 156) are determined as search targets for the position. The search target for the position and the search target for the time are output together as search conditions.

As described above, according to the lightning prediction apparatus according to the fifth embodiment of the present invention, past lightning cases in a region where lightning is most likely to occur in current weather conditions are searched. It is possible to predict lightning.

Embodiment 6 FIG. FIG. 19 shows a past case search unit 1 in the lightning prediction device according to the embodiment of the present invention.
1 is a block diagram showing a configuration of FIG. In the past case search unit 11 shown in FIG. 19, the same parts as those in the first embodiment shown in FIG. As a new code, 31 is a search condition weighting device, and 32 is a weighted similarity calculator.

Next, an operation according to the sixth embodiment of the present invention will be described. The search condition weighting unit 31 inputs the search condition and the specified position or the specified time from the search condition determiner 14, and calculates the similarity between the search object for the time, the search object for the position, or both. Set the weight to perform. For example, if the designated position is (50, 150) and the search target for the position is (4
In the case of (0, 160), the weight is determined using, for example, an equation such as Equation 3.

[0071]

(Equation 3)

The determined weight is output as a search condition weight.

The weighted similarity calculator 32 receives the similarity calculation target data, the lightning determination element data, and the search condition weight, and inputs the similarity calculation target data and the lightning determination element data. The similarity for each element is calculated for each similarity calculation target data using, for example, the following equation. RW = R × WW

The subsequent operation is the same as that of the similarity calculator 17 in the first embodiment.

As described above, according to the lightning prediction apparatus according to the sixth embodiment of the present invention, under the current weather conditions,
It is possible to search for past case data around the original selected position while considering the original selected position, and perform lightning prediction.

Embodiment 7 FIG. 20 shows a past case search unit 1 in the lightning prediction device according to the embodiment of the present invention.
1 is a block diagram showing a configuration of FIG. In the past case search unit 11 shown in FIG. 20, the same parts as those in the first embodiment shown in FIG. As a new code, reference numeral 33 denotes a similarity calculation target all position data extractor.

Next, an operation according to the seventh embodiment of the present invention will be described. The similarity calculation target all data extractor 33 is
With the search condition as input, all data in the search target for the position is extracted from the weather case data file 16 regardless of the search target value for the time. The extracted data is output as similarity calculation target data.

As described above, according to the lightning prediction apparatus according to Embodiment 7 of the present invention, under the current weather conditions,
It is possible to search for all past cases that occurred in the area where lightning could occur, and to perform lightning prediction.

Embodiment 8 FIG. FIG. 21 is a past case search unit 1 in the lightning prediction device according to the embodiment of the present invention.
1 is a block diagram showing a configuration of FIG. In the past case search unit 11 shown in FIG. 21, the same parts as those in the first embodiment shown in FIG. As a new code, 34 is an extraction target time determiner, and 35 is a similarity calculation target data selector based on time.

Next, an operation according to the eighth embodiment of the present invention will be described. The extraction target time determiner 34 receives the specified time as input, determines how much data before and after the specified time is to be searched, and outputs the determination result as the extraction target time. Similarity calculation target data selector 3 based on time
5 inputs the extraction target time and the search condition, and extracts all data having the extraction target time from the weather case data file 16 irrespective of the search target value for the position. The extracted data is output as similarity calculation target data.

As described above, according to the lightning prediction apparatus according to the eighth embodiment of the present invention, all past case data that occurred at a time similar to the current weather condition are searched to predict lightning. It becomes possible.

Embodiment 9 FIG. FIG. 22 shows a past case search unit 1 in the lightning prediction device according to the embodiment of the present invention.
1 is a block diagram showing a configuration of FIG. In the past case search unit 11 shown in FIG. 22, the same parts as those in the first embodiment shown in FIG. As a new code, 36 is a similarity calculation target data compulsory designator.

Next, an operation according to the ninth embodiment of the present invention will be described. Similarity calculation target data compulsory designator 36
Input the search conditions, consider the search target for the position and the search target for the time, judge which data should be extracted by the operator etc. from the weather case data file, and select the past case to be the similarity calculation target I do. The selected past case data is output as similarity calculation target data.

As described above, according to the lightning prediction apparatus according to the ninth embodiment of the present invention, the past case data similar to the current situation is searched while satisfying the intention of the operator or the like, and the lightning prediction is performed. Can be performed.

Embodiment 10 FIG. FIG. 23 is a block diagram showing a configuration of the past case search unit 11 in the lightning prediction device according to Embodiment 10 of the present invention. In the past case search unit 11 shown in FIG. 23, the same parts as those in the first embodiment shown in FIG. As a new code, 37 is a maximum similarity calculator and 38 is a maximum similarity search result display.

Next, an operation according to the tenth embodiment of the present invention will be described. The maximum similarity calculator 37 receives the search result as input, and calculates the maximum value of the similarity of each search case in the search result as the similarity of the search case. The maximum similarity use search result display 38 displays the search results using the similarity of the above search cases.

As described above, the tenth embodiment of the present invention is described.
Makes it possible to display search results and predict lightning with reference to past case data most similar to the current situation.

Embodiment 11 FIG. FIG. 24 is a block diagram showing a configuration of the past case search unit 11 in the lightning prediction device according to the embodiment of the present invention. In the past case search unit 11 shown in FIG. 24, the same parts as those in the first embodiment shown in FIG. As a new code, reference numeral 39 denotes a minimum similarity calculator, and reference numeral 40 denotes a minimum similarity utilization search result display.

Next, an operation according to the eleventh embodiment of the present invention will be described. The minimum similarity calculator 39 receives the search result as input, and calculates the minimum value of the similarity of each search case in the search result as the similarity of the search case. The search result display unit 40 using the minimum similarity displays the search result using the similarity of the search case.

As described above, Embodiment 11 of the present invention
Makes it possible to search for past cases in which all data in the same case is similar to the current situation, and predict lightning.

Embodiment 12 FIG. FIG. 25 is a block diagram showing a configuration of the past case search unit 11 in the lightning prediction device according to Embodiment 12 of the present invention. In the past case search unit 11 shown in FIG. 25, the same parts as those in the first embodiment shown in FIG. As a new code, 41 is a search condition weight file, 42 is a search condition weight consideration similarity calculator, and 43 is a search condition weight consideration search result display.

Next, an operation according to the twelfth embodiment of the present invention will be described. The search condition weight file 41 stores in advance search condition weights indicating how much similarity is considered for each search condition. The method of setting the weight of the search condition is, for example, to be able to set up to N storage amounts of the image memory 3 so that the larger the time is closer to the current situation and the earlier the time is, the smaller it is in the current weather condition. .
The search condition weight consideration similarity calculator 42 calculates the similarity of each search case using the above search condition weight. The search condition display 43 considering the search condition weight displays the search results using the similarity of the search results.

As described above, the twelfth embodiment of the present invention is described.
Makes it possible to search for past cases and predict lightning in consideration of the importance of each search condition.

Embodiment 13 FIG. FIG. 26 is a block diagram showing a configuration of the past case search unit 11 in the lightning prediction device according to Embodiment 13 of the present invention. In the past case search unit 11 shown in FIG. 26, the same parts as those in the first embodiment shown in FIG. As a new code, 44 is a position index file, 45
Is a similarity calculation target data selector using a position index.

Next, an operation according to the thirteenth embodiment of the present invention will be described. In the position index file 44,
Advance past case data of any position is created in advance the position index of whether it is stored anywhere in the weather case data file 16. FIG. 27 shows an example of the position index. Here, the figure on the left side shows the X coordinate and the Y coordinate representing each position, and the figure on the right side shows the address of the position in the weather case data file where the data corresponding to the position is located. In the case of FIG. 27, for example, X
The address containing the data whose coordinate is 1 and the Y coordinate is 1 is address 3, address 1004 and address 1056. Also, FIG. 27 shows a position index in which the X coordinate and the Y coordinate are put together. However, a position index group that represents the X coordinate and a position index group that represents the Y coordinate are separately created and used. Is also possible.

The similarity calculation target data selector 45 using the position index uses the above search condition as input, and uses the above position index from the weather case data file.
The necessary past case data is extracted, and the past case data is output as similarity calculation target data.

As described above, the thirteenth embodiment of the present invention is described.
According to the apparatus for predicting lightning, similarity calculation target data can be quickly extracted using a position index.

Embodiment 14 FIG. FIG. 28 is a block diagram showing a configuration of the past case search unit 11 in the lightning prediction device according to Embodiment 14 of the present invention. In the past case search unit 11 shown in FIG. 28, the same parts as those in the first embodiment shown in FIG. As a new code, 46 is a time index file, 47
Is a similarity calculation target data selector using a time index.

Next, an operation according to the fourteenth embodiment of the present invention will be described. The time index file 46 includes
Advance past case data of which time is created in advance the time index of whether it is stored anywhere in the weather case data file 16. FIG. 29 shows an example of the time index. In FIG. 29, the left side shows the time, and the right side shows the address of the position of the data corresponding to the time in the weather case data file. In the case of FIG. 29, for example, addresses containing data from 0:00 to 6:00 on January 1 are 549 and 133
Address 2 and address 5098 are shown. FIG.
In, the date is used as the time, but the time index collects the dates of multiple days, or
It may be created like a time unit.

The similarity calculation target data selector 47 using the time index uses the above search condition as an input and uses the above time index from the weather case data file.
The necessary past case data is extracted, and the past case data is output as similarity calculation target data.

As described above, the fourteenth embodiment of the present invention
According to the apparatus for predicting lightning, similarity calculation target data can be quickly extracted using a time index.

[0102]

As described above, according to the lightning prediction device of the present invention, the output of the echo intensity image display, the output of the lightning risk image display, and the weather status data file are stored. A past case search unit that outputs a past weather condition similar to the current lightning judgment element based on the lightning judgment element, an output of an echo intensity image display, an output of a lightning risk image display, It is equipped with a past case considering lightning prediction judging unit that outputs a lightning prediction result that considers past cases based on past weather conditions from the past case search unit and the output of the graph display. Not only the situation but also past lightning cases similar to the current weather situation can be searched and used as a reference for prediction, so that more accurate lightning prediction can be performed. In addition, the past case search unit
Output of co intensity image display or lightning risk image display
Specified position to search on the image based on the output of
An image / position selector for selecting a designated time and an image / position selector
Input the designated position and designated time from the selector, and
Search target and the search target for time
Search condition determiner that determines
A weather case data file that stores past case data
The search conditions from the search condition determination unit and the weather conditions data file
Lightning judgment element data stored in the
Input past case data stored in data file
And calculate similarity from past case data based on search conditions.
The data to be output is extracted and the data for similarity calculation is extracted.
Similarity calculation pair output as a lightning judgment element
From the elephant data selector and the similarity calculation data selector
The similarity between the calculation target data and the lightning judgment element data
Similarity calculated and output for each similarity calculation target data
Similarity calculator and data to be calculated with similarity from the similarity calculator
Data based on data similarity to create past case search results.
Search result creator to generate and output, and past weather image data
Image data file for storing search results and search result creator
Of past case search results and weather image data files
Image corresponding to the past case search result based on the elephant image
Expressed as past weather conditions similar to the current lightning judgment factor
And a search result display for displaying and outputting
This makes it possible to smoothly perform the past case search process.

[0103]

The above-mentioned past case search section selects an image / position for selecting a designated position and a designated time on the image to be searched based on the output of the echo intensity image display or the output of the lightning risk image display. A selector, a search condition determiner that receives a designated position and a designated time from the image / position selector as input, and determines a search target for the position and a search target for the time as search conditions, and a search in which similarity target data is not extracted. A single search condition extractor that receives a remaining condition as an input and extracts a single search condition from the remaining condition, a search condition from the search condition determiner, and a lightning strike stored in a weather status data file A sequential processing searcher that executes sequential processing of past case searches based on the determination element data and a single search condition from a single search condition extractor and outputs search results and remaining conditions; The image corresponding to the search result based on the search result from the search result creator and the weather image data of the weather image data file similar to the current lightning judgment element is stored. A search result display for displaying and outputting past weather conditions, and the sequential processing searcher is configured to store a past case data as weather case data, a weather case data file, and a single search condition extractor. The single search condition, the lightning judgment element data stored in the weather status data file, and the past case data stored in the weather case data file are input. A similarity calculation target that extracts data from which similarity should be calculated and outputs the similarity calculation target data together with the lightning judgment element Data selector, and the similarity between the similarity calculation target data from the similarity calculation data selector and the lightning determination element data are calculated for each similarity calculation target data, and the similarity-added past case data is calculated. A similarity calculator that outputs search conditions, a search result file that stores search results that are output to the search result display, past case data with similarity from the similarity calculator, and search results from the search result file A sequential search result creator that outputs a new search result created by modifying the search result based on the search condition, a search condition from the search condition determiner, and a single search condition from a single search condition extractor , Determine whether the remaining condition remains based on the search result from the sequential search result creator, and if so, output the remaining condition to a single search condition extractor; if not, Search results table Since the search completion judging unit for outputting to the indicator is provided, it is possible to smoothly perform the past case search processing by the computer.

The above-mentioned past case search section selects an image / position for selecting a specified position and a specified time to search on an image based on the output of the echo intensity image display or the output of the lightning risk image display. A selector, a specified position and a specified time from the image / position selector are input, and a search condition determiner that determines a search target for the position and a search target for the time as search conditions, and stores past case data as weather case data. Based on the weather case data file to be searched, the search conditions from the search condition determiner, the lightning judgment element data stored in the weather status data file, and the past case data stored in the weather case data file. A parallel processing searcher that processes and outputs search results for each search in parallel, and a search that integrates and outputs search results for each search condition from the parallel searcher A result integrator, a weather image data file for storing past weather image data, and an image corresponding to the search result based on the search result from the search result integrator and the weather image data in the weather image data file. A search result display for displaying and outputting as a past weather condition similar to the lightning determination element, and the parallel processing search device is provided with a search condition from the search condition determination unit and a weather condition data file. Lightning judgment element data
The past case data stored in the weather case data file is used as input, and the data for which the similarity should be calculated is extracted from the past case data based on the search conditions. A similarity calculation target data selector to be output, and a similarity between the similarity calculation target data from the similarity calculation data selector and the lightning determination element data are calculated and output for each similarity calculation target data. A search result for each search condition by arranging a plurality of similarity calculators and a plurality of search result creators for generating and outputting past case search results by processing data to be calculated with similarity from the similarity calculators by data similarity Can be executed in parallel, and the search process can be executed for each condition, so that the past case search process can be executed quickly and the lightning can be predicted.

The search condition deciding unit receives the designated position and the designated time from the image / position selector and outputs the specified position corresponding to the designated position as one of the search conditions. The similarity calculation target data selector is a search condition from the designated coordinate search condition determination unit, lightning judgment element data stored in the weather status data file, and stored in the weather case data file. By inputting past case data that has been input, extracting data for which similarity should be calculated from the past case data based on search conditions, and outputting it as similarity calculation target data together with the lightning judgment element, the current weather Closer to the conditions,
It is possible to carefully search and search past lightning cases and perform lightning prediction.

The search condition determiner receives the designated position and the designated time from the image / position selector, and outputs a search condition candidate. Input the designated time, the search condition candidate from the search condition candidate determiner, the lightning risk data from the lightning risk image display and the echo intensity data from the echo intensity image display, and output the search condition The similarity calculation target data selector is the search condition from the maximum feature search condition determiner, the lightning judgment element data stored in the weather status data file, and the weather case data. The past case data stored in the file is input, and the data for which the similarity should be calculated is extracted from the past case data based on the search condition, and the extracted data is used as the similarity calculation target data. By outputting with lightning determination element, retrieves the past lightning cases most likely of lightning areas in the current weather conditions, it is possible to make predictions of lightning.

[0108] The past case search unit is configured to select a designated position and a designated time to search on an image based on the output of the echo intensity image display or the output of the lightning risk image display. A selector, a specified position and a specified time from the image / position selector are input, and a search condition determiner that determines a search target for the position and a search target for the time as search conditions, and stores past case data as weather case data. Input the weather case data file to be searched, the search condition from the search condition determiner, the lightning judgment element data stored in the weather status data file, and the past case data stored in the weather case data file. Extract the data for which similarity should be calculated from the past case data according to the conditions, and use the lightning judgment element as the similarity calculation target data. And a similarity calculation target data selector for outputting also an input search conditions from the search condition determiner,
A search condition weighter that weights the search condition and outputs a search condition weight, and a similarity calculation target data from the similarity calculation target data selector and a search condition weight from the search condition weighter as inputs, A weighted similarity calculator that calculates and outputs a weighted similarity for each element between the similarity calculation target data and the lightning determination element data for each similarity calculation target data, and a weighted similarity calculator. A search result creator that creates and outputs past case search results by processing the calculation target data with similarity based on data similarity, a weather image data file that stores past weather image data, and a past data from the search result creator. Based on the case search result and the weather image in the weather image data file, an image corresponding to the past case search result is displayed and output as a past weather condition similar to the current lightning judgment element. Since a result indicator, the current weather conditions, taking into account the original selected position, and searches the past case data of its periphery becomes possible to perform lightning prediction.

The similarity calculation target data selector receives the search condition and the lightning determination element data as input, extracts past case data at all positions corresponding to the search condition, and calculates similarity. By being a similarity calculation target all data extractor that outputs as target data,
Under the current weather conditions, it is possible to search for all past cases that occurred in a region where lightning could occur, and to perform lightning prediction.

Also, the search condition from the search condition determiner and the designated time from the image / position selector are input.
The apparatus further includes an extraction target time determiner that determines and outputs an extraction target time, wherein the similarity calculation target data selector includes a search condition from the search condition determiner and a source stored in the weather status data file. The lightning judgment element data, the past case data stored in the weather case data file, and the extraction target time from the extraction target time determiner are input, and all the data having the extraction target time are similar from the past case data. A similarity calculation target data selector for extracting the degree as a target to be calculated and outputting the similarity calculation target data together with the lightning determination element, and the similarity calculator calculates the similarity based on the time. The similarity between the similarity calculation target data based on the time from the calculation data selector and the lightning determination element data is calculated for each similarity calculation target data. By force, to search for all of the past case data that occurred at a time that is similar to the current weather situation, it is possible to make predictions of lightning.

The similarity calculation target data selector receives the search condition and the lightning judgment element data as inputs and manually calculates the similarity while considering the search condition and the lightning judgment element data. Searching past case data similar to the current situation and predicting lightning while satisfying the intentions of the operator etc. by being a similarity calculation target data compulsory designator that selects and outputs target data Becomes possible.

Further, there is further provided a maximum similarity calculator which receives a search result from the search result creator as input and outputs a maximum value of the similarity of the search case in the search result as the similarity of the search case. The result display takes the similarity of the search case as an input, and is a search result display that uses the maximum similarity to display the search result. By referring to the past case data most similar to the current situation, the search result Is displayed, and lightning can be predicted.

Further, the apparatus further comprises a minimum similarity calculator which receives a search result from the search result creator as input, and outputs a minimum value of the similarity of the search case in the search result as the similarity of the search case. The result display unit receives the similarity of the search case as an input, and is a search result display unit that uses the minimum similarity to display the search result, so that all data in the same case is similar to the current situation. It is possible to search past cases and predict lightning strikes.

A search condition weight file for storing search condition weights, a search result from the search result generator and a search condition weight stored in the search condition weight file are input, and the search condition weight is used. A search condition weight consideration similarity calculator for calculating and outputting the similarity of each search case, and wherein the search result display is configured to input the similarity of the search case as input and to consider search condition weights. By using the search condition display that considers the search condition weight, the past cases are searched,
It is possible to predict lightning.

[0115] In addition, historical case data in advance which position further comprises a location index file to store the position index of whether the stored anywhere weather case data file, the similarity calculation target data selector, said The position index file, the search condition and the lightning determination element are input, and the similarity calculation target data selector using the position index that outputs the similarity calculation target data is used. Calculation target data can be quickly extracted.

[0116] Further, in advance past case data for which time, where further comprises a time index file storing time indices that are either stored, the similarity calculation target data selector weather case data file, the The time index file, the search condition and the lightning determination element are input, and the similarity calculation target data selector using the time index that outputs the similarity calculation target data is used. Calculation target data can be quickly extracted.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a lightning prediction device according to Embodiment 1 of the present invention.

FIG. 2 is a block diagram showing a configuration of a past case search unit in the lightning prediction device according to Embodiment 1 of the present invention.

FIG. 3 is an explanatory diagram showing an example of selecting a position to be searched from an image.

FIG. 4 is an explanatory diagram showing past case data.

FIG. 5 is an explanatory diagram showing similarity calculation target data;

FIG. 6 is an explanatory diagram showing past case data with similarity;

FIG. 7 is an explanatory diagram showing search results.

FIG. 8 is a flowchart showing a flow of creating a search result.

FIG. 9 is an explanatory diagram showing a search list.

FIG. 10 is an explanatory diagram showing lightning prediction.

FIG. 11 is a block diagram showing a configuration of a past case search unit in a lightning prediction device according to Embodiment 2 of the present invention.

FIG. 12 is an explanatory diagram showing an example of creating a new search result.

FIG. 13 is a block diagram showing a configuration of a past case search unit in the lightning prediction device according to Embodiment 3 of the present invention.

FIG. 14 is an explanatory diagram showing an example of integrating search results.

FIG. 15 is a block diagram showing a configuration of a past case search unit in a lightning prediction device according to Embodiment 4 of the present invention.

FIG. 16 is an explanatory diagram showing an output of a designated position coordinate search condition determiner.

FIG. 17 is a block diagram showing a configuration of a past case search unit in a lightning prediction device according to Embodiment 5 of the present invention.

FIG. 18 is an explanatory diagram illustrating an example of determining a search target.

FIG. 19 is a block diagram showing a configuration of a past case search unit in a lightning prediction device according to Embodiment 6 of the present invention.

FIG. 20 is a block diagram showing a configuration of a past case search unit in a lightning prediction device according to Embodiment 7 of the present invention.

FIG. 21 is a block diagram showing a configuration of a past case search unit in a lightning prediction device according to Embodiment 8 of the present invention.

FIG. 22 is a block diagram showing a configuration of a past case search unit in a lightning prediction device according to Embodiment 9 of the present invention.

FIG. 23 is a block diagram showing a configuration of a past case search unit in a lightning prediction device according to Embodiment 10 of the present invention.

FIG. 24 is a block diagram showing a configuration of a past case search unit in a lightning prediction device according to Embodiment 11 of the present invention.

FIG. 25 is a block diagram showing a configuration of a past case search unit in the lightning prediction device according to Embodiment 12 of the present invention.

FIG. 26 is a block diagram showing a configuration of a past case search unit in a lightning prediction device according to Embodiment 13 of the present invention.

FIG. 27 is an explanatory diagram showing a position index.

FIG. 28 is a block diagram showing a configuration of a past case search unit in the lightning prediction device according to Embodiment 14 of the present invention.

FIG. 29 is an explanatory diagram showing a time index.

FIG. 30 is a diagram showing a conventional example.

[Explanation of symbols]

1 weather sensor, 2 radar echo analyzer, 3 image memory, 4 echo intensity image display, 5 lightning judgment element calculator, 6 lightning risk calculator, 7 lightning risk image display, 8 weather status data File, 9 Graph display, 10 Lightning prediction judge, 11 Past case search section, 1
2 Past Case Consideration Lightning Prediction Judge, 13 Image / Position Selector, 14 Search Condition Determiner, 15 Similarity Calculation Target Data Selector, 16 Meteorological Case Data File, 17 Similarity Calculator, 18 Search Result Creator, 19 search result display device, 20 weather image data file, 21 single search condition extractor, 22 sequential processing search device, 23 sequential search result creator, 24 search result file, 25 search end judgment device, 26 parallel processing search device , 27 search result integrator,
28 designated coordinate search condition determiner, 29 search condition candidate determiner, 30 maximum feature search condition determiner, 31 search condition weighter, 32 weighted similarity calculator, 33 similarity calculation target all position data extractor, 34 Extraction target time determiner, similarity calculation target data selector based on 35 times, 3
6 Similarity calculation target data compulsory designator, 37 Maximum similarity calculator, 38 Maximum similarity use search result display, 39
Minimum similarity calculator, 40 Minimum similarity use search result display, 41 Search condition weight file, 42 Search condition weight consideration similarity calculator, 43 Search condition weight consideration search result display, 44 Position index file, 45 Position index A similarity calculation target data selector using
46 Time index file, 47 Similarity calculation target data selector using time index.

Continued on the front page (72) Inventor Yoshio Kosuge 2-3-2 Marunouchi, Chiyoda-ku, Tokyo Inside Mitsubishi Electric Corporation (72) Inventor Fukuyoshi Asano 2-3-2 Marunouchi, Chiyoda-ku, Tokyo Mitsubishi Electric Corporation (56) References JP-A-1-321393 (JP, A) JP-A-11-160453 (JP, A) JP-A-11-38154 (JP, A) JP-A-7-110378 (JP, A) Table 6-503421 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G01S ⁷ /00-7/42 G01S 13/00-13/95 G01W 1/00-1/18 G06F 15/403

Claims (14)

(57) [Claims] 1. A weather sensor for observing and outputting a lightning situation, a radar echo analyzer for outputting echo intensity data and three-dimensional echo intensity distribution data based on a lightning situation from the weather sensor, and the radar echo An image memory for storing echo intensity data from the analyzer, an echo intensity image display for inputting and displaying the echo intensity data stored by the image memory, and a three-dimensional echo height distribution data from the radar echo analyzer. A lightning judgment element calculator that calculates and outputs a lightning judgment element based on the lightning judgment element calculator that calculates and outputs lightning risk data based on the lightning judgment element from the lightning judgment element calculator; A lightning risk image display for inputting and displaying lightning risk data from the lightning risk calculator; and a lightning judgment from the lightning determination factor calculator. A weather status data file for storing elements, and a graph display for inputting lightning judgment element data stored in the weather condition data file and outputting past lightning judgment element data, for predicting and judging lightning In the lightning prediction device, a current lightning is generated based on an output of the echo intensity image display, an output of the lightning risk image display, and a lightning determination element stored in the weather status data file. A past case search unit that outputs a past weather condition similar to the determination element; an output of the echo intensity image display; an output of the lightning risk image display; and a past weather from the past case search unit. status and, a past case consideration lightning predictor for outputting a lightning prediction result of considering the past case on the basis of the output of the graph indicator, the past case search unit, the echo intensity image Display output or above lightning hazard
Finger to search on the image based on the output of the image display
An image / position selector for selecting a fixed position and a specified time, and a specified position and a specified time from the image / position selector are entered.
Force, the search object for the position and the search
A search condition determiner that determines an elephant as a search condition, and a weather event that stores past case data as weather case data
Example data file, search conditions from the search condition determination unit, and the weather status data
Lightning judgment element data stored in the data file,
Past cases stored in the above weather case data file
Past case data based on the above search conditions
Extract data for which similarity should be calculated from
And, as the similarity calculation target data,
A similarity calculation target data selector that outputs the similarity calculation target data from the similarity calculation target data selector.
Calculation of the similarity between the
A similarity calculator that calculates and outputs each data to be output and data to be calculated with similarity from the similarity calculator
Creates and outputs past case search results based on data similarity
And a weather image data file that stores past weather image data.
And the past case search results from the search result generator and the weather
Past case inspection based on weather image in image data file
The image corresponding to the search result is similar to the current lightning judgment element
And a search result display for displaying output as past weather conditions
Lightning prediction apparatus characterized by comprising. 2. A weather sensor for monitoring and outputting a lightning situation.
And the echo intensity data based on the lightning situation from the weather sensor
Output three-dimensional echo intensity distribution data
And Daeko analyzer, the echo intensity data from the radar echo analyzer Case
An image memory for pay, the echo intensity data stored by the image memory
Echo intensity image display for input and display, and three-dimensional echo height distribution from the radar echo analyzer
Lightning judgment that calculates and outputs the lightning judgment element based on the data
Element calculator , based on the lightning determination element from the lightning determination element calculator
Input the lightning risk calculator that calculates and outputs the lightning risk data and the lightning risk data from the lightning risk calculator
The lightning risk image display to be displayed and the lightning judgment element from the lightning judgment element calculator are stored.
Meteorological condition data file and lightning judgment element stored in the meteorological condition data file
Data input and output of past lightning judgment element data
A lightning prediction device that has a rough indicator and predicts and determines lightning
In the output of the echo intensity image display, the lightning risk Degas
Image display output and stored in the above-mentioned weather status data file
Current lightning judgment is required based on
Past case retrieval unit that outputs past weather conditions similar to element
And the output of the echo intensity image display and the lightning risk
The output of the image display and the past
Based on the image situation and the output of the above graph display,
Past Case Considered Lightning that Outputs Lightning Prediction Result Considering Examples
A prediction judgment unit , wherein the past case search unit outputs the echo intensity image display or the lightning risk
Finger to search on the image based on the output of the image display
An image / position selector for selecting a fixed position and a specified time, and a specified position and a specified time from the image / position selector are entered.
Force, the search object for the position and the search
A search condition determiner that determines an elephant as a search condition, and a remaining condition where the similarity target data is a search condition that has not been extracted.
Input and extract a single search condition from the remaining conditions
Search condition extractor, the search condition from the search condition determiner, and the weather condition data
Lightning judgment element data stored in the data file,
Based on the single search condition from the single search condition extractor
Based on the results of the past case search.
A sequential processing searcher that outputs the remaining conditions, and a weather image data file that stores past weather image data
And the search results from the search result generator and the weather image data
Search results based on the weather image data
Corresponding image of past weather similar to current lightning judgment factor
I and a search result display unit for displaying the output STATUS
And the sequential processing searcher stores the past case data as the weather case data.
Example data file and the single search condition from the single search condition extractor above
Lightning judgment stored in the current weather data file is required.
Raw data and stored in the above weather case data file
Past case data as input, and
Targets for calculating similarity from past case data
Is extracted as the similarity calculation target data
Similarity calculation target data selection to be output together with the lightning judgment element
Selector and the similarity calculation target data from the similarity calculation data selector.
Calculation of the similarity between the
Calculated for each data to be output
A similarity calculator that outputs search conditions and a search that stores search results output to the search result display.
Search result file, past case data with similarity from the similarity calculator, and
Search results based on search results from search result files
Sequential search that outputs a new search result created by modification
Result creator, search condition from the search condition determiner, and the single search
A single search condition from the condition extractor and the above sequential search results
Based on the search results from the creator
Judge whether or not there is any
Output to a single search condition extractor, if there is no remaining
A search end determiner that outputs search results to a search result display;
A lightning prediction device comprising: 3. A weather sensor for observing and outputting a lightning situation.
And the echo intensity data based on the lightning situation from the weather sensor
Output three-dimensional echo intensity distribution data
And Daeko analyzer, the echo intensity data from the radar echo analyzer Case
An image memory for pay, the echo intensity data stored by the image memory
Echo intensity image display for input and display, and three-dimensional echo height distribution from the radar echo analyzer
Lightning judgment that calculates and outputs the lightning judgment element based on the data
Element calculator , based on the lightning determination element from the lightning determination element calculator
Input the lightning risk calculator that calculates and outputs the lightning risk data, and the lightning risk data from the lightning risk calculator
The lightning risk image display to be displayed and the lightning judgment element from the lightning judgment element calculator are stored.
Meteorological condition data file and lightning judgment element stored in the meteorological condition data file
Data input and output of past lightning judgment element data
A lightning prediction device that has a rough indicator and predicts and determines lightning
In the output of the echo intensity image display, the lightning risk Degas
Image display output and stored in the above-mentioned weather status data file
Current lightning judgment is required based on
Past case retrieval unit that outputs past weather conditions similar to element
And the output of the echo intensity image display and the lightning risk
The output of the image display and the past
Based on the image situation and the output of the above graph display,
Past Case Considered Lightning that Outputs Lightning Prediction Result Considering Examples
A prediction judgment unit , wherein the past case search unit outputs the echo intensity image display or the lightning risk
Finger to search on the image based on the output of the image display
An image / position selector for selecting a fixed position and a specified time, and a specified position and a specified time from the image / position selector are entered.
Force, the search object for the position and the search
A search condition determiner that determines an elephant as a search condition, and a weather event that stores past case data as weather case data
Example data file, search condition from the search condition determiner and weather condition data
Lightning judgment element data stored in the
Past case data stored in the meteorological case data file
Search results for each search condition are processed in parallel based on
Integrated with the parallel processing searcher that outputs the search results and the search results for each search condition from the parallel processing searcher
And a weather image data file that stores past weather image data.
And the search results from the search result integrator and the weather image data
Search results based on the weather image data
Corresponding image of past weather similar to current lightning judgment factor
I and a search result display unit for displaying the output STATUS
In addition, the parallel processing search unit is configured to search the search condition from the search condition determination unit and the weather condition data.
Lightning judgment element data stored in the data file,
Past cases stored in the above weather case data file
Past case data based on the above search conditions
Extract data for which similarity should be calculated from
And, as the similarity calculation target data,
A similarity calculation target data selector that outputs the similarity calculation target data from the similarity calculation target data selector.
Calculation of the similarity between the
A similarity calculator that calculates and outputs each data to be output and data to be calculated with similarity from the similarity calculator
Creates and outputs past case search results based on data similarity
Multiple search result creators to search for each search condition.
A lightning prediction device characterized in that search results are processed in parallel and output . 4. The search condition determining device according to claim 1 , wherein
Input the designated position and designated time from the selector, and
Output a specific position corresponding to the position as one of the search conditions
Is a designated coordinate search condition determiner, and the similarity calculation
The elephant data selector is the
Search conditions and the data stored in the weather status data file
And the weather case data file
The past case data stored in the
Similarity should be calculated from past case data according to search conditions.
The data to be extracted is extracted, and the similarity calculation target data is
Output together with the lightning judgment element.
The lightning prediction device according to claim 1 . 5. The search condition determiner according to claim 1 , wherein the image / position
Input the designated position and designated time from the selector, and
Search condition candidate determiner that outputs
Designation and designation time from the position selector, and the above search condition candidates
Display of search condition candidates from the determiner and the above-mentioned lightning risk image
Display of data on the risk of lightning strike from the detector and the echo intensity image
Input echo intensity data from the instrument and output search conditions
The maximum feature search condition determiner that performs
The elephant data selector selects the maximum feature search condition
Search conditions and the data stored in the weather status data file
And the weather case data file
The past case data stored in the
Similarity should be calculated from past case data according to search conditions.
The data to be extracted is extracted, and the similarity calculation target data is
Output together with the lightning judgment element.
The lightning prediction device according to claim 1 . 6. A weather sensor for monitoring and outputting a lightning situation.
And the echo intensity data based on the lightning situation from the weather sensor
Output three-dimensional echo intensity distribution data
And Daeko analyzer, the echo intensity data from the radar echo analyzer Case
An image memory for pay, the echo intensity data stored by the image memory
Echo intensity image display for input and display, and three-dimensional echo height distribution from the radar echo analyzer
Lightning judgment that calculates and outputs the lightning judgment element based on the data
Element calculator , based on the lightning determination element from the lightning determination element calculator
Input the lightning risk calculator that calculates and outputs the lightning risk data, and the lightning risk data from the lightning risk calculator
The lightning risk image display to be displayed and the lightning judgment element from the lightning judgment element calculator are stored.
Meteorological condition data file and lightning judgment element stored in the meteorological condition data file
Data input and output of past lightning judgment element data
A lightning prediction device that has a rough indicator and predicts and determines lightning
In the output of the echo intensity image display, the lightning risk Degas
Image display output and stored in the above-mentioned weather status data file
Current lightning judgment is required based on
Past case retrieval unit that outputs past weather conditions similar to element
And the output of the echo intensity image display and the lightning risk
The output of the image display and the past
Based on the image situation and the output of the above graph display,
Past Case Considered Lightning that Outputs Lightning Prediction Result Considering Examples
A prediction judgment unit , wherein the past case search unit outputs the echo intensity image display or the lightning risk
Finger to search on the image based on the output of the image display
An image / position selector for selecting a fixed position and a specified time, and a specified position and a specified time from the image / position selector are entered.
Force, the search object for the position and the search
A search condition determiner that determines an elephant as a search condition, and a weather event that stores past case data as weather case data
Example data file, search condition from the search condition determiner, and weather condition data
Lightning judgment element data stored in the data file,
Past cases stored in the above weather case data file
Past case data based on the above search conditions
Extract data for which similarity should be calculated from
And, as the similarity calculation target data,
The similarity calculation target data selector, which is also output to the user, and the search condition from the search condition determiner are input, and the search
Search conditions that weight conditions and output search condition weights
Weighting device and similarity calculation target from the similarity calculation target data selector
Data and the search condition weight from the above search condition weighter
As input, similarity calculation target data and lightning judgment element data
The weighted similarity of each element between
Weighted similarity calculator that calculates and outputs data for each data, and similarity calculation target data from the weighted similarity calculator.
Data based on data similarity to create past case search results.
A search result creator that creates and outputs a weather image data file that stores past weather image data
And the past case search results from the search result generator and the weather
Past case inspection based on weather image in image data file
The image corresponding to the search result is similar to the current lightning judgment element
And a search result display for displaying output as past weather conditions
Emitting lightning prediction apparatus characterized by comprising. 7. The similarity calculation target data selector includes:
The search conditions and the above lightning judgment element data
Past case data at all locations that match the search conditions.
Data is extracted and output as similarity calculation target data
A contractor characterized by comprising all data extractors for similarity calculation
The lightning prediction device according to claim 1 . 8. A search condition from the search condition determiner,
Extract specified time from the above image / position selector and extract
The extraction target time determiner that determines and outputs the target time
The similarity calculation target data selector is configured to perform the search
Search conditions from the condition determiner and the above
Lightning judgment element data stored in the
Past case data and case data stored in the case data file
And the extraction target time from the extraction target time determiner
And extract all data with the extraction target time from the past case data.
Is extracted as an object to calculate the similarity from
When outputting with the above lightning judgment element as target data
Data selector for similarity calculation by
The similarity calculator calculates the similarity calculation data based on the time.
Similarity calculation target data and lightning judgment by time from selector
The similarity between element data is calculated for each similarity
The lightning prediction device according to claim 1 , wherein the lightning prediction device calculates and outputs the calculated value every time . 9. The data selector for calculating similarity according to claim 1 ,
The search conditions and the above lightning judgment element data
Considering the search conditions and the lightning judgment element data,
Similarity calculation to select and output similarity calculation target data
Claims characterized in that it comprises an output target data compulsory designator
1. The lightning prediction device according to 1 . 10. A search result from the search result creator is provided.
As input, the maximum similarity of search cases in the search results
Maximum similarity calculator that outputs values as similarities of search cases
And the search result display includes a class of the search case.
Using the similarity as an input, the maximum similarity
2. The lightning prediction device according to claim 1, wherein the device is a search result display . 11. A search result from the search result creator is provided.
As input, minimum similarity of search cases in the search results
Minimum similarity calculator that outputs values as similarities of search cases
And the search result display includes a class of the search case.
Using the similarity as an input, the minimum similarity
2. The lightning prediction device according to claim 1, wherein the device is a search result display . 12. Search condition weights for storing search condition weights
File and search results from the search result creator
Enter the search condition weight stored in the search condition weight file.
Similarity of each search case using the above search condition weight as power
And a search condition weighted similarity calculator that calculates and outputs degrees
The search result display further includes a search result similar to the search case.
Display search results considering search condition weights with degree as input
Featured by a search result display that considers search conditions
The lightning prediction device according to claim 1, wherein 13. A pre-past case data of any position
Is stored in the weather case data file
Position index that stores the position index
File for the similarity calculation.
The selector selects the position index file and the search condition
And the above-mentioned lightning judgment element as input, and the similarity calculation target data
Similarity calculation target using position index that outputs
2. The lightning prediction device according to claim 1, comprising a data selector . Past case data of 14. The advance what time
Is stored in the weather case data file
Time index that stores the time index
File for the similarity calculation.
The selector is the time index file and the search condition
And the above-mentioned lightning judgment element as input, and the similarity calculation target data
Similarity calculation target using the time index that outputs
2. The lightning prediction device according to claim 1, comprising a data selector .