TW201812339A - Epicentral distance estimating device, epicentral distance estimating method, and computer-readable recording medium - Google Patents

Abstract

This epicentral distance estimating device (10) is provided with: an earthquake information acquiring unit (11) which acquires waveform data relating to an earthquake that has occurred; and an estimation processing unit (12) which estimates an epicentral distance by applying the acquired waveform data to a learning model obtained by learning a relationship between the waveform data and the epicentral distance of an earthquake.

Description

Epicenter distance estimation device, epicenter distance estimation method, and computer-readable recording medium

The present invention relates to an epicenter distance estimation device and an epicenter distance estimation method for estimating an epicenter distance when an earthquake occurs, and also to a computer-readable recording medium that records a program used to implement the foregoing method.

In the event of an earthquake, in order to estimate the magnitude of the earthquake and the time of arrival of the main shock, the epicenter distance must be quickly determined. In general, the depth of the epicenter distance is determined by the magnitude detected by the seismometer at multiple locations.

However, if the location of the earthquake source is on the sea floor or in an area where the density of the seismometer is relatively low, it takes too much time in order to obtain the magnitude measured by multiple seismometers, and the epicenter distance cannot be determined in time. Therefore, in recent years, a technique for determining the epicenter distance using only the magnitude measured by a single seismometer has been developed.

As such a technique, a technique that estimates the epicenter distance by using "the intensity of the seismic waveform data at the time of the arrival of the earthquake is closer to the epicenter, and the farther the earthquake is, the more gentle" is known. Literature 1).

Specifically, in the technique disclosed in Patent Document 1, the absolute value of the time series data obtained from the seismometer is set to y (t), the time is set to t, and the time when the seismometer detects an earthquake is set to t = 0, the waveform shape of the initial motion portion of the earthquake obtained by the seismometer is fitted by a function shown in the following mathematical formula 1. In the following mathematical formula 1, A is a parameter related to the maximum amplitude of the initial motion part, and B is a parameter related to the time variation of the initial motion amplitude of the seismic waveform. Moreover, in fact, in the fitting process, human experience and intuition must be relied on in order to reflect the complex individual location attributes in the influencing factors and the parameters A and B of unknown methods.

[Mathematical formula 1]

Then, in the technique disclosed in Patent Document 1, parameters A and B are obtained by a least square method. It can be seen that although there is a correlation between the parameter B and the epicenter distance, this correlation will not be affected by the magnitude of the earthquake. Therefore, if the relationship between the parameter B and the epicenter distance is formulated in advance, the epicenter distance can be determined by calculating the parameter B using Mathematical Formula 1 from the waveform shape of the initial motion portion of the earthquake. According to the technique disclosed in Patent Document 1, the epicenter distance can be quickly determined from the waveform shape of the initial motion portion of the earthquake. [Prior Art Literature] [Patent Literature]

[Patent Document 1] Japanese Patent Laid-Open No. 2002-277557

[Problems to be Solved by the Invention] However, in the technique disclosed in Patent Document 1, in some cases, the coefficients A and B cannot be calculated, and there is also a problem of insufficient reliability. In addition, the technique disclosed in Patent Document 1 has a problem that it is not easy to shorten the time required to calculate the epicenter distance.

An object of the present invention is to provide an epicenter distance estimation device, an epicenter distance estimation method, and a computer-readable recording medium that can solve the above-mentioned problems and stably calculate the epicenter distance and shorten the calculation time. [Means for solving problems]

In order to achieve the above object, a type of epicenter distance estimation device of the present invention is characterized by having: an earthquake information acquisition unit that acquires waveform data of an earthquake that has occurred; and an estimation processing unit that learns waveform data of an earthquake and The learning model obtained from the relationship between the epicenter distances applies the previously obtained waveform data to estimate the epicenter distances.

In addition, in order to achieve the above-mentioned object, a method of estimating the epicenter distance of the present invention is characterized by having the following steps: (a) a step of obtaining waveform data of an earthquake that has occurred; (b) learning waveform data of an earthquake and The learning model obtained from the relationship between the epicenter distances applies the previously obtained waveform data to estimate the epicenter distances.

In addition, in order to achieve the above-mentioned object, a type of computer-readable recording medium of the present invention is characterized in that a program is recorded, and the program includes a command for executing the following steps in the computer: (a) obtaining the earthquake Steps of waveform data; (b) For the learning model obtained by studying the relationship between the waveform data of the earthquake and the epicenter distance, the previously obtained waveform data is applied to estimate the epicenter distance. [Inventive effect]

As described above, according to the present invention, the epicenter distance can be calculated stably, and the calculation time can be shortened.

(Embodiment Mode 1) Hereinafter, the epicenter distance estimation device, the epicenter distance estimation method, and the program according to Embodiment 1 of the present invention will be described with reference to FIGS. 1 to 5.

[Apparatus configuration] First, a schematic configuration of the epicenter distance estimation apparatus according to the first embodiment will be described with reference to FIG. 1. FIG. 1 is a block diagram showing a schematic configuration of an epicenter distance estimation device according to a first embodiment of the present invention.

As shown in FIG. 1, the epicenter distance estimation device 10 of the first embodiment of the present embodiment is a device for estimating the epicenter distance from waveform data measured when an earthquake occurs. As shown in FIG. 1, the epicenter distance estimation device 10 includes a seismic data acquisition unit 11 and an estimation processing unit 12.

The seismic data acquisition unit 11 acquires waveform data of an earthquake that has occurred. The estimation processing unit 12 applies waveform data obtained by the seismic data acquisition unit 11 to the learning model to estimate the epicenter distance. The learning model is obtained in advance by learning the relationship between the waveform data of the earthquake and the epicenter distance.

In this way, the first embodiment is different from the conventional technology in that the epicenter distance can be estimated without fitting the waveform data to a function, so the epicenter distance can be calculated stably. In addition, in the first aspect of the present embodiment, it is not necessary to perform the calculation process by using the least square method, so the calculation time can be shortened.

Next, the configuration of the epicenter distance estimation device according to the first embodiment of the present invention will be described in more detail with reference to FIG. 2. FIG. 2 is a block diagram specifically showing the configuration of the epicenter distance estimation device according to the first embodiment of the present invention.

As shown in FIG. 2, in the first embodiment, the epicenter distance estimation device 10 is connected to an integrated monitoring system 30 such as an earthquake detection device 20 and an earthquake activity via a network. Among them, the seismic detection device 20 is provided with a seismometer, and when the seismic wave is detected by the seismometer, the waveform data of the detected seismic wave is transmitted to the epicenter distance estimation device 10. In the first embodiment, the seismic detection device 20 is the acquisition source of the waveform data of the seismic data acquisition unit 11.

In the example of FIG. 2, although only a single seismic detection device 20 is exemplified, the number of seismic detection devices 20 to which the epicenter distance estimation device 10 is connected is not particularly limited. However, the seismic detection device 20 as the acquisition source of the seismic data acquisition unit 11 may be any one of them.

The integrated monitoring system 30 such as seismic activity is a system maintained by the Meteorological Agency in Japan. When an earthquake occurs, the system calculates the magnitude of the meteorological agency's earthquake, and then estimates the tsunami height based on the calculated magnitude of the meteorological agency's earthquake. In addition, the integrated monitoring system 30 such as an earthquake activity transmits the calculated meteorological agency earthquake scale and predicted tsunami height to various media in the form of an emergency earthquake alert.

In addition, in the first embodiment, the epicenter distance estimation device 10 inputs the estimated epicenter distance to a comprehensive monitoring system 30 such as seismic activity. Therefore, the integrated monitoring system 30 such as seismic activity uses the epicenter distance estimated by the epicenter distance estimation device 10 to calculate the magnitude of the meteorological agency earthquake and the predicted tsunami height.

As shown in FIG. 2, in the first embodiment, the epicenter distance estimation device 10 includes a learning data acquisition unit 13, a learning unit 14, and a storage unit 15 in addition to the above-mentioned seismic data acquisition unit 11 and estimation processing unit 12. In addition, FIG. 2 shows an example of the epicenter distance estimation device 10. The learning data acquisition unit 13, the learning unit 14, and the storage unit 15 may be located in a device other than the epicenter distance estimation device 10.

The learning material acquisition unit 13 acquires the waveform data which becomes input data during the learning of the learning unit 14 described later, and the epicenter distance, which is also a positive solution data during learning, and inputs them to the learning unit 14. Moreover, the source of input data and correct solution data is not particularly limited to this.

The learning unit 14 uses the waveform data of the earthquake as input data, uses the epicenter distance of the earthquake as positive solution data, learns the relationship between the waveform data and the epicenter distance, and then generates a learning model 16 representing the learning result. In addition, the learning unit 14 stores the generated learning model 16 in the storage unit 15.

FIG. 3 is a diagram showing an example of input data and positive solution data for learning in the first embodiment of the present invention. A plurality of waveform data with different epicenter distances are shown in FIG. 3. Each waveform data shown in FIG. 3 is waveform data of earthquakes observed in the past. In addition, the epicenter distance corresponding to each waveform data is the positive solution data. The learning unit 14 uses the waveform data shown in FIG. 3 as input data, and uses the epicenter distance as the positive solution data for learning.

In addition, in the first embodiment, as the forward solution data, data published by the Meteorological Agency can be used. The data published by the Meteorological Agency includes the epicenter distance and source factors of each observation point. These data are the detections calculated by the Meteorological Agency using a unified system (http://www.data.jma.go.jp/svd/eqev/data/ bulletin / deck.html). In addition, in the first embodiment, the input data and correct data used for learning are preferably calculated by an earthquake whose magnitude is greater than or equal to 値 (eg, magnitude 4).

In addition, in the first embodiment, the learning unit 14 may construct a neural network through mechanical learning, and then set the neural network as the learning model 16. Specifically, the learning unit 14 generates learning by using input data and positive solution data in a hierarchical neural network including an input layer, an intermediate layer, and an output layer, and adjusting the coupling between nodes in adjacent layers. model.

In addition, in the first embodiment, the “learning” performed by the learning unit 14 means a so-called “mechanical learning”. In addition, the "learning" performed by the learning unit 14 is not limited to deep learning using the neural network described above, and may be learning using logistic regression, learning using a support vector machine, learning using a decision tree, and heterogeneous hybrid learning.

The seismic data acquisition unit 11 receives waveform data of an earthquake that has occurred from a single seismic detection device 20 in the first embodiment. The seismic data acquisition unit 11 transmits the received waveform data to the estimation processing unit 12.

The estimation processing unit 12 accesses the storage unit 15 in the first embodiment, acquires the learning model 16, and applies the waveform data transmitted from the learning data acquisition unit 13 to the acquired learning model 16 to estimate the epicenter distance.

In addition, in the first embodiment, the learning unit 14 may further use the depth of the earthquake source as the positive solution data, so as to learn the relationship between the waveform data, the epicenter distance, and the depth of the earthquake source to generate a learning model 16. At this time, the estimation processing unit 12 can estimate the depth of the seismic source in addition to the epicenter distance.

In addition, in the first embodiment, in addition to the waveform data as the input data, the learning unit 14 can also use the location data of the location where the waveform data has been obtained as the input data. At this time, the learning unit 14 learns the relationship between the waveform data and location data, and the epicenter distance (or the epicenter distance and the depth of the epicenter), and generates a learning model 16.

Here, the place where the waveform data has been obtained refers to a place where a seismic wave has been observed as a source of the waveform data. In addition, as the location data, for example, the surface area increase rate, the data indicating the state of the plate, the data indicating the volcano existing near the place where the seismic wave was observed, the thickness of the crust, and the thickness of the lithosphere. In this way, if the learning model 16 is generated using both waveform data and location data as input data, the accuracy of the estimation processing can be improved.

When using location data as input data for learning, in addition to acquiring waveform data of an earthquake that has occurred, the seismic data acquisition unit 11 will also acquire location data of the location where the waveform data has been obtained, that is, the location of the location where the seismic detection device 20 is installed. Location information.

In addition, the location data may be stored in advance in the storage unit 15 of each seismic detection device 20. In this state, whenever the seismic data acquisition unit 11 acquires waveform data, the corresponding location data is acquired from the storage unit 15. In addition, the location data may be transmitted from the seismic detection device 20 together with the waveform data. In this state, the seismic data acquisition unit 11 will obtain the location data together with the waveform data.

In addition, when using location data as input data for learning, the estimation processing unit 12 applies the acquired waveform data and location data to the learning model 16 generated by the learning unit 14, and estimates the epicenter distance (or the epicenter distance and the source of the epicenter). depth).

Moreover, in the first embodiment, the input data and the positive solution data are not limited to the above examples. Data other than waveform data and location data can be used as input data. In addition, data other than the epicenter distance and the focal depth can be used as the positive solution data.

[Device Operation] Next, the operation of the epicenter distance estimation device 10 according to the first embodiment will be described with reference to FIGS. 4 and 5. In the following description, FIGS. 1 to 3 will be appropriately referred to. In addition, in the first embodiment, the epicenter distance estimation device 10 is operated to implement the epicenter distance estimation method. Therefore, the description of the epicenter distance estimation method in the first embodiment will replace the following description of the operation of the epicenter distance estimation device 10.

In the first embodiment, the epicenter distance estimation device 10 mainly performs learning processing and estimation processing. First, the learning process will be described. FIG. 4 is a flowchart showing an operation when the learning process of the epicenter distance estimation device according to the first embodiment of the present invention is executed.

As shown in FIG. 4, first, the learning material acquisition unit 13 obtains input data and correct solution data (step A1). Specifically, in step A1, in addition to the waveform data as the input data, the learning data acquisition unit 13 also acquires the location data as the input data, in addition to the epicenter distance as the positive solution data, and the depth of the seismic source as the positive solution data.

Next, the learning unit 14 determines whether the learning model 16 already exists (step A2). Specifically, the learning unit 14 determines whether or not the learning model 16 is stored in the storage unit 15.

If the result of the determination in step A2 is that the learning model 16 does not yet exist, the learning unit 14 learns the relationship between the waveform data and location data, the epicenter distance and the depth of the epicenter, and then newly generates a learning model 16 representing the learning result (step A3 ).

Specifically, in step A3, the learning unit 14 constructs a neural network through learning and uses it as a learning model 16. The learning unit 14 stores the created learning model 16 in the storage unit 15.

In addition, when the result of the determination in step A2 is that the learning model 16 already exists, the learning unit 14 uses the input data and positive solution data obtained in step A1 to update the existing learning model 16 (step A4). Specifically, the learning unit 14 uses the input data and the positive solution data obtained in step A1 to update the coupling between the nodes.

By executing steps A1 to A4, a learning model can be created or updated. Then, the estimation process is executed using the created or updated learning model. FIG. 5 is a flowchart showing an operation when the estimation process of the epicenter distance estimation device according to the first embodiment of the present invention is executed.

As shown in FIG. 5, first, when waveform data of an earthquake that has occurred is transmitted from the seismic detection device 20, the seismic data acquisition unit 11 receives the transmitted waveform data (step B1).

Next, the seismic data acquisition section 11 acquires, from the storage section 15, the location data of the place where the seismic detection device 20 of the transmitted waveform data is installed (step B2). When the location data and the waveform data are transmitted, the seismic data acquisition unit 11 receives the transmitted location data.

Next, the estimation processing unit 12 applies the waveform data received in step B1 and the location data acquired in step B2 to the learning model 16 created or updated by the learning processing shown in FIG. 4, and estimates the epicenter distance and The depth of the source (step B3).

By performing steps B1 to B3, the epicenter distance and the focal depth are estimated based on the waveform data obtained from the single seismic detection device 20.

[Effects of Implementation Mode 1] As described above, according to the implementation mode 1, it is possible to estimate the epicenter distance and the focal depth from a single waveform data without fitting the waveform data to a function. In addition, since the estimation process is performed by the learning model 16, the epicenter distance and the depth of the epicenter can be stably calculated in a short time.

In other words, in the first embodiment, unlike the conventional method disclosed in Patent Document 1, it is not necessary to manually perform the operation of reflecting complex individual site attributes on influencing factors and unknown parameters. According to the first embodiment, data that can be used to obtain the available accuracy can be obtained only by objective waveform data and mechanical learning. The epicenter distance estimation device 10 according to the first embodiment can be introduced into multiple locations and multiple regions.

In addition, as described above, in this embodiment, the depth of the source can also be estimated from a single waveform data, but the technology disclosed in the above Patent Document 1 cannot estimate the depth of the source. When the technique disclosed in the aforementioned Patent Document 1 is used, in order to measure the depth of the seismic source, the measurement results of a plurality of seismometers must be used.

[Modification 1] Next, a modification of the first embodiment will be described. First, in Modification 1, the learning unit 14 generates a learning model 16 for each of the set waveform amounts. Specifically, the waveform amount is displayed in the time elapsed since the earthquake occurred. Therefore, the learning unit 14 cuts the waveform data of only the elapsed time component from the waveform data obtained by the learning data acquisition unit 13 at each set elapsed time, and then uses the cut waveform data as input data, and then performs Learning results in learning models 16. Thereby, the learning model 16 is generated for each waveform amount.

In addition, in Modification 1, the estimation processing unit 12 calculates the waveform amount of the waveform data of the earthquake that has occurred, and then selects the learning model to be used from among the plurality of generated learning models 16 based on the calculated waveform amount. . Then, the estimation processing unit 12 applies the waveform data of the earthquake that has occurred to the selected learning model 16 and estimates the epicenter distance (or the epicenter distance and the depth of the epicenter).

In general, the estimation of the epicenter distance and focal depth used in emergency earthquake breaking news can be performed even when the amount of waveform data is small. Therefore, the waveform data obtained by the seismic data acquisition unit 11 is not necessarily constant, and the waveform amount of the waveform data used in the generation of the learning model may not be consistent with the waveform amount of the waveform data of the earthquake that has occurred, making the estimated accuracy reduce. However, according to the first modification, the learning model 16 is selected in accordance with the waveform amount of the waveform data of the earthquake that has occurred, so that the above-mentioned problem of reduced estimation accuracy can be avoided.

[Modification 2] In Modification 2, the learning unit 14 generates a learning model 16 for each of the observation points of the waveform data that becomes the input data. Specifically, the learning unit 14 generates a learning model 16 using only the waveform data obtained by each of the seismic detection devices (each of the seismometers).

In addition, in Modification 2, the estimation processing unit 12 confirms the observation points of the waveform data of the earthquake that has occurred (that is, the seismic detection device 20 as the transmission source of the waveform data), and based on the confirmed observation points, Among the plurality of learning models 16, a learning model to be used is selected. Then, the estimation processing unit 12 applies the waveform data of the earthquake that has occurred to the selected learning model 16 and estimates the epicenter distance (or the epicenter distance and the depth of the epicenter).

According to Modification Example 2, even if learning without using the characteristics of each observation point (that is, location data), it is possible to perform estimation processing that matches the characteristics of the observation point. Moreover, it is difficult to perform complete learning at observation points that cannot sufficiently ensure input data, which makes it difficult to generate a learning model for such observation points.

[Modification 3] In Modification 3, the learning unit 14 generates a learning model 16 for each of the site characteristics of the observation points of the waveform data that becomes the input data. Specifically, for example, the observation points (earthquake detection device 20) are classified into different groups in accordance with site characteristics (such as location data 値) such as the site increase rate. At this time, the learning unit 14 uses only the waveform data obtained in the group for each group to generate a learning model 16.

In addition, in Modification 3, the estimation processing unit 12 confirms the site characteristics of the observation points of the waveform data of the earthquake that has occurred, and then selects the one to be used from among the plurality of learning models 16 based on the confirmed site characteristics. Learning model. Then, the estimation processing unit 12 applies the waveform data of the earthquake that has occurred to the selected learning model 16 and estimates the epicenter distance (or the epicenter distance and the depth of the epicenter).

According to Modification 3, even if there are observation points for which input data is not completely obtained, it is not necessary to perform learning of the use site data, that is, to perform estimation processing that matches the characteristics of the observation points.

[Program] The program of the implementation mode 1 may be a program that executes steps A1 to A4 shown in FIG. 4 and steps B1 to B3 shown in FIG. 5 in a computer. By installing this program on a computer and executing it, the epicenter distance estimation device 10 and the epicenter distance estimation method of the first embodiment can be realized. At this time, the CPU (Central Processing Unit) of the computer can perform the functions of the seismic data acquisition unit 11, the estimated processing unit 12, the learning data acquisition unit 13, and the learning unit 14.

In addition, the program of the embodiment 12 can be executed by a computer system constructed by a plurality of computers. At this time, each of the computers can function as one of the seismic data acquisition unit 11, the estimation processing unit 12, the learning data acquisition unit 13, and the learning unit 14. In addition, the storage unit 15 may be provided on a computer different from a computer that executes the program of the embodiment.

(Embodiment Mode 2) Next, the epicenter distance estimation device, the epicenter distance estimation method, and the program according to Embodiment 2 of the present invention will be described with reference to FIGS. 6 to 8.

[Device Configuration] First, the configuration of the epicenter distance estimation device according to the second embodiment of the present invention will be described with reference to FIG. 6. FIG. 6 is a block diagram specifically showing the configuration of the epicenter distance estimation device according to the second embodiment of the present invention.

As shown in FIG. 6, the epicenter distance estimation device 40 according to the second embodiment is provided with a waveform pre-processing unit 41. This point is different from the epicenter distance estimation device 10 according to the first embodiment shown in FIGS. 1 and 2. . The differences from the first embodiment will be mainly described below.

The waveform pre-processing unit 41 performs pre-processing on the waveform data used as input data in the learning unit 14 and the waveform data obtained by the seismic data acquisition unit 11. Examples of the pre-processing include image conversion processing, envelope conversion processing, band-pass conversion processing, differential conversion processing, and Fourier conversion processing.

Specifically, the image conversion process is a process of converting waveform data into image data of an image displayed in a chart. It is generally considered that if the image conversion process is performed, the learning unit 14 performs learning based on the image data, making the learning process easier.

The envelope conversion processing is processing for smoothing the waveform of the waveform data. When the envelope processing is performed, it is easy to confirm the rising characteristics of the seismic wave, and a learning model 16 reflecting the rising characteristics of the seismic wave is generated.

Bandpass conversion processing is processing that highlights a waveform of a specific period. If the processing is performed in accordance with the bandpass conversion, the characteristics of the seismic wave will be highlighted, and a learning model 16 reflecting the characteristics of the seismic wave will be generated.

In addition, the differential conversion process is a process in which waveform data is differentiated and then converted into acceleration data. When the differential conversion process is performed, it is easy to confirm the rising characteristics of the seismic wave, and a learning model 16 reflecting the rising characteristics of the seismic wave is generated.

The Fourier transform process is a process of obtaining the frequency distribution of the waveform data. If the Fourier transform is followed, the period difference of each waveform data will be prominent, so a learning model 16 that reflects the period of the seismic wave will be generated.

The waveform pre-processing section 41 may perform any one or more of image conversion processing, envelope conversion processing, band-pass conversion processing, differential conversion processing, and Fourier conversion processing.

[Device Operation] Next, the operation of the epicenter distance estimation device 40 according to the second embodiment will be described with reference to FIGS. 7 and 8. In the following description, FIGS. 1 to 6 are appropriately referred to. In addition, in the second embodiment, the epicenter distance estimation device 40 is operated to implement the epicenter distance estimation method. Therefore, the description of the epicenter distance estimation method according to the second embodiment replaces the following description of the operation of the epicenter distance estimation device 40.

First, the learning process will be described. FIG. 7 is a flowchart showing the operation of the epicenter distance estimation device according to the second embodiment of the present invention when learning processing is executed.

As shown in FIG. 7, first, the learning material obtaining unit 13 obtains input data and correct solution data (step A11). The learning material acquisition unit 13 inputs the acquired data to the waveform pre-processing unit 41.

Next, the waveform pre-processing unit 41 performs pre-processing on the waveform data included in the input data obtained in step A11 (step A12). Then, the waveform pre-processing unit 41 inputs the pre-processed waveform data, other input data (location data), and correct solution data to the learning unit 14.

Next, the learning unit 14 determines whether the learning model 16 already exists (step A13).

As a result of the determination of step A13, when the learning model 16 does not yet exist, the learning unit 14 learns the relationship between the waveform data and location data, the distance between the epicenter and the depth of the epicenter, and then newly generates a learning model 16 representing the learning result (step A14 ).

In addition, when the result of the determination in step A13 is that the learning model 16 already exists, the learning unit 14 uses the input data and the positive solution data to update the existing learning model 16 (step A15).

By performing steps A11 to A15, the learning model 16 is updated or created. After that, the learning model 16 that has been created or updated is used to perform the estimation process. FIG. 8 is a flowchart showing an operation when the learning process of the epicenter distance estimation device according to the second embodiment of the present invention is executed.

As shown in FIG. 8, first, when waveform data of an earthquake that has occurred is transmitted from the seismic detection device 20, the seismic data acquisition unit 11 receives the transmitted waveform data (step B11).

Next, the seismic data acquisition unit 11 acquires, from the storage unit 15, the location data of the place where the seismic detection device 20 of the transmitted waveform data is installed (step B12). When the location data and the waveform data are transmitted, the seismic data acquisition unit 11 receives the transmitted location data.

Next, the waveform pre-processing section 41 performs pre-processing on the waveform data received in step B11 (step B13). Then, the waveform pre-processing section 41 inputs the pre-processed waveform data and location data to the estimation processing section 12.

Next, the estimation processing unit 12 applies the waveform data preprocessed in step B13 and the location data obtained in step B12 to the learning model 16 created or updated by the learning processing shown in FIG. 7 to estimate the epicenter Distance and depth of the source (step B14).

By performing steps B11 to B14, the second embodiment is also the same as the first embodiment, and based on the waveform data obtained from the single seismic detection device 20, the epicenter distance and the depth of the source are estimated.

[Effects of Implementation Mode 2] As described above, in the implementation mode 2, the preprocessing performed by the waveform preprocessing unit 41 suppresses noise and highlights features of the waveform data used for learning. Therefore, according to the second embodiment, the accuracy of the learning model is improved, and as a result, the estimation accuracy can be improved.

(Physical Configuration) Here, a computer implementing the epicenter distance estimation device will be described using FIG. 9 by executing the programs of the implementation modes 1 and 2. FIG. 9 is a block diagram showing an example of a computer that implements the epicenter distance estimation device according to the first and second embodiments of the present invention.

As shown in FIG. 9, the computer 110 includes a CPU 111, a main memory 112, a storage device 113, an input interface 114, a display controller 115, a data reader / writer 116, and a communication interface 117. The above units are connected via a bus 121 in such a manner that data can be transmitted to each other.

The CPU 111 is stored in the storage device 113, and the programs (encoding) of this embodiment are developed in the main memory 112, and then these programs are executed in a predetermined order to perform various calculations. The main memory 112 is generally a volatile storage device such as DRAM (Dynamic Random Access Memory). The program according to this embodiment is provided in a state of being stored in a computer-readable recording medium 120. In addition, the program in this embodiment may be a program circulating on a network connected through the communication interface 117.

In addition, as a specific example of the storage device 113, in addition to a hard disk, a semiconductor storage device such as a flash memory may be mentioned. The input interface 114 serves as an intermediary to assist data transfer between the input device 118 such as the CPU 111, the keyboard, and the mouse. The display controller 115 is connected to the display device 119 and controls the display of the display device 119.

The data reader / writer 116 serves as an intermediary to assist the data transfer between the CPU 111 and the recording medium 120, and reads programs from the recording medium 120 and writes the processing results of the computer 110 into the recording medium 120. The communication interface 117 acts as an intermediary to assist the data transfer between the CPU 111 and other computers.

Specific examples of the recording medium 120 include general-purpose semiconductor recording devices such as CF (Compact Flash (registered trademark)) and SD (Secure Digital), magnetic recording media such as flexible disks, and CD-ROM ( Compact Disk Read Only Memory).

In addition, the epicenter distance estimation devices 10 and 40 of the embodiments 1 and 2 can be implemented not only by a computer having a program installed, but also by using a hard disk corresponding to each unit. In addition, the epicenter distance estimation devices 10 and 40 may be configured to be partially implemented by a program, and the remaining portions may be implemented by a hard disk.

Part or all of the above-mentioned embodiments can be expressed by (Appendix 1) to (Appendix 15) described below, but are not limited to the following description.

(Remark 1) An epicenter distance estimation device, comprising: an earthquake information acquisition unit that acquires waveform data of an earthquake that has occurred; and an estimation processing unit that learns the relationship between the waveform data of the learned earthquake and the epicenter distance The obtained learning model applies the previously obtained waveform data to estimate the epicenter distance.

(Note 2) If the epicenter distance estimation device of Note 1 further includes a learning section, the learning section takes the waveform data of the earthquake as input data, and uses the epicenter distance of the aforementioned earthquake as positive solution data, and learns the relationship between the waveform data and the epicenter distance Then, a learning model representing a learning result is generated, and the estimation processing unit applies the obtained waveform data to the learning model generated by the learning unit to estimate the epicenter distance.

(Note 3) If the epicenter distance estimation device of Note 2 is used, in addition to the aforementioned waveform data as input data, the aforementioned learning unit also uses as input data the location data of the place where the aforementioned waveform data has been obtained, and learns the aforementioned waveform data and the aforementioned The relationship between the location data and the epicenter distance is then used to generate the aforementioned learning model. In addition to the waveform information, the seismic information acquisition unit obtains the location information of the location where the waveform data of the earthquake that has occurred has been obtained. The learning model not only applies the waveform data, but also applies the obtained location data to estimate the epicenter distance.

(Note 4) If the epicenter distance estimation device of note 2 or 3 is used, the aforementioned learning processing unit further uses the depth of the epicenter of the earthquake as the forward solution data, and learns the relationship between the waveform data and the epicenter distance and the depth of the epicenter. Then, the learning model is generated, and the estimation processing unit estimates the depth of the epicenter in addition to the epicenter distance.

(Note 5) The epicenter distance estimation device according to any one of notes 2 to 4, wherein the learning processing unit constructs a neural network by learning, and uses the neural network as a learning model.

(Note 6) If the epicenter distance estimation device according to any one of notes 2 to 5, wherein the aforementioned learning unit generates the aforementioned learning model for each of the waveform amounts of the waveform data which becomes the input data, the aforementioned estimation processing unit will calculate the acquired Based on the calculated waveform amount, the waveform amount of the aforementioned waveform data is used to select the aforementioned learning model to be used from each of the aforementioned learning models that have been generated, and then apply the previously acquired waveform to the previously selected learning model Data, and the aforementioned epicenter distance is estimated.

(Note 7) If the epicenter distance estimation device according to any one of notes 2 to 5, wherein the aforementioned learning unit generates the aforementioned learning model for each observation point of the waveform data which becomes the input data, the aforementioned estimation processing unit confirms that the previously acquired The observation points of the waveform data are based on the confirmed observation points, and from among the previously generated learning models, the learning model to be used is selected, and the previously obtained waveform data is applied to the selected learning model. , And the aforementioned epicenter distance is estimated.

(Remark 8) If the epicenter distance estimation device of any of Remarks 2 to 5, wherein the aforementioned learning section generates the aforementioned learning model for each of the site characteristics of the observation point of the waveform data which becomes the input data, the aforementioned estimation processing section confirms that The site characteristics of the observation points of the obtained waveform data are based on the confirmed site characteristics. From among the previously generated learning models, the aforementioned learning model to be used is selected, and the previously selected learning model is applied. The aforementioned waveform data has been obtained, and the aforementioned epicenter distance is estimated.

(Note 9) If the epicenter distance estimation device according to any one of notes 1 to 8 further includes a waveform pre-processing section, the waveform pre-processing section applies waveform data used as the aforementioned input data in the aforementioned learning section, and The waveform data obtained by the seismic information acquisition department is pre-processed.

(Note 10) As in the epicenter distance estimation device of Note 9, wherein the aforementioned waveform pre-processing section executes at least one processing from image conversion processing, envelope conversion processing, band-pass conversion processing, differential conversion processing, and Fourier conversion processing as the The aforementioned pretreatment.

(Note 11) A method for estimating the epicenter distance, which is characterized by the following steps: (a) the step of obtaining the waveform data of an earthquake that has occurred; (b) the relationship between the waveform data of the earthquake and the epicenter distance is obtained The learning model is based on the previously obtained waveform data, and the epicenter distance is estimated.

(Note 12) If the method for estimating the epicenter distance of Note 11 also has the following steps: (c) Using the waveform data of the earthquake as input data, using the epicenter distance of the aforementioned earthquake as the positive solution data, and learning the relationship between the waveform data and the epicenter distance Then, a learning model representing the learning result is generated. In the step (b), the learning model generated in the step (c) is applied with the obtained waveform data to estimate the epicenter distance.

(Note 13) If the method for estimating the epicenter distance of Note 12 is used, in the step (c) above, in addition to using the aforementioned waveform data as input data, the location data of the place where the aforementioned waveform data has been obtained as input data will be studied. The relationship between the aforementioned waveform data and the aforementioned location data and the epicenter distance, and then the aforementioned learning model is generated. In step (a), in addition to obtaining the aforementioned waveform data, the location of the location where the waveform data of the earthquake that has occurred has been obtained. For the location data, in step (b) above, in addition to applying the waveform data to the learning model, the obtained location data is also applied to estimate the epicenter distance.

(Note 14) For the method for estimating the epicenter distance of Note 12 or 13, wherein: in the step (c), the depth of the epicenter of the earthquake is used as the positive solution data, and the waveform data and the distance between the epicenter and the epicenter are learned. Then, the aforementioned learning model is generated. In step (b), in addition to estimating the epicenter distance, the depth of the source is also estimated.

(Note 15) The epicenter distance estimation method according to any one of notes 12 to 14, wherein: in step (c) above, a neural network is constructed by learning, and the aforementioned neural network is used as a learning model.

(Note 16) The method for estimating the epicenter distance according to any one of notes 12 to 15, wherein: in the step (c) above, the aforementioned learning model is generated for each of the waveform amounts of the waveform data that becomes the input data, and in the aforementioned ( Step b), calculate the waveform amount of the previously obtained waveform data, and then select the learning model to be used from among the previously generated learning models based on the calculated waveform amount, and then select the previously selected previously described learning model. The learning model applies the previously obtained waveform data and estimates the epicenter distance.

(Note 17) The method for estimating the epicenter distance according to any one of notes 12 to 15, wherein: in step (c) above, the aforementioned learning model is generated for each of the observation points of the waveform data that becomes the input data, and in the aforementioned ( Step b), confirm the observation points of the previously obtained waveform data, and then select the learning model to be used from among the previously generated learning models based on the confirmed observation points, and then select the previously described previously selected learning models. The learning model applies the previously obtained waveform data and estimates the epicenter distance.

(Note 18) If the epicenter distance estimation method according to any one of notes 12 to 15, wherein: in step (c) above, the aforementioned learning model is generated for each of the site characteristics of the observation points of the waveform data that becomes the input data, In the step (b) above, confirm the site characteristics of the observation points of the waveform data obtained, and then select the learning model to be used from each of the aforementioned learning models based on the confirmed site characteristics, Then, for the selected learning model, the obtained waveform data is applied to estimate the epicenter distance.

(Note 19) If the method for estimating the epicenter distance of any of Notes 11 to 18, it further has the following steps: (d) in the step (c) above, the waveform data used as the aforementioned input data, and the aforementioned The waveform data obtained in step (a) is pre-processed.

(Note 20) In step (d) above, at least one of image conversion processing, envelope conversion processing, band-pass conversion processing, differential conversion processing, and Fourier conversion processing is performed as the aforementioned pre-processing.

(Note 21) A computer-readable recording medium for recording programs, which includes a command to perform the following steps in the computer: (a) the step of obtaining waveform data of an earthquake that has occurred; (b) the waveform data of learning earthquakes The learning model obtained from the relationship between the epicenter distance and the previously obtained waveform data is used to estimate the epicenter distance.

(Note 22) If the computer-readable recording medium of Note 21, it also performs the following steps in the aforementioned computer: (c) Using the waveform data of the earthquake as the input data, using the epicenter distance of the aforementioned earthquake as the positive solution data, and learning The relationship between the waveform data and the epicenter distance generates a learning model representing the learning result. In step (b) above, the previously obtained waveform data is applied to the learning model generated by step (c) above. The epicenter distance is estimated.

(Note 23) If the computer-readable recording medium of Note 22, wherein: in the step (c) above, in addition to using the aforementioned waveform data as input data, the location data of the location obtained from the aforementioned waveform data is used as input. Data, and learn the relationship between the aforementioned waveform data and the aforementioned location data and the epicenter distance, and then generate the aforementioned learning model. In step (a), in addition to obtaining the aforementioned waveform data, we also obtain the waveform of the earthquake that has occurred. For the location data of the location of the data, in step (b) above, in addition to applying the aforementioned waveform data to the aforementioned learning model, the previously acquired location data is also applied to estimate the epicenter distance.

(Note 24) If the computer-readable recording medium of Note 22 or 23, wherein: in the step (c) above, the depth of the source of the earthquake is used as the positive solution data, and the distance between the waveform data and the epicenter is learned. And the relationship between the depth of the source, and then the aforementioned learning model is generated. In step (b), in addition to estimating the epicenter distance, the depth of the source is also estimated.

(Note 25) The computer-readable recording medium according to any one of Notes 22 to 24, wherein: in step (c) above, a neural network is constructed by learning, and the aforementioned neural network is used as a learning model .

(Note 26) If the computer-readable recording medium of any one of notes 22 to 25, wherein: in the step (c) above, the aforementioned learning model is generated for each of the waveform amounts of the waveform data that becomes the input data, In the step (b), the waveform amount of the obtained waveform data is calculated, and based on the calculated waveform amount, the learning model to be used is selected from each of the generated learning models, and then the existing learning model is used. The selected learning model applies the obtained waveform data to estimate the epicenter distance.

(Note 27) The computer-readable recording medium of any one of Notes 22 to 25, wherein: in the step (c) above, the aforementioned learning model is generated for each of the observation points of the waveform data that becomes the input data, In step (b) above, the observation points of the waveform data obtained are confirmed, and based on the confirmed observation points, from among the generated learning models, the learning model to be used is selected, and The selected learning model applies the obtained waveform data to estimate the epicenter distance.

(Remark 28) The computer-readable recording medium of any one of Remarks 22 to 25, wherein: in step (c) above, for each of the site characteristics of the observation point of the waveform data that becomes the input data, the aforementioned In the learning model, in the step (b) above, confirm the site characteristics of the observation points of the waveform data that have been obtained, and then based on the confirmed site characteristics, select the aforementioned one to be used from each of the aforementioned learning models that have been generated. A learning model, and then applying the previously obtained waveform data to the selected learning model to estimate the epicenter distance.

(Remark 29) If the computer-readable recording medium of any one of Remarks 21 to 28, the following steps are also performed on the aforementioned computer: (d) In step (c) above, the data used for becoming the aforementioned input data is used. The waveform data and the waveform data obtained in the step (a) above are pre-processed.

(Note 30) The computer-readable recording medium according to any one of Note 29, wherein: in step (d) above, image conversion processing, envelope conversion processing, band-pass conversion processing, differential conversion processing, and Fourier are performed. At least one of the conversion processes is used as the aforementioned pre-processing.

As mentioned above, although the invention of this application was demonstrated with reference to embodiment, the invention of this application is not limited to the said embodiment. Various changes that can be understood by those skilled in the relevant field within the scope of the invention of this application can be made to the constitution or details of the invention of this application.

This application claims priority based on Japanese Voluntary Request No. 2016-136310 filed on July 8, 2016, and the entire disclosure thereof is incorporated herein by reference. [Industrial availability]

As described above, according to the present invention, it is possible to stably calculate the epicenter distance and shorten the calculation time. The present invention can be applied to a system in which earthquake-related data must be transmitted as early as possible when an earthquake occurs.

10‧‧‧ Epicenter distance estimation device (implementation type 1)

11‧‧‧ Earthquake Information Acquisition Department

12‧‧‧ Presumed Processing Department

13‧‧‧Learning Information Acquisition Department

14‧‧‧Learning Department

15‧‧‧Storage Department

16‧‧‧ learning model

20‧‧‧earthquake detection device

30‧‧‧Integrated surveillance system such as seismic activity

40‧‧‧ Epicenter distance estimation device (implementation type 2)

41‧‧‧Waveform preprocessing department

110‧‧‧Computer

111‧‧‧CPU

112‧‧‧Main memory

113‧‧‧Storage device

114‧‧‧Input interface

115‧‧‧Display Controller

116‧‧‧Data Reader / Writer

117‧‧‧ communication interface

118‧‧‧ input machine

119‧‧‧display device

120‧‧‧Recording Media

121‧‧‧Bus

[Fig. 1] Fig. 1 is a block diagram showing a schematic configuration of an epicenter distance estimation device according to a first embodiment of the present invention. [Fig. 2] Fig. 2 is a block diagram specifically showing a configuration of a epicenter distance estimation device according to Embodiment 1 of the present invention. FIG. 3 is a diagram showing an example of input data and positive solution data for learning in Embodiment 1 of the present invention. 4] FIG. 4 is a flowchart showing an operation when a learning process of the epicenter distance estimation device according to Embodiment 1 of the present invention is executed. 5] FIG. 5 is a flowchart showing an operation when an estimation process of the epicenter distance estimation device according to Embodiment 1 of the present invention is executed. [FIG. 6] FIG. 6 is a block diagram specifically showing a configuration of an epicenter distance estimation device according to a second embodiment of the present invention. [FIG. 7] FIG. 7 is a flowchart showing an operation when the learning process of the epicenter distance estimation device according to Embodiment 2 of the present invention is executed. [FIG. 8] FIG. 8 is a flowchart showing an operation when an estimation process of the epicenter distance estimation device according to Embodiment 2 of the present invention is executed. [FIG. 9] FIG. 9 is a block diagram showing an example of a computer that implements the epicenter distance estimation device according to the embodiments 1 and 2 of the present invention.

Claims (12)

An epicenter distance estimation device includes: an earthquake information acquisition unit that acquires waveform data of an earthquake that has occurred; and an estimation processing unit that obtains a learning model for learning the relationship between the waveform data of an earthquake and the epicenter distance , Apply the obtained waveform data to estimate the epicenter distance. For example, the epicenter distance estimation device for the scope of patent application No. 1 further includes a learning unit that takes the waveform data of the earthquake as input data and uses the epicenter distance of the earthquake as positive solution data, and learns the relationship between the waveform data and the epicenter distance. Then, a learning model representing a learning result is generated, and the estimation processing unit applies the acquired waveform data to the learning model generated by the learning unit to estimate the epicenter distance. For example, the epicenter distance estimation device of the second scope of the patent application, in addition to using the waveform data as input data, the learning unit also uses the location data of the place where the waveform data has been obtained as input data to learn the waveform data and The relationship between the location data and the epicenter distance generates the learning model. In addition to the waveform data, the seismic information acquisition unit also obtains the location data of the location where the waveform data of the earthquake has occurred. The presumption processing unit In addition to applying the waveform data to the learning model, the acquired location data is also used to estimate the epicenter distance. For example, the epicenter distance estimation device for the second or third item of the patent application scope, wherein the learning processing unit further uses the depth of the epicenter of the earthquake as the positive solution data to learn the distance between the waveform data and the epicenter and the depth of the epicenter This estimation model generates the learning model. In addition to estimating the epicenter distance, the estimation processing unit estimates the depth of the epicenter. For example, the epicenter distance estimation device of the second or third item of the patent application scope, wherein the learning processing unit constructs a neural network by learning and uses the neural network as a learning model. For example, the epicenter distance estimation device of the second or third item of the patent application scope, wherein the learning section generates the learning model for each waveform amount of the waveform data that becomes the input data, and the estimation processing section calculates the acquired waveform data Based on the calculated waveform amount, the learning model to be used is selected from each of the learning models that have been generated, and then the obtained waveform data is applied to the selected learning model, and Estimate the epicenter distance. For example, the epicenter distance estimation device of the second or third item of the patent application scope, wherein the learning section generates the learning model for each observation point of the waveform data that becomes the input data, and the estimation processing section confirms the acquired waveform data Based on the confirmed observation points, the learning model to be used is selected from each of the learning models that have been generated, and then the obtained waveform data is applied to the selected learning model, and Estimate the epicenter distance. For example, if the epicenter distance estimation device of the second or third item of the patent application is applied, the learning section generates the learning model for each of the site characteristics of the observation points of the waveform data that becomes the input data, and the estimation processing section confirms that the acquired The site characteristics of the observation points of the waveform data are based on the confirmed site characteristics. From among the learning models that have been generated, the learning model to be used is selected, and the obtained learning model is applied to the obtained learning model. This waveform data is used to estimate the epicenter distance. For example, the epicenter distance estimation device of any one of claims 1 to 3 of the patent application scope further includes a waveform pre-processing section that processes waveform data used as input data in the learning section, and The waveform data obtained by the seismic information acquisition unit is subjected to pre-processing. For example, the epicenter distance estimation device for item 9 of the patent application, wherein the waveform pre-processing section executes at least one process from image conversion processing, envelope conversion processing, band-pass conversion processing, differential conversion processing, and Fourier conversion processing as the The pre-processing. An epicenter distance estimation method is characterized by having the following steps: (a) a step of obtaining waveform data of an earthquake that has occurred; (b) applying a learning model obtained by learning the relationship between the waveform data of an earthquake and the epicenter distance, and applying This waveform data has been obtained, and the epicenter distance is estimated. A computer-readable recording medium that records a program, the program includes a command to perform the following steps in the computer: (a) the step of obtaining waveform data of an earthquake that has occurred; (b) the distance between the waveform data and the epicenter of the learned earthquake The learning model obtained from the relationship between them is applied to the obtained waveform data to estimate the epicenter distance.