KR101923166B1 - Method for correcting amplitude magnitude of seismic signal extracted from seismic ambient noise - Google Patents

Abstract

The present invention relates to a method using seismic background noise to extract phase and amplitude information of a seismic signal, and reflecting a spatiotemporal distribution of the seismic background noise to correct the magnitude of an extracted amplitude. According to the present invention, to solve a problem of a conventional technology with limitation, which does not provide a technology capable of extracting both of phase and amplitude information of a seismic signal to utilize only the phase information of the seismic signal, a series of processes are performed by a computer or exclusive hardware, wherein the processes comprise: a process of using a seismic interferometry technology to extract phase and amplitude information of impulse response functions (IRF) from seismic background noise normally recorded in a seismological observatory regardless of seismicity; and a process of reflecting spatiotemporal distribution of the seismic background noise in nine seismic signal components (ZZ, ZR, ZT, RZ, RR, RT, TZ, TR, TT) extracted from the seismic background signal to correct the magnitude of the extracted amplitude omnidirectionally. Accordingly, data extracted from the seismic background noise is effectively applicable to analysis and research of seismic ground motion in a low seismic zone.

Description

[0001] The present invention relates to a method of correcting an amplitude magnitude of a seismic signal extracted using an earthquake background noise,

The present invention relates to a method for extracting phase and amplitude information of an earthquake signal using seismic ambient noise and correcting the extracted amplitude information for use in simulating an earthquake ground motion, and more particularly, A technique for extracting both phase and amplitude information of an earthquake signal can not be provided and seismic interferometry technology is used to solve the problem of the prior art in which only the phase information of the seismic signal is limited. The phase and amplitude information of impulse response functions (IRF) are extracted from seismic background noise, which is regularly recorded at seismic stations regardless of seismic activity, and 9 - component seismic signals ZZ, ZR, ZT, RZ, RR, RT, TZ, TR, and TT) It relates to an azimuth angle (360 °) set of processing the amplitude correction method of seismic signals extracted by the seismic background noise is configured to be performed by the hardware of the computer or only the correction on.

Also, the present invention can efficiently extract the phase and amplitude information of the IRF from the background noise as described above, and can correct the magnitude of the extracted amplitude by reflecting the spatio-temporal distribution of the background noise of the earthquake The present invention relates to a method for correcting an amplitude magnitude of an earthquake signal extracted using an earthquake background noise configured to effectively utilize data extracted from an earthquake background noise even in a low earthquake region for analysis and study of the earthquake ground motion.

Recently, natural disasters such as extreme weather are increasing due to climate change such as global warming. In addition, earthquakes are increasing worldwide, including Korea.

In other words, it is important to anticipate and prepare for the occurrence of earthquakes in advance, as earthquakes can cause more damage to lives and property than other natural disasters. In Korea, there is no large-scale earthquake in the past, Earthquake of 5.8 magnitude occurred in Gyeongju on September 12, 2016 and 5.4 magnitude earthquake occurred in Pohang on November 15, 2017, and the earthquake continued until recently. The risk is increasing.

Therefore, in order to prepare for the increasing risk of earthquakes, there have been various studies to detect or predict the occurrence of earthquakes in the past. It has been suggested.

More specifically, as described above, as an example of the prior art for detecting and predicting earthquake occurrence, for example, according to Korean Patent Registration No. 10-1184382, a free space including a horizontal or slope condition The seismic measuring apparatus and the installation method of the apparatus are provided so as to be able to accurately and accurately measure the results of the earthquake measurement on the ground because the stable and reasonable installation and operation are possible on all the surfaces of the condition.

Also, as an example of the prior art for detecting and predicting the occurrence of an earthquake as described above, for example, in Korean Patent Registration No. 10-1523355 and Korean Patent Registration No. 10-1028779, And a P-wave is detected based on the variation in the time-frequency domain and the threshold value in which the background noise is reflected, thereby to reduce the false detection rate due to the background noise. And a method for detecting the seismic early warning and accurate seismic factor analysis is disclosed in Korean Patent Publication No. 10-1520399 by removing the measurement noise and background noise at the time of collecting the seismic data The present invention relates to an apparatus and a method for removing noise of a seismic wave using micro-positional alignment polymerization.

As described above, various techniques for detecting and predicting the occurrence of the earthquake have been proposed. However, the conventional techniques have the following problems.

In other words, the recent magnitude and frequency of earthquake occurrence has increased the necessity of studies on the earthquake occurrence environment and the characteristics of earthquake disasters on the Korean peninsula. However, since the average seismic activity is low in the Korean peninsula region, There was a problem in that it was relatively insufficient.

If earthquake background noise recorded at seismic stations at any time can be used as seismic observation data regardless of seismic activity, it is expected to help secure the base technology for studying seismic propagation characteristics and crustal structure of the Korean peninsula. The prior art contents of the prior art mainly concentrate on improving the accuracy of seismic detection by separating seismic waves and background noise and eliminating noise. However, it is not considered in construction of seismic observation data necessary for analysis of seismic occurrence There was a problem.

In addition, since the seismic signal extracted based on the background noise of the earthquake is influenced by the seasonal change and the spatial distribution characteristic of the background noise source, understanding the regional and temporal distribution characteristics of the background noise source causes the reliability of the extracted seismic signal However, the above-mentioned prior art contents merely focus on removing the background noise, and the meaning of such an earthquake background noise has not been considered.

In addition, the study on the correlation analysis of the background noise of the earthquake has recognized the possibility of applying the seismic data in the area with the small earthquake, and since the mid 2000s, the interest from the identification of the cause of the background noise to the derivation of the velocity structure based on the observation data However, in the past, only the extracted phase information of the seismic signal has been utilized. Furthermore, the technique capable of extracting the amplitude information in addition to the phase information of the seismic signal has also been studied as a seismic wave propagation characteristic including both the perception speed and the damping structure It is still in its infancy, and no standardization work has yet been achieved.

Furthermore, since the information extracted from the background noise of the earthquake is a state in which the influence of the spatio-temporal distribution of the background noise used in the extraction is already reflected in the relative magnitude of the amplitude according to the pair of the observation station, In addition to extraction from noise, it is required to correct the extracted amplitude magnitude by reflecting the spatio - temporal distribution characteristics of the background noise. However, there is no apparatus or method that satisfies all such requirements yet.

[Prior Art Literature]

1. Korean Registered Patent No. 10-1184382 (2012.09.13.)

2. Korean Patent Registration No. 10-1523355 (May 20, 2015)

3. Korean Patent Registration No. 10-1028779 (April 05, 2011)

4. Korean Patent Registration No. 10-1520399 (Aug.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a technique for extracting both phase information and amplitude information of an earthquake signal, In addition, there is a limitation in the prior art in which an amplitude error magnitude extracted from an earthquake background noise is generated due to the temporal and spatial distribution of an earthquake background noise, but the method of correcting the amplitude is not presented. The phase and amplitude information of impulse response functions (IRF) are extracted from seismic background noise recorded at seismic stations at all times regardless of seismic activity using seismic interferometry, (ZZ, ZR, ZT, RZ, RR, RT, TZ, TR and TT) of the nine components extracted from the background noise A series of processes for correcting the magnitude of the extracted amplitude by 360 ° in accordance with the spatio-temporal distribution of the background noise are extracted by using a computer or dedicated hardware, And to provide a method for correcting amplitude magnitude of an earthquake signal.

It is another object of the present invention to provide a method and apparatus for efficiently extracting phase and amplitude information of an IRF from an earthquake background noise as described above and also to correct the magnitude of the extracted amplitude by reflecting the temporal- The amplitude amplitude correction method of the seismic signal extracted using the earthquake background noise which is constructed so that the data extracted from the background noise of the earthquake region can be utilized effectively for the analysis and study of the earthquake ground motion, .

In order to achieve the above object, according to the present invention, phase information and amplitude information of an earthquake signal are calculated from seismic ambient noise which is always recorded in an earthquake observation station regardless of an earthquake activity, A method for correcting an amplitude magnitude of an earthquake signal extracted by using an earthquake background noise configured to compensate a relative magnitude of amplitude information calculated by reflecting temporal and spatial distribution of an earthquake background noise, A data collecting step of collecting data on an earthquake signal including an earthquake background noise; An impulse response function (IRF) between each seismic station using a seismic interferometry technique is applied to the seismic background noise collected at each seismic station using the seismic signal data collected in the data acquisition step, Extracting an impulse response function; Calculating phase information and amplitude information of the impulse response function (IRF) extracted at the impulse response function extracting step; And an amplitude correction step of performing a correction for the relative amplitude magnitude according to an azimuth angle of a pair of source-receiving station pairs with respect to the amplitude information calculated in the phase and amplitude information calculation step, The amplitude magnitude of the seismic signal extracted by using the seismic background noise is provided.

Here, the method may further include a step of performing a process of displaying the result obtained through the data collection step, the impulse response function extraction step, the phase and amplitude information calculation step, and the amplitude correction step through display means including display And a display step.

The data collection step may include collecting data on seismic signals including seismic background noise recorded at seismic stations installed in each area and storing the collected data for each observation station and period to construct a database Is performed.

The extraction of the impulse response function (IRF) may be performed using a deconvolution method using the following equation. ≪ EMI ID = 1.0 >

Where xr is the receiver station, Xs is the source station, v is the background noise rate, ω is the frequency, i and j are the i-th component of the receiving station, is the complex conjugate, ε is the water level defined as 1% of the average spectral power, F ^-1 is the inverse Fourier transform, ≪> represents time domain stacking)

Moreover, the phase and amplitude information calculating step, using the following equation, the maximum amplitude ratio (PGV _ratio for the causal portion (causal part) of the IRF and ratio widow (acausal part) derived from the impulse response function extraction step ) Is calculated and azimuthal variation according to the position of each seismic station is calculated to calculate the phase information and the amplitude information reflecting the correction of the azimuth according to the position of each seismic station .

(Wherein, PGV _causal maximum amplitude, PGV _acausal the causal portion (causal part) of the extracted IRF represents the maximum amplitude ratio of the widow (acausal part) of each of the extracted IRF)

)을 구하고, (b) 거리감쇠가 보상된 최대진폭값들의 평균(

)을 계산하며, (c) 상기 (b)에서 구해진 평균값에 대한 상기 (a)의 거리감쇠가 보정된 최대진폭값의 비율로부터 방위각에 따른 상대적 진폭을 보정하기 위한 진폭보정계수를 산출하고, (d) 상기 임펄스 응답함수 추출단계에서 추출된 임펄스 응답함수에 상기 (c)에서 산출된 진폭보정계수를 곱하는 것에 의해 소스-수신 관측소쌍의 방위각에 따른 상대적 진폭 크기를 보정하는 처리가 수행되도록 구성되는 것을 특징으로 한다. The amplitude correction step may include: (a) multiplying the maximum amplitude value (PGV _noise ) of the impulse response function (IRF) extracted in the step of extracting the impulse response function by a distance R between seismic stations, Amplitude value (

(B) the average of the maximum amplitude values at which the attenuation is compensated (

(C) calculating an amplitude correction coefficient for correcting the relative amplitude according to the azimuth angle from the ratio of the maximum amplitude value corrected by the distance attenuation of (a) to the average value obtained in (b) d) a process of correcting the relative amplitude magnitude according to the azimuth angle of the source-receiving station pair by multiplying the impulse response function extracted in the impulse response function extraction step with the amplitude correction coefficient calculated in (c) .

The amplitude correction step may further include a step of calculating an earthquake signal from the azimuth x and the maximum amplitude value PGV _noise of the seismic signal (causal part of the impulse response function) A regression equation for obtaining a ratio value y for correcting the relative amplitude magnitude of the input signal is defined as a mathematical expression including a trigonometric function using the following equation,

(Where y is a ratio value for correcting the relative amplitude magnitude of the seismic signal, x is the azimuth angle, and a, p, and b are the regression coefficients)

)과 거리감쇠가 보상된 최대진폭값의 평균(

)으로부터 산출되는 값이 상기 회귀식을 만족하도록 하는 회귀계수(a, p, b)를 회귀분석을 통하여 각각 도출하는 것에 의해 상기 진폭보정계수(y)를 산출하는 처리가 수행되도록 구성되는 것을 특징으로 한다. Using the following equation, the distance attenuation is calculated as the compensated maximum amplitude value (

) And the average of the maximum amplitude values at which the attenuation is compensated (

(A, p, b) for calculating the amplitude correction coefficient (y) by calculating a regression coefficient (a, p, b) .

은 거리감쇠가 보상된 최대진폭값,

은 거리감쇠가 보상된 최대진폭값의 평균을 각각 나타냄) (Where PGV _noise is the maximum amplitude of the extracted IRF, R is the distance between seismic stations,

Is the maximum amplitude value at which the attenuation is compensated,

Represents the average of the maximum amplitude values at which the attenuation is compensated)

Further, in the step (d), the amplitude correction step may be configured to perform a process of correcting a relative amplitude magnitude according to an azimuth angle of the source-receiving station pair using the following equation.

(Where IRF _corrected is the impulse response function after the amplitude correction according to the azimuth change, IRF is the impulse response function before the amplitude correction according to the azimuth change, and f (x) is the ratio value for the amplitude correction according to the azimuth angle, respectively )

According to another aspect of the present invention, there is provided an earthquake signal analysis method comprising: receiving an earthquake signal from each earthquake observation station; And analyzing the seismic signal received in the seismic signal receiving step, wherein the step of analyzing the seismic signal comprises the step of analyzing the amplitude magnitude of the seismic signal extracted using the seismic background noise described above, Wherein the analysis of the seismic signal is performed using the method of the present invention.

According to another aspect of the present invention, there is provided an earthquake analysis system for analyzing an earthquake signal, comprising: an earthquake signal receiving unit for receiving an earthquake signal from each earthquake observing station; And an earthquake signal analyzing unit for analyzing the earthquake signal received through the earthquake signal receiving unit, wherein the earthquake signal analyzing unit analyzes an amplitude amplitude correction method of the earthquake signal extracted using the earthquake background noise described above And an analysis of the seismic signal is performed using the analysis result.

As described above, according to the present invention, the seismic interferometry technique can be used to detect the phase of impulse response functions (IRF) from seismic background noise, which is regularly recorded at seismic stations regardless of seismic activity, And the amplitude information is extracted and a series of processes for correcting the magnitude of the extracted amplitude reflecting the spatio-temporal distribution of the background noise are performed by a computer or dedicated hardware, and the extracted earthquake background noise There is a limitation in using only the phase information of the seismic signal because the technique of extracting both the phase information and the amplitude information of the seismic signal is not provided by the provision of the amplitude magnitude correction method of the seismic signal, The amplitude magnitude extracted from the background noise of the earthquake caused an error. However, It is possible to solve all of the problems of the prior art in which there is a limitation in not providing a method of correcting the error.

According to the present invention, the phase and amplitude information of the IRF can be efficiently extracted from the background noise of the earthquake as described above, and the nine-component seismic signals (ZZ, ZR, ZT (RZ, RR, RT, TZ, TR, TT) using the seismic background noise, which is constructed to correct the magnitude of the extracted amplitude in relation to the omnidirectional angle By providing the amplitude amplitude correction method of the extracted seismic signal, it is possible to contribute to the analysis and study of the seismic ground motion using the data extracted from the background noise even for the low seismic region.

FIG. 1 is a view schematically showing broadband observation stations used in the earthquake interferometry method and amplitude amplitude correction method of the present invention in the Korea Institute of Geoscience & Mineral Resources (KIGAM) and Korea Meteorological Administration (KMA) network.
FIG. 2 is a schematic view showing the location of a broadband observation station observed and recorded at a mine collapse event occurring at Samcheok at 08:53 am on Jan. 31, 2015.
FIG. 3 is a diagram showing the results of moment tensor inversion for a mine collapse event occurring in Samcheok at 08:53 am on Jan. 31, 2015. FIG.
FIG. 4 is a graphical representation of trends in winter (from December 2015 to February 2016), spring (March 2016 to May 2016) and summer (June 2016 to 2016) And a seasonal change of the source distribution of the background noise from the viewpoint of the PGV _ratio for one year from September 2015 to August 2016.
5 is a graph showing the azimuthal dependence of the background noise source distribution in the Korean peninsula region when the KNUD observatory is used as a transmission station.
Figure 6 shows the relative amplitude magnitude of the seismic signal for the nine component seismic signals (ZZ, ZR, ZT, RZ, RR, RT, TZ, TR, TT) extracted from the seismic background noise according to the azimuth angle of the pair of stations. And a triangular function for correcting for an omnidirectional angle (360 deg.).
FIG. 7 is a diagram comparing vertical, radial and tangential components of an event record with an IRF whose amplitude magnitude is corrected using a deconvolution-based seismic interferometry technique for a virtual source observatory (KNUD).
FIG. 8 is a flowchart schematically illustrating a general configuration of a method of extracting an earthquake signal using an earthquake background noise according to an embodiment of the present invention and correcting the amplitude magnitude of the earthquake signal.

Hereinafter, a method of correcting amplitude magnitude of an earthquake signal extracted using an earthquake background noise according to the present invention will be described with reference to the accompanying drawings.

Hereinafter, it is to be noted that the following description is only an embodiment for carrying out the present invention, and the present invention is not limited to the contents of the embodiments described below.

In the following description of the embodiments of the present invention, parts that are the same as or similar to those of the prior art, or which can be easily understood and practiced by a person skilled in the art, It is important to bear in mind that we omit.

Next, with reference to the drawings, the details of the amplitude magnitude correction method of the seismic signal extracted using the earthquake background noise according to the present invention will be described.

1 to 3, there is shown schematically a broadband observation station applied to the seismic interferometry method and the amplitude magnitude correction method of the present invention in the KIGAM (Korea Institute of Geoscience and Mineral Resources) and the Korea Meteorological Administration (KMA) network And Figs. 2 and 3 are diagrams showing the results of the moment tensor inversion for the collapse event and the location of the broadband observatory where the mine collapse event occurred in Samcheok at 08:53 am on Jan. 31, 2015 .

More particularly, the present invention relates to a method of extracting an earthquake signal using an earthquake background noise, and more particularly, to a method of extracting an earthquake signal by using broadband Rayleigh waves and their dispersion characteristics using a cross correlation analysis It has been shown that it can be extracted from ambient noise, which is also known as seismic interferometry. Conventionally, a short-period surface wave group measurement velocity measurements were used to construct high-resolution tomographic images of California.

It is also possible to apply the impulse response sum (IRF) extracted from an earthquake background noise to predict a strong ground motion, that is, from a deconvolution of an earthquake background noise region to a complex underground structure it is possible to extract reliable phase and amplitude information for the IRF, including the influence of subsurface structures.

In addition, we apply both depth and focal mechanism corrections to earthquakes and extend these methods to simulate ground motion in southern California and Japan's moderate offshore subduction earthquake And the long-term ground motion simulations generated by these simulations.

Here, the seismic interferometry technique can be classified into three types of deconvolution, coherency, and cross correlation. First, the deconvolution is applied to the seismic wave amplification It is used to simulate long-term ground motion caused by research and large earthquakes.

In addition, the coherence technique is applied to recover the anelastic attenuation estimates. In order to numerically investigate the recovery of the IRF attenuation, conventionally, the spatial distribution and attenuation of the source are identified, The theoretical basis for calculating the crosscorrelation that represents the quantification of the amplitude is presented. The amplification effect of the elastic subsurface structure, the crust and the damping effect of the upper mantle and the building As shown in FIG.

As described above, the background noise technique is widely used to identify the velocity structure of the earth, in particular, the crustal structure of the Korean peninsula, and the seismic interferometry method has a high-magnitude earthquake activity, , It is particularly useful in low-seismicity regions, but high-density seismic networks are required to record background noise.

Accordingly, in the present invention, a method of efficiently extracting the phase and amplitude information of the IRF from the background noise data of the low seismic region through the excited ground motion by the limestone mine collapse event (Mw 4.2) in Korea Respectively.

More specifically, a limestone mine collapse occurred in Samcheok at 08:53 am on Jan. 31, 2015 as a Korean standard time. As shown in FIG. 1, the Korea Institute of Geoscience and Mineral Resources KIGAM) and the Korea Meteorological Administration (KMA), respectively, were recorded for each of the broadband stations.

There were no casualties since the mine had been abandoned a few years before the incident occurred, and by the Korea Institute of Geoscience and Mineral Resources, there are five 3-component regional broadband The moment tensor solution was inverted from the data of the recordings and the decay occurred at a depth of several hundred meters as shown in Fig. 3 was evaluated as being similar to the negative tensile crack .

In other words, when a shallow-depth collapse occurs, a singular vertical force can be approximated by a tensile fault, and this model assumes that the source depth is shallower than the wavelength, It is valid only when recorded by remote observation, and the limestone mine collapse event applied to the present invention satisfies both of these conditions.

Here, from a theoretical point of view, a centroid single force inversion may be more effective in understanding the source mechanism, but with the contents shown in FIG. 3, the source mechanism of the event is equivalent to a single downward force .

In addition, the field survey revealed that the collapsed area corresponding to a moment magnitude (Mw) of 4.2 was spread over a width of several hundred meters, and the time for mine collapse was estimated to be within a few seconds, IRF can be modeled as a short-term impulse for the target period band and the IRF extracted using the seismic interferometry can be directly compared to the event record without depth or mechanism correction.

More specifically, the seismic interferometry technique can be applied according to a method of correlation analysis on background noise data. Conventionally, in order to extract IRF from background noise data, the power of source station data Deconvolution of normalized receiver station data using the spectrum has been applied.

More specifically, the deconvolution method can be expressed as Equation (1) below, which shows the background noise velocity v for two seismic stations located at Xr (receiver station) and Xs Is performed in the frequency domain omega with simultaneous data segments representing < RTI ID = 0.0 >

[Equation 1]

In Equation (1), i and j are the i-th component of the receiving station and the j-th component of the virtual source station, respectively, and * denotes a complex conjugate.

In Equation (1), the numerator corresponds to the cross-correlation of the simultaneous background noise data segments in the frequency domain, and is normalized using the power spectrum of the source station.

In order to maintain the stability of deconvolution during the numerical calculation in the above Equation 1, in the present invention, the water level? Defined as 1% of the average spectral power F ^-1 indicates an inverse Fourier transform, parentheses <> indicate time domain stacking, and the time-domain stacking over a long time series increases the signal-to- .

In the present invention, as shown in FIG. 1, an IRF is extracted using a background noise seismic interferometry using a KNUD observation station as a virtual source observation station, a KIGAM and a KMA network as a reception station, The seismic background noise data was collected over a period of one year from September 2015 to August 2016.

Also, prior to applying the seismic interferometry, successive recordings of background noise data are divided into nonoverlapping time segments of 1 hour, raw data being instrumental responses And the trend and mean are removed from the initial noise data using a Seismic Analysis Code. ≪ RTI ID = 0.0 > [0031] < / RTI &

In addition, all time segments are downsampled to 10 samples per second to reduce the computation of the data processing, and segments having spikes greater than five times the standard deviation of one hour long data to minimize the effects of potential seismic events Is discarded.

Furthermore, the three components of the receiver station background noise data are correlated with the polarity-changed vertical components of the virtual source station background noise data to identify the downward direction of the mine collapse, After the background noise time series data is transformed using the transform (FFT), the correlation of each segment in the frequency domain is calculated.

Thereafter, the inverse FFT is applied to the correlated result to accumulate the time series segments over a period of one year in the time domain, and the azimuth of the receiving station with respect to the virtual source station (KNUD) The horizontal components ZN and ZE of the rotated IRF are rotated to radial and tangential components and the causal part of the rotated IRF ) Is regarded as a source-to-arrival impulse response with a source mechanism consisting of a single downward force and the IRF extracted using the seismic interferometry technique is considered to be a ground motion observed for the mine collapse event The data are compared in a period of 2 to 10 seconds using a fifth-order Butterworth filter.

In order to compare these seismic interferometry results, the amplitude of the IRF is calculated for each of the three components (vertical, radial and tangential components) in order to preserve the relative amplitude information of the receiving station, Is calculated by the ratio of the average peak amplitude of the IRF to the ground motion data recorded with a constant scaling factor for each station, which applies equally to all stations.

Subsequently, to determine the uncertainty of the origin time and focal depth estimates, a constant temporal shift factor is applied to the ground motion data recorded for each receiving station, , And KIGAM's automatic seismic detection system. For the most effective waveform comparison, this time shift was set to 2 seconds according to trial and error method.

Further, in the present invention, the causal part (positive delay) of the extracted IRF is the source-receiving station impulse response, and the acausal part (negative lag) It is assumed that the source of the background noise is not uniformly distributed in the direction of the source, but the distribution of the background source of the background noise in the Korean peninsula is not yet known.

In order to confirm the difference and the level of the spatial distribution of the background noise before correcting the relative amplitude difference according to the azimuth angle, in the present invention, as shown in the following formula (2) Peak Ground Velocity (PGV) and anti-causal part (negative lag) in the causal part (positive lag) of the impulse response function (IRF) The azimuthal variation was investigated by using the peak ground velocity ratio (PGV _ratio ), which is the _ratio of the maximum amplitude (PGV)

&Quot; (2) "

4, there is shown a graph showing the results of a comparison between the autumn (September 2015 to November 2015), winter (December 2015 to February 2016), spring (March 2016 to May 2016) Is a diagram showing the azimuth and seasonal variations of the source distribution of background noise in terms of PGV _ratio for one year from September 2015 to August 2016, including summer (June 2016 to August 2016) .

As shown in FIG. 4, a significant azimuthal change and a relatively weak seasonal variation of the background noise source level are observed together.

More specifically, referring to FIG. 5, FIG. 5 is a diagram illustrating an azimuth dependency of a background noise source distribution in a Korean peninsula region when a KNUD station is used as a transmission station.

In Fig. 5, the causal part (blue) and the non-widow part (red) of IRF (vertical component) obtained together with the name of the station are displayed on the left side of each panel, and the numbers under the name of the station are displayed on the virtual source station KNUD And the number on the right side of each waveform is a peak ground velocity ratio (PGV _ratio ) defined in Equation 2, and a positive PGV _ratio value is defined as a value obtained by subtracting the PGV of the causal part from the non- It means strength, and vice versa.

As shown in FIG. 5, it can be seen that the causal part (blue) generally has a larger PGV when the receiving station is located to the north than the source station. In order to extract both the amplitude and phase information of the IRF, It is important to consider the temporal and spatial variation of the distribution.

Accordingly, in the present invention, to accumulate waveforms for one year in order to average seasonal variations, data is generated, and amplitude correction is applied to the azimuth angle to identify changes in the azimuth angle.

6, the earthquake signals (ZZ, ZR, ZT, RZ, RR, RT, TZ, TR, TT) of the nine components extracted from the background noise of the earthquake And a trigonometric function for correcting the magnitude of the relative amplitude of the seismic signal with respect to the omnidirectional angle (360 deg.).

More specifically, as shown in FIGS. 4 and 5, the temporal and spatial distribution characteristics of seismic background noise can be confirmed from the PGV _ratio , and the impulse response functions extracted from the source- Delay) corresponds to the earthquake ground motion signal between source (transmission) and receiving station.

Since the impulse response function extracted from the background noise of the earthquake has already affected the relative magnitude of the amplitude of the earthquake background noise used in the extraction to the pair of the station, In order to be able to use it for ground motion simulation, the relative amplitude magnitude of the seismic signal should be corrected by reflecting the spatio - temporal distribution characteristics.

That is, according to the results of the spatiotemporal distribution characteristic analysis confirmed from the above description with reference to FIG. 4 and FIG. 5, it is confirmed that the distribution of the earthquake background noise effectively differs according to the azimuth angle of the pair of stations, Based on these results, a function for correcting the magnitude of the relative amplitude of the seismic signal according to the azimuth angle of the pair of observation stations was derived by the following method.

At this time, the azimuth must be paired with the location of the station where it can evenly cover 360 degrees.

)으로부터 삼각함수를 포함하는 형태의 회귀식을 도출하기 위해 이하의 [수학식 3]과 같은 관계식을 가정한다. More specifically, the relative amplitude correction function is obtained in a regression form from the azimuth along the pair of observations and the maximum amplitude value (PGV _noise ) of the seismic signal (causal part of the impulse response function) Maximum amplitude value (

) To derive a regression equation of a form including a trigonometric function, as shown in the following Equation (3).

&Quot; (3) "

In Equation (3), y represents a ratio value for correcting the relative amplitude magnitude of the seismic signal, x represents an azimuth angle, and a, p and b represent regression coefficients.

In order to take an azimuth angle as an x value to be used in the regression equation, it is appropriate to assume a relational expression in the form of a trigonometric function as in Equation (3), and the y value is a ratio for correcting the relative amplitude magnitude of the seismic signal, Variable.

)들로부터 구성되는 [수학식 4]의 값이 방위각(x)값들과 함께 [수학식 3]의 회귀식을 만족하도록 하는 회귀계수 a, p, b를 각각 도출한다. Further, by using the following expression (4), the distance attenuation is calculated by using the compensated maximum amplitude value (

(X, y) and the azimuth angle (x) values of the equation (4), which are constituted by the azimuth angle?

&Quot; (4) "

은 거리감쇠가 보상된 최대진폭값,

는 거리감쇠가 보상된 진폭값의 평균을 각각 나타낸다. In Equation (4) above,

Is the maximum amplitude value at which the attenuation is compensated,

Represents the average of the amplitude values at which the attenuation is compensated.

That is, Equation (4) represents the maximum amplitude value compensated by the distance attenuation and a ratio of the compensated maximum amplitude value to the values. The value of greater than 0 indicates that the earthquake signal extracted from the background noise of the background is the temporal and spatial distribution characteristic The amplitude of the extracted seismic signal is smaller than the amplitude to be implemented as an impulse response function due to the spatio-temporal distribution characteristic. In other words, do.

Therefore, the regression coefficients a, p, and b are derived to reconstruct the y value of Equation (3) according to the azimuth angle (x) based on the ratio value of Equation (4).

) 값의 집합)를 만족할 수 있는 3개의 변수(a, p, b)를 회귀분석을 통하여 도출하는 것에 의해 구해질 수 있으며, 이때, 상기한 회귀식에 대한 회귀분석을 위해 사용되는 데이터((x, y) 값의 집합)는 가능한 전방위각(360°)을 포함할 수 있는 관측소쌍으로 구성되는 것이 바람직하다. That is, the regression coefficient (a, p, b) satisfying the above-described expression (3) is obtained by multiplying a plurality of data ((x, y; x value is an azimuth angle,

(A, p, b) that can satisfy the above-described regression equation (a set of values) (a, p, b) x, y) values) is preferably made up of a pair of stations that may include a possible omni-directional angle (360 deg.).

Here, as described above, more detailed contents of the process of obtaining the three regression coefficients a, p, and b through the regression analysis are obvious to those skilled in the art through the literature of the prior art. Accordingly, It should be noted that the detailed description of the contents that can be readily understood and practiced by those skilled in the art through the prior art documents and the like is omitted.

In addition, the seismic signals (ZZ, ZR, ZT, RZ, RR, RT, TZ, TR, TT) of nine components extracted from the temporal and spatial distribution of the background noise The regression equation obtained in the form of FIG. 6 is as shown in FIG.

More specifically, when a seismic signal extracted from an earthquake background noise is corrected, a ratio value for relative amplitude correction to an azimuth angle (x) from Equation (3) using a regression coefficient derived through the above- (f (x)) and corrects it according to the following formula (5).

&Quot; (5) "

In Equation (5), IRF _corrected is the impulse response function after the amplitude correction according to the azimuthal change, IRF is the impulse response function before the amplitude correction according to the azimuthal change, and f (x) Respectively.

Since the IRF in the amplitude correction function expressed in the form of [Equation 3] includes the relative amplification of the amplitude according to the azimuth angle of the impulse response function extracted from the background noise, the ratio value f (x ) Must be corrected to a form that is multiplied by 10- ^{f (x)} as in Equation (5 ⁾ .

)을 구하고, (2) (1)에서 거리감쇠가 보상된 최대진폭값들의 평균(

)을 계산하며, (3) [수학식 3] 및 [수학식 4]에 나타낸 바와 같이 하여, (2)의 평균값에 대한 (1)의 거리감쇠가 보정된 최대진폭값의 비율로부터 방위각(x)에 따른 상대적 진폭 보정을 위한 비율값(y)의 관계를 상기한 [수학식 3]과 같은 삼각함수를 포함하는 형태의 회귀식으로 도출하고, (4) [수학식 5]에 나타낸 바와 같이, 관측소 쌍에 따라 추출된 임펄스 응답함수에 대해 (3)의 회귀식을 만족하는 진폭보정계수(10^-f(x))를 곱하는 것에 의해 방위각에 따른 상대적인 진폭 크기의 보정이 수행될 수 있다. That is, as shown in FIG. 6 and the above-mentioned equations (3) to (5), the method of correcting the amplitude according to the azimuthal angle change is as follows. (1) From the extracted impulse response function, Maximum amplitude value (

(2) In (1), the distance attenuation is the average of the compensated maximum amplitude values

(3) From the ratio of the maximum amplitude value corrected for the distance attenuation of (1) to the average value of (2) as shown in [Expression 3 and Expression 4], the azimuth angle x (Y) for the relative amplitude correction according to the above equation (3) is derived by a regression equation of a form including a trigonometric function as in the above-mentioned equation (3), and (4) , A correction of the amplitude magnitude relative to the azimuth angle can be performed by multiplying the extracted impulse response function according to the pair of observation stations by the amplitude correction coefficient 10 ^{-f (x)} satisfying the regression equation (3).

Next, the inventors compared the waveforms of the IRF obtained using the ground motion recorded from the mine collapse event and the seismic interferometry technique of the deconvolution type, and the vertical, radial and tangential components are shown in FIG.

7, FIG. 7 is extracted using the deconvolution seismic interferometry technique for the virtual source observatory KNUD and is shown in FIG. 6 and the above equations (3) to (5) And the vertical, radial and tangential components of the size-corrected IRF (red) and the event recording (gray).

Therefore, it is possible to implement a method of calibrating amplitude magnitude of an earthquake signal extracted using the earthquake background noise according to the present invention based on the above-mentioned contents. That is, referring to FIG. 8, FIG. FIG. 3 is a flowchart schematically showing an overall configuration of an amplitude magnitude correction method of an earthquake signal extracted using an earthquake background noise; FIG.

8, an amplitude magnitude correction method of an earthquake signal extracted by using an earthquake background noise according to an embodiment of the present invention is roughly divided into an earthquake ground noise estimation method using an earthquake background noise, (S10) for collecting data on a signal and constructing a database, and a data collection step (S10) for collecting data on a seismic background noise signal collected at each seismic observation station by using a seismic interferometry technique of a deconvolution method, An impulse response function extraction step (S20) for extracting a response function (IRF), a phase and amplitude information calculation step (S30) for calculating phase and amplitude information of the seismic signal from the extracted impulse response function, And an amplitude correction step (S40) for correcting a relative amplitude magnitude according to an azimuth angle of the pair of source-receiver stations, Or may be configured to be performed by dedicated hardware.

1, the above-described database building step (S10) collects seismic signal data including seismic background noise from KIGAM and KMA from regular seismic stations operated in the Korean peninsula region , Preprocessing processes such as removing trends and averages from the collected seismic signal data and correcting device characteristics, and then arranging and storing the preprocessed data according to each observation station and period to construct a database.

In addition, the impulse response function extraction step S20 may be configured to extract an impulse response function IRF between each observation station by applying the deconvolution method as described above with reference to Equation (1) have.

The phase and amplitude information extracting step S30 and the amplitude correcting step S40 may be performed by using the impulse response function (S30) and the amplitude correction step S40 as described above with reference to Equations (3) to (5) And calculates the ratio value for correcting the relative amplitude magnitude according to the azimuth angle of the source-receiving station pair with respect to the calculated amplitude information, And may be configured to calculate an impulse response function (IRF _corrected ).

Furthermore, the above-described method may further comprise a result display step of displaying the results obtained through the above-described steps through display means such as a display.

Therefore, the amplitude magnitude correction method of the seismic signal extracted using the earthquake background noise according to the present invention as described above can be implemented.

According to another aspect of the present invention, there is provided an apparatus for analyzing an earthquake signal using an amplitude magnitude correction method for an earthquake signal extracted using an earthquake background noise according to the present invention, It is possible to provide an earthquake signal analysis method and an earthquake analysis system configured to be capable of analyzing an earthquake signal.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present invention as defined by the appended claims. It is a matter of course.

Claims (10)

The phase information and the amplitude information of the seismic signal are calculated from the seismic ambient noise which is always recorded in the seismic observation station regardless of the seismic activity and the relative information of the amplitude information calculated by reflecting the spatio- A method for correcting an amplitude magnitude of an earthquake signal extracted using an earthquake background noise configured to correct a magnitude,
A data collecting step of collecting data on an earthquake signal including an earthquake background noise recorded at a seismic observing station installed in each area;
An impulse response function (IRF) between each seismic station using a seismic interferometry technique is applied to the seismic background noise collected at each seismic station using the seismic signal data collected in the data acquisition step, Extracting an impulse response function;
Calculating phase information and amplitude information of the impulse response function (IRF) extracted at the impulse response function extracting step; And
And a process of correcting the relative amplitude magnitude according to the azimuth angle of the pair of source-reception station pairs with respect to the amplitude information calculated in the phase and amplitude information calculation step is performed is performed by a computer or a dedicated hardware The amplitude magnitude of the seismic signal extracted using the earthquake background noise.
The method according to claim 1,
The method comprises:
Further comprising a result displaying step in which a process of displaying the results obtained through the data collecting step, the impulse response function extracting step, the phase and amplitude information calculating step and the amplitude correcting step through display means including a display is performed Wherein the amplitude magnitude of the seismic signal extracted using the seismic background noise is corrected.
The method according to claim 1,
Wherein the data collection step comprises:
A process for collecting data on an earthquake signal including an earthquake background noise recorded at an earthquake observing station installed in each area and storing the collected data for each observation station and period to construct a database is performed A method for calibrating amplitude magnitude of seismic signals extracted by using background noise.
The method according to claim 1,
Wherein the extracting of the impulse response function comprises:
Wherein the processing of extracting the impulse response function (IRF) using a deconvolution method is performed using the following equation: < EMI ID = Size correction method.

Where xr is the receiver station, Xs is the source station, v is the background noise rate, ω is the frequency, i and j are the i-th component of the receiving station, is the complex conjugate, ε is the water level defined as 1% of the average spectral power, F ^-1 is the inverse Fourier transform, ≪> represents time domain stacking)
The method according to claim 1,
Wherein the phase and amplitude information calculation step comprises:
Using the following equation, to calculate the maximum amplitude ratio (PGV _ratio) for the causal portion (causal part) of the IRF and ratio widow (acausal part) derived from the impulse response function, the extraction step according to the positions of the respective seismic stations And calculating phase information and amplitude information reflecting the correction of azimuth angle according to the position of each seismic observation station by calculating an azimuthal variation. A method for calibrating the magnitude of an earthquake signal.

(Wherein, PGV _causal maximum amplitude, PGV _acausal the causal portion (causal part) of the extracted IRF represents the maximum amplitude ratio of the widow (acausal part) of each of the extracted IRF)
The method according to claim 1,
Wherein the amplitude correction step comprises:
(a) multiplying the maximum amplitude value (PGV _noise ) of the impulse response function (IRF) extracted in the step of extracting the impulse response function by the distance (R) between the seismic observing stations to obtain a maximum amplitude value

),
(b) the average of the maximum amplitude values at which the attenuation is compensated (

),
(c) calculating an amplitude correction coefficient for correcting a relative amplitude with respect to an azimuth angle from a ratio of the maximum amplitude value corrected by the distance attenuation of (a) to the average value obtained in (b)
(d) multiplying the impulse response function extracted in the impulse response function extraction step by the amplitude correction coefficient calculated in (c) to perform a process of correcting the relative amplitude magnitude according to the azimuth angle of the source- The amplitude magnitude of the earthquake signal extracted using the earthquake background noise.
The method according to claim 6,
Wherein the amplitude correction step comprises:
In (c), to correct the magnitude of the relative amplitude of the seismic signal from the azimuth (x) according to the source-receiver station pair and the maximum amplitude value (PGV _noise ) of the seismic signal (causal part of the impulse response function) A regression equation for obtaining the ratio value (y) is defined by a mathematical expression including a trigonometric function using the following equation,

(Where y is a ratio value for correcting the relative amplitude magnitude of the seismic signal, x is the azimuth angle, and a, p, and b are the regression coefficients)

Using the following equation, the distance attenuation is calculated as the compensated maximum amplitude value (

) And the average of the maximum amplitude values at which the attenuation is compensated (

(A, p, b) for calculating the amplitude correction coefficient (y) by calculating a regression coefficient (a, p, b) A method for calibrating amplitude magnitude of an earthquake signal extracted using an earthquake background noise.

(Where PGV _noise is the maximum amplitude of the extracted IRF, R is the distance between seismic stations,

Is the maximum amplitude value at which the attenuation is compensated,

Represents the average of the maximum amplitude values at which the attenuation is compensated)
8. The method of claim 7,
Wherein the amplitude correction step comprises:
(D), the process of correcting the relative amplitude magnitude according to the azimuth angle of the source-receiving station pair is performed using the following equation: < EMI ID = Size correction method.

(Where IRF _corrected is the impulse response function after the amplitude correction according to the azimuth change, IRF is the impulse response function before the amplitude correction according to the azimuth change, and f (x) is the ratio value for the amplitude correction according to the azimuth angle, respectively )
In an earthquake signal analysis method,
Receiving an earthquake signal from each of the seismic stations; And
And an earthquake signal analysis step of analyzing the earthquake signal received in the earthquake signal receiving step,
In the seismic signal analysis step,
A method of analyzing an earthquake signal using the amplitude amplitude correction method of an earthquake signal extracted using the earthquake background noise according to any one of claims 1 to 8.
1. An earthquake analysis system for performing analysis of an earthquake signal,
An earthquake signal receiver for receiving an earthquake signal from each of the seismic stations; And
And an earthquake signal analyzer for analyzing the earthquake signal received through the earthquake signal receiver,
Wherein the seismic signal analyzer comprises:
The earthquake analysis system according to claim 1, wherein the analysis of the seismic signal is performed using the magnitude amplitude correction method of the seismic signal extracted using the earthquake background noise according to any one of claims 1 to 8.

Images