• 한국지구물리탐사학회:학술대회논문집
• /
• 2006.06a
• /
• pp.181-186
• /
• 2006

An imaging method of seismic sources using time-reversal wave propagation is presented. The method is based on the time-reversal invariance and the spatial reciprocity of the wave equation. Time-reversal wave propagation has been used to image anomalous features of a midium in medical imaging, non destructive testing and waveform tomography. Seismogram is the record whose energy is propagated from the seismic source. If time-reversed seismogram propagates back into the medium, seismic energy is concentrated at the origin time of the event and at the source location. In this work, a staggered-grid finite-difference method of the elastic wave equation is parallelized for 3-D wave propagation simulation. With numerical experiments, we show that the time-reversal imaging will enable us to explore the spatio-temporal history of complex earthquake.

• Geophysics and Geophysical Exploration
• /
• v.24 no.2
• /
• pp.53-66
• /
• 2021

Velocity model building is an essential procedure in seismic data processing. Conventional techniques, such as traveltime tomography or velocity analysis take longer computational time to predict a single velocity model and the quality of the inversion results is highly dependent on human expertise. Full-waveform inversions also depend on an accurate initial model. Recently, deep neural network techniques are gaining widespread acceptance due to an increase in their integration to solving complex and nonlinear problems. This study investigated cases of seismic velocity model building using deep neural network techniques by classifying items according to the neural networks used in each study. We also included cases of generating training synthetic velocity models. Deep neural networks automatically optimize model parameters by training neural networks from large amounts of data. Thus, less human interaction is involved in the quality of the inversion results compared to that of conventional techniques and the computational cost of predicting a single velocity model after training is negligible. Additionally, unlike full-waveform inversions, the initial velocity model is not required. Several studies have demonstrated that deep neural network techniques achieve outstanding performance not only in computational cost but also in inversion results. Based on the research results, we analyzed and discussed the characteristics of deep neural network techniques for building velocity models.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.24 no.2
• /
• pp.23-32
• /
• 2020

The compression sensing technology, CAFB, was developed to obtain the raw signal of the target structure by compressing it into a signal of the intended frequency range. At this point, for compression sensing, the CAFB can be optimized for various reference signals depending on the desired frequency range of the target structure. In addition, optimized CAFB should be able to efficiently compress the effective structural answers of the target structure even in sudden/dangerous conditions such as earthquakes. In this paper, the targeted frequency range for efficient structural integrity monitoring of relatively flexible structures was set below 10Hz, and the optimization method of CAFB for this purpose and the seismic response performance of CAFB in seismic conditions were evaluated experimentally. To this end, in this paper, CAFB was first optimized using Kobe seismic waveform, and embedded it in its own wireless IDAQ system. In addition, seismic response tests were conducted on two span bridges using Kobe seismic waveform. Finally, using an IDAQ system with built-in CAFB, the seismic response of the two-span bridge was wirelessly obtained, and the compression signal obtained was cross-referenced with the raw signal. From the results of the experiment, the compression signal showed excellent response performance and data compression effects in relation to the raw signal, and CAFB was able to effectively compress and sensitize the effective structural response of the structure even in seismic situations. Finally, in this paper, the optimization method of CAFB was presented to suit the intended frequency range (less than 10Hz), and CAFB proved to be an economical and efficient data compression sensing technology for instrumentation-monitoring of seismic conditions.

• The Journal of Engineering Geology
• /
• v.19 no.4
• /
• pp.509-515
• /
• 2009

Three earthquakes with local magnitude ($M_L$) greater than 3.0 occurred on April 24, June 2 and September 12 in 1999 nearby the Gyeongju area. Redetermined epicenters were located within the radius of 1 km. We carried out waveform inversion analysis to estimate focal mechanism of June 2 event, and P and S wave polarity and their amplitude ratio analysis to estimate focal mechanisms of April 24 and September 12 events. June 2 and September 12 events had similar fault plane solutions each other. The fault plane solution of April 24 event included those of other 2 events, but its distribution range was relatively broad. Focal mechanisms of those events had a strike slip faulting with a small normal component. P-axes of those events were ENE-WSW which were similar to previous studies on the P-axis of the Korean Peninsula. Considering distances between epicenters, similarities of seismic waves and sameness of polarities of seismic data recorded at common seismic stations, these events might occurred at the same fault. The seismic moment of June 2 event was estimated to be $3.9\;{\times}\;10^{14}\;N{\cdot}m$ and this value corresponded to the moment magnitude ($M_W$) 3.7. The moment magnitude estimated by spectral analysis was 3.8, which was similar to that estimated by waveform inversion analysis. The average stress drop was estimated to be 7.5 MPa. Moment magnitudes of April 24 and September 12 events were estimated to be 3.2 and 3.4 by comparing the spectrum of those events recorded at common single seismic station.

• Journal of the Korean earth science society
• /
• v.37 no.5
• /
• pp.277-285
• /
• 2016

We introduce a depth scaling strategy to improve the accuracy of frequency-domain elastic full waveform inversion (FWI) using the new pseudo-Hessian matrix for seismic data without low-frequency components. The depth scaling strategy is based on the fact that the damping factor in the Levenberg-Marquardt method controls the energy concentration in the gradient. In other words, a large damping factor makes the Levenberg-Marquardt method similar to the steepest-descent method, by which shallow structures are mainly recovered. With a small damping factor, the Levenberg-Marquardt method becomes similar to the Gauss-Newton methods by which we can resolve deep structures as well as shallow structures. In our depth scaling strategy, a large damping factor is used in the early stage and then decreases automatically with the trend of error as the iteration goes on. With the depth scaling strategy, we can gradually move the parameter-searching region from shallow to deep parts. This flexible damping factor plays a role in retarding the model parameter update for shallow parts and mainly inverting deeper parts in the later stage of inversion. By doing so, we can improve deep parts in inversion results. The depth scaling strategy is applied to synthetic data without lowfrequency components for a modified version of the SEG/EAGE overthrust model. Numerical examples show that the flexible damping factor yields better results than the constant damping factor when reliable low-frequency components are missing.

• The Journal of the Acoustical Society of Korea
• /
• v.34 no.6
• /
• pp.487-492
• /
• 2015

A short-range underwater target detection and identification techniques using mid- and high-frequency bands have been highly developed. However, nowadays the long-range detection using the low-frequency band is requested and one of the most challengeable issues. The waveform inversion technique is widely used and the hottest technology in both academia and industry of the seismic exploration. It is based on the numerical analysis tool, and could construct more than a few kilometers of the subsurface structures and model-parameters such as P-wave velocity using a low-frequency band. By applying this technique to the underwater acoustic circumstance, firstly application of underwater target detection is verified. Furthermore, subsurface structures and it's parameters of the war-field are well reconstructed. We can confirm that this technique greatly reduces the false-alarm rate for the underwater targets because it could accurately reproduce both the shape and the model-parameters at the same time.

• Geophysics and Geophysical Exploration
• /
• v.1 no.1
• /
• pp.49-56
• /
• 1998

Seismic tomography has been widely used as high resolution subsurface imaging techniques in engineering applications. Although most of the techniques have been using travel time inversion, waveform method is being driven forward owing to the progress of computational environments. Although full-waveform inversion method has been known as the best method in terms of model resolving power without high-frequency restriction and weak scattering approximation, it has practical disadvantage that it is apt to get stuck in local minimum if the initial guess is far from the actual model and it consumes so much time to calculate. In this study, 2-D full-waveform inversion algorithm in acoustic medium is developed, which uses result of traveltime tomography as initial model. From the application on synthetic data, it is proved that this approach can efficiently reduce the problem of conventional approaches: our algorithm shows much faster convergence rate and improvement of model resolution. Result of application on physical modeling data also shows much improvement. It is expected that this algorithm can be applicable to real data.

• Proceedings of the KSEG Conference
• /
• 2005.04a
• /
• pp.107-110
• /
• 2005

The seismological observation of Korea began in 1905, and has been run with continuous earthquake network of observation, expanding to the advanced country, but still has some problems in accuracy and speed for report. There are many problems to issue the early warning system for earthquakes and Tsunami in the East Sea because most events in the East Sea occur outside the seismic network. Therefore multi-waveform data conversion and composition from the surrounding countries such as Korea, Japan and Far East Russia are requested in order to more accurate determination of earthquake parameters. We used FESNET(Far East Seismic Network) technology to analyze 2004 May 29th Uljin Earthquake and 2005 March 20th Japanese Fukuoka Earthquake in this research, using the data of KMA, Japan stations and IRIS(4 station).

• Geophysics and Geophysical Exploration
• /
• v.10 no.4
• /
• pp.223-228
• /
• 2007

First arrival signals of multi-frequency crosswell seismic data, acquired in wells drilled in granitic rock, were analyzed to investigate the characteristic behavior of the signals at the shear zones. Dominant frequencies of the sources were; 10-, 20-, 40-, 56-, and 80 kHz. No obvious changes in the waveform at the shear zones were found; however, at the shear zones, some degree of velocity reductions were observed in the signals of all frequency sources. The 80 kHz signal is slightly faster than 10 kHz signal in the survey region, and the velocity difference between the two signals were found largest at the shear zone where the permeability measured greatest in the survey interval.

• Korean Journal of Agricultural Science
• /
• v.50 no.3
• /
• pp.469-484
• /
• 2023

In this study, three types of structures-stepped gabion retaining walls, vertical gabion retaining walls, and parapets-were installed on the dam floor crest to prevent the overflow of deteriorative homogeneous reservoirs. The acceleration response, displacement behavior, and pore water pressure ratio behavior were compared and evaluated using shaking-table model tests. The experimental conditions were set to 0.154 g in consideration of the domestic standard and the seismic acceleration range according to the magnitude of the earthquake, and the input waveform was applied with Pohang, Gongen, and artificial earthquake waves. The acceleration response according to the design ground acceleration increased as the height of the embankment increased, and the observed value were larger in the range of 1.1 to 2.1 times the input acceleration for all structures. The horizontal and vertical displacements exhibited maximum values on the upstream slope, and the embankment was evaluated as stable and included within the allowable range for all waveforms. The settlement ratio considering the similarity law exhibited the least change in the case of the parapet structure. The amplification ratio was 1.1 to 1.5 times in all structures, with the largest observed in the dam crest. The maximum excess pore water pressure ratio was in the range of 0.010 - 0.021, and the liquefaction evaluation standard was within 1.0, which was considered very stable.