• Geophysics and Geophysical Exploration
• /
• v.23 no.3
• /
• pp.157-167
• /
• 2020

With the recent advancement of computer hardware and the contribution of open source libraries to facilitate access to artificial intelligence technology, the use of machine learning (ML) and deep learning (DL) technologies in various fields of exploration geophysics has increased. In addition, ML researchers have developed complex algorithms to improve the inference accuracy of various tasks such as image, video, voice, and natural language processing, and now they are expanding their interests into the field of automatic machine learning (AutoML). AutoML can be divided into three areas: feature engineering, architecture search, and hyperparameter search. Among them, this paper focuses on hyperparamter search with Bayesian optimization, and applies it to the problem of facies classification using seismic data and well logs. The effectiveness of the Bayesian optimization technique has been demonstrated using Vincent field data by comparing with the results of the random search technique.

• Geophysics and Geophysical Exploration
• /
• v.9 no.1
• /
• pp.93-101
• /
• 2006

In this study, we propose a joint inversion method, using genetic algorithms, to determine the shear-wave velocity structure of deep sedimentary layers from receiver functions and surface-wave phase velocity. Numerical experiments with synthetic data indicate that the proposed method can avoid the trade-off between shear-wave velocity and thickness that arises when inverting the receiver function only, and the uncertainty in deep structure from surface-wave phase velocity inversion alone. We apply the method to receiver functions obtained from earthquake records with epicentral distances of about 100 km, and Rayleigh-wave phase velocities obtained from a microtremor array survey in the Kanto Plain, Japan. The estimated subsurface structure is in good agreement with the previous results of seismic refraction surveys and deep borehole data.

• Steel and Composite Structures
• /
• v.17 no.3
• /
• pp.215-236
• /
• 2014

The flexural behaviour of steel beams significantly affects the structural performance of the steel frame structures. In particular, the flexural overstrength (namely the ratio between the maximum bending moment and the plastic bending strength) that steel beams may experience is the key parameter affecting the seismic design of non-dissipative members in moment resisting frames. The aim of this study is to present a new formulation of flexural overstrength factor for steel beams by means of artificial neural network (NN). To achieve this purpose, a total of 141 experimental data samples from available literature have been collected in order to cover different cross-sectional typologies, namely I-H sections, rectangular and square hollow sections (RHS-SHS). Thus, two different data sets for I-H and RHS-SHS steel beams were formed. Nine critical prediction parameters were selected for the former while eight parameters were considered for the latter. These input variables used for the development of the prediction models are representative of the geometric properties of the sections, the mechanical properties of the material and the shear length of the steel beams. The prediction performance of the proposed NN model was also compared with the results obtained using an existing formulation derived from the gene expression modeling. The analysis of the results indicated that the proposed formulation provided a more reliable and accurate prediction capability of beam overstrength.

• Geophysics and Geophysical Exploration
• /
• v.6 no.2
• /
• pp.95-100
• /
• 2003

In this study, we developed reflection tomography inversion algorithm using Straight Ray Technique (SRT) which can calculate travel time easily and fast for complex geological structure. The inversion process begins by setting the initial velocity model as a constant velocity model that hat only impedance boundaries. The inversion process searches a layer-interface structure model that is able to explain the given data satisfactorily by inverting to minimize data misfit. For getting optimal solution, we used Gauss-Newton method that needed constructing the approximate Hessian matrix. We also applied the Marquart-Levenberg regularization method to this inversion process to prevent solution diverging. The ability of the method to resolve typical target structures was tested in a synthetic salt dome inversion. Using the inverted velocity model, we obtained the migration image close to that of the true velocity model.

• Journal of the Korean Geotechnical Society
• /
• v.21 no.9
• /
• pp.65-75
• /
• 2005

Maximum shear modulus of soil is a principal parameter for the design of earth structures under static and dynamic loads. In this study, the statistical data of maximum shear moduli of silty sands in Songdo area in the west coast of Korea evaluated by various field and laboratory tests - standard penetration test (SPT), cone penetration test (CPT), self-boring pressuremeter test (SBPT), downhole test (DH), seismic cone penetration test (SCPT) and resonant column test (RC) were analyzed. Based on the measurement of shear moduli using DH which is known as maximum value at very small strain, the new empirical correlations between shear moduli and SPT or CPT values were proposed. Predictions of maximum shear moduli using the proposed correlations were compared with the data obtained from DH. The good agreement confirmed that the proposed correlations reasonably predicted the maximum shear moduli of silty sands in the area.

• Ecology and Resilient Infrastructure
• /
• v.5 no.2
• /
• pp.59-71
• /
• 2018

Lake Enriquillo, the largest lake in the Dominican Republic, recently has undergone unusual water area changes since 2001 thus it has been affected seriously by local community's livelihood. Earthquakes and seismic activities of Hispaniola plate tectonic coupled with human activities and climate change are addressed as factors causing the increasing. Thus, a thorough study on relationship between lake area changing, and those factors is needed urgently. To do so, this study applied MESMA on MODIS data to extract water area of Lake Enriquillo during 2001 and 2012 bimonthly, with six issues 12-year. MODIS provides high temporal resolution, and its coarse spatial resolution is compensated by MESMA fraction map. The increase in water area was $142.2km^2$, and the maximum lake area was $338.0km^2$ (in 2012). Water areas extracted by two Landsat scenes at two different times with three image classification approaches (ISODATA, MNDWI, and TCW) were used to assess accuracy of MODIS and MESMA results; it indicated that MESMA water areas are same as ISODATA's, less than 0.4%, while the highest difference is between MESMA and TCW, 2.4%. A number of previously formulated hypotheses of lake area change were investigated based on the outcomes of the present study, though none of them could fully explain the changes.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.19 no.4
• /
• pp.405-414
• /
• 2001

Plate tectonics is a dominant paradigm in modern geophysics. Because of its geological mechanism, Korea has a possibility of earthquake according to plate motion. Besides the disaster of earthquake grows rapidly, the importance of recognition for earthquake has been emphasized. This study attempts to decide crustal movement parameters with GPS data, analysed baseline after processing data with GIPSY-OASIS II S/W, observed from 6 stations in and around the Korean peninsula, and obtained from selected 11 stations in Korea. As a results, maximum shear strain was $0.04{\mu}/yr$ and the mean azimuth of the maximum compression axes$(A_{z2})$ is estimated as $97.75^{\circ}$ in and around the Korean peninsula. The average rate of the maximum shear strain($({\gamma}_max)$) is $0.17{\mu}/yr$. The mean azimuth of the maximum compression axes$(A_{z2})$ is estimated as $70.25^{\circ}$ in Korea. Such a pattern of strain distribution is harmonious with that of seismic activity in Korea both historically as well as today.

• Tunnel and Underground Space
• /
• v.18 no.5
• /
• pp.386-392
• /
• 2008

Recently, an earthquake-resistant has become essential to the large-scale structures at Gyeonggi and Gyeongsangnam province in Korea, but it is generally compared the measured data with foreign references because of the lack of the research data. It will be presented the characteristics of suitable dynamic elastic modulus in Korean geology, which characteristics are characterized the seismic wave velocities($V_p,\;V_s$) and correlation with dynamic elastic modulus($E_d,\;G_d\;K_d$) by each rock type of Korean granite, because it is very different between the values of foreign references and Korean geological characteristics.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.6
• /
• pp.59-68
• /
• 2008

In this paper, we propose a method for selecting meaningful event packets from which can receive before anything else from seismograph according to allotted priority and estimate epicenter using selected packets. Event packets which received from each station will be evaluated with their onset time, signal period and SNR by statistical method and will be selected packets related with real earthquake's P-wave. And estimated epicenters using by 'Application of epicenter estimation using first P arrivals'. With local earthquakes occurred in 2007 were announced by KMA, collected event packets on earthquake happened date and selected p-wave related packets and estimated epicenter. After result of experiment, if an earthquake occurred within seismic networks, can estimate epicenter with small misfits just after event packets arrived from above 4 stations. Considering average distance of each station, in case of using all stations' data include other organization, can estimate and alert rapidly. It show this method is useful when construct a local earthquake early warning system later.

• Geophysics and Geophysical Exploration
• /
• v.20 no.4
• /
• pp.232-240
• /
• 2017

Normal moveout correction is one of the main procedures of seismic reflection data processing and a crucial pre-processing step for AVO analysis. Unfortunately, stretch phenomenon, which is the intrinsic problem of NMO correction, degrades the quality of stack section and reliability of AVO analysis. Although muting is applied to resolve this problem, it makes far-offset traces more useful to develop an advanced NMO correction technique without stretch. In this paper, easy and detailed explanations are provided on the definition and methodology of NMO correction, and then the cause of stretch is explained with its characteristics. A graphical explanation for NMO correction is given for the intuitive understanding of stretch phenomenon. Additionally, the theoretical formulation is derived to quantitatively understand the NMO correction. Through explaining the muting process to remove NMO stretch, the limitations of conventional methods are investigated and the need for a new resolution comes to discussion. We describe a stretch-free NMO correction based on inverse theory among many different stretch-free NMO corrections. Finally, the stretch-free NMO correction is verified through synthetic example and real data.