• Journal of the Korean Geotechnical Society
• /
• v.38 no.11
• /
• pp.87-95
• /
• 2022

The geotechnical information constructed in the National Geotechnical Information DB System has been extensively used in design, construction, underground safety management, and disaster assessment. However, it is necessary to refine the geotechnical information because it has nearly 300,000 established cases containing a lot of missing or incorrect information. This research proposes a method for automatic quality control of geotechnical information using a fully connected neural network. Significantly, the anomalies in geotechnical information were detected using a database combining the standard penetration test results and strata information of Seoul. Consequently, the misclassification rate for the verification data is confirmed as 5.4%. Overall, the studied algorithm is expected to detect outliers of geotechnical information effectively.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.3C
• /
• pp.97-104
• /
• 2009

In this study, artificial neural network was performed using the data of soils characteristic value, standard penetration test, and field permeability test of the 12 embankment that are located in the near Nak-dong and Kum-ho river to estimate the coefficient of field permeability of river embankment. The 89 data of total 108, 82% was used in learning step, and the other 19 data was used in estimation step. Also the results of generally used empirical equations were compared with those of artificial neural network for evaluation of application. As results, all of the coefficient of field permeability by empirical equation showed below 0.4 in terms of the coefficient of correlation with the measured values, but the coefficient of correlation of the predicted results by artificial neural network was up 0.8 in the all case. Therefore artificial neural network could predict more the precise field permeability well than the empirical equations.

• Journal of the Korean Geotechnical Society
• /
• v.37 no.6
• /
• pp.5-20
• /
• 2021

In this study, a modified empirical formula for estimating the bearing capacity of the steel pipe prebored and precast pile was proposed by performing 20 cases of real-scale field pile loading tests. The proposed formula will be based on expanded SPT N-value in order to consider the realistic condition of the surrounding soil. The formula is proposed based on a statistic approach of the data points from the field pile loading test, in order to ensure safe engineering practice while finding a reliable formula. The statistical analysis of the data points from the loading test indicated that the existing formula has been underestimated the bearing capacity of the prebored and precast pile. The proposed formula estimates 15% and 20% higher pile End bearing capacity (q_t=230P_driven(kN/m²)) and the shaft resistance (f_max=3.0N_s^E(kN/m²)) compared to the existing formula. The accuracy and the stability of the proposed formula was verified by comparing the estimated results with additional field test data. The verification process showed that the proposed formula was estimated to be more accurate than the existing formula.

• Geotechnical Engineering
• /
• v.9 no.4
• /
• pp.37-44
• /
• 1993

Model pile tests in calibration chamber are performed in order to study the two factors that the pile bearing capacity is significantly influenced by. Those factors are the critical depth concept and the scale effect caused by pile diameters. Firstly, the predicted values of end bearing capacity from the various static formulae were compared with the measured ones from model pile tests. Secondly, the critical depth concept and the scale effect were investigated by using two different soil conditions in a series of calibration chamber tests : the one is uniform sand : and the other is weathered granites overlayered by sand. Main results obtained from the model tests can be summarized as follows : (1) The end bearing capacity was increased with pile penetration depth up to penetration ratio of 7 to 8 when the cell pressure is high, and the critical depth was observed in the current chamber tests with uniform sand layer , (2) The predicted end bearing capacities were mostly lager than the measured, and it was found that the differences between the predicted and the measured values became smaller as the pile penetration ratio was increased : (3) The end bearing capacity of the small diameter pile in weathered granites layer was mostly less than that of the larger pile, while in uniform sand layer it was vice.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2009.05a
• /
• pp.2033-2037
• /
• 2009

호소내의 흐름은 중력의 영향을 받는 하천의 흐름과는 달리 열수지에 의한 밀도류 및 바람 등에 영향을 크게 받아 특유의 유동특성을 나타낸다. 본 연구에서는 EFDC(Environmental fluid Dynamics Code, 미국 버지니아 해양연구소 개발) 모델을 이용하여 호소 내 수치모의를 실시하였다. 복잡한 지형을 형성하고 있는 호수 내의 유동특성의 정밀도를 향상시키기 위해서는, 흐름특성을 고려한 grid작성이 중요하다. 본 연구에서는 1:3000, 1:5000 축척의 수치지도를 이용하여 장흥호의 하상단면 값을 추정하였으며, GIS를 이용하여 (${\Delta}x$, ${\Delta}y$)=100m로 DEM을 생성하였다. 그리고 구성된 DEM값을 이용하여 정밀도가 높은 최종 Cartesian grid(active cell=706)를 구축하였다. EFDC를 이용한 장흥호의 수치모의 결과 호수내의 흐름은 수면을 통한 열 교환에 의한 수온 밀도류를 형성하고 있으며, 수온성층의 형성과 파괴가 호수 흐름을 형성하는 큰 인자로 작용한다는 것을 확인할 수 있었다. 그리고, 호에 유입되는 하천부분에서는 유입되는 수온에 따라 흐름특성이 변하며, 유입수온에 따라 호수 내 관입위치가 변하여 표면흐름, 내부흐름, 저부흐름 등의 흐름특성을 보인다 이상과 같이, 호수내의 흐름특성을 열수지, 외력에 의해서 크게 변동하며 해석대상역의 지역특성을 반영한 수리현상의 정확한 재현 없이는, 이 유동해석결과를 입력치로 하는 수질모델링 결과는 신뢰도에 심각한 문제를 발생시킨다고 할 수 있다.

• Geotechnical Engineering
• /
• v.9 no.3
• /
• pp.35-48
• /
• 1993

This paper was focused on the design and construction of stone columns as a method of reinforcements in slightly soft clay or loose silty sand overlying the firm strata. It was intended to find out the effects of soil improvement by field tests in Kwangmyoung Haan Site by comparing the test results performed before and after the installation of stone columns . The results of this experimental study are as follows : the hearing capacity was increased by 3.33 times and the SPT N -valued was increased by 3.9 times by installing the stone columns : the stone column method can cut down the construction cost almost 35% compared to the precast concrete pile method in economic viewpoints.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.7
• /
• pp.1668-1672
• /
• 2009

In this study, the method by Robertson which has been most commonly used for classifying soils, using piezocone test results, was compared with that by Schneider which was most recently proposed. Both methods were applied to the soils in Gyeonggi province and the classifying results were investigated. It has been found that the difference between the results according to the methods was not so large and Schneider's method showed slightly better results for clay region and vice versa. Such factors as large field database, normalized tip resistance, pore water pressure, and drain condition were found to need further research for more reliable soil classification.

• Korean Journal of Soil Science and Fertilizer
• /
• v.44 no.3
• /
• pp.344-346
• /
• 2011

This study was conducted in order to compare between Yamanaka hardness (YA), hand-push type, and penetrometer cone index (PE), on 16 upland sites. Yamanaka hardness had significant correlation with penetrometer cone index. The regression equation was $PE=1.80^*YA+0.16$ ($R^2=0.91^{***}$, N=16) with the applicative YA range of 0.1-1.3 MPa.

• Journal of the Korean Geotechnical Society
• /
• v.26 no.9
• /
• pp.5-13
• /
• 2010

Recently, the use of barrette pile has remarkably increased for high-rise building and bridge foundations. However, relatively few studies have been made for analyzing barrette pile behavior by considering shear load transfer on interface between pile and soils. Therefore, in this paper, an empirically derived equation is proposed. This equation correlates the load transfer curve of barrette piles with the N value from field standard penetration test based on full-scale load tests. The results from all procedures are presented. In addition, the effect of interface on pile-soil is evaluated using 3-D non-linear finite element method and verified with the field data.

• Economic and Environmental Geology
• /
• v.2 no.4
• /
• pp.23-71
• /
• 1969

Sediment samples were taken at about half-mile intervals from all the inajor rivers draining the St. Austell granite mass. The minus 80 mesh(B.S.S.) fraction of each sample was analysed, using semiquantitative methods, for sodium, potassium, lithium, phosphorus, nickel, chromium, tin, tungsten, arsenic copper, zinc and lead. The work was carried out with the view to gaining further information as to the geographical distribution of such different granite facies as might axist, and to investigate the geochemical dispersion of these elements with relation to mineralisation in this area. The sesults confirm Exley's suggestion that the mass consists of two major granite intrusions, the earlier undifferentiated one is joined on the west by a later differentiated intrutive. During the work grid deviation maps proved particularly useful in obtaining data concerning the nature of the granite but frequency diagrams were not particularly helpful. All the known lode areas were associated with stream sediments containing anomalously high concentrations of lode metals and it is concluded that these high concentrations are due premarily to lode material transferred to the streams in the form of tailings lost during milling operations.