• 한국지반공학회논문집
• /
• 제20권3호
• /
• pp.53-65
• /
• 2004

본 연구에서는, 굴착 공사로 인한 주변 건물 손상 평가의 일환으로 지표 변위유형의 효과적인 예측 방안을 제시하였다. 먼저 검증된 유한요소 모델을 국내에서 행해지는 다양한 경우의 굴착특성으로 해석한 결과를 통해 인접 지반의 거동에 대한 매개 변수 연구를 수행하였고, 인공신경망 엔진의 학습을 위한 데이터베이스를 구축하였다. 최적의 구조로 학습된 신경망 엔진은 간단한 굴착 특성으로 다양하게 나타나는 지표 변위 유형 예측에 효율적임을 검증하였다.

• 한국지반환경공학회 논문집
• /
• 제10권3호
• /
• pp.63-71
• /
• 2009

대도시에서의 공사는 외부방향이 아닌 상부로 진행되기 때문에 현장의 벽면 움직임은 매우 중요하다. 고층 구조물은 주차장을 위한 여분의 공간 확보뿐 아니라 건물의 잠재적 침하를 줄이기 위하여 일반적으로 깊은 굴착을 수반한다. 이러한 대형 굴착은 깊은 심도에 따른 횡방향 지중압력에 견디기 위한 견고한 브레이싱 시스템을 필요로 한다. 벽체 움직임은 잠재적인 인접 구조물의 침하를 허용하기 때문에 sheetpile 이나 diaphragm wall과 같은 옹벽구조물의 변형을 예측하는 방법은 매우 중요하다. 사례들을 분석하고 측정된 벽체 변형은 경험적 도표로부터 예측된 값들과 비교되었다.

• 한국지반공학회논문집
• /
• 제16권2호
• /
• pp.23-30
• /
• 2000

The use of strut-preloading method is gradually increasing in braced excavations in Korea. And it is necessary to analyze the effects of strut preloading on the wall deflection, wall bending moment and strut axial force, etc. In this study, by using the analysis method of beams on elasto-plastic foundations, measured data and calculated results of 2 sites are compared and parametric studies of correlation between preloading and earth retaining structures in sandy soils are carried out in strut preloading application. As results, about 50%~75% of design strut load is effective as preloading force in considering the displacement and member forces of earth retaining structures. And the effective stiffiness of strut should be at least 25% of th ideal value in order to restrain the excessive increase of wall deflection and bending moments. As one of some methods to prevent excessive movements in braced excavation, to preload the strut is confirmed as more effective way than to increase the stiffiness of strut in braced wall, if the excessive axial force of strut due to preloading can be avoided.

• 한국지반공학회논문집
• /
• 제22권2호
• /
• pp.19-28
• /
• 2006

굴착 공정에 따른 주변지반의 거동 특성을 파악하기 위하여 굴착 깊이가 15m$\∼$29m인 3개 굴착현장에서 장기간에 걸쳐 계측을 실시하고 그 결과를 분석하였다. 이 논문에서는 본 굴착뿐만 아니라 토류벽 설치 및 지보 해체 작업이 굴착 주변지반의 거동에 미치는 영향을 정량적으로 평가하기 위하여 전체 굴착 공정을 굴토 전 준비 공정, 본 굴착 공정, 지보 해체의 세 공정으로 구분하여 계측을 수행하였다. 3개 굴착현장에서 이루어진 계측결과에 의하면, 지보 해체로 인하여 추가적으로 발생한 벽체 수평변위는 본 굴착단계에서 발생한 변위의 약 40$\%$ 내외였으며, 지표침하의 경우에는 본 굴착 시 발생한 침하체적의 $18\%\∼33\%$가 발생하였다. 이와 같은 현장계측 결과를 통하여, 굴착 전 및 지보 해체 공정 중에도 무시할 수 없을 정도의 지반 변위가 발생할 수 있음을 정량적으로 확인하였다.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2002년도 가을 학술발표회 논문집
• /
• pp.147-179
• /
• 2002

The important role of observational method in geotechnical engineering are emphasized together with the direction of future development, concerning successful application of the technique on the site investigation, design and feed back at various construction stages. Case histories on the application of feed back are introduced in order to achieve the most economical and reliable construction for tunnel, rock slope and deep excavations through feed back system at design and construction stages. Also the limitations and advantages of the observational method and the role of feed back system are discussed for construction of tunnel, rock slope and deep excavation in hard ground such as layered ground conditions including weathered, soft and hard rocks.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2002년도 봄 학술발표회 논문집
• /
• pp.263-270
• /
• 2002

During deep excavation, changes in the state of stress in the ground mass around the excavation and subsequent ground losses inevitably occur. These changes in the stress and ground losses are reflected on surrounding ground in the form of ground movements, which eventually Impose strains onto nearby structures through translation, rotation, distortion, and possibly damage. A substantial portion of the cost of deep excavations in urban environments is, therefore, devoted to prevent ground movements. Prediction of ground movements and assessment of the risk of damage to adjacent structures has become an essential part of the planning, design, and construction of a deep excavation project in the urban environments. This paper presents excavation-induced ground movement characteristics as well as important issues related to excavation-induced building damage assessment.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2009년도 세계 도시지반공학 심포지엄
• /
• pp.1294-1300
• /
• 2009

Recently, excavations in highly congest urban area have been increased. For the excavations conducted in extremely narrow spaces, we have been developing a novel soil reinforcement system of temporary retaining walls by using deep cement mixing method. The developing method installs largerdiameter ($\Phi$=300~500mm) and shorter reinforcement blocks than previous reinforcement system for mobilizing friction with soils, therefore it has advantages of not only shortening the length of reinforcement system but also reducing the amount of reinforcement. In this study, we performed a numerical analysis of the new reinforcement system by using a commercial finite element program, and evaluated the behavior of the reinforced retaining wall system under various conditions of the length, the diameter, the spacing, and the angle of the reinforcement system.

• 한국지반공학회지:지반
• /
• 제9권1호
• /
• pp.59-68
• /
• 1993

In deep excavations for creation of underground spaces, it would be difficult to predict earth pressure, especially multilayered ground including rock strata. The earth pressures and displacements on the retention walls are measured by load cell, strain gauge and inclinometer which were installed at struts or anchors at 4 deep excavation sites in Seoul area. In this paper, the measured earth pressure from the struts or anchors are compared with Peck's empirical values, and the coefficient of the earth pressures for each strata and horizontal wall displacement are investigated. The coefficient of earth pressure distribution, a(0.65zka), in the flexible and the rigid walls was about 74% and 88% of Peck's value respecitively. The measured earth pressure distributions for the 4 sites showed about 70%∼80% of Peck's empirical values and the average earth pressure coefficients based on the measured data were 0.3 for the felted layer, 0.23 for the weathered rock and 0.19 for the weak rock. The maximum w리1 displacements were found to be less 0.2% of excavation depth.

• Geomechanics and Engineering
• /
• 제23권5호
• /
• pp.405-418
• /
• 2020

Deep excavation may have impact on the adjacent tunnels. It is obvious that the excavation will adversely affect and even damage the existing tunnels if the induced deformation exceeds the capacity of tunnel structures. It hence creates a high necessity to predict tunnel displacement induced by nearby excavation to ensure the safety of tunnel. In this paper, a simplified method to evaluate the heave of the underlying tunnel induced by adjacent excavation is presented and verified by field measurement results. In the proposed model, the tunnel is represented by a series of short beams connected by tensile springs, compressional springs and shear springs, so that the rotational effect and shearing effect of the joints between lining rings can be captured. The proposed method is compared with the previous modelling methods (e.g., Euler-Bernoulli beam, a series of short beams connected only by shear springs) based on a field measured longitudinal deformation of subway tunnels. Results of these case studies show a reasonable agreement between the predictions and observations.

• Geomechanics and Engineering
• /
• 제37권6호
• /
• pp.629-641
• /
• 2024

The presence of excavations or cavities beneath the foundations of a building can have a significant impact on their stability and cause extensive damage. Traditional methods for calculating the bearing capacity and subsidence of foundations over cavities can be complex and time-consuming, particularly when dealing with conditions that vary. In such situations, machine learning (ML) and deep learning (DL) techniques provide effective alternatives. This study concentrates on constructing a prediction model based on the performance of ML and DL algorithms that can be applied in real-world settings. The efficacy of eight algorithms, including Regression Analysis, k-Nearest Neighbor, Decision Tree, Random Forest, Multivariate Regression Spline, Artificial Neural Network, and Deep Neural Network, was evaluated. Using a Python-assisted automation technique integrated with the PLAXIS 2D platform, a dataset containing 272 cases with eight input parameters and one target variable was generated. In general, the DL model performed better than the ML models, and all models, except the regression models, attained outstanding results with an R² greater than 0.90. These models can also be used as surrogate models in reliability analysis to evaluate failure risks and probabilities.