• Computers and Concrete
• /
• 제32권3호
• /
• pp.327-337
• /
• 2023

Concrete is the most widely used building material, with various types including high- and ultra-high-strength, reinforced, normal, and lightweight concretes. However, accurately predicting concrete properties is challenging due to the geotechnical design code's requirement for specific characteristics. To overcome this issue, researchers have turned to new technologies like machine learning to develop proper methodologies for concrete specification. In this study, we propose a highly accurate deep learning-based predictive model to investigate the compressive strength (UCS) of lightweight concrete with natural aggregates (pumice). Our model was implemented on a database containing 249 experimental records and revealed that water, cement, water-cement ratio, fine-coarse aggregate, aggregate substitution rate, fine aggregate replacement, and superplasticizer are the most influential covariates on UCS. To validate our model, we trained and tested it on random subsets of the database, and its performance was evaluated using a confusion matrix and receiver operating characteristic (ROC) overall accuracy. The proposed model was compared with widely known machine learning methods such as MLP, SVM, and DT classifiers to assess its capability. In addition, the model was tested on 25 laboratory UCS tests to evaluate its predictability. Our findings showed that the proposed model achieved the highest accuracy (accuracy=0.97, precision=0.97) and the lowest error rate with a high learning rate (R2=0.914), as confirmed by ROC (AUC=0.971), which is higher than other classifiers. Therefore, the proposed method demonstrates a high level of performance and capability for UCS predictions.

• 한국지반신소재학회논문집
• /
• 제17권2호
• /
• pp.33-40
• /
• 2018

지반구조물에서는 흙과 구조물 사이의 상호 마찰 특성이 구조물에 작용하는 하중과 저항 분석에 큰 영향을 미친다. 일반적으로 지반구조물의 주재료로서 콘크리트 또는 강재가 많이 사용되며, 흙 입자 크기와 비교하여 마찰면의 거칠기 정도에 따라 경계면 마찰 특성이 큰 차이를 보인다. 특히 비신장성 보강재를 사용하는 보강토 옹벽, 말뚝 주면, 옹벽면 등의 지반구조물에 대해서는 흙과 구조물 사이의 마찰특성을 합리적으로 분석하여 구조물의 안전성을 확보하여야 한다. 구조물면과 사질토 사이의 마찰특성에 대하여 보다 정확한 예측식을 제시하고 이 예측식이 가지고 있는 불확실성을 제시하는 것은 지반-지반구조물의 신뢰성 분석에 필수적이다. 따라서 본 연구에서는 기존 양질의 연구 문헌들에 보고된 흙과 거칠기가 다른 면 사이의 마찰실험결과를 분석하여, 강재와 사질토 사이의 첨두 경계면 마찰각을 구하는 식을 제안하였다. 첨두 경계면 마찰각을 결정하는 주요인자는 경계면의 상대조도, 흙의 상대밀도, 구속압, 한계상태 경계면 마찰각이다. 본 연구에서 개발한 첨두 경계면 마찰각 예측식에 대한 불확실성을 통계적으로 Goodness-of-fit 시험 결과를 반영하여 평가하였다.

• 한국지반공학회논문집
• /
• 제31권10호
• /
• pp.5-16
• /
• 2015

본 연구에서는 국가건설기준이 한계상태설계법으로 전환됨에 따라 현재 연구기반이 미약한 쏘일네일링의 LRFD 설계법 개발의 일환으로 인발저항에 대한 저항편향계수를 분석하였다. 쏘일네일링의 설계 및 시공 기술수준 등 지역적 특성을 반영하기 위하여 국내 비탈면과 굴착공사에서 수행된 인발시험 자료를 수집하고, 지반특성, 쏘일네일의 제원, 인발시험 결과 등으로 구성된 데이터베이스를 구축하였다. 쏘일네일링의 인발저항에 대한 저항계수를 보정하기 위하여 국내에서 가장 많이 이용되며 비교적 자료수가 충분한 중력식과 가압식 그라우팅공법의 인발저항에 대한 저항편향계수를 산정하였다. 저항편향계수는 1.144~1.325으로 실제 극한인발저항력을 비교적 보수적으로 예측하고 있으며 미국의 연구사례(NCHRP Report)보다 안전측으로 설계가 이루어지고 있으나, 인발저항력에 대한 불확실성은 $COV_R$=0.27~0.43으로 비교적 큰 것으로 나타났다. 또한 가압식 그라우팅공법은 중력식 그라우팅공법에 비해 내재된 안전여유가 많으며, 실제 발휘되는 극한인발저항력의 불확실성이 낮은 것으로 분석되었다.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2010년도 춘계 학술발표회
• /
• pp.891-894
• /
• 2010

Steep slope disasters accompany economic loss along with casualties, so the evaluation and the systematic management on the regions with slope collapse danger are required. A lot of manpower, time, and economic cost are needed to accumulate disaster history of steep slope areas by the national and small-sized region. As the method for this, it construed location data about each area with disaster occurrence by maknd elocation data of collapsed steep areas through high-resolution satellite image and collectnd edata on the regions with disasters through media and literature data such as a disaster annual report and a disaster comprehensive report. The study selected three shortest routes includnd ethe area with disaster in Jeolla province on literature and the collapsed area found by the image data, and constructed the results of the field survey as database.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2006년도 춘계 학술발표회 논문집
• /
• pp.1287-1294
• /
• 2006

A pile is compressed with settlements when loading and bearing capacity is altered along relative displacement of pile/soil on settlement and compression. Settlements of pile displaying limit skin friction is different from displaying tip resistance. Therefore, it is an error in traditional method that bearing capacity of pile is estimated from the sum of limit skin fraction and tip resistance. Accordingly, development of design method considering behavior of load-settlement is needed. In this study, we would like to establish the base for development of design method considering bearing capacity altering along displacement on settlement and compression. For this, we established system and substance of design method. And in order to establish relationship of load-settlement of pile on the type of soil, we analyzed and arranged existing database and pile loading test. On design method, settlement is assumed gradually on each capacity level being assumed gradually. Bearing capacity developing on the pile is obtained on each settlement level. Until the obtained bearing capacity will be equal to assumed capacity, this process is continued with increasing settlement. Load-settlement curve for soil classification is sketched in the process computing settlement on assumed capacity. This design method will be materialized by computation program.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2005년도 춘계 학술발표회 논문집
• /
• pp.603-610
• /
• 2005

The end bearing behavior of piles socketed in weathered/soft rock is generally dependent upon the mass conditions of rock with fractures rather than the strength of intact rock. However, there are few available data and little guidance in the prediction of the end bearing capacity of drilled shafts socketed in weathered/soft rock, considering rock mass weathering. Therefore, a database of 13 load tests was constructed first, and new empirical relationships between the base reaction modulus of piles in rock and rock mass properties were developed. No correlation was found between the compressive strengths of intact rock and the base reaction modulus of weathered/soft rock. The ground investigation data regarding the rock mass conditions(e.g. Em, Eur, RMR, RQD) was found to be highly correlated with the base reaction modulus, showing the coefficients of correlation greather than 0.7 in most cases. Additionally, the applicability of existing methods for the end bearing capacity of piles in rock was verified by comparison with the field test data.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2005년도 춘계 학술발표회 논문집
• /
• pp.191-198
• /
• 2005

Slope failures during heavy rainfall have resulted in death of life and economic loss. In recent years, the research on slope damage assessment using Geographical Information System(GIS) has been actively carried out by researchers of several goverment organizations and schools. The researchers in Highway and Transportation Technology Institute (HTTI) of Korea Highway Corporation has developed the GIS database(DB), including highway, rainfalls, soil or rock geometry, types of damage, etc. and have been working on the damage assessment of highway slopes. The DB has been established and summarized in two different ways, such as highway routes and administrative districts. Grid of rainfall intensity generated by maximum rainfalls of each administrative district has been devloped. It shows good correlation of slope damage with heavy rainfalls. Most of damaged slopes were found in the amount of 100 mm to 300 mm rainfalls.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2009년도 춘계 학술발표회
• /
• pp.667-672
• /
• 2009

Gathering information and systemization of infrastructure disaster management is to reduce uncertainties in making decisions and maximize the number of alternations. The key objects of a sensor-based progress report and propagation automation systems are to provide objective data, realize and support decision making and deliver them to a certain area, department, manager and other people rapidly. The major findings and results of this study are as follows. 1) Application of international standard-based alerting protocol(CAP; Common Alerting Protocol). 2) Development of database of existing progress report and propagation manual in order to achieve networking of safety management on major social infrastructure of the nation. 3) Development middleware application programs to progress report and propagation data using SMS, FAX, EMS, VMS, MMS.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2009년도 춘계 학술발표회
• /
• pp.456-463
• /
• 2009

The preconsolidation pressure has been commonly determined by oedometer test. However, it can also be determined by in-situ test, such as piezocone test with theoretical and(or) empirical correlations. Recently, Neural Network(NN) theory was applied and some models were proposed to estimate the preconsolidation pressure or OCR. However, since the optimization process of synaptic weights of NN model is dependent on the initial synaptic weights, NN models which are trained with different initial weights can't avoid the variability on prediction result for new database even though they have same structure and use same transfer function. In this study, Committee Neural Network(CNN) model is proposed to improve the initial weight dependency of multi-layered neural network model on the prediction of preconsolidation pressure of soft clay from piezocone test result. It was found that even though the NN model has the optimized structure for given training data set, it still has the initial weight dependency, while the proposed CNN model can improve the initial weight dependency of the NN model and provide a consistent and precise inference result than existing NN models.

• Geomechanics and Engineering
• /
• 제21권2호
• /
• pp.153-164
• /
• 2020

Tunnel excavation in shallow soft ground conditions of urban areas experiences inevitable surface settlements that threaten the stability of nearby infrastructures. Surface settlements during shield TBM tunneling are related to a number of factors including geotechnical conditions, tunnel geometry and excavation methods. In this paper, a database collected from a construction section of Hong Kong subway was used to analyze the correlation of settlement-inducing factors and surface settlements monitored at different locations of a transverse trough. The Pearson correlation analysis result revealed a correlation between the factors in consideration. Factors such as the face pressure, advance speed, thrust force, cutter torque, twin tunnel distance and ground water level presented a modest correlation with the surface settlement, while no significant trends between the other factors and the surface settlements were observed. It can be concluded that an integrated effect of the settlement-inducing factors should be related to the magnitude of surface settlements.