• Journal of Korean Tunnelling and Underground Space Association
• /
• v.21 no.3
• /
• pp.347-362
• /
• 2019

Underground excavation using TBM machines has been increasing to reduce complaints caused by noise, vibration, and traffic congestion resulted from the urban underground construction in Korea. However, TBM excavation design and construction still need improvement because those are based on standards of the technologically advanced countries (e.g., Japan, Germany) that do not consider geological environment in Korea at all. Above all, although TBM performance is a main factor determining the TBM machine type, duration and cost of the construction, it is estimated by only using UCS (uniaxial compressive strength) as the ground parameters and it often does not match the actual field conditions. This study was carried out as part of efforts to predict penetration rate suitable for Korean ground conditions. The effective parameters were defined through the correlation analysis between the penetration rate and the geotechnical parameters or TBM performance parameters. The effective parameters were then used as variables of the multiple regression analysis to derive a regression model for predicting TBM penetration rate. As a result, the regression model was estimated by UCS and joint spacing and showed a good agreement with field penetration rate measured during TBM excavation. However, when this model was applied to another site in Korea, the prediction accuracy was slightly reduced. Therefore, in order to overcome the limitation of the regression model, further studies are required to obtain a generalized prediction model which is not restricted by the field conditions.

• Geomechanics and Engineering
• /
• v.34 no.5
• /
• pp.547-559
• /
• 2023

We evaluated the applicability of machine learning techniques and the Kuz-Ram model for predicting the mean fragmentation size in open-pit mines. The characteristics of the in-situ rock considered here were uniaxial compressive strength, tensile strength, rock factor, and mean in-situ block size. Seventy field datasets that included these characteristics were collected to predict the mean fragmentation size. Deep neural network, support vector machine, and extreme gradient boosting (XGBoost) models were trained using the data. The performance was evaluated using the root mean squared error (RMSE) and the coefficient of determination (r²). The XGBoost model had the smallest RMSE and the highest r² value compared with the other models. Additionally, when analyzing the error rate between the measured and predicted values, XGBoost had the lowest error rate. When the Kuz-Ram model was applied, low accuracy was observed owing to the differences in the characteristics of data used for model development. Consequently, the proposed XGBoost model predicted the mean fragmentation size more accurately than other models. If its performance is improved by securing sufficient data in the future, it will be useful for improving the blasting efficiency at the target site.

• Tunnel and Underground Space
• /
• v.30 no.3
• /
• pp.214-225
• /
• 2020

Uniaxial compressive strength (UCS) of rock is one of the important factors to determine the advance speed during shield TBM tunnel excavation. UCS can be obtained through the Geotechnical Data Report (GDR), and it is difficult to measure UCS for all tunneling alignment. Therefore, the purpose of this study is to predict UCS by utilizing TBM machine driving data and machine learning technique. Several machine learning techniques were compared to predict UCS, and it was confirmed the stacking model has the most successful prediction performance. TBM machine data and UCS used in the analysis were obtained from the excavation of rock strata with slurry shield TBMs. The data were divided into 8:2 for training and test and pre-processed including feature selection, scaling, and outlier removal. After completing the hyper-parameter tuning, the stacking model was evaluated with the root-mean-square error (RMSE) and the determination coefficient (R²), and it was found to be 5.556 and 0.943, respectively. Based on the results, the sacking models are considered useful in predicting rock strength with TBM excavation data.

• Journal of the Society of Naval Architects of Korea
• /
• v.48 no.3
• /
• pp.281-287
• /
• 2011

This paper explains the basic theory and a new development of for the residual strength prediction program of the asymmetrically damaged ships, being capable of searching moment-curvature relations considering neutral axis mobility. It is noted that moment plane and neutral axis plane should be separately defined for asymmetric sections. The validity of the new program is verified by comparing moment-curvature curves of 1/3 scaled frigate model where the results from new algorithm well coincide with experimental and nonlinear FEA results for intact condition and with nonlinear FEA results for damaged condition. Applicability of new algorithm is also verified by applying VLCC model to the newly developed program. It is proved that reduction of residual strengths is visually presented using the new algorithm when damage specifications of ABS, DNV and IMO are applied. It is concluded that the new algorithm shows very good performance to produce moment-curvature relations with neutral axis mobility on the asymmetrically damaged ships. It is expected that the new program based on the developed algorithm can largely reduce design period of FE modeling and increase user conveniences.

• Steel and Composite Structures
• /
• v.33 no.4
• /
• pp.489-507
• /
• 2019

In order to obtain high bearing capacity and good ductility simultaneously, a structural column with hybrid normal and high strength steel (HNHSS) welded box section has been developed. Residual stress is an important factor that can influence the behaviour of a structural member in steel structures. Accordingly, the magnitudes and distributions of residual stresses in HNHSS welded box sections were investigated experimentally using the sectioning method. In this study, the following four box sections were tested: one normal strength steel (NSS) section, one high strength steel (HSS) section, and two HNHSS sections. Based on the experimental data from previous studies and the test results of this study, the effects of the width-to-thickness ratio of plate, yield strength of plate, and the plate thickness of the residual stresses of welded box sections were investigated in detail. A unified residual stress model for NSS, HSS and HNHSS welded box sections was proposed, and the corresponding simplified prediction equations for the maximum tensile residual stress ratio (${\sigma}_{rt}/f_y$) and average compressive residual stress ratio (${\sigma}_{rc}/f_y$) in the model were quantitatively established. The predicted magnitudes and distributions of residual stresses for four tested sections in this study by using the proposed residual stress model were compared with the experimental results, and the feasibility of this proposed model was shown to be in good agreement.

• Steel and Composite Structures
• /
• v.51 no.6
• /
• pp.679-695
• /
• 2024

This paper presents a novel technique that combines machine learning (ML) with moth-flame optimization (MFO) methods to predict the axial compressive strength (ACS) of concrete filled double skin steel tubes (CFDST) columns. The proposed model is trained and tested with a dataset containing 125 tests of the CFDST column subjected to compressive loading. Five ML models, including extreme gradient boosting (XGBoost), gradient tree boosting (GBT), categorical gradient boosting (CAT), support vector machines (SVM), and decision tree (DT) algorithms, are utilized in this work. The MFO algorithm is applied to find optimal hyperparameters of these ML models and to determine the most effective model in predicting the ACS of CFDST columns. Predictive results given by some performance metrics reveal that the MFO-CAT model provides superior accuracy compared to other considered models. The accuracy of the MFO-CAT model is validated by comparing its predictive results with existing design codes and formulae. Moreover, the significance and contribution of each feature in the dataset are examined by employing the SHapley Additive exPlanations (SHAP) method. A comprehensive uncertainty quantification on probabilistic characteristics of the ACS of CFDST columns is conducted for the first time to examine the models' responses to variations of input variables in the stochastic environments. Finally, a web-based application is developed to predict ACS of the CFDST column, enabling rapid practical utilization without requesting any programing or machine learning expertise.

• Computers and Concrete
• /
• v.15 no.4
• /
• pp.503-514
• /
• 2015

Fibre-added concretes are frequently used in large site applications such as slab and airports as well as in bearing system elements or prefabricated elements. It is very difficult to determine the mechanical properties of the fibre-added concretes by experimental methods in situ. The purpose of this study is to develop an artificial neural network (ANN) model in order to predict the compressive and bending strengths of hybrid fibre-added and non-added concretes. The strengths have been predicted by means of the data that has been obtained from destructive (DT) and non-destructive tests (NDT) on the samples. NDTs are ultrasonic pulse velocity (UPV) and Rebound Hammer Tests (RH). 105 pieces of cylinder samples with a dimension of $150{\times}300mm$, 105 pieces of bending samples with a dimension of $100{\times}100{\times}400mm$ have been manufactured. The first set has been manufactured without fibre addition, the second set with the addition of %0.5 polypropylene and %0.5 steel fibre in terms of volume, and the third set with the addition of %0.5 polypropylene, %1 steel fibre. The water/cement (w/c) ratio of samples parametrically varies between 0.3-0.9. The experimentally measured compressive and bending strengths have been compared with predicted results by use of ANN method.

• Structural Engineering and Mechanics
• /
• v.26 no.2
• /
• pp.211-229
• /
• 2007

An experimental investigation to determine the transfer length of a seven-wire prestressing strand in different concretes is presented in this paper. A testing technique based on the analysis of bond behaviour by means of measuring the force supported by the prestressing strand on a series of specimens with different embedment lengths has been used. An analytical bond model to calculate the transfer length from an inelastic bond stress distribution along the transfer length has been obtained. A relationship between the plastic bond stress for transfer length and the concrete compressive strength at the time of prestress transfer has been found. An equation to predict the average and both the lower bound and the upper bound values of transfer length is proposed. The experimental results have not only been compared with the theoretical prediction from proposed equations in the literature, but also with experimental results obtained by several researchers.

• Geomechanics and Engineering
• /
• v.30 no.1
• /
• pp.75-91
• /
• 2022

This paper aims to show how to use several Machine Learning (ML) methods to estimate the TBM penetration rate systematically (TBM-PR). To this end, 1125 datasets including uniaxial compressive strength (UCS), Brazilian tensile strength (BTS), punch slope index (PSI), distance between the planes of weakness (DPW), orientation of discontinuities (alpha angle-α), rock fracture class (RFC), and actual/measured TBM-PRs were established. To evaluate the ML methods' ability to perform, the 5-fold cross-validation was taken into consideration. Eventually, comparing the ML outcomes and the TBM monitoring data indicated that the ML methods have a very good potential ability in the prediction of TBM-PR. However, the long short-term memory model with a correlation coefficient of 0.9932 and a route mean square error of 2.68E-6 outperformed the remaining six ML algorithms. The backward selection method showed that PSI and RFC were more and less significant parameters on the TBM-PR compared to the others.

• Journal of the Korea Concrete Institute
• /
• v.26 no.6
• /
• pp.707-714
• /
• 2014

CFT (Concrete-Filled Tube) is a type of steel column comprised of steel tube and concrete. Steel tube holds concrete and the concrete inside tube takes charge of compressive load. This study presents structural performance of the CFT column which has 73~100 MPa high strength concrete inside. Fluidity, mechanical compression, pump pressure test in flexible pipe were conducted for understanding properties of the high strength concrete. Material properties were achieved by various experimental tests, such as slump, slump flow, air content, U-box, O-Lot, L-flow. In addition, mock-up tests were conducted to monitor concrete filling, hydration heat, compressive strength. From construction sites in Sang-am dong and University of Seo-kang, long-term behaviors could be effectively predicted in terms of ACI 209 material model considering elastic deformation, shrinkage and creep.