• Journal of the Earthquake Engineering Society of Korea
• /
• v.18 no.4
• /
• pp.201-212
• /
• 2014

The plastic hinge region of RC pier ensures its nonlinear behavior during strong earthquake events. It is assumed that the piers secure sufficient strength and ductility in order to prevent the collapse of the bridge during strong earthquake. However, the presence of a lap-splice of longitudinal bars in the plastic hinge region may lead to the occurrence of early bond failure in the lap-splice zone and result in significant loss of the seismic performance. The current regulations for seismic performance evaluation limit the ultimate strain and displacement ductility considering the eventual presence of lap-splice, but do not consider the lap-splice length. In this study, seismic performance test and analysis are performed according to the cross-sectional size and the lap-splice length in the case of longitudinal bars with lap-splice located in the plastic hinge region of existing RC bridge columns with circular cross-section. The seismic behavioral characteristics of the piers are also analyzed. Based upon the results, this paper presents a more reasonable seismic performance evaluation method considering the lap-splice length and the cross-sectional size of the column.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2005.11a
• /
• pp.723-726
• /
• 2005

The two major obstacles in the application of IPMC to flapping actuators operated in the air are solvent loss and actuation force. In this paper, solvent loss of various IPMCs made of Nafion$^{TM}$117(183$\mu$m thickness) has been experimentally investigated to find out the best combination of cation and solvent for minimal solvent loss in IPMCs and higher actuation force. For this purpose. experiments for the internal solvent loss measurement of IMPCs have been conducted for various combinations of cation and solvent. From the experiments, it was found that heavy water showed improvement in the operating time up to more than two minutes. in the tip force measurement of IPMCs, it was found that smaller and thicker IPMCs produced larger tip forces. However, the shorter IPMCs generated reduced actuation displacements and created flapping motion with decreased natural frequency. For the design of flapping device actuated by 5mm wide, 10mm long, 0.2mm thick IPMCs were used in the stacked form. Since the actuation force is a few gram-force, we stacked five IPMCs to improve actuation force. To amply the actuation force, rack-and-pin ion type hinge was used for the flapping device and insect (Cicadidae) wing was attached to the stacked IPMC actuator. In the flapping test, the device could generate flapping angle of 15$^{\circ}$ at 6Hz excitation by 2.5 voltage square wave input.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2008.11a
• /
• pp.101-104
• /
• 2008

Recently, the serious damage by earthquakes is increased around the world. SOC fo city is established to minimize the loss of lives and assets by earthquakes, which an objective standard is required. Generally, bridges damage by earthquakes occurred the inelastic hinge under the column. Nonlinear element model of inelastic hinge have been used to Bilinear model, but Takeda model for material characterization of concrete is a little. In this study, railway bridge was performed seismic fragility analysis for Takeda model and Bilinear model comparatively. This analysis shows that damage probability of Takeda model is larger than Bilinear model. And analysis of Takeda model in longitudinal direction and transverse direction are different. Therefore developed analysis for concrete column of bridge is expected to apply to material characterization.

• Communications for Statistical Applications and Methods
• /
• v.26 no.6
• /
• pp.635-644
• /
• 2019

Class probability is a fundamental target in classification that contains complete classification information. In this article, we propose a class probability estimation method when the predictor is functional. Motivated by Wang et al. (Biometrika, 95, 149-167, 2007), our estimator is obtained by training a sequence of functional weighted support vector machines (FWSVM) with different weights, which can be justified by the Fisher consistency of the hinge loss. The proposed method can be extended to multiclass classification via pairwise coupling proposed by Wu et al. (Journal of Machine Learning Research, 5, 975-1005, 2004). The use of FWSVM makes our method model-free as well as computationally efficient due to the piecewise linearity of the FWSVM solutions as functions of the weight. Numerical investigation to both synthetic and real data show the advantageous performance of the proposed method.

• Structural Engineering and Mechanics
• /
• v.18 no.5
• /
• pp.541-563
• /
• 2004

For reaching large inelastic deformations without a substantial loss in strength, the potential plastic hinge regions of the reinforced concrete structural members should be confined by adequate transverse reinforcement. Therefore, simple and realistic representation of confined concrete behaviour is needed for inelastic analysis of reinforced concrete structures. In this study, a trilinear stress-strain model is proposed for the axial behaviour of confined concrete. The model is based on experimental work that was carried out on nearly full size specimens. During the interpretation of experimental data, the buckling and strain hardening of the longitudinal reinforcement are also taken into account. The proposed model is used for predicting the stress-strain relationships of confined concrete specimens tested by other researchers. Although the proposed model is simpler than most of the available models, the comparisons between the predicted results and experimental data indicate that it can represent the stress-strain relationship of confined concrete quite realistically.

• Communications for Statistical Applications and Methods
• /
• v.20 no.6
• /
• pp.481-490
• /
• 2013

Classification is an important research area as data can be easily obtained even if the number of predictors becomes huge. The support vector machine(SVM) is widely used to classify a subject into a predetermined group because it gives sound theoretical background and better performance than other methods in many applications. The SVM can be viewed as a penalized method with the hinge loss function and penalty functions. Instead of $L_2$ penalty function Fan and Li (2001) proposed the smoothly clipped absolute deviation(SCAD) satisfying good statistical properties. Despite the ability of SVMs, they have drawbacks of non-robustness when there are outliers in the data. We develop a robust SVM method using a weight function with the SCAD penalty function based on the local quadratic approximation. We compare the performance of the proposed SVM with the SVM using the $L_1$ and $L_2$ penalty functions.

• Coupled systems mechanics
• /
• v.3 no.2
• /
• pp.213-232
• /
• 2014

Pounding is a major cause of bridge damage during earthquakes. In an extreme situation, it can even contribute to the unseating of bridge girders. Long-span bridges will inevitably experience spatially varying ground motions. Soil-structure interaction (SSI) may play a significant role in the structural response of these structures. The objective of this research is to experimentally investigate the effect of spatially varying ground motions on the response of a three-segment bridge considering SSI and pounding. To incorporate SSI, the model was placed on sand contained in sandboxes. The sandboxes were fabricated using soft rubber in order to minimise the rigid wall effect. The spatially varying ground motion inputs were simulated based on the New Zealand design spectra for soft soil, shallow soil and strong rock conditions, using an empirical coherency loss function. The results show that with pounding, SSI can amplify the pier bending moments and the relative opening displacements.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.91-94
• /
• 2021

As deep learning technologies becoming developed, realistic fake videos synthesized by deep learning models called "Deepfake" videos became even more difficult to distinguish from original videos. As fake news or Deepfake blackmailing are causing confusion and serious problems, this paper suggests a novel model detecting Deepfake videos. We chose Residual Convolutional Neural Network (Resnet50) as an extraction model and Long Short-Term Memory (LSTM) which is a form of Recurrent Neural Network (RNN) as a classification model. We adopted cosine similarity with hinge loss to train our extraction model in embedding the features of Deepfake and original video. The result in this paper demonstrates that temporal features in the videos are essential for detecting Deepfake videos.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.93-97
• /
• 2020

본 연구는 딥 러닝 기반 의존 구문 분석에서, 학습에 적용하는 손실 함수에 따른 성능을 평가하였다. Pointer Network를 이용한 Left-To-Right 모델을 총 세 가지의 손실 함수(Maximize Golden Probability, Cross Entropy, Local Hinge)를 이용하여 학습시켰다. 그 결과 LH 손실 함수로 학습한 모델이 선행 연구와 같이 MGP 손실 함수로 학습한 것에 비해 UAS/LAS가 각각 0.86%p/0.87%p 상승하였으며, 특히 의존 거리가 먼 경우에 대하여 분석 성능이 크게 향상됨을 확인하였다. 딥러닝 의존 구문 분석기를 구현할 때 학습모델과 입력 표상뿐만 아니라 손실 함수 역시 중요하게 고려되어야 함을 보였다.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2005.11a
• /
• pp.89-92
• /
• 2005

Recently there have been occurred many loss of life and extensive damage to social infrastructures due to moderate and strong earthquakes all over the world. In this research, major design factors have been evaluated for the establishment of the rational seismic design code of rectangular RC bridge piers. It was concluded from this study that the axial force ratio and the longitudinal steel ratio should be the most important influencing design parameter for the seismic displacement ductility. However those parameters are not considered in the confinement steel ratio of the KHBDS. Thus, the objective of this study is to propose a rational design equation for transverse reinforcements of rectangular RC bridge piers. New confinement steel ratio is proposed by introducing the effect of the axial force and the longitudinal steel to the current KHBDS. It is thought that these new codes could release the rebar congestion problem in the plastic hinge region of RC bridge piers which contribute to the enhancement of constructibility and economization for RC bridge construction.