• Proceedings of the Korea Water Resources Association Conference
• /
• 2007.05a
• /
• pp.1701-1705
• /
• 2007

국부세굴은 수공구조물 주변에서의 국부적인 흐름변화가 그 원인이라 할 수 있으며 이러한 변화는 교각 또는 교대의 하상재료를 굴착하여 이동시키는 침식작용의 결과이다. 이러한 교량의 교각 또는 교대, 그리고 교각을 지지하는 구조물에서 발생하는 국부세굴은 교량의 수명과 안정성에 좋지 않은 영향을 미쳐 재산상의 피해를 줄뿐만 아니라 교량 붕괴 사태 등으로 인한 대형 인명사고를 초래하는 원인이 될 수도 있다. 따라서 국부세굴의 경향을 예측하고 분석하는 것은 교량 구조물이나 수중 구조물의 설계에 있어서 매우 중요한 과업이다. 이러한 국부세굴을 예측하고 분석하는데 있어서 국내에서는 일정한 흐름조건에 대한 최종세굴심을 예측하는데 초점이 맞추어져 있다. 하지만 국부세굴은 하상재료에 따라 세굴이 진행되는 속도가 다르며 세굴의 진행은 시간 의존적인데 반해 국내에서의 국부세굴의 예측과 분석은 최종세굴심 산정에 초점이 맞추어져 있다. 이에 본 연구에서는 보다 실질적이고 경제적인 국부세굴 경향을 예측할 수 있도록 다양한 하상재료에 대한 시간에 따른 세굴 변화의 특성을 파악하고 분석하였다. 이를 위해 하상재료에 따라 수리실험을 실시하였고, 이를 통해 하상재료에 따른 세굴심의 시간적 변화 곡선을 구하였다. 이렇게 구한 하상재료에 따른 세굴심의 시간적 변화 곡선을 통해 하상재료별 평형세굴심 산정하였고, 산정한 평형세굴심과 기존 세굴심 산정공식을 이용하여 입자별 세굴심을 산정하여 비교 분석하였다.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2009.10a
• /
• pp.187-188
• /
• 2009

On this study, the artificial neural network, one of the nonlinear forecasting methods, is compared with ARIMA model through performing a forecast of container traffic. The existing studies have been used the rule of thumb in topology design for network which had a great effect on forecasting performance of the artificial neural network. However, this study applied the genetic algorithm, known as the effectively optimal algorithm in the huge and complex sample space, as the alternative.

• The Korean Journal of Applied Statistics
• /
• v.26 no.2
• /
• pp.253-265
• /
• 2013

It has been interested in developing useful information variables that are able to predict the future movement of final objects to attain the specific policy and strategic target. Term structure of interest rates is known as an important variable to predict future business and economic activity, yet there is little empirical work on the predictability of future changes in real output. This study attempts to develop the statistical model and examine whether domestic term structure of interest rates can predict variations of future cumulative changes in real growth on a long time horizon.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.35 no.4
• /
• pp.207-214
• /
• 2022

A prediction model is proposed for a blast load acting on a column component because of an external explosion. The model can predict the pressure-time histories acting on a column using the fitting curves established from a database composed of finite-element (FE) analysis results. To this end, 70 numerical simulations using the commercial software AUTODYN were performed by changing the column width. To confirm the performance of the proposed model, pressure-time histories estimated from an existing empirical formula and the proposed model were compared based on the FE analysis results. It was verified that the proposed model can more precisely predict the pressure-time histories compared with the existing model.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.347-347
• /
• 2022

최근 데이터 사이언스의 비약적인 발전과 함께 다양한 형태의 딥러닝 알고리즘이 개발되어 수자원 분야에도 적용되고 있다. 이 연구에서는 LSTM(Long Short-Term Memory) 네트워크와 BO-LSTM이라는 베이지안 최적화(BO) 기술을 결합하여 일단위 앙상블 미래 댐유입량을 projection하는 딥 러닝 모델을 제안하였다. BO-LSTM 하이퍼파라미터 및 손실 함수는 베이지안 최적화 기법을 통해 훈련 및 최적화되며, BO 접근법은 모델의 하이퍼파라미터와 손실 함수를 높은 정확도로 빠르게 최적화할 수 있었다(R=0.92 및 NSE=0.85). 또한 미래 댐 유입량을 예측하기 위한 LSTM의 구조는 Forecasting 모형과 Proiection 모형으로 구분하여 두 모형의 장단점을 분석하였으며, 본 연구의 결과로부터 데이터 처리 단계가 모델 훈련의 효율성을 높이고 노이즈를 줄이는 데 효과적이고 미래 예측에 있어 LSTM 구조에 따른 영향을 확인할 수 있었다. 본 연구는 소양강 유역, 2020-2100년 기간 동안의 미래 예측에 적용되었다. 전반적으로, CIMIP6 데이터에 따르면 10%에서 50%의 미래 유입량 증가가 발생하는 것으로 확인되었으며, 이는 미래 강수량의 증가의 폭과 유사함을 확인하였다. 유입량 산정에 있어 신뢰할 수 있는 예측은 저수지 운영, 계획 및 관리에 있어 정책 입안자와 운영자에게 도움이 될 것입니다.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.20 no.3
• /
• pp.255-264
• /
• 2007

In the current domestic and overseas standards concerning seismic design, especially on the capacity & demand spectra in the multi-story building, failure is caused more by story drift than by displacement; and the existing capacity spectrum method (CSM) does not make a close estimate of story drift because response is derived using displacement. Therefore, this paper proposes an improved CSM to estimate story drift and its direct effect on the collapse of structures, yet still maintaining the same advantage and convenience of the existing CSM about a most basic model of multi-story building: shear building. To establish its reliability, the proposed method is applied to an example model and results are then compared with those obtained through nonlinear time-history analysis.

• Applied Chemistry for Engineering
• /
• v.4 no.4
• /
• pp.709-714
• /
• 1993

The cetane number is a measure of ignition quality, specifically ignition delay, of diesel fuel. It is an engine measure of kinetic phenomena. The ignition quality such as kinetic behavior does correlate with the molecular structure, the carbon type structural composition. In fact, we use the group additivity rule to dissect the molecular structures and predict cetane number. In this study, the use of $^{13}C-Nuclear$ Magnetic Resonance spectroscopic measuring the molecular structure and group additivity rule at different diesel fuels, whose cetane numbers were determined on a number of standard cetane rating engines is proposed to predict cetane numbers that relate the carbon type structural composition. The effect of the molecular structures on the cetane numbers has been studied.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.12A
• /
• pp.1974-1984
• /
• 1999

In this paper, we introduce a highly efficient data structure that effectively captures the multipath phenomenon needed for accurate propagation modeling and fast propagation prediction. The proposed object representation procedure is called 'circular representation (CR)' of microwave masking objects such as buildings, to improve over the conventional vector representation (VR) form in fast ray tracing. The proposed CR encapsulates a building with a circle represented by a center point and radius. In this configuration, the CR essentially functions as the basic building block for higher geometric structures, enhancing the efficiency more than when VR is used alone. The simulation results indicate that the proposed CR scheme reduces the computational load proportionally to the number of potential scattering objects while its hierarchical structure achieves about 50% of computational load reduction in the hierarchical octree structure.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.13 no.2
• /
• pp.189-196
• /
• 2000

The objective of this paper is to propose an analysis technique to predict the behavior of PC wall structures subjected to cyclic load. While PC wall panel is idealized by finite elements, the joints at which PC walls are connected each other are idealized by nonlinear spring elements. Axial and shear spring elements are developed for simulating shear, compression and tension behaviors of joints. The strength and stiffness of each spring elements we presented from the previous research results and incorporated into the computer program of DRAIN-2DX. The proposed analysis technique is evaluated by analyzing specimens previously tested and comparing with those. On the strength, stiffness, energy dissipation and lateral drift, analytical results show good agreements with test results. This means the proposed technique is effective to predict the response of the PC wall structures.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.23 no.3
• /
• pp.136-142
• /
• 2019

Nuclear power plant concrete structures are in contact with the coast, and durability due to chloride attack is very important because it is used as cooling water by taking seawater. For this purpose, a 3-year long-term saltwater immersion test was carried out to evaluate chloride ion diffusion coefficient and age apponent (m) The m values of the foundation with 4,000 class was 0.35 ~ 0.39, similar to KCI or ACI suggested values. essential service water constructions and tunnels of 5,000 class were 0.44 ~ 0.53 and 6,000 class, and 0.62 of reactor containment buildings were similar to the proposed values of FIB. As a result of the prediction of the service life with the measured age coefficient, all the safety related concrete structures of the nuclear power plants satisfied the service life of more than 60 years.