• Structural Monitoring and Maintenance
• /
• 제6권2호
• /
• pp.125-145
• /
• 2019

Bridge scour is one of the predominant causes of bridge failure. Current climate deterioration leads to increase of flooding frequency and severity and thus poses a higher risk of bridge scour failure than before. Recent studies have explored extensively the vibration-based scour monitoring technique by analyzing the structural modal properties before and after damage. However, the state-of-art of this area lacks a systematic approach with sufficient robustness and credibility for practical decision making. This paper attempts to develop a data-driven methodology for bridge scour monitoring using support vector machines. This study extracts features from the bridge dynamic responses based on a generic sensitivity study on the bridge's modal properties and selects the features that are significantly contributive to bridge scour detection. Results indicate that the proposed data-driven method can quantify the bridge scour damage with satisfactory accuracy for most cases. This paper provides an alternative methodology for bridge scour evaluation using the machine learning method. It has the potential to be practically applied for bridge safety assessment in case that scour happens.

• 한국빅데이터학회지
• /
• 제8권1호
• /
• pp.35-47
• /
• 2023

It is crucial to develop effective and efficient big data analytics methods for problem-solving in the field of business in order to improve the performance of data analytics and reduce costs and risks in the analysis of customer data. In this study, a big data-driven data analysis system using artificial intelligence techniques is designed to increase the accuracy of big data analytics along with the rapid growth of the field of data science. We present a key direction for big data analysis systems through missing value imputation, outlier detection, feature extraction, utilization of explainable artificial intelligence techniques, and exploratory data analysis. Our objective is not only to develop big data analysis techniques with complex structures of business data but also to bridge the gap between the theoretical ideas in artificial intelligence methods and the analysis of real-world data in the field of business.

• Journal of Astronomy and Space Sciences
• /
• 제29권1호
• /
• pp.47-49
• /
• 2012

Massive binary stars lose mass by two mechanisms: jet-driven mass loss during periods of active mass transfer and by wind-driven mass loss. Beta Lyrae is an eclipsing, semi-detached binary whose state of active mass transfer provides a unique opportunity to study how the evolution of binary systems is affected by jet-driven mass loss. Roche lobe overflow from the primary star feeds the thick accretion disk which almost completely obscures the mass-gaining star. A hot spot predicted to be on the edge of the accretion disk may be the source of beta Lyrae's bipolar outflows. I present results from spectropolarimetric data taken with the University of Wisconsin's Half-Wave Spectropolarimeter and the Flower and Cook Observatory's photoelastic modulating polarimeter instrument which have implications for our current understanding of the system's disk geometry. Using broadband polarimetric analysis, I derive new information about the structure of the disk and the presence and location of a hot spot. These results place constraints on the geometrical distribution of material in beta Lyrae and can help quantify the amount of mass lost from massive interacting binary systems during phases of mass transfer and jet-driven mass loss.

• 한국IT서비스학회지
• /
• 제7권4호
• /
• pp.231-246
• /
• 2008

This paper suggests an e-learning system model, a goal-driven personalized e-learning system, which increase the effectiveness of learning. An e-learning system following this model makes the learner choose the learning goal. The learner's choice would lead learning. Therefore, the system enables a personalized adaptive learning, which will raise the effectiveness of learning. Moreover, this paper proposes a SCORM standard, which modifies SCORM 2004 that has been insufficient to implement the "goal driven personalized e-learning system." We add a data model representing the goal that motivates learning, and propose a standard for statistics on learning objects usage. We propose each standard for contents model and sequencing information model which are parts of "goal driven personalized e-learning system." We also propose that manifest file should be added for the standard for contents model, and the file which represents the information of hierarchical structure and general learning paths should be added for the standard for sequencing information model. As a result, the system could sequence and search learning objects. We proposed an e-learning system and modified SCORM standards by considering the many factors of adaptive learning. We expect that the system enables us to optimally design personalized e-learning system.

• 한국정보처리학회논문지
• /
• 제3권6호
• /
• pp.1418-1432
• /
• 1996

분산 실시간 트랜잭션 처리 시스템에서 데이타의 복제는 유용성의 향상 및 여러 노드에서 발생한 트랜잭션들을 병행적으로 처리할 수 있기 때문에 성능을 향상시킬 수 있다. 데이타의 일관성과 실시간적 제약성을 충족하기 위해서 동시실행 제어 및 원자적 종료 프로토콜이 시간 구동형 스케쥴링 과정에 통합될 필요가 있다. 기존의 동시실행 제어 프로토콜들에게 발생하는 중단 현상은 트랜잭션들의 종료시한을 만족 할 수 있는 스케쥴링이 어렵기 때문에 시간 구동형 스케쥴링에 적합하지 않다. 복제 된 데이타의 일관성 유지 그리고 스케쥴링 가능성 및 예측가능성을 향상시키기 위하 여 제시된 시간 구동형 스케쥴링 기법은 중단 시간을 최소화하고 우선순위에 따른 직 렬가능성 순서를 생성하는 낙관적 동시실행 제어 프로토콜을 통합하였다. 분산 환경 에서 트랜잭션의 성공적인 종류를 보장하는 원자성도 유지된다. 시간 구동형 스케쥴 링 기법의 성능 분석 결과 및 구체적인 과정들이 기술된다.

• 한국측량학회지
• /
• 제27권1호
• /
• pp.723-733
• /
• 2009

This study proposes a sequence of procedures to extract building boundaries and planar patches through segmentation of rasterized lidar data. Although previous approaches to building extraction have been shown satisfactory, there still exist needs to increase the degree of automation. The methodologies proposed in this study are as follows: Firstly, lidar data are rasterized into grid form in order to exploit its rapid access to neighboring elevations and image operations. Secondly, propagation of errors in raw data is taken into account for in assessing the quality of gradients-driven properties and further in choosing suitable parameters. Thirdly, extraction of planar patches is conducted through a sequence of processes: histogram analysis, least squares fitting, and region merging. Experimental results show that the geometric components of building models could be extracted by the proposed approach in a streamlined way.

• 대한음성학회:학술대회논문집
• /
• 대한음성학회 2003년도 10월 학술대회지
• /
• pp.149-152
• /
• 2003

In this paper, data-driven temporal filter methods[1] are investigated for robust feature extraction. A principal component analysis technique is applied to the time trajectories of feature sequences of training speech data to get appropriate temporal filters. We did recognition experiments on the Korean connected digit telephone speech database released by SITEC, with data-driven temporal filters. Experimental results are discussed with our findings.

• 자동차안전학회지
• /
• 제5권1호
• /
• pp.5-10
• /
• 2013

This paper describes a Forward Collision Warning (FCW) system based on the radar driven fusion with camera. The objective of FCW system is to provide an appropriate alert with satisfying the evaluation scenarios of US-NCAP and a driver acceptance. For this purpose, this paper proposed a data fusion algorithm and a collision warning algorithm. The data fusion algorithm generates information of fusion target depending on the confidence of camera sensor. The collision warning algorithm calculates indexes and determines an appropriate alert-timing by using analysis results of manual driving data. The FCW system with the proposed data fusion and collision warning algorithm was investigated via scenarios of US-NCAP and a real-road driving. It is shown that the proposed FCW system can improve the accuracy of an alarm-timing and reduce the false alarm in real roads.

• 대한환경공학회지
• /
• 제36권6호
• /
• pp.421-428
• /
• 2014

상수관망의 수리학적 거동을 해석함에 있어 현재까지 제안된 일반적인 방법은 유량기반 수리해석(Demand Driven Analysis, DDA)과 압력기반 수리해석(Pressure Driven Analysis, PDA)이다. DDA는 상수관망이 정상적으로 운영되는 상태에서 수리해석 결과를 빠르게 제시할 수 있으며, PDA는 관파괴 및 관내 저압력 상태시 DDA 방법으로는 모의할 수 없는 각 절점의 실질적인 공급가능유량을 제시할 수 있다. PDA 수행에 있어 우선적으로 각 절점에서 특정 압력마다 공급가능유량을 산정할 수 있는 관계식(Head-Outflow Relationship, HOR)을 결정해야 한다. 기존 연구들에서는 상수관망을 구성하는 각 절점에서 HOR이 다르고 각 절점의 HOR 검증을 위해서는 현장데이터의 실측이 필요하다고 지적하였다. 또한 다중 관파괴 상황과 같이 수리학적 조건이 변경된 상태에서 각 절점의 HOR 특성을 파악하기 위해서는 모든 절점에 모니터링 시스템이 구축되어 있어야 하는 제약조건이 있다. 본 연구는 실제 상수관망 형태의 모의 상수관망을 제작하여 상수관망의 정상상태 및 현장에서 측정이 어려운 비정상상태 운영조건을 시나리오별로 구성하고 각 조건 하에서 절점들의 압력 및 유량을 실시간으로 확인하였다. 또한 측정된 실험결과와 기존 HOR 관계식을 비교 평가하여 최소 오차 결과를 보이는 HOR을 제시하였으며, 특정 절점에서 다중 관파괴 발생시 HOR 특성을 파악하였다. 이로부터 관파괴 위치가 다르거나 다중 관파괴가 발생하더라도 특정 절점의 압력(수두) 저하에 따른 공급가능유량의 축소 현상은 본래 대상 절점이 가지고 있는 지수함수 형태인 고유한 HOR을 따라 감소하는 것으로 확인되었다. 본 연구에서 실험을 통해 측정한 압력-유량 관계와 기존 HOR 식들을 비교한 결과 최소 오차를 보이는 지수함수 형태의 Wagner 등1)이 제안한 식을 최적의 HOR 관계식으로 제안하며, 해당 식의 지수 m은 3.0으로 적용할 것을 제안한다.

• 한국환경과학회지
• /
• 제11권1호
• /
• pp.41-56
• /
• 2002

In order to estimate the quantitative roles of the tide induced residual currents, density currents and wind driven currents on the residual currents in Kwangynag Bay Korea, numerical experiments with a diagnostic multi-level model were carried out. Density currents were calculated from the temperature and salinity data observed in January, May, August and November 1998. Anti-clockwise circulations are formed at the western inner part of the bay, the location from the east of Myodo Is. to the south of POSCO Co. and the place between Yeosu peninsula and Namhae Is. from the calculation results of the tide induced residual currents. Velocities of the density currents are less than 3cm/s at the western inner part of the bay and about 5cm/s at the southern entrance of the bay. Density currents get strong in summer and weak in autumn. Wind driven currents at the surface layer flow in the directions of the given winds which are the daily mean winds when the temperature and salinity observations are carried out. In the middle and lower layers, however the wind driven currents flow in the opposite direction to the surface currents as supplementary currents. The surface wind driven currents are greater than the tide induced residual currents or the density currents. The calculated residual currents including the tide induced residual currents, density currents and wind driven currents agree with the results of the current observations approximately. In the Bay, the wind driven currents affect on the residual currents greatestly and tide induced residual currents and density currents do in the second place and the third place.