• Journal of the Korean Society for Nondestructive Testing
• /
• v.16 no.2
• /
• pp.95-108
• /
• 1996

In ultrasonic testing, flaw signal from which quantitative information on flaws is determined is influenced by 3 factors : (1) the incident wavefield.produced by the transducer, (2) the scattered waves produced by flaws, and (3) the reception of the scattered waves back at the transducer. So even small changes in transducer performance due to aging or unexpected damages can produce the changes in the characteristics of flaw signal and finally the changes in the quantitative information on flaws. Thus a reliable calibration method of transducer performance is desired. Recently, theoretical models for ultrasonic testing have been employed as reference standards for the calibration of transducers which are considered as circular planar piston sources in the most of cases. But this simplification cannot be applied to partially damaged transducer which has lost their symmetry in performance, even not in appearance. Unfortunately there has been no reliable practical model which can be used for the calibration of partially damaged transducers. Here a pulse-echo testing model for partially damaged ultrasonic transducers was developed with experimental verification. The experimental responses agree very well with the theoretical prediction. So we expect that this model can be served as a theoretical reference standards for transducer calibration.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.11 no.2
• /
• pp.108-120
• /
• 2008

This paper discusses the prediction of deforestation areas using probability models from forest census database, Geographic information system (GIS) database and the land cover database. The land cover data was analyzed using remotely-sensed (RS) data of the Landsat TM data from 1989 to 2001. Over the analysis period of 12 years, the deforestation area was about 40ha. Most of the deforestation areas were attributable to road construction and residential development activities. About 80% of the deforestation areas for residential development were found within 100m of the road network. More than 20% of the deforestation areas for forest road construction were within 100m of the road network. Geographic factors and vegetation change detection (VCD) factors were used in probability models to construct deforestation occurrence map. We examined the size effect of area partition as training area and validation area for the probability models. The Bayes model provided a better deforestation prediction rate than that of the regression model.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.383-383
• /
• 2023

오늘날 수문학 분야에서는 유역에 대한 강우-유출 시뮬레이션을 머신 러닝(ML: Machine Learning)을 활용하여 다양한 연구를 실행하고 있다. 본 연구에서는 시간별 강우-유출 예측 모델인 GR4H(Génie Rural à 4 paramètres Horaires)를 사용하여 충주댐 유역을 대상으로 연구를 수행하였다. 유역의 속성에 따라서 모델의 성능이 어떻게 달라지는지 비교하여 특성에 맞는 모델을 알아내고. 또한 이 과정에서 기상 및 유출 데이터의 보정 길이를 가지고 어느 정도의 데이터 기간이 모델에서 좋은 성능을 보이는지 파악하였다. 뿐만 아니라 모델에 필요한 선행기간의 데이터가 있는 경우와 없는 경우를 비교하여 어떠한 차이를 보이는지, 그리고 선행기간은 얼마나 필요한지 연구를 통하여 알아냈다. 본 연구를 통하여 충주댐 유역에 대한 모델의 적용성 및 성능을 파악하고 수문 모형 구축에 제한이 있는 유역에 대해서도 사용이 가능한지 판단한다. 실험 유역의 관측 값을 모델에 입력한 후 각 모델에 해당하는 매개변수의 최적값을 찾아내는 과정을 거쳐 시뮬레이션을실 행했다. 본 연구에서 사용한 강우-유출 모델인 GR4H는 프랑스의 INRAE-Antony(Institut National de la recherche agronomique-Antony)에서 만들어진 airGR의 일종으로, 시간별 강우-유출 예측을 위해 개발된 공정 기반(process-based)의 집중적, 개념적 수문학 모델이다. 4개의 매개변수(parameter)가 있으며 이는 유역의 특정 속성을 나타낸다. GR4H를 시뮬레이션 하는 과정에서 매개변수의 최적화를 위해 적절한 보정 길이를 파악하여야 한다. 이러한 과정은 4년, 5년, 6년 등 1년씩 데이터의 양을 늘려가며 매개변수를 최적화한다. 이 과정에서 기상 및 유출 데이터의 적절한 보정 길이를 찾아낸다. 시뮬레이션을 통해 얻은 데이터를 관측 값과 비교하여 모델의 성능을 평가하고 다른 관측 값을 통해 시뮬레이션을 실행하여 검증을 거친다.

• Journal of KIISE:Software and Applications
• /
• v.37 no.4
• /
• pp.323-339
• /
• 2010

The complexity of embedded systems and the effort to develop them has been rising in proportion with their importance. Also, the heterogeneity of the hardware and software parts in embedded systems makes it more challenging to develop. Errors caused by hardware/software interfaces, especially, account for up to 13 percent of failures with an increasing trend. Therefore, verifying the interface between hardware and software in embedded system is one of the most important research areas. However, current approaches such as co-simulation method and model checking have explicit limitations. In this paper, we propose the synthesizable interface co-verification framework for hardware/software co-design. Firstly, we introduce the separate interface specifications for the heterogeneous components to describe hardware design and software design. Our specifications are expressive enough to describe both. We also provide the transformation rules from the software specification to the hardware specification so that the whole system can be described from the software view. Secondly, we address the solution of verifying the interface of the software and hardware design by adopting and extending existing verification-techniques and extending them. In hardware interface verification, we exploit the model checking technique and provide more efficient verification by closing the hardware design from the assumption of the software behavior which is ensured by software verification step. Lastly, we generate the interface codes such as device APIs, device driver, and device controller from the specification so that verified hardware and software codes can be synthesized without extra efforts.

• Information Systems Review
• /
• v.25 no.4
• /
• pp.265-290
• /
• 2023

Even though crowdfunding has become popular as a novel means of raising capital for early-stage ventures and startups through an Internet-based platform, it is unclear how a funder's characteristics, such as motivation and ability, influence their information processing and pledging decision. This study aims to propose and test a research model for determining the relationships between a funder's personal attributes, information processing style, and funding intention. To test the research model, we collected data from 139 Amazon Mechanical Turk participants through an online questionnaire survey. The findings indicate that a funder's self-efficacy has a positive effect on heuristic processing but has no significant effect on systematic processing. By contrast, a funder's personal relevance positively influences both systematic and heuristic processing. Furthermore, heuristic processing, as well as perceived value and perceived risk, influence pledging intentions positively. Our findings potentially contribute to improving the design of crowdfunding platforms to better support a funder's information needs. Based on our findings, we discuss the implications of our study as well as the directions for future research.

• Journal of Intelligence and Information Systems
• /
• v.12 no.2
• /
• pp.167-183
• /
• 2006

Since data mining attempts to find unknown facts or rules by dealing with also vaguely-known data sets, it always suffers from high error rate. In order to reduce the error rate, many researchers have employed multiple models in solving a problem. In this research, we present a new type of multiple models, called DyMoS, whose unique feature is that it classifies the input data and applies the different model developed appropriately for each class of data. In order to evaluate the performance of DyMoS, we applied it to a real customer churn problem of an automobile insurance company, The result shows that the DyMoS outperformed any model which employed only one data mining technique such as artificial neural network, decision tree and case-based reasoning.

• Journal of Korea Society of Industrial Information Systems
• /
• v.29 no.3
• /
• pp.79-87
• /
• 2024

Recently, the importance of pre-trained language models (PLM) has been emphasized for natural language processing (NLP) such as text classification, sentiment analysis, and question answering. Korean PLM shows high performance in NLP in general-purpose domains, but is weak in domains such as finance, medicine, and law. The main goal of this study is to propose a language model learning process and method to build a financial-specific language model that shows good performance not only in the financial domain but also in general-purpose domains. The five steps of the financial-specific language model are (1) financial data collection and preprocessing, (2) selection of model architecture such as PLM or foundation model, (3) domain data learning and instruction tuning, (4) model verification and evaluation, and (5) model deployment and utilization. Through this, a method for constructing pre-learning data that takes advantage of the characteristics of the financial domain and an efficient LLM training method, adaptive learning and instruction tuning techniques, were presented.

• Journal of Aerospace System Engineering
• /
• v.18 no.4
• /
• pp.10-17
• /
• 2024

Solar-powered unmanned aerial vehicles(SPUAV), which are being actively developed domestically and internationally, generally feature high aspect ratio(AR) wings. These high AR wings necessitate a lightweight design as their weight increases, rendering them susceptible to flutter. Consequently, flutter analysis is critical from the initial design phase. Typically, flutter analysis is conducted using a standard section wing or more precisely through a 3D model. However, due to the extended analysis time required by 3D models, this study opts for a 2D aircraft model. The 2D model computes faster than the 3D model and intuitively secures the flutter boundary. In this study, a structural/aerodynamic force model of the 2D aircraft was established, and the findings were compared with those from a 3D half model. The results showed that the flutter analysis between the 2D model and the 3D half model was similar, within about a 3% margin, thus validating the proposed 2D model's effectiveness.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2012.05a
• /
• pp.159-159
• /
• 2012

유역의 도시화는 지체시간, 첨두유량 및 총 유출량 등 홍수 유출특성 뿐만 아니라 개발행위에 따른 토사유출특성 변화를 초래하는 등 다양한 문제를 유발하고 있다. 또한 인구증가, 산업발달, 교통량 증가로 인한 화석연료의 소비증가로 대기 중의 이산화탄소의 농도가 급증하여 기후변화에도 큰 영향을 미치고 있다. 이러한 지나친 도시화 진행을 억제하고 웰빙과 건강에 대한 관심 증대에 따라 저탄소 녹색사업의 일환으로 도시 녹지조성 계획이 진행되고 있다. 따라서 본 연구에서는 도시하천인 중랑천 ($288km^2$) 유역을 대상으로 도시화 진행에 따른 토지이용 변화가 강우-유출 특성에 미치는 영향을 분석하였다. 이를 위해, 중랑천 유역의 과거 (1975, 1980, 1985, 1990, 1995, 2000) 토지 이용도로부터 각 항목별 면적변화 추이를 분석한 결과, 17.8%의 도시지역 면적 증가가 나타났다. 미래토지이용 예측을 위하여 CLUE-s (Conversion of Land Use change and its Effects) 모델을 이용하였다. 과거 토지이용 변화 특성을 분석하여 토지피복의 변화와 전이 특성값을 결정하였고, 이를 바탕으로 토지면적 시나리오, 변화 제한지역, 회귀식 결과와 토지이용 변화 특성에 따른 CLUE-s 모델을 이용하여 미래 토지이용변화 (2040, 2080) 모의를 실시하였다. 이러한 토지이용 변화에 따른 유출특성변화 모의를 위해, 물리적 기반의 준분포형 강우-유출 모형인 SWAT (Soil and Water Assessment Tool) 모형을 이용하였다. 모형의 적용성 평가를 위해, 매개변수 민감도 분석에 따른 도시유역 최적의 유출관련 매개변수를 선정하고 중랑교 지점의 일별 유출량자료(2000~2009)를 이용하여 보정 및 검증을 실시하였다. SWAT 모형의 검보정 후, 예측된 미래 토지이용도를 적용하여 과거와 현재, 미래 토지이용변화에 따른 유출특성변화를 비교분석하였다.

• Proceedings of the KIPE Conference
• /
• 2016.07a
• /
• pp.209-210
• /
• 2016

프로토타입 컨버터 구현 시 전압/전류 제어기의 안정성, 제어 알고리즘, 정상적인 게이트 전압 제어가 보장되지 않은 경우 이를 수정하기 위해 사용하는 기존의 Trial and Error 방식은 조정에 많은 시간 및 비용이 소모된다. 본 논문은 이러한 시간적 비용 저감을 위하여 Control Hardware-in-the-Loop Simulation(CHILS)를 이용하여 제어기의 정상 동작 여부와 성능을 실시간으로 검증하는 방법을 제안한다. 이를 위해 LLC 공진형 컨버터를 CHILS로 구현하여, 개발된 DSP 제어기의 성능을 검증하고자 한다. 제안된 실시간 모의시험에서는 LLC 공진형 컨버터를 Matlab/Simulink에서 모델링 하여, 실제 DSP 제어기의 신호를 컨버터 모델에서 입력받아 모의시험장치의 출력 결과를 관찰함으로써 제어기의 동작 특성을 확인하였다.