• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.319-319
• /
• 2019

기후변화로 인한 극한 기후 상황의 증가로 홍수기 홍수피해와 갈수기 가뭄피해가 심화되고 있으며, 수자원 관리에 대한 어려움이 발생하고 있다. 효율적인 수자원 관리를 위해 국내에는 약 1,8000여개의 댐을 운영하고 있으며, 댐의 유입량과 저수량을 감안하여 물을 적절하게 방류하는 것을 목적으로 한다. 그러기 위해서는 유입량이 우선적으로 확보되어야 하며, 더 나아가 유입량을 미리 예측할 수 있다면 더욱 효율적인 댐 운영이 가능할 것이다. 기존에는 수위나 유량을 예측하기 위해서는 주로 물리적 모형이 사용되어 왔으며, 물리적 모형은 매개변수 결정을 위한 많은 자료를 필요로 하고 그 과정에서 많은 불확실성을 포함하고 있기 때문에 계산 과정을 거치는 동안 다양한 오차가 반복 누적되는 단점이 있다. 반면에 시계열 데이터 예측을 위한 알고리즘 LSTM(Long Short-Term Memory)은 입력된 데이터와 출력된 데이터를 동시에 이용하여 보다 정확한 예측 값을 얻을 수 있다. 따라서 본 연구는 다목적댐의 유입유량 예측을 위해 구글에서 제공하는 딥러닝 오픈소스 라이브러리를 활용하여 LSTM모형을 구축하고 댐 유입유량을 예측하였다. 분석 자료로는 wamis에서 제공하는 용담댐의 2006년부터 2018년까지의 시간당 유입량 자료를 사용하였으며, 입력 데이터로 모형을 학습한 후 2018년의 유입량을 예측하였다. 예측 값의 정확도를 판단하기 위해 2018년의 실제 유입량 자료와 비교하였다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.2
• /
• pp.611-618
• /
• 2020

The reliability of the results of LCC analysis is determined by accurate analytical procedures and energy data from which the uncertainty is removed. Until now, systems that can automatically measure these energy data and produce databases have not been commercialized. Therefore this paper proposes a concept model of an S-LCC platform that can automatically collect and analyze electric energy consumption data of equipment systems using the IOT, which is the core tool in the Fourth Industrial Revolution and operates the equipment system efficiently using the analyzed results. The proposed concept model was developed by the convergence of existing BLCS and IOT and was comprised of five modules: Facility Control Module, LCC Analysis Module, Energy Consumption Control Module, Efficiency Analysis Module, and Maintenance Standard Reestablishment Module. Using the results of LCC analysis deduced from this system, the deterioration condition of an equipment system can be identified in real-time. The results can be used as the baseline data to re-establish standards for the maintenance factor, replacement frequency, and lifetime of existing equipment, and establish new maintenance standards for new equipment. If the S-LCC platform is established, it would increase the reliability of LCC analysis, reduce the labor force for entering data and improve accuracy, and would also change disregarded data into big data with high potential.

• Knowledge Management Research
• /
• v.25 no.2
• /
• pp.193-218
• /
• 2024

The current study investigated the factors influencing the buyer's repurchase intention for second-hand products. This study first identified perceived risk and purchase value as the two primary influencing variables. Additionally, some exogenous variables influencing these two variables were examined. Statistical analysis using Partial Least Squares (PLS) revealed that product uncertainty, seller uncertainty, and site trust had statistically significant relationships with perceived transaction risk. However, while economic benefit showed a significant impact on purchase value, product scarcity and resale value did not exhibit a significant relationship with purchase value. Perceived transaction risk was found to have an insignificant relationship with repurchase intention, but indirectly influenced repurchase intention through purchase value. Purchase value was identified as having a significant influence on repurchase intention. Therefore, it was concluded that purchase value is the most important factor influencing repurchase intention in the purchase of second-hand products, while transaction risk indirectly influences repurchase intention through purchase value. The study indicates that product uncertainty and economic benefit are the most significant exogenous factors influencing transaction risk and purchase value, respectively.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.197-197
• /
• 2021

최근 기후변화와 하천 제방의 노후화로 인해 수재해 위험이 지속적으로 증가하고 있다. 그러나 기존의 재래적인 하천 제방의 점검은 많은 인력과 예산 소모로 비효율적이며 제방 전구간 점검의 한계, 객관성의 한계 등 많은 한계점들이 존재하여 효과적인 홍수 대응을 위해 새로운 모니터링과 예/경보 시스템의 구축이 반드시 필요한 상황이다. 따라서 본 연구는 인공위성을 이용한 하천 제방 변위 산출과 수문학적 요인과의 관계 분석을 통해 하천 제방 건강상태 모니터링 시스템 방안을 제안하고자 한다. Sentinel-1 SAR 영상과 유럽 우주국(ESA)의 위성 영상 전처리 도구인 SNAP을 이용하여 2020년 여름 붕괴된 남원시의 금곡교 제방의 봄(4~5월), 여름(7~8월)의 변위를 산출하였고, 제방의 위험도 산정을 위해 토양수분관계를 분석하였다. 선행 연구(김상우,2019)에서는 농촌진흥청에서 제공하는 TDR(Time Domain Reflectrometry) 관측값과 Sentinel-1 SAR의 후방 산란계수의 토양수분관계가 일치하는 경향을 제시하여, 본 연구에서는 이를 이용하여 제방 후 방산란계수를 산출하고 변위와 토양수분도의 상관관계를 분석하여 변위 추세와 토양수분도의 추세가 일치하는 경향을 확인하였다. 본 연구 결과를 통해 향후에는 위성을 이용하여 산출한 제방의 변위와 토양수분도의 불확실성을 보완하고 기온, 수위, 토양도, 지하수위와 같은 수문기상학적 데이터의 분석을 통해 초정밀, 자동화 하천 제방 건강상태 모니터링 시스템이 구현 가능할 것으로 기대한다.

• Journal of Digital Convergence
• /
• v.13 no.6
• /
• pp.157-164
• /
• 2015

An efficient retrieval of useful information is a prerequisite of an optimal decision making system. Hence, A research of data mining techniques finding useful patterns from the various forms of data has been progressed with the increase of the application of Big Data for convergence and integration with other industries. Each technique is more likely to have its drawback so that the generalization of retrieving useful information is weak. Another integrated technique is essential for retrieving useful information. In this paper, a uncertainty measure of information is calculated such that algebraic probability is measured by Bayesian theory and then information entropy of the probability is measured. The proposed measure generates the effective reduct set (i.e., reduced set of necessary attributes) and formulating the core of the attribute set. Hence, the optimal decision rules are induced. Through simulation deciding contact lenses, the proposed approach is compared with the equivalence and value-reduct theories. As the result, the proposed is more general than the previous theories in useful decision-making.

• Knowledge Management Research
• /
• v.22 no.1
• /
• pp.57-84
• /
• 2021

In this paper, we explore market and environmental factors which affecting organization's priority claim application which is more powerful and prompt strategic method to protect technology from competitors under uncertain and volatile environments. This study empirically examines why organizations strategically choose the priority claim application which is more strong tactics to protect technology as the source of sustainable competitive advantage. We suggest that market and environmental factors, such as exogenous shock, volatility, and uncertainty, may also affect strategic decision that organization take patent application with claiming priority. The results of our analysis of priority claim application in the Korean high-tech electronics industry from 1994 to 2008 showed that these three strategic factors affected the technology protection decision and organization's status also moderate theses effects, as predicted in our hypotheses.

• Proceedings of the KIEE Conference
• /
• 2001.11b
• /
• pp.256-258
• /
• 2001

전력 계통의 운용 계획을 최적화 하기 위해서 수요예측에 관한 연구가 활발히 진행되고 있다. 기존의 수요예측 기법의 최대 오차는 특수일이 토요일과 월요일인 경우와 연휴인 경우에 발생한다. 이 중 특수일이 토요일과 월요일인 경우는 퍼지 선형회귀분석법과 상대계수법을 이용하여 우수한 결과를 도출한 바 있다. 구정과 추석은 특수일 중 평일과의 부하 차이가 가장 큰 특수일이며 약 $45{\sim}50%$ 정도가 감소된다. 이러한 부하의 감소 폭은 서서히 줄어서 연휴 당일 4일 후에는 완전히 복구가 되며 연휴 전 부하가 낮아지는 시점은 연휴 당일 3일 전이다. 연휴 예측의 불확실성은 연휴 기간의 길이 변동 및 기타 다양한 변수들에 의한 유동성에 기인한다. 특히 추석의 경우 과거 데이터 이용에 더욱 신중해야 하며 타 특수일에 비해 부하 값의 예측이 힘들다. 또한 직전 평일 대비 추석 연휴의 부하는 변화가 심하게 나타나며 본 논문에서는 퍼지 선형회귀분석법을 기본으로 변형된 알고리즘으로 향상된 예측도를 제시한다.

• The KIPS Transactions:PartB
• /
• v.17B no.2
• /
• pp.139-148
• /
• 2010

This paper shows a reinforcement post-processing method and feedback algorithm for improvement of assigning method in classification. Especially, we focused on complex documents that are generally considered to be hard to classify. A basis factors in traditional classification system are training methodology, classification models and features of documents. The classification problem of the documents containing shared features and multiple meanings, should be deeply mined or analyzed than general formatted data. To address the problems of these document, we proposed a method to expand classification scheme using decision boundary detected automatically in our previous studies. The assigning method that a document simply decides to the top ranked category, is a main factor that we focus on. In this paper, we propose a post-processing method and feedback algorithm to analyze the relevance of ranked list. In experiments, we applied our post-processing method and one time feedback algorithm to complex documents. The experimental results show that our system does not need to change the classification algorithm itself to improve the accuracy and flexibility.

• Environmental and Resource Economics Review
• /
• v.21 no.2
• /
• pp.301-319
• /
• 2012

The primary purpose of Renewable Portfolio Standard (RPS) is to facilitate investment in renewable energy technology. Since emission trading program has similar purpose, it is conceivable to attempt to link RPS and emission trading program through interlinked markets. RPS in Korea with single REC and emission allowance markets has particular advantages for constructing linkages between two markets. This paper provides a real option model to examine investment effects of linkage of RPS to the trading program. Emission permit price and REC price are assumed to follow stochastic processes and renewable investment is irreversible. The result shows that linked market provides further incentive for renewable investment by raising managerial flexibility for power companies.

• Journal of Korea Port Economic Association
• /
• v.22 no.3
• /
• pp.43-57
• /
• 2006

Responsiveness to the uncertainty of SCM system shows its competitiveness. In order to secure SCM competitiveness, RFID-related projects aim to enhance both system visibility and process automation. Nowadays, we conduct RFID technology-oriented researches very actively; however, quantitative ROI analysis model from the perspective of SCM does not exist yet, which helps decide the introduction of technology. Therefore, our study suggests a ROI analysis model for the adoption of RFID technology, and we demonstrate its usefulness using the real world data that is taken from one of the government-funded RFID projects in Korea.