강우자료 형태에 따른 인공신경망의 일유입량 예측 정확도 평가

강문성,김계웅,황순호,박지훈,이재남,강문성 한국농공학회 2019 한국농공학회논문집 Vol.61 No.2

The objective of this study was to evaluate the influence of rainfall observation network on daily dam inflow using artificial neural networks(ANNs). Chungju Dam and Soyangriver Dam were selected for the study watershed. Rainfall and dam inflow data were collected as input data for constructionof ANNs models. Five ANNs models, represented by Model 1 (In watershed, point rainfall), Model 2 (All in the Thiessen network, point rainfall),Model 3 (Out of watershed in the Thiessen network, point rainfall), Model 1-T (In watershed, area mean rainfall), Model 2-T (All in the Thiessennetwork, area mean rainfall), were adopted to evaluate the influence of rainfall observation network. As a result of the study, the models that usedall station in the Thiessen network performed better than the models that used station only in the watershed or out of the watershed. The models thatused point rainfall data performed better than the models that used area mean rainfall. Model 2 achieved the highest level of performance. The modelperformance for the ANNs model 2 in Chungju dam resulted in the R2 value of 0.94, NSE of 0.94 NSEln of 0.88 and PBIAS of –0.04 respectively. The model-2 predictions of Soyangriver Dam with the R2 and NSE values greater than 0.94 were reasonably well agreed with the observations. Theresults of this study are expected to be used as a reference for rainfall data utilization in forecasting dam inflow using artificial neural networks.

하수처리수의 재이용을 위한 벼 재배시험

강문성,박승우,김상민,성충현,Kang, Moon-Seong,Park, Seung-Woo,Kim, Sang-Min,Seong, Chung-Hyun 한국농공학회 2004 한국농공학회논문집 Vol.46 No.1

The objective of the research is to develop agricultural resue technologies of reclaiming the effluents from a municipal wastewater treatment plant and reusing for irrigated rice paddies. The Suwon wastewater treatment plant was selected for wastewater reuse tests. The control was the plots with groundwater irrigation (TR#1), the treatment (TR#2) using polluted stream water as it was, and three others using wastewater after treatment. Three levels of wastewater treatments were employed: the effluent from the wastewater treatment plant (TR#3), sand filtering after treatment plant(TR#4), and ultra-violet treatment after sand filtering (TR#5). The randomized block method was applied to wastewater application to paddy rice with five treatments and six replica. The effects of various wastewater treatment levels on water quality, paddy soil, crop growth, yields, and the health hazards were investigated. The primary results indicate that cultivating rice with reclaimed wastewater irrigation did not cause a problem to adverse effects on crop growth and yields. Overall, wastewater could be used as a practical alternative measure for reclaimed wastewater irrigation. However, long-term monitoring is recommended on the effects on soil chemical characteristics and its related health concerns.

An Analysis of Economic Impacts of FTAs on Strategic Industries in Jordan

강문성 한국외국어대학교 국제지역연구센터 2011 International Area Studies Review Vol.14 No.4

This paper analyzes the economic impacts of Jordan’s regional trade agreements on its major strategic industries. We calculate each individual industry’s revealed comparative advantage (RCA) index to identify Jordan’s strategic industries showing comparative advantages. And then we analyze each individual industry’s export performance before and after its regional trade agreements have entered into force. This paper found that the bilateral free trade agreement (FTA) between Jordan and the United States has shown a positive impact on the Jordanian export performance, mainly in the textiles and clothing industry. However, it has not been quite as effective in enhancing Jordanian export opportunities in other industries, such as chemicals, vegetable products, paper and paperboard, and prepared foodstuffs, although there has been a growing positive impact on precious stones. The positive impacts of the FTA with the United States on Jordan’s textiles and clothing industry have been heavily dependent upon its initiatives and programs of QIZs. Pan Arab Free-Trade Area(PAFTA) has shown a relatively positive impact on several industries, such as chemicals, vegetable products, paper and paperboard, precious stones, and prepared foodstuffs, although chemicals and vegetable products have shown a steady export performance no matter whether PAFTA enters into force. However, the bilateral FTA with the EC has failed to show any significant impact on the Jordanian export performances of its strategic industries having comparative advantages.

소유역에서의 수계환경관리 및 평가시스템의 개발(I) - 시스템의 개발 및 구성 -

강문성,박승우,임상준,Kang, Moon-Seong,Park, Seung-Woo,Im, Sang-Jun 한국농촌계획학회 2001 농촌계획 Vol.7 No.1

In an effort to effectively manage and evaluate a water environment at a small watershed, a decision support system for a water environment management and evaluation has been developed. This paper described the overall features and functions of the water environment management and evaluation systems (WEMES) for environmental management, conservation, and evaluation at a small watershed. WEMES consisted of fore subsystems: data, simulation model, evaluation model, and user interface. Each of the systems were briefly described. And special features like simulation and evaluation models were also introduced.

정삼투 응용을 위한 수분산성 마그네틱 나노입자의 제조 및 멤브레인 개질

강문성,김도형,연경호 대한환경공학회 2013 대한환경공학회 학술발표논문집 Vol.2013 No.6

퀀텀닷 어셈블리의 박막 트랜지스터 이용

강문성,Gang, Mun-Seong 한국전기전자재료학회 2016 전기전자재료 Vol.29 No.5

이온성 전도체 전극을 활용한 초고신축성 교류전압구동 발광전자소자 개발

강문성 한국고분자학회 2016 고분자 과학과 기술 Vol.27 No.6

[주제발표 2] 「선진통상국가의 개념과 실천방안」

강문성 한국무역학회 2005 무역학자 전국대회 발표논문집 Vol.2005 No.8

인공신경망 이론을 이용한 단기 홍수량 예측

강문성,박승우 한국농공학회 2003 한국농공학회논문집 Vol.45 No.2

An artificial neural network model was developed to analyze and forecast Short-term river runoff from the Naju watershed, in Korea. Error back propagation neural networks (EBPN) of hourly rainfall and runoff data were found to have a high performance in forecasting runoff. The number of hidden nodes were optimized using total error and Bayesian information criterion. Model forecasts are very accurate (i.e., relative error is less than 3% and R2 is greater than 0.99) for calibration and verification data sets. Increasing the time horizon for application data sets, thus making the model suitable for flood forecasting, decreases the accuracy of the model. The resulting optimal EBPN models for forecasting hourly runoff consists of ten rainfall and four runoff data(ANN0410 model) and ten rainfall and ten runoff data(ANN1010 model). Performances of the ANN0410 and ANN1010 models remain satisfactory up to 6 hours (i.e., R2 is greater than 0.92). 본 연구에서는 모멘트법과 학습적응률을 고려한 오류역전파 알고리즘에 의한 홍수량예측모형을 총 9개 구성하였고, 영산강 유역의 나주지점에 적용하여 보정 및 검증하였으며, 홍수량의 추정을 위한 예보시간별 예측을 수행하였고, 그 결과를 비교 평가하였다. 본 연구의 결과를 정리하면 다음과 같다. 1) 신경망 이론의 학습률과 모멘텀 계수를 고려한 오류역전파 알고리즘을 이용한 총 9개의 유출예측모형을 구성하였다. 2) 각 모형별 은닉층의 노드수에 따른 모형의 학습 결과, ANN0410과 ANN1010 모형이 은닉층의 노드수에 따라 총오차가 가장 작게 나타남으로서 학습이 가장 효과적으로 이루어졌다. 3) 모형별 은닉층의 최적노드수에 따라 보정 자료기간에 대하여 모형을 적용하여 통계적 변량을 비교한 결과, ANN0410과 ANN1010 모형이 학습정도가 뛰어났으며, 유출량만을 입력층으로 구성한 ANN1000T 모형이 학습정도가 가장 떨어지는 것으로 나타났다. 4) 2000년의 폭우사상에 대한 모형별 검증 결과, ANN0410와 ANN0410T 모형이 검증 결과가 뛰어난 것으로 나타났으며, 유출량만을 입력층으로 구성한 ANN1000T 모형이 검증 정도가 가장 떨어지는 것으로 나타났다.5) 1시간에서 6시간까지 시간별 홍수량을 예측하여 모형에 대한 응용 결과, 예측 시간이 길어질수록 실측치의 재현 능력이 떨어졌으며, ANN0410과 ANN1010 모형이 실측치를 잘 재현하고 있는 것으로 나타났으며, 이는 ANN0410과 ANN1010 모형이 유역 홍수 도달시간인 10시간 전의 강우량을 입력층으로 사용했기 때문으로 사료된다. 또한, 면적가중평균 강우량을 사용한 모형보다 각 강우측점의 강우량을 그대로 사용한 모형이 더 우수한 결과를 보였다. 6) 각 모형별로 6시간 후의 홍수량 예측 결과와 실측치의 통계적 변량을 비교한 결과, RB는 0.46∼21.31%, RMSE는 270.82∼638.68 m3/s, RMAE는 0.16∼0.49 m3/s, EI는 0.95∼0.99, 그리고 R2은 0.6517∼0.9273의 범위를 나타냈으며, ANN0410과 ANN1010 모형이 6시간 시간별 예측이 실측치를 가장 잘 반영하는 것으로 나타났고, 시우량 자료를 제외하고 유출량 자료만을 입력층으로 사용한 ANN1000T 모형이 상대적으로 실측치의 재현 정도가 떨어지는 것으로 나타났다.

미국 경제 : 최근 장기호황과 신경제

강문성 사회과학원 2000 계간 사상 Vol.47 No.-