• 한국물환경학회지
• /
• 제20권3호
• /
• pp.223-230
• /
• 2004

The QUAL2E and Box-Jenkins time series model were applied to the Miho river, a main tributary of the Geum river, to predict water quality. The models are widely used to predict water quality in rivers and watersheds because of its accuracy. As results of the study, we concluded as follows: Pollutant loadings in upper stream of Miho river were determined to 57,811 kgBOD/d, 19,350 kgTN/d, and 5,013 kgTP/d. The loading of TN in Mushim river was 19,450 kgTN/d, respectively. As the mass loadings were compared with pollutant sources, it concluded that the farming livestock contributed highly to mass emissions of BOD and TP and the population contributed to TN mass loading. The observed water quality values were applied to the models to verify and the models were used to predict the water quality. The QUAL2E Model predicted the concentrations of DO, BOD, TN and TP with high accuracy, but not for E-Coli. The Box-Jenkins time series model also showed high prediction for DO, BOD and TN. However, the concentrations of TP and E-Coli were poorly predicted. The result shows that the QUAL2E model is more applicable in Miho basin for prediction of water quality compared to Box-Jenkins time series model.

• 품질경영학회지
• /
• 제35권4호
• /
• pp.61-66
• /
• 2007

The performance of a building material degrades as time goes by and the failure of the material is often defined as the point at which the performance of the material reaches a pre-specified degraded level. Based on a stochastic deterioration model, a performance based service life prediction method for building materials and components is developed. As a stochastic degradation model, a gamma process is considered and lifetime distribution and service life of a material are predicted using the degradation model. A numerical example is provided to illustrate the use of the proposed service life prediction method.

• 한국시뮬레이션학회논문지
• /
• 제30권2호
• /
• pp.41-48
• /
• 2021

농촌 인구의 감소와 고령화가 지속되면서 농업 생상성 향상의 중요성이 높아지고 있는 가운데, 농작물 품질에 대한 조기 예측은 농업 생산성 및 수익성 향상에 중요한 역할을 할 수 있다. 최근 CNN 기반의 딥러닝 기술 및 전이 학습을 활용하여 농작물의 질병을 분류하거나 수확량을 예측하는 연구가 활발하게 진행되고 있지만, 수확 후 농작물의 품질을 식재단계에서 조기에 예측하는 연구는 찾아보기 힘들다. 본 연구에서는 건강 기능성 식품으로 주목받고 있는 새싹삼을 대상으로, 수확 후 새싹삼의 품질을 식재단계에서 조기에 예측하는 모델을 제안한다. 이를 위하여 묘삼의 이미지를 촬영한 후 수경재배를 통해 새싹삼을 재배하였고, 수확 후 새싹삼의 품질을 분류하여 실험 데이터를 수집하였다. 다수의 CNN 기반의 사전 학습된 모델을 활용하여 새싹삼 조기 품질 예측 모델을 구축하고, 수집된 데이터를 이용하여 각 모델의 학습 및 예측 성능을 비교 분석하였다. 분석 결과 모든 예측 모델에서 80% 이상의 예측 정확도를 보였으며, 특히 ResNet152V2 기반의 예측 모델에서 가장 높은 정확도를 보였다. 본 연구를 통해 인력에 의존하던 기존의 묘삼 선별 작업을 자동화하여 새싹삼의 품질을 높이고 생산량을 증대시켜 농가의 수익창출에 기여할 수 있을 것으로 기대된다.

• 한국항만학회지
• /
• 제12권2호
• /
• pp.313-322
• /
• 1998

The distributions of chemical oxygen demand (COD) and suspended solid (SS) in Ulsan Bay were simulated and reproduced by a numerical ecosystem model for the practical application to the management of marine water quality and the prediction of water quality change due to coastal developments or the constructions of breakwater and marine facilities. Comparing the computed with the observed data of COD and SS in Ulsan bay the results of simulation were found to be good enough to satisfy the practical applications.

• 한국농공학회:학술대회논문집
• /
• 한국농공학회 2005년도 학술발표논문집
• /
• pp.527-533
• /
• 2005

Recently reservoir is polluted by concentrative development of urbanization. Accordingly, the prediction of water quality has import meaning for protecting of water quality pollution. This study was carried out to predict water quality of Gyung Cheon reservoir by WASP5. We have established an integrated system on the basis of web, which predicts the future quality of water through water quality model, WASP5 based on information of water environment in a reservoir for agriculture, uniting expert system which supports the determination to set up measures for improving the quality of water to cope with the result.

• 물과 미래
• /
• 제17권1호
• /
• pp.31-36
• /
• 1984

This study has performed to investigate and evaluate the simulation model of steam Water Quality and the simulated results have 매내 been compared with the observed data in the Han River. The predicted BOD, Total-N, Coliform concentrations in the downstream of the Chungrang-Cheon are 8.6m/1, 4.5mg/1 and $3.7X10^5$ respectively. It is interesting to note that the results simulated based on the WQRRS model are extremely in good agreement and also are very much comparable with those observed data reported previously references.

• Environmental Engineering Research
• /
• 제10권2호
• /
• pp.62-70
• /
• 2005

Pollution loading of Combined Sewer Overflows (CSOs) is frequently over the capacity of a wastewater treatment plant (WWTP) receiving the water. The objectives of this study are to investigate water quality of CSOs in Anmyun-ueup, Tean province and to apply Storm Water Management Model to predict flow rate and water quality of the CSOs. The capacity of a local WWTP was also estimated according to rainfall duration and intensity. Eleven water quality parameters were analyzed to characterize overflows. SWMM model was applied to predict the flow rate and pollutant load of CSOs during rain event. Overall, profile of the flow and pollutant load predicted by the model well followed the observed data. Based on model prediction and observed data, CSOs frequently occurs in the study area, even with light precipitation or short rainfall duration. Model analysis also indicated that the local WWTP’s capacity was short to cover the CSOs.

• 품질경영학회지
• /
• 제21권1호
• /
• pp.152-160
• /
• 1993

In developing a quadratic regression relationship, independent variable is frequently measured with error. In this paper the integrated mean square error of prediction is developed for a quadratic functional relationship model as a measure of the effect of measurement error of the independent variable on the predicted values. The amount of the effect of error is presented and illustrated with an example.

• 산업경영시스템학회지
• /
• 제18권34호
• /
• pp.107-114
• /
• 1995

The objective of this study is to design a one-stage dynamic prediction model with Kalman state space model. For a model verification, it is compared with EWMA(Exponentially Weighed Moving Average) model. The model designed in this research can be extended to process prevention control and quality monitoring.

• 한국컴퓨터정보학회논문지
• /
• 제21권8호
• /
• pp.57-64
• /
• 2016

Initial prevention activities and rapid propagation conditions is the most important to prevent diffusion of water pollution. If water pollutants flow into streams river or main stresm located in environmental conservation area or water intake facilities, we must predict immediately arrival time and the diffusion concentration to the proactive. National Institute of Environmental Research developed water pollution incident response prediction system linking dam and movable weir. the system is mathematical model which is updated daily. Therefore it can quickly predict the arrival time and the diffusion concentration when there are accident of oil spills and hazardous chemicals. Also we equipped with mathematical model and toxicity model of EFDC(Environmental Fluid Dynamics Code) to calculate the arrival time and the diffusion concentration. However these systems offer the services of an offline manner than real-time control services. we have ensured the reliability of data collection and have developed a real-time water quality measurement data transmission device by using the data linkage utilizing a mode bus communication and a commercial SCADA system, in particular, we implemented to be able to do real-time water quality prediction through information infrastructure of the water quality integrated management business created by utilizing the construction of the real-time prediction system that utilizes the data collected, the Open map, the visual representation using charts API and development of integrated management system development based on web maps.