• 한국멀티미디어학회논문지
• /
• 제24권8호
• /
• pp.1122-1137
• /
• 2021

PM_2.5 concentration in Seoul could be predicted by deep neural network model. In this paper, the contribution of input factors to the model's prediction results is analyzed using the LRP(Layer-wise Relevance Propagation) technique. LRP analysis is performed by dividing the input data by time and PM concentration, respectively. As a result of the analysis by time, the contribution of the measurement factors is high in the forecast for the day, and those of the forecast factors are high in the forecast for the tomorrow and the day after tomorrow. In the case of the PM concentration analysis, the contribution of the weather factors is high in the low-concentration pattern, and that of the air quality factors is high in the high-concentration pattern. In addition, the date and the temperature factors contribute significantly regardless of time and concentration.

• Nuclear Engineering and Technology
• /
• 제24권1호
• /
• pp.30-39
• /
• 1992

본 연구에서는 붕산주입 및 회석운전동안에 노심의 붕산농도를 변화시키기 위한 보충수 유량을 예측하고 화학 및 체적제어계통을 포함한 원자로 냉각재계통내에 있는 각종 계통에서 붕산농도 거동분석을 위한 종합적 붕산주입 및 희석모델(INBAD)이 제안되었다. 이 모델은 기존의 노심코드와 새로 개발된 붕산주입 및 희석모델로 구성되어 있으며 붕산주입 및 희석모델은 단일 cell 모델 및 다중 cell모델을 이용하여 본 연구목적에 맞게 개발되었다. 또한, 본 모델에서는 보다 실제적인 붕산농도 거동분석을 위하여 가변적 가압기 가열기 출력 및 선택적인 보충수 운전형태 (직접주입 또는 간접주입)가 모사되었다. 이 모델의 유용성을 증명하기 위하여 영광 3,4호기 설계자료를 이용하여 각종 계통에서 직접주입 및 간접주입운전에 대한 붕산농도 거동분석을 수행하였고, 노심의 붕산농도에 대한 가압기 가열기의 영향을 검토하였다. 그 결과 본 모델은 붕산주입 및 희석운전시에 각종 계통에서 붕산농도 변화를 정확히 예측할 수 있음을 보여 주었다.

• Journal of Radiation Protection and Research
• /
• 제28권3호
• /
• pp.239-245
• /
• 2003

Every nuclear power plant measured concentrations of tritium in groundwater and surface water around the plants periodically. It was not easy to predict the tritium concentration only with these measurement data in case of various release scenarios. KAERI developed a new approach to find the relationship between the tritium release rate and tritium concentration in the environment. The approach was based upon a dynamic compartment model. In this paper the dynamic compartment model was modified to predict the tritium behavior more accurately. The mechanisms considered for the transfer of tritium between the compartments were evaporation, groundwater flow, infiltration, runoff, and hydrodynamic dispersion. Time dependent source terms of the compartment model were introduced to refine the release scenarios. Also, transfer coefficients between the compartments were obtained using realistic geographical data. In order to illustrate the model various release scenarios were developed, and the change of tritium concentration in groundwater and surface water around the nuclear power plants was estimated.

• 한국멀티미디어학회논문지
• /
• 제22권9호
• /
• pp.1069-1079
• /
• 2019

In this study, we developed a $PM_{10}$ forecasting model using DNN and Membership Function, and improved the forecasting performance. The model predicts the $PM_{10}$ concentrations of the next 3 days in the Seoul area by using the weather and air quality observation data and forecast data. The best model(RM14)'s accuracy (82%, 76%, 69%) and false alarm rate(FAR:14%,33%,44%) are good. Probability of detection (POD: 79%, 50%, 53%), however, are not good performance. These are due to the lack of training data for high concentration $PM_{10}$ compared to low concentration. In addition, the model dose not reflect seasonal factors closely related to the generation of high concentration $PM_{10}$. To improve this, we propose Julian date membership function as inputs of the $PM_{10}$ forecasting model. The function express a given date in 12 factors to reflect seasonal characteristics closely related to high concentration $PM_{10}$. As a result, the accuracy (79%, 70%, 66%) and FAR (24%, 48%, 46%) are slightly reduced in performance, but the POD (79%, 75%, 71%) are up to 25% improved compared with those of the RM14 model. Hence, this shows that the proposed Julian forecast model is effective for high concentration $PM_{10}$ forecasts.

• 한국대기환경학회지
• /
• 제10권4호
• /
• pp.224-232
• /
• 1994

A new neural network model has been developed to predict short-term air pollution concentration. In addition, a multiple regression model widely used in statistical analysis was tested. These models were applied for prediction of daily maximum ozone concentration in Seoul during the summer season of 1991. The time periods between May and September 1989 and 1990 were utilized to train set of learning patterns in neural network model, and to estimate multiple regression model. To evaluate the results of the different models, several Performance indices were used. The results indicated that the multiple regression model tended to underpredict the daily maximum ozone concentration with small r$^{2}$(0.38). Also, large errors were found in this model; 21.1 ppb for RMSE, 0.324 for NMSE, and -0.164 for MRE. On the other hand, the results obtained from the neural network model were very promising. Thus, we can know that this model has a prominent efficiency in the adaptive control for the non-linear multi- variable systems such as photochemical oxidants. Also, when the recent new information was added in the neural network model, prediction accuracy was increased. From the new model, the values of RMSE, NMSE and r$^{2}$ were 13.2ppb, 0.089, 0.003 and 0.55 respectively.

• 한국대기환경학회지
• /
• 제7권3호
• /
• pp.156-168
• /
• 1991

The purpose of this paper is to develop a stochastic model for daily sulphur dioxide $(SO_2)$ concentrations prediction in urban area (Seoul). For this, the influence of the meteorological parameters on the $SO_2$ concentrations is investigated by a statistical analysis of the 24-hr averaged $SO_2$ levels of Seoul area during 1989 $\sim$ 1990. The annual fluctuations of the regression trend, periodicity and dependence of the daily concentration are also analyzed. Based on these, a nonlinear regression transfer function model for the prediction of daily $SO_2$ concentrations is derived. A statistical procedure for using the model to predict the concentration level is also proposed.

• 한국대기환경학회지
• /
• 제7권2호
• /
• pp.127-136
• /
• 1991

This paper is concerned with the development and application of a stochastic model for daily sulphur dioxide $(SO_2)$ concentrations in urban area (Seoul). For this, the characteristics of the regression trend, periodicity and dependence of the daily $SO_2$ concentration are investigated by a statistisical analysis of the daily average $SO_2$ values measured in Seoul area during 1989 $\sim$ 1990. Based on these, nonlinear regression time series model for the prediction of daily $SO_2$ concentrations is derived. A statistical procedure for using the model to predict the concentration level is also proposed.

• 한국대기환경학회지
• /
• 제4권2호
• /
• pp.67-71
• /
• 1988

To estimate the probability of short term concentration of air pollution using long term arithmetic average concentration, the procedure was developed and added to Texas Climatological Model version 2. In the procedure, such statistical characteristics that frequency distribution of short term concentration may be approximated by a lognormal distribution, were applied. This procedure is capable of estimating not only highest concentration for a variety of averaging times but also concentrations for arbitrary occurrence frequency. Evaluation of the procedure with the results of short term concentrations calculated by Texas Episodic Model version 8 using the meteorological data and emission data in Seoul shows that the procedure estimates concentrations fairly well for wide range of percentiles.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2022년도 가을 학술논문 발표대회
• /
• pp.53-54
• /
• 2022

There is an increasing demand for prediction of asbestos concentration which has an fatal effect on human body. While demolishing asbestos, the dust scatters and makes workers be exposed to danger. Up to this date, however, factors that particularly influences have not considered in predicting asbestos concentration. Most of the studies could not quantify the distribution of asbestos. Also, they did not use nominal data on buildings as important factors. Therefore, this study aims to build an asbestos concentration prediction model by quantifying distribution of asbestos and using nominal data of buildings based on Artificial Neural Network (ANN). This model can give significant contribution of improving the safety of workers and be useful for finding effective ways to demolish asbestos in planning.

• 생태와환경
• /
• 제41권4호
• /
• pp.472-484
• /
• 2008

The calibrated Andong Reservoir hydro-dynamic module (PART I) of the 2-dimensional hydrodynamic and water quality model, CE-QUAL-W2 [v3.2], was applied to examine the dynamics of total phosphorus, and chlorophyll $\alpha$ concentration within Andong Reservoir. The modeling effort was supported with the data collected in the field for a five year period. In general, the model achieved a good accuracy throughout the calibration period for both chlorophyll ${\alpha}$ and total phosphorus concentration. The greatest deviation in algal concentration occurred on $10^{th}$ October, starting at the layer just beneath the surface layer and extending up to the depth of 35 m. This deviation is principally attributed to the effect of temperature on the algal growth rate. Also, on the same date, the model over-predicts hypolimnion and epilimnion total phosphorus concentration but under-predicts the high concentrated plume in the metalimnion. The large amount of upwelling of finer suspended solid particles, and re-suspension of the sediments laden with phosphorus, are thought to have caused high concentration in the epilimnion and hypolimnion, respectively. Nevertheless, the model well reproduced the seasonal dynamics of both chlorophyll a and total phosphorus concentration. Also, the model tracked the interflow of high phosphorus concentration plume brought by the turbid discharge during the Asian summer monsoon season. Two different hypothetical discharge scenarios (discharge from epilimnetic, and hypolimnetic layers) were analyzed to understand the response of total phosphorus interflow plume on the basis of differential discharge gate location. The simulated results showed that the hypolimnetic discharge gate operation ($103{\sim}113\;m$) was the most effective reservoir structural control method in quickly discharging the total phosphorus plume (decrease of in-reservoir concentration by 219% than present level).