• Journal of Korea Multimedia Society
• /
• v.22 no.9
• /
• 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.

• The Korean Journal of Applied Statistics
• /
• v.24 no.3
• /
• pp.535-546
• /
• 2011

This study develops statistical models for the binary forecast of Asian dust days over South Korea in the winter season. For this study, we used three kinds of data; the rst one is the observed Asian dust days for a period of 31 years (1980 to 2010) as target values, the second one is four meteorological factors(near surface temperature, precipitation, snowfall, ground wind speed) in the source regions of Asian dust based on the NCEP reanalysis data and the third one is the large-scale climate indices. Four kinds of statistical models(multiple regression models, logistic regression models, decision trees, and support vector machines) are applied and compared based on skill scores(hit rate, probability of detection and false alarm rate).

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.15 no.26
• /
• pp.111-117
• /
• 1992

In the stock marketing. investor needs speedy and accurate decision making for the investment. A stock exchange index provides the important index of the early of 1993 in Korea using Fuzzy Delphi Method(F. D. M) which is widely used to a mid and long range forecasting in decision making problem. In the Fuzzy Delphi method, considerably qualified experts an first requested to give their opinion seperately and without intercommunication. The forecasting data of experts consist of Triangular Fuzzy Number (T.F.N) which represents the pessimistic, moderate, and optimistic forecast of a stock exchange index. A statistical analysis and dissemblance index are then made of these subject data. These new information are then transmitted to the experts once again, and the process of reestimation is continued until the process converges to a reasonable stable forecast of stock exchange index. The goal of this research is to forecast the stock exchange index using F.D.M. in which subjective data of experts are transformed into quasi -objective data index by some statistical analysis and fuzzy operations. (a) A long range forecasting problem must be considered as an uncertain but not random problem. The direct use of fuzzy numbers and fuzzy methods seems to be more compatible and well suited. (b) The experts use their individual competency and subjectivity and this is the very reason why we propose the use of fuzzy concepts. (c) If you ask an expert the following question: Consider the forecasting of the price index of stocks in the near future. This experts wi11 certainly be more comfortable giving an answer to this question using three types of values: the maximum value, the proper value, and the minimum value rather than an answer in terms of the probability.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.55 no.10
• /
• pp.426-432
• /
• 2006

This paper describes the concepts of Static Line Rating (SLR) and Dynamic Line Rating (DLR) and the computational methods to demonstrate them. Calculation of the line capacity needs the heat balance equation which is also used for computing the reduced tension in terms of line aging. SLR is calculated with the data from the worst condition of weather throughout the year. Even now, the utilization ratio is obtained from this SLR data in Korea. DLR is the improved method compared to SLR. A process for DLR reveals not only improved line ratings but also more accurate allowed line ratings based on line aging and real time conditions of weather. In order to reflect overhead transmission line aging in DLR, this paper proposes the method that considers the amount of decreased tension since the lines have been installed. Therefore, the continuous allowed temperature for remaining life time is newly acquired. In order to forecast DLR, this paper uses weather forecast models, and applies the concept of Thermal Overload Risk Probability (TORP). Then, the new concept of Dynamic Utilization Ratio (DUR) is defined, replacing Static Utilization Ratio (SUR). For the case study, the two main transmission lines which are responsible for the north bound power flow in the Seoul metropolitan area are chosen for computing line rating and utilization ratio. And then line rating and utilization ratio are analyzed for each transmission line, so that comparison of the present and estimated utilization ratios becomes available. Finally, this paper proves the validity of predictive DUR as the objective index, with simulations of emergency state caused by system outages, overload and so on.

• Environmental and Resource Economics Review
• /
• v.28 no.2
• /
• pp.177-200
• /
• 2019

Energy demand forecast which serves as an essential input in energy policy is exposed to multiple factors of uncertainty such as GDP and weather forecast uncertainty. The Master Plan of Electricity Market in Korea which is biennially prepared is critically based on fluctuating energy demand forecast whereas its resulting proposal on electricity generation mix is substantially irreversible. The paper provides a real options model to evaluate energy transition policy by considering Knightian uncertainty as a measure to study multiple uncertainties with multiple set of probability distributions. Our finding is that the current energy transition policy under the master plan is not robust in terms of securing stable management of electricity demand and supply system.

• Journal of Korea Water Resources Association
• /
• v.52 no.7
• /
• pp.475-482
• /
• 2019

Subseasonal-to-Seasonal (S2S) prediction information which have 2 weeks to 2 months lead time are expected to be used through many parts of industry fields, but utilizability is not reached to expectation because of lower predictability than weather forecast and mid- /long-term forecast. In this study, we used multi-layer perceptron (MLP) which is one of machine learning technique that was built for regression training in order to improve predictability of S2S precipitation data at South Korea through post-processing. Hindcast information of ECMWF was used for MLP training and the original data were compared with trained outputs based on dichotomous forecast technique. As a result, Bias score, accuracy, and Critical Success Index (CSI) of trained output were improved on average by 59.7%, 124.3% and 88.5%, respectively. Probability of detection (POD) score was decreased on average by 9.5% and the reason was analyzed that ECMWF's model excessively predicted precipitation days. In this study, we confirmed that predictability of ECMWF's S2S information can be improved by post-processing using MLP even the predictability of original data was low. The results of this study can be used to increase the capability of S2S information in water resource and agricultural fields.

• Journal of Korean Institute of Industrial Engineers
• /
• v.38 no.3
• /
• pp.208-219
• /
• 2012

As the number of end-of-life vehicles (ELVs) has kept increasing, the management of ELV has also become one of the academic research focuses and European Union recently adopted the directive on ELVs. For the stakeholders has become a principle agent of dealing with all about ELVs, it is relevant investment decision to set up and to decide high-cost ELVs entity locations and to forecast future ELVs' amount in advance. In this paper, transition probability matrixes between months are made by using Markov Chain and the number of ELVs is predicted with them. This study will perform a great role as a fundamental material in Korea where just started having interests about recycling resources and studies related to the topic. Moreover, the forecasting method developed for this research can be adopted for other enhancements in different but comparable situations.

• Journal of Korean Society of Rural Planning
• /
• v.14 no.3
• /
• pp.37-51
• /
• 2008

This study aims to build a model dealing with the location decision of new manufacturing firms and their land demand. The model is composed with 1) the binary logit model structure identifying a future probability of manufacturing firms to locate in a city and their land demand; and 2) the land use suitability of the land demand. The model was empirically tested in the case of Anseong City. We used establishment-level data for the manufacturing industry from the Report on Mining and Manufacturing Survey. 48 industry groups were scrutinized to find the location probability in the city and their land demand via logit model with the dependent variables: number of employment, land capital, building capital, total products, and value-added for a new industry since 2001. It is forecasted that the future land areas (to 2025) for the manufacturing industries in the city are $5.94km^2$ and additional land demand for clustering the existing industries scattered over the city is $2.lkm^2$. Five industrial complex locations were identified through the land use suitability analysis.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.50 no.3
• /
• pp.71-79
• /
• 2008

Generally the price of agricultural products has much different characteristics from that of manufacturing products. If products have the limitation of long-term storage and the short period of cultivation, the price of products can be more unstable. Moreover, the price forecasting is very difficult because it doesn't follow any cycle or trend. However price can be regarded as risk instead of uncertainty if we can calculate the probability of price. Reliability analysis techniques are used for forecasting the price change of Chinese cabbage. This study aims to show the usability of reliability analysis for price forecasting. A price-forecasting model was developed based on weather data of the first 10 days of the full cultivating cycle (80 days) 70 days and the average price and standard deviation of wholesale market prices from 1996 to 2001 and applied to forecast the boom price, or the orice which is over the tolerance of market prices, of upland Chinese cabbage in 2002 and 2003. Applied results showed the possibility of boom price forecasting using reliability analysis techniques.

• The Bulletin of The Korean Astronomical Society
• /
• v.37 no.1
• /
• pp.88.2-88.2
• /
• 2012

We investigate solar flare occurrence rate and daily flare probability depending on McIntosh sunspot classification, its area, and its area change. For this we use the NOAA active region and GOES solar flare data for 15 years (from January 1996 to December 2010). We consider the most flare-productive 10 sunspot classification: 'Dko', 'Dai', 'Eai', 'Fai', 'Dki', 'Dkc', 'Eki', 'Ekc', 'Fki', and 'Fkc'. Sunspot area and its change can be a proxy of magnetic flux and its emergence/cancellation, respectively. we classify each sunspot group into two sub-groups: 'Large' and 'Small'. In addition, for each group, we classify it into three sub-groups according to sunspot group area change: 'Decrease', 'Steady', and 'Increase'. As a result, in the case of compact groups, their flare occurrence rates and daily flare probabilities noticeably increase with sunspot group area. We also find that the flare occurrence rates and daily flare probabilities for the 'Increase' sub-groups are noticeably higher than those for the other sub-groups. In case of the (M+X)-class flares of 'Dkc' group, the flare occurrence rate of the 'Increase' sub-group is three times higher than that of the 'Steady' sub-group. Mean flare occurrence rates and flare probabilities for all sunspot regions increase with the following order: 'Steady', 'Decrease', and 'Increase'. Our results statistically demonstrate that magnetic flux and its emergence enhance major solar flare occurrence. We are going to forecast solar flares based on these results and NOAA scale.