• Journal of the Korean Solar Energy Society
• /
• v.30 no.6
• /
• pp.66-72
• /
• 2010

In this study, we analyzed the characteristics of wind speed and wind direction at different locations in Jeju area using past 10 years observed data and used them in our wind power forecasting model. Generally the strongest hourly wind speeds were observed during daytime(13KST~15KST) whilst the strongest monthly wind speeds were measured during January and February. The analysis with regards to the available wind speeds for power generation gave percentages of 83%, 67%, 65% and 59% of wind speeds over 4m/s for the locations Gosan, Sungsan, Jeju site and Seogwipo site, respectively. Consequently the most favorable periods for power generation in Jeju area are in the winter season and generally during daytime. The predicted wind speed from the forecast model was in average lower(0.7m/s) than the observed wind speed and the correlation coefficient was decreasing with longer prediction times(0.84 for 1h, 0.77 for 12h, 0.72 for 24h and 0.67 for 48h). For the 12hour prediction horizon prediction errors were about 22~23%, increased gradually up to 25~29% for 48 hours predictions.

• Journal of Korea Port Economic Association
• /
• v.27 no.4
• /
• pp.1-11
• /
• 2011

In much the same way as the US Lehman crisis of 2008-2009 severely impacted the European economy through financial market dislocation, a European banking crisis would materially impact the US economy through a generalized increase in global risk aversion. A deepening of the European crisis could very well derail the US economic recovery and have a harmful impact on the Asian economies. This kind of vicious circle could be a bad news to the shipping companies. The purpose of the study is to predict the Baltic Dry Index representing the shipping business during the period of 2012 using the ARIMA-type models. This include the ARIMA and Intervention-ARIMA models. This article introduces the four ARIMA models and six Intervention-ARIMA models. The monthly data cover the period January 2000 through October 2011. The out-of-sample forecasting performance is also calculated. Forecasting performance is measured by three summary statistics: root mean squared percent error, mean absolute percent error and mean percent error. The root mean squared percent errors, however, are somewhat higher than normally expected. This reveals that it is very difficult to predict the BDI The ARIMA-type models show that the shipping market will be bearish in 2012. These pessimistic ex-ante forecasts are supported by the Hodrick-Prescott filtering technique.

• Journal of the Korean Institute of Gas
• /
• v.23 no.5
• /
• pp.52-58
• /
• 2019

In this study, we have developed a forecasting model for city- gas acceptance. City-gas corporations have to report about city-gas sale volume next year to KOGAS. So it is a important thing to them. Factors influenced city-gas have differences corresponding to usage classification, however, in city-gas acceptence, it is hard to classificate. So we have considered tha outside temperature as factor that influence regardless of usage classification and the model development was carried out. ARIMA, one of the traditional time series analysis, and LSTM, a deep running technique, were used to construct forecasting models, and various Ensemble techniques were used to minimize the disadvantages of these two methods.Experiments and validation were conducted using data from JB Corp. from 2008 to 2018 for 11 years.The average of the error rate of the daily forecast was 0.48% for Ensemble LSTM, the average of the error rate of the monthly forecast was 2.46% for Ensemble LSTM, And the absolute value of the error rate is 5.24% for Ensemble LSTM.

• Korean Journal of Agricultural Science
• /
• v.47 no.4
• /
• pp.1029-1037
• /
• 2020

The machine learning algorithm has been widely used in water-related fields such as water resources, water management, hydrology, atmospheric science, water quality, water level prediction, weather forecasting, water discharge prediction, water quality forecasting, etc. However, water quality prediction studies based on the machine learning algorithm are limited compared to other water-related applications because of the limited water quality data. Most of the previous water quality prediction studies have predicted monthly water quality, which is useful information but not enough from a practical aspect. In this study, we predicted the dissolved oxygen (DO) using recurrent neural network with long short-term memory model recurrent neural network long-short term memory (RNN-LSTM) algorithms with hourly- and daily-datasets. Bugok Bridge in Oncheoncheon, located in Busan, where the data was collected in real time, was selected as the target for the DO prediction. The 10-month (temperature, wind speed, and relative humidity) data were used as time prediction inputs, and the 5-year (temperature, wind speed, relative humidity, and rainfall) data were used as the daily forecast inputs. Missing data were filled by linear interpolation. The prediction model was coded based on TensorFlow, an open-source library developed by Google. The performance of the RNN-LSTM algorithm for the hourly- or daily-based water quality prediction was tested and analyzed. Research results showed that the hourly data for the water quality is useful for machine learning, and the RNN-LSTM algorithm has potential to be used for hourly- or daily-based water quality forecasting.

• The Journal of the Korea Contents Association
• /
• v.7 no.3
• /
• pp.140-151
• /
• 2007

In this study, we attempted to develop a watershed runoff index subject to main control points by dividing the Geum River basin into 14 sub-basins. The Yongdam multipurpose dam Daecheong multipurpose dam and Gongju gage station were selected to serve as the main control points of the Geum River basin, and the observed flow of each control point was calculated by the discharge rating curve, whereas the simulated flow was estimated using the Rainfall Runoff Forecasting System (RRFS), user-interfaced software developed by the Korea Water Corporation, based on the Streamflow Synthesis and Reservoir Regulation (SSARR) model developed by the US Army Corps of Engineers. This study consisted of the daily unit observed flow and the simulated flow of the accumulated moving average flow by daily, 5-days, 10-days, monthly, quarterly and annually, and normal monthly/annually flow. We also performed flow duration analysis for each of the accumulated moving average and the normal monthly/annually flows by unit period, and abundant flow, ordinary flow, low flow and drought flow estimated by each flow duration analysis were utilized as watershed runoff index by main control points. Further, as we determined the current flow by unit period and the normal monthly/annually flow through the drought and flood flow analysis subject to each flow we were able to develop the watershed runoff index in a system that can be used to determine the abundance and scarcity of the flow at the corresponding point.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.21 no.3
• /
• pp.19-34
• /
• 2018

The purpose of this study is to estimate monthly agricultural reservoir storage with multiple linear regression model(MLRM) based on reservoir storage and meteorological data. The regression model was developed using 15 years(2002 to 2016) of 3,067 reservoirs by KRC(Korea Rural Community) and 63 meteorological stations by KMA (Korean Meteorological Administration), and the MLRM showed the determination coefficient($R^2$) of 0.51~0.95. The MLRM was applied to 9 selected reservoirs among the whole reservoirs and validated with $R^2$ of 0.44~0.81. The ROC(Receiver Operating Characteristics) analysis of Reservoir Drought Index(RDI) classified by comparing the present reservoir storage with normal year(1976~2005 average) reservoir storage showed average value of 0.64 for 2 years(2015~2016) with the highest value of 0.70 for winter period, lowest value of 0.58 for summer period. If 1 to 3 months weather forecasting data such as Glosea5 produced by KMA are applied, the predicted monthly reservoir storage from the MLRM can be a useful information for agricultural drought pre-preparation.

• Journal of the Korean Society for Marine Environment & Energy
• /
• v.17 no.2
• /
• pp.122-130
• /
• 2014

Seasonal variations and long term linear trends of SST (Sea Surface Temperature) at Yeosu Coast ($127^{\circ}37.73^{\prime}E$, $34^{\circ}37.60^{\prime}N$) in Korea were studied performing the harmonic analysis and the regression analysis of the monthly mean SST data of 46 years (1965-2010) collected by the Fisheries Research and Development Institute in Korea. The mean SST and the amplitude of annual SST variation show $15.6^{\circ}C$ and $9.0^{\circ}C$ respectively. The phase of annual SST variation is $236^{\circ}$. The maximum SST at Yeosu Coast occurs around August 26. Climatic changes in annual mean SST have had significant increasing tendency with increase rate $0.0305^{\circ}C/Year$. The warming trend in recent 30 years (1981-2010) is more pronounced than that in the last 30 years (1966-1995) and the increasing tendency of winter SST dominates that of the annual SST. The time series model that could be used to forecast the SST on a monthly basis was developed applying Box-Jenkins methodology. $ARIMA(1,0,0)(2,1,0)_{12}$ was suggested for forecasting the monthly mean SST at Yeosu Coast in Korea. Mean absolute percentage error to measure the accuracy of forecasted values was 8.3%.

• Journal of Korea Water Resources Association
• /
• v.35 no.5
• /
• pp.583-595
• /
• 2002

The purposes of this study are to perform the management and monitoring of droughts for Mokpo area via the monthly Palmer index(PDSI), the data is obtained from the Mokpo meteorological station, and the used data are in the period of 1906 to 1999. Monthly Palmer index is classified into 7 stochastic classes and its dynamic change of monthly transition probability estimated by Markov chain is investigated. We also estimate the steady state probability of the classified PDSI. The 4th class shows the highest frequency of 49.6% out of 7 classes and the 7th class which is the most extreme drought show that a stochastic transition probability is more or less larger than an empirical one. Also, we found that the monthly steady state probability could be used for the forecasting of changing pattern of drought magnitude for the study area.

• Journal of Korea Port Economic Association
• /
• v.26 no.1
• /
• pp.222-233
• /
• 2010

This paper aims at predicting the BDI from Jan. to Dec. 2010 using such econometric techniues of the univariate time series as stochastic ARIMA-type models and Hodrick-Prescott filtering technique. The multivariate cause-effect econometric model is not employed for not assuring a higher degree of forecasting accuracy than the univariate variable model. Such a cause-effect econometric model also fails in adjusting itself for the post-sample. This article introduces the two ARIMA models and five Intervention-ARIMA models. The monthly data cover the period January 2000 through December 2009. The out-of-sample forecasting performance is compared between the ARIMA-type models and the random walk model. Forecasting performance is measured by three summary statistics: root mean squared error (RMSE), mean absolute error (MAE) and mean error (ME). The RMSE and MAE indicate that the ARIMA-type models outperform the random walk model And the mean errors for all models are small in magnitude relative to the MAE's, indicating that all models don't have a tendency of overpredicting or underpredicting systematically in forecasting. The pessimistic ex-ante forecasts are expected to be 2,820 at the end of 2010 compared with the optimistic forecasts of 4,230.

• Journal of Korean Institute of Industrial Engineers
• /
• v.39 no.1
• /
• pp.30-45
• /
• 2013

Predicting monthly electricity price has been a significant factor of decision-making for plant resource management, fuel purchase plan, plans to plant, operating plan budget, and so on. In this paper, we propose a sophisticated prediction model in terms of the technique of modeling and the variety of the collected variables. The proposed model hybridizes the semi-supervised learning and the artificial neural network algorithms. The former is the most recent and a spotlighted algorithm in data mining and machine learning fields, and the latter is known as one of the well-established algorithms in the fields. Diverse economic/financial indexes such as the crude oil prices, LNG prices, exchange rates, composite indexes of representative global stock markets, etc. are collected and used for the semi-supervised learning which predicts the up-down movement of the price. Whereas various climatic indexes such as temperature, rainfall, sunlight, air pressure, etc, are used for the artificial neural network which predicts the real-values of the price. The resulting values are hybridized in the proposed model. The excellency of the model was empirically verified with the monthly data of electricity price provided by the Korea Energy Economics Institute.