• Journal of agriculture & life science
• /
• v.45 no.4
• /
• pp.29-35
• /
• 2011

A forest fire was one of the huge disasters and damaged human lifes and a properties. Therefore, many countries operated forest fire forecasting systems which developed from forest fire records, weather data, fuel models and etc. And many countries also estimated future state of forest fire using a long-term climate forecasting like GCMs and prepared resources for future huge disasters. In this study, we analyzed relationships between forest fire occurrence and meteorological factors (the minimum temperature ($^{\circ}C$), the relative humidity (%), the precipitation (mm), the duration of sunshine (hour) and etc.) for developing a estimating tools, which could forecast forest fire regime under future climate change condition. Results showed that forest fires in this area were mainly occurred when the maximum temperature was $10{\sim}200^{\circ}C$, when the relative humidity was 40~60%, and when the average wind speed was under 2m/s. And forest fires mainly occurred at 2~3 day after rainfall.

• Journal of Climate Change Research
• /
• v.5 no.4
• /
• pp.331-338
• /
• 2014

The continued efforts to reduce GHG emission by international cooperation and each country are in progress. As part of these efforts, Korea's ETS is enforced in 2015. This was the time to make strategies for each company to respond Korea's ETS. This study was performed to suggest a draft of basic strategies for electronic component industry in current Korea's ETS stage are as follows; - Analyzing the nature of electronic component industry - Identifying needs for corresponding ETS of electronic component industry - Analyzing basic countermeasures for each stage of ETS - Suggesting drafts of basic strategies for electronic component industry in current Korea's ETS stage The result of this study, the current stage of Korea's ETS is moving from implementation of the scheme become determined and prepare the minimum corresponding to direct corresponding to the regulation and market change. Electronic component industry has many GHG emission growth(or change) factor, and it will be make electronic component industry as a buyer when Korea's ETS is enforced. Korea's ETS will be clearly act as a regulation rather than new business for electronic component industry. Therefore, identifying the Korea's ETS as a regulation is resonable strategy for corresponding the scheme. The basic strategies of electronic component industry th responding Korea's ETS are as follows; - Building internal organization and decision-making system before enforcement the Korea's ETS - Establishing internal basic corresponding strategies according to carbon price forecast scenarios - Considering the energy consumption and GHG emissions in design phase and preparing the global ETS market in mid or long term.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.40-40
• /
• 2017

최근 2014년 마른장마의 영향으로 중부 지방에 가뭄이 발생하였으며, 장마철 강수부족은 2015년까지 영향을 미친바 있다. 이로 인해 소양강 댐은 역대 최저수위를 기록하였으며, 일부 지역에서는 제한급수, 농업용수 부족 등의 피해가 발생하였다. 일반적으로 가뭄은 발생순서에 따라 기상학적, 농업적, 수문학적 가뭄 등으로 분류하고 있다 (Wilhite and Grantz, 1985). 기상학적 가뭄은 농업 및 수문학적 가뭄에 영향을 미치는 가뭄의 시작 단계를 의미하며, 가뭄을 판단하는데 있어 중요한 요소라 할 수 있다. 기상학적 가뭄을 정량적으로 판단하기 위해 SPI, PDSI, PN 등이 활용되고 있으며, 특히 강수량 기반의 SPI는 계산과정이 쉽고, 다양한 지속시간(3, 6, 9, 12개월 등)에 따라 가뭄을 객관적으로 판단할 수 있어 가장 활발하게 이용되고 있다(Mckee et al., 1993). 최근 기상청은 대기와 해양-해빙 모델을 접합한 GloSea5의 장기예보자료를 활용하여 월 내지 계절 가뭄전망을 위한 기상학적 가뭄지수를 현업에 활용하고 있다. 다만 국내에서는 주로 단기가뭄(1~3개월)이 빈번하게 발생함에 따라 짧은 예보선행시간을 갖는 가뭄전망에 대한 평가에 집중되어 왔다. 2014, 15년에는 이례적으로 2년 연속 가뭄이 지속된바 있으며, 장기가뭄(3개월 이상)에 대한 전망정보의 필요성이 증가하고 있다. 본 연구에서는 장기예보자료 기반의 기상학적 가뭄전망정보를 산정하고, 2015년 가뭄을 대상으로 활용성을 평가하였다. 이를 위해 ASOS 59개 지점의 관측강수량, GloSea5의 미래예측(Foreacst) 및 과거재현(Hindcast) 자료를 활용하였으며, 다양한 지속시간(3, 6, 9, 12개월)에 대한 SPI를 산정하였다. 또한 예보선행시간(1~6개월)에 따른 SPI와 관측자료 기반의 SPI 간의 통계적 분석(상관계수, 평균제곱근오차)을 수행하여 전망정보의 정확도를 평가하였다.

• Journal of the Korea Society for Simulation
• /
• v.30 no.4
• /
• pp.1-8
• /
• 2021

Deep learning is one of the most widely accepted methods for the forecasting of time series data which have the complexity and non-linear behavior. In this paper, we investigate the modification of a state-of-art WaveNet deep learning architecture and walk forward validation (WFV) in order to forecast electric power consumption data 24-hour-ahead. WaveNet originally designed for raw audio uses 1D dilated causal convolution for long-term information. First of all, we propose a modified version of WaveNet which activates real numbers instead of coded integers. Second, this paper provides with the training process with tuning of major hyper-parameters (i.e., input length, batch size, number of WaveNet blocks, dilation rates, and learning rate scheduler). Finally, performance evaluation results show that the prediction methodology based on WFV performs better than on the traditional holdout validation.

• Journal of Climate Change Research
• /
• v.7 no.1
• /
• pp.85-93
• /
• 2016

This study the efficiency of greenhouse gas reduction of 'low carbon car collaboration fund' and its alternative 'control of average fuel efficiency and greenhouse gas', and 'improving driving behavior' were analyzed by using LEAP, long term energy analysis model. Total 4 scenarios were set, baseline scenario, without energy-saving activity, 'low carbon car collaboration fund' scenario, 'fuel efficiency improving scenario', and 'improving driving behavior' scenario. The contents of analysis were forecast of energy demand by scenario and application as well as reduction of greenhouse gas emission volume, and the period taken for analysis was every 1 year during 2015~2030. Baseline scenario, greenhouse gas emission volume in 2015 would be 7,935,697 M/T and 13,081,986 M/T in 2030, increased 64.8%. The analysis result was average annual increase rate of 3.4%. The expected average annual increase rate of other scenarios was, 'low carbon car collaboration fund' scenario 1.7%, 'fuel efficiency improving' scenario 3.0%. and 'improving driving behavior' scenario 3.4%. and these were each 1.7%, 0.3%. 0.3% reduce from baseline scenario. The largest reduction was 'low carbon car collaboration fund' scenario, and there after were 'fuel efficiency improving scenario', and 'improving driving behavior' scenario.

• Journal of the Korea Convergence Society
• /
• v.12 no.4
• /
• pp.249-258
• /
• 2021

Bitcoin prices have been soaring recently as investors flock to cryptocurrency exchanges. The purpose of this study is to predict the Bitcoin price using a deep learning model and analyze whether Bitcoin is profitable through investment strategy. LSTM is utilized as Bitcoin prediction model with nonlinearity and long-term memory and the profitability of MA cross-over strategy with predicted prices as input variables is analyzed. Investment performance of Bitcoin strategy using LSTM forecast prices from 2013 to 2021 showed return improvement of 5.5% and 46% more than market price MA cross-over strategy and benchmark Buy & Hold strategy, respectively. The results of this study, which expanded to recent data, supported the inefficiency of the cryptocurrency market, as did previous studies, and showed the feasibility of using the deep learning model for Bitcoin investors. In future research, it is necessary to develop optimal prediction models and improve the profitability of Bitcoin investment strategies through performance comparison of various deep learning models.

• Asia-Pacific Journal of Business
• /
• v.11 no.4
• /
• pp.189-203
• /
• 2020

Purpose - This study aims in analyzing the dynamic relationship between household loans and housing prices according to the characteristics of depository institutions after the financial crisis, identifying the recent trends between them, and making policy suggestions for stabilizing house prices. Design/methodology/approach - The monthly data used in this study are household loans, household loan interest rates, and housing prices ranging from January 2012 to May 2020, and came from ECOS of the Bank of Korea and Liiv-on of Kookmin Bank. This study used vector auto-regression, generalized impulse response function, and forecast error variance decomposition with the data so as to yield analysis results. Findings - The analysis of this study no more shows that the household loan interest rates in both deposit banks and non-bank deposit institutions had statistically significant effects on housing prices. Also, unlike the previous studies, there was statistically significant bi-directional causality between housing prices and household loans in neither deposit banks nor non-bank deposit institutions. Rather, it was found that there is a unidirectional causality from housing prices to household loans in deposit banks, which is considered that housing prices have one-sided effects on household loans due to the overheated housing market after the financial crisis. Research implications or Originality - As a result, Korea's housing market is closely related to deposit banks, and housing prices are acting as more dominant information variables than interest rates or loans under the long-term low interest rate trend. Therefore, in order to stabilize housing prices, the housing supply must be continuously made so that everyone can enjoy housing services equally. In addition, the expansion and reinforcement of the social security net should be realized systematically so as to stop households from being troubled with the housing price decline.

• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.1176-1179
• /
• 2020

Vessel path prediction methods generally predict the latitude and longitude of a future location directly. However, in the case of direct prediction, errors could be large since the possible output range is too broad. In addition, error accumulation could occur since recurrent neural networks-based methods employ previous predicted data to forecast future data. In this paper, we propose a vessel path prediction method that does not directly predict the longitude and latitude. Instead, the proposed method predicts the acceleration of the vessel. Then the acceleration is employed to generate the velocity and direction, and the values decide the longitude and latitude of the future location. In the experiment, we show that the proposed method makes smaller errors than the direct prediction method, while both methods employ the same model.

• Journal of Korean Society on Water Environment
• /
• v.27 no.3
• /
• pp.264-272
• /
• 2011

The objective of this study was to develop a model for predicting long-term runoff in a basin using the ensemble streamflow prediction (ESP) technique and review its reliability. To achieve the objective, this study improved not only the ESP technique based on the ensemble scenario analysis of historical rainfall data but also conventional ESP techniques used in conjunction with qualitative climate forecasting information, and analyzed and assessed their improvement effects. The model was applied to the Geum River basin. To undertake runoff forecasting, this study tried three cases (case 1: Climate Outlook + ESP, case 2: ESP probability through monthly measured discharge, case 3: Season ESP probability of case 2) according to techniques used to calculate ESP probabilities. As a result, the mean absolute error of runoff forecasts for case 1 proposed by this study was calculated as 295.8 MCM. This suggests that case 1 showed higher reliability in runoff forecasting than case 2 (324 MCM) and case 3 (473.1 MCM). In a discrepancy-ratio accuracy analysis, the Climate Outlook + ESP technique displayed 50.0%. This suggests that runoff forecasting using the Climate Outlook +ESP technique with the lowest absolute error was more reliable than other two cases.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.3
• /
• pp.20-29
• /
• 2021

In the Industry 4.0 era, artificial intelligence has attracted considerable interest for learning mass data to improve the accuracy of forecasting and classification. On the other hand, the current method of detecting anomalies relies on traditional statistical methods for a limited amount of data, making it difficult to detect accurate anomalies. Therefore, this paper proposes an artificial intelligence-based anomaly detection methodology to improve the prediction accuracy and identify new data patterns. In particular, data were collected and analyzed from the point of view that sensor data collected at vehicle idle could be used to detect abnormalities. To this end, a sensor was designed to determine the appropriate time length of the data entered into the forecast model, compare the results of idling data with the overall driving data utilization, and make optimal predictions through a combination of various sensor data. In addition, the predictive accuracy of artificial intelligence techniques was presented by comparing Convolutional Neural Networks (CNN) and Long Short Term Memory (LSTM) as the predictive methodologies. According to the analysis, using idle data, using 1.5 times of the data for the idling periods, and using CNN over LSTM showed better prediction results.