• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2009년도 추계학술대회
• /
• pp.720-722
• /
• 2009

경제 성장에 따른 국내 산업분야의 발달 및 국민 생활수준의 향상으로 전력 소비가 지속적으로 증가하고 있다. 전력을 안정적으로 공급하기 위해서는 전력 수요에 대한 중 단기 예측이 중요하며, 정확한 예측에 따라 안정적인 수급 계획을 확립할 수 있다. 본 논문에서는 부산시에서 공급되는 부산지역의 전력 데이터와 기후 관련 자료를 1995년 1월부터 2007년 12월까지의 측정치를 가지고 시계열 데이터를 수집하여 분석하고 신경회로망의 구조를 설계하여 실험을 통하여 실제 데이터와 예측 데이터를 비교 분석하고 평가한다.

• 대기
• /
• 제31권1호
• /
• pp.73-83
• /
• 2021

Numerical weather prediction (NWP) models play an essential role in predicting weather factors, but using them is challenging due to various factors. To overcome the difficulties of NWP models, deep learning models have been deployed in weather forecasting by several recent studies. This study adapts long short-term memory (LSTM), which demonstrates remarkable performance in time-series prediction. The combination of LSTM model input of meteorological features and activation functions have a significant impact on the performance therefore, the results from 5 combinations of input features and 4 activation functions are analyzed in 9 Automated Surface Observing System (ASOS) stations corresponding to cities/islands/mountains. The optimized LSTM model produces better performance within eight forecast hours than Local Data Assimilation and Prediction System (LDAPS) operated by Korean meteorological administration. Therefore, this study illustrates that this LSTM model can be usefully applied to very short-term weather forecasting, and further studies about CNN-LSTM model with 2-D spatial convolution neural network (CNN) coupled in LSTM are required for improvement.

• 한국산업융합학회 논문집
• /
• 제26권4_2호
• /
• pp.715-724
• /
• 2023

Solar energy forecasting is essential for (1) power system planning, management, and operation, requiring accurate predictions. It is crucial for (2) ensuring a continuous and sustainable power supply to customers and (3) optimizing the operation and control of renewable energy systems and the electricity market. Recently, research has been focusing on developing solar energy forecasting models that can provide daily plans for power usage and production and be verified in the electricity market. In these prediction models, various data, including solar energy generation and climate data, are chosen to be utilized in the forecasting process. The most commonly used climate data (such as temperature, relative humidity, precipitation, solar radiation, and wind speed) significantly influence the fluctuations in solar energy generation based on weather conditions. Therefore, this paper proposes a hybrid forecasting model by combining the strengths of the Prophet model and the GRU model, which exhibits excellent predictive performance. The forecasting periods for solar energy generation are tested in short-term (2 days, 7 days) and medium-term (15 days, 30 days) scenarios. The experimental results demonstrate that the proposed approach outperforms the conventional Prophet model by more than twice in terms of Root Mean Square Error (RMSE) and surpasses the modified GRU model by more than 1.5 times, showcasing superior performance.

• 한국군사과학기술학회지
• /
• 제21권3호
• /
• pp.403-412
• /
• 2018

In this study, a three-dimensional variational(3DVAR) data assimilation system was developed for the operational numerical weather prediction(NWP) system at the Republic of Korea Air Force Weather Group. The Air Force NWP system utilizes the Weather Research and Forecasting(WRF) meso-scale regional model to provide weather information for the military service. Thus, the data assimilation system was developed based on the WRF model. Experiments were conducted to identify the nested model domain to assimilate observations and the period appropriate in estimating the background error covariance(BEC) in 3DVAR. The assimilation of observations in domain 2 is beneficial to improve 24-h forecasts in domain 3. The 24-h forecast performance does not change much depending on the estimation period of the BEC in 3DVAR. The results of this study provide a basis to establish the operational data assimilation system for the Republic of Korea Air Force Weather Group.

• 한국태양에너지학회 논문집
• /
• 제30권2호
• /
• pp.46-53
• /
• 2010

In this paper the feasibility of wind power forecasting from MOS(Model Output Statistics) was evaluated at Gosan area in Jeju during February to Octoberin 2008. The observed wind data from wind turbine was compared with 24 hours and 48 hours forecasting wind data from MOS predicting. Coefficient of determination of measured wind speed from wind turbine and 24 hours forecasting from MOS was around 0.53 and 48 hours was around 0.30. These determination factors were increased to 0.65 from 0.53 and 0.35 from 0.30, respectively, when it comes to the prevailing wind direction($300^{\circ}\sim60^{\circ}$). Wind power forecasting ratio in 24 hours of MOS showed a value of 0.81 within 70% confidence interval and it also showed 0.65 in 80% confidence interval. It is suggested that the additional study of weather conditions be carried out when large error happened in MOS forecasting.

• 응용통계연구
• /
• 제31권1호
• /
• pp.139-153
• /
• 2018

최근 정부의 친환경 정책에 따라 태양광 발전 설비가 지속적으로 증가하고 있다. 태양광 발전량은 에너지원인 태양의 특성상 계절에 따라 하루 중 발전이 이루어지는 시간이 일정하지 않다. 이러한 특성으로 인해 태양광 발전량 예측에서는 연속된 시간간격으로 수집된 자료에 적용할 수 있는 시계열 모형 적용에 어려움이 있다. 본 논문에서 제안하는 방법은 연속된 시간자료를 각 시간대 별로 분리, 재구성하여 24개의 (1시-24시) 일별 자료 형태로 예측에 활용하는 방법이다. 강원도 영암 태양광 발전소의 시간별 발전량 자료를 공공데이터포털에서 수집하여 연구하였다. 기존방법과 제안된 방법의 성능차이를 비교하기 위해 ARIMAX, 신경망(neural network model) 모형을 동일한 모형과 변수를 가지는 환경에서 성능차이를 확인하였다.

• 전기학회논문지
• /
• 제66권4호
• /
• pp.621-628
• /
• 2017

This paper presents an artificial neural network (ANN) based model with a back-propagation algorithm for short-term load forecasting in microgrid power systems. Owing to the significant weather factors for such purpose, relevant input variables were selected in order to improve the forecasting accuracy. As remarked above, forecasting is more complex in a microgrid because of the increased variability of disaggregated load curves. Accurate forecasting in a microgrid will depend on the variables employed and the way they are presented to the ANN. This study also shows numerically that there is a close relationship between forecast errors and the number of training patterns used, and so it is necessary to carefully select the training data to be employed with the system. Finally, this work demonstrates that the concept of load forecasting and the ANN tools employed are also applicable to the microgrid domain with very good results, showing that small errors of Mean Absolute Percentage Error (MAPE) around 3% are achievable.

• Journal of Biosystems Engineering
• /
• 제17권2호
• /
• pp.177-191
• /
• 1992

The MAFSAP(Monitoring and Forecasting System of Air Pollution) was developed to measure the weather and air pollution data automatically, then make them input to microcomputer and analyze them for monitoring and forecasting air pollution at all times. And the air pollution telemetering systems installed at Young-Dong Thermal Power Plant was analyzed and an ideal telemetering system utilizing MAFSAP was suggested.

• 한국지능시스템학회:학술대회논문집
• /
• 한국지능시스템학회 2008년도 춘계학술대회 학술발표회 논문집
• /
• pp.415-417
• /
• 2008

기상자료는 매순간 방대한 양으로 쏟아져 나온다. 본 논문은 이 방대한 양의 자료를 토대로 신경망을 학습시켜 정보(예보)를 도출시키는 데 얼마나 적합한지 확인하고자 함에 있다. 과거 의사결정나무를 통해서 위와 같은 연구가 진행된 바 있으나, 현재 우리나라에서 신경망을 통한 분석은 전무한 상태이다. 따라서 우리나라 3개지역을 선정 96년도부터 05년까지의 10년간의 9, 10월 지상자료를 토대로 안개예보에 신경망이 적합한지에 대해 연구하였다.

• 한국신재생에너지학회:학술대회논문집
• /
• 한국신재생에너지학회 2006년도 춘계학술대회
• /
• pp.323-324
• /
• 2006

In the present paper a forecasting system of wind power generation for Walryong Site, Jejudo is presented, which has been developed and evaluated as a first step toward establishing Korea Forecasting Model of Wind Power Generation. The forecasting model, KIER forecaster is constructed based on statistical models and is trained with wind speed data observed at Gosan Weather Station nearby Walryong Si to. Due to short period of measurements at Walryong Site for training statistical model, Gosan wind data were substituted and transplanted to Walryong Site by using Measure-Correlate-Predict technique. Three-hour advanced forecast ins shows good agreement with the measurement at Walryong site with the correlation factor 0.88 and MAE(mean absolute error) 15% under.