Atmosphere (대기)

Korean Meteorological Society (한국기상학회)
권호 표지
DOI QR코드

DOI QR Code

Comparison of Solar Power Generation Forecasting Performance in Daejeon and Busan Based on Preprocessing Methods and Artificial Intelligence Techniques: Using Meteorological Observation and Forecast Data

전처리 방법과 인공지능 모델 차이에 따른 대전과 부산의 태양광 발전량 예측성능 비교: 기상관측자료와 예보자료를 이용하여

  • Chae-Yeon Shim (Department of Earth and Environmental Sciences, Jeonbuk National University) ;
  • Gyeong-Min Baek (Department of Earth and Environmental Sciences, Jeonbuk National University) ;
  • Hyun-Su Park (Department of Earth and Environmental Sciences, Jeonbuk National University) ;
  • Jong-Yeon Park (Department of Earth and Environmental Sciences, Jeonbuk National University)
  • 심채연 (전북대학교 지구환경과학과) ;
  • 백경민 (전북대학교 지구환경과학과) ;
  • 박현수 (전북대학교 지구환경과학과) ;
  • 박종연 (전북대학교 지구환경과학과)
  • Received : 2024.03.21
  • Accepted : 2024.04.25
  • Published : 2024.05.31
Download PDF
⟨ Previous Next ⟩

Abstract

As increasing global interest in renewable energy due to the ongoing climate crisis, there is a growing need for efficient technologies to manage such resources. This study focuses on the predictive skill of daily solar power generation using weather observation and forecast data. Meteorological data from the Korea Meteorological Administration and solar power generation data from the Korea Power Exchange were utilized for the period from January 2017 to May 2023, considering both inland (Daejeon) and coastal (Busan) regions. Temperature, wind speed, relative humidity, and precipitation were selected as relevant meteorological variables for solar power prediction. All data was preprocessed by removing their systematic components to use only their residuals and the residual of solar data were further processed with weighted adjustments for homoscedasticity. Four models, MLR (Multiple Linear Regression), RF (Random Forest), DNN (Deep Neural Network), and RNN (Recurrent Neural Network), were employed for solar power prediction and their performances were evaluated based on predicted values utilizing observed meteorological data (used as a reference), 1-day-ahead forecast data (referred to as fore1), and 2-day-ahead forecast data (fore2). DNN-based prediction model exhibits superior performance in both regions, with RNN performing the least effectively. However, MLR and RF demonstrate competitive performance comparable to DNN. The disparities in the performance of the four different models are less pronounced than anticipated, underscoring the pivotal role of fitting models using residuals. This emphasizes that the utilized preprocessing approach, specifically leveraging residuals, is poised to play a crucial role in the future of solar power generation forecasting.

Keywords

Acknowledgement

본 논문의 개선을 위해 좋은 의견을 제시해 주신 두 분의 심사위원께 감사를 드립니다. 이 연구는 한국연구재단(NRF)의 지원(RS-2023-00207866, 2020R1C1C1008631)으로 수행되었습니다.

References

  1. Chai, T., and R. R. Draxler, 2014: Root mean square error (RMSE) or mean absolute error (MAE)? Arguments against avoiding RMSE in the literature. Geosci. Model Dev., 7, 1247-1250, doi:10.5194/gmd-7-1247-2014.
  2. IPCC, 2021: Climate Change 2021: The Physical Science Basis. Cambridge University Press, Intergovernmental Panel on Climate Change, 2391 pp.
  3. Kang, B.-B., and J.-H. Yun, 2023: A study on the prediction model of photovoltaic power generation using deep learning algorithm. J. Institute of Electron. Inf. Engineers, 60, 119-125, doi:10.5573/ieie.2023.60.2.119.
  4. Kang, K.-M., Y.-G. Choi, Y.-J. Kim, S.-J. Min, D.-S. Choi, K.-Y. Kim, and D.-Y. Lee, 2021: The classification of climate types and the delineation of their climatic characteristics using new normals (1991~ 2020) in the Republic of Korea. J. Climate Res., 16, 179-195, doi:10.14383/cri.2021.16.3.179.
  5. Ki, J.-H., S.-J. Baek, J.-Y. So, H.-G. Eom, and J.-H. Shin, 2023: Solar power generation prediction using machine learning and study of power generation using solar tracking panels. Trans. Korean Soc. Mech. Eng. B, 47, 55-62, doi:10.3795/ksme-b.2023.47.1.055.
  6. KEA, 2021: 2021 Renewable Energy Distribution Statistics, Korea Energy Agency, Accessed October 23, 2023, [Available online at https://www.knrec.or.kr/biz/pds/statistic/view.do?no=170].
  7. KIER, 2023: Analyzing 2022 Power Generation in South Korea through Electricity Statistics. Accessed December 20, 2023, Korea Institute of Energy Research Techpolicy Platform, [Available online at https://www.kier.re.kr/tpp/tppBoard/view/25?menuId=MENU00962].
  8. Kim, B.-C., S.-H. Jung, M.-S. Kim, J. G. Kim, H.-S. Lee, and S.-S. Kim, 2020: Solar power generation forecasting based on LSTM considering weather conditions. J. Korean Institute of Intelligent Syst., 30, 7-12, doi:10.5391/jkiis.2020.30.1.7.
  9. Kim, J.-W., 2019: A solar power prediction scheme based on machine learning algorithm from weather forecasts. J. Korean Institute of Inf. Technol., 17, 83-89, doi:10.14801/jkiit.2019.17.9.83.
  10. KPX, 2020: Amendment to the Electricity Market Operation Rules, Accessed November 21, 2023, Korea Power Exchange, [Available online at https://new.kpx.or.kr/board.es?mid=a10205010000&bid=0030&act=view&list_no=68951&tag=&nPage=1].
  11. Lee, D.-H., and G.-H. Kim, 2019: Deep learning based prediction method of long-term photovoltaic power generation using meteorological and seasonal information. J. Soc. e-Business Studies, 24, 1-16, doi:10.7838/jsebs.2019.24.1.001.
  12. Lee, K.-H., H.-G. Son, and S. Kim, 2018: A study on solar energy forecasting based on time series models. J. Appl. Stat., 31, 139-153, doi:10.5351/KJAS.2018.31.1.139.
  13. Mekhilef, S. , R. Saidur, and M. Kamalisarvestani, 2012:Effect of dust, humidity and air velocity on efficiency of photovoltaic cells. Renewable and Sustainable Energy Rev., 16, 2920-2925, doi:10.1016/j.rser.2012.02.012.
  14. NOAA, 2020: Despite pandemic shutdowns, carbon dioxide and methane surged in 2020. Accessed July 10, 2023, National Oceanic and Atmospheric Administration, [Available online at https://research.noaa.gov/2021/04/07/despite-pandemic-shutdowns-carbon-dioxideand-methane-surged-in-2020/].
  15. Son, W.-B., 2023: Solar power generation forecasting technology status for anticipating the expansion of renewable energy deployment. J. Korean Institute of Communication Sciences, 40, 20-26.
  16. Voyant, C., G. Notton, J.-L. Duchaud, J. Almorox, and Z. M. Yaseen, 2020: Solar irradiation prediction intervals based on Box-Cox transformation and univariate representation of periodic autoregressive model. Renewable Energy Focus, 33, 43-53, doi:10.1016/j.ref.2020.04.001.
  17. Yu, J.-H., S.-J. Kim, B.-H. Jang, and S.-M. Woo, 2022: Renewable energy prediction market participation type photovoltaics prediction model development. J. Korean Solar Energy Soc., 42, 1-13, doi:10.7836/kses.2022.42.6.001.