DOI QR코드

DOI QR Code

Wind power forecasting based on time series and machine learning models

시계열 모형과 기계학습 모형을 이용한 풍력 발전량 예측 연구

  • Park, Sujin (Department of Applied Statistics, University of Chung-Ang) ;
  • Lee, Jin-Young (Department of Applied Statistics, University of Chung-Ang) ;
  • Kim, Sahm (Department of Applied Statistics, University of Chung-Ang)
  • 박수진 (중앙대학교 응용통계학과) ;
  • 이진영 (중앙대학교 응용통계학과) ;
  • 김삼용 (중앙대학교 응용통계학과)
  • Received : 2021.06.17
  • Accepted : 2021.08.11
  • Published : 2021.10.31

Abstract

Wind energy is one of the rapidly developing renewable energies which is being developed and invested in response to climate change. As renewable energy policies and power plant installations are promoted, the supply of wind power in Korea is gradually expanding and attempts to accurately predict demand are expanding. In this paper, the ARIMA and ARIMAX models which are Time series techniques and the SVR, Random Forest and XGBoost models which are machine learning models were compared and analyzed to predict wind power generation in the Jeonnam and Gyeongbuk regions. Mean absolute error (MAE) and mean absolute percentage error (MAPE) were used as indicators to compare the predicted results of the model. After subtracting the hourly raw data from January 1, 2018 to October 24, 2020, the model was trained to predict wind power generation for 168 hours from October 25, 2020 to October 31, 2020. As a result of comparing the predictive power of the models, the Random Forest and XGBoost models showed the best performance in the order of Jeonnam and Gyeongbuk. In future research, we will try not only machine learning models but also forecasting wind power generation based on data mining techniques that have been actively researched recently.

빠르게 발전하고 있는 재생에너지 중 하나인 풍력에너지는 기후변화 대응에 맞추어 개발 및 투자가 이루어지고있다. 신재생에너지 정책과 발전소 설치가 추진됨에 따라 국내 풍력 보급이 점차 확대되어 수요를 정확히 예측하기 위한 시도들이 확대되고 있다. 본 논문에서는 전남지역과 경북지역의 풍력 발전량 예측을 위하여 시계열 기법인 ARIMA, ARIMAX 모형과 기계학습 모형인 SVR, Random Forest, XGBoost 모형들을 비교 분석하였다. 모형의 예측 결과를 비교하기 위한 지표로서 mean absolute error (MAE)와 mean absolute percentage error (MAPE)를 사용하였다. 2018년 1월 1일부터 2020년 10월 24일까지의 시간별 원 데이터를 차분한 후 모형을 훈련시켜 2020년 10월 25일부터 2020년 10월 31일까지의 168시간에 대한 풍력 발전량을 예측하였다. 모형의 예측력 비교 결과, Random Forest와 XGBoost 모형이 전남지역, 경북지역 순으로 가장 우수한 성능을 보였다. 향후 연구에서는 기계학습뿐 아니라 최근 활발한 연구가 이루어지는 데이터 마이닝 기법 기반의 풍력 발전량 예측을 시도할 것이다.

Keywords

Acknowledgement

이 논문은 2021년도 정부(교육부)의 재원으로 한국연구재단의 지원을 받아 수행된 기초연구사업임(No. 2016R1D1A1B01014954).

References

  1. Aditya C, Akash S, Ayush K, Karan D, and Neeraj K (2020). Short term wind power forecasting using machine learning techniques, Journal of Statistics and Management Systems, 23.1, 145-156. https://doi.org/10.1080/09720510.2020.1721632
  2. Ahmadi A, Nabipour M, Mohammadi-ivatloo B, Rho SM, and Piran J (2020). Long-term wind power forecasting using tree-based learning algorithms, IEEE Access, 8, 151511-151522. https://doi.org/10.1109/access.2020.3017442
  3. Anastasiades G and McSharry P (2013). Quantile forecasting of wind power using variability indices, Journal of Applied Meteorology and Climatology, 6.2, 662-695.
  4. Brown BG, Richard WK, and Allan HM (1984). Time series models to simulate and forecast wind speed and wind power, Journal of Applied Meteorology and Climatology, 23.8, 1184-1195. https://doi.org/10.1175/1520-0450(1984)023<1184:TSMTSA>2.0.CO;2
  5. Cadenas E and Wilfrido R (2010). Wind speed forecasting in three different regions of Mexico, using a hybrid ARIMA-ANN model, Renewable Energy, 35.12, 2732-2738. https://doi.org/10.1016/j.renene.2010.04.022
  6. Catalao, Joao Paulo da Silva, Hugo Miguel Inacio Pousinho, and Victor Manuel Fernandes Mendes (2011). Short-term wind power forecasting in Portugal by neural networks and wavelet transform, Renewable energy, 36.4, 1245-1251. https://doi.org/10.1016/j.renene.2010.09.016
  7. Halil D, Ahmet SD, Alper E, and Murat G (2019). Wind power forecasting based on daily wind speed data using machine learning algorithms, Energy Conversion and Management, 198, 111823. https://doi.org/10.1016/j.enconman.2019.111823
  8. Hossain A, Chakrabortty RK, Elsawah S, and Ryan MJ (2020). Hybrid deep learning model for ultra-short-term wind power forecasting, 2020 IEEE International Conference on Applied Superconductivity and Electromagnetic Devices (ASEMD).
  9. Hu YL and Liang C (2018). A nonlinear hybrid wind speed forecasting model using LSTM network, hysteretic ELM and differential evolution algorithm, Energy conversion and management, 173, 123-142. https://doi.org/10.1016/j.enconman.2018.07.070
  10. Ko MS, Lee KS, Kim JK, Hong CW, Dong ZY, and Hur K (2020). Deep concatenated residual network with bidirectional LSTM for one-hour-ahead wind power forecasting, IEEE Transactions on Sustainable Energy, 12, 1321-1335.
  11. Lahouar A and Ben Hadj Slama J (2017). Hour-ahead wind power forecast based on random forests, Renewable energy, 109, 529-541. https://doi.org/10.1016/j.renene.2017.03.064
  12. Li S, Peng W, and Lalit G (2015). Wind power forecasting using neural network ensembles with feature selection, IEEE Transactions on sustainable energy, 6.4, 1447-1456. https://doi.org/10.1109/TSTE.2015.2441747
  13. Liu H, Erdem E, and Shi J (2011). Comprehensive evaluation of ARMA-GARCH (-M) approaches for modeling the mean and volatility of wind speed, Applied Energy, 88.3, 724-732. https://doi.org/10.1016/j.apenergy.2010.09.028
  14. Park SH and Kim S (2016). A study on short-term wind power forecasting using time series models, The Korean Journal of Applied Statistics , 29.7, 1373-1383. https://doi.org/10.5351/KJAS.2016.29.7.1373
  15. Pinson P (2012). Very-short-term probabilistic forecasting of wind power with generalized logit-normal distributions, Journal of the Royal Statistical Society: Series C (Applied Statistics), 61.4, 555-576. https://doi.org/10.1111/j.1467-9876.2011.01026.x
  16. Priya B and Arulanand N (2021). Univariate and multivariate models for Short-term wind speed forecasting. Materials Today: Proceedings, 2214-7853.
  17. Suh YM, Son HG, and Kim S (2018). Solar radiation forecasting by time series models, The Korean Journal of Applied Statistics , 31.6, 785-799. https://doi.org/10.5351/KJAS.2018.31.6.785
  18. Yu J, Chen X, Yu K, and Liao Y (2017). Short-term wind power forecasting using hybrid method based on enhanced boosting algorithm, Journal of Modern Power Systems and Clean Energy, 5.1, 126-133. https://doi.org/10.1007/s40565-015-0171-6
  19. Zheng H and Wu Y (2019). A xgboost model with weather similarity analysis and feature engineering for short-term wind power forecasting, Applied Science, 9.15, 3019.
  20. Zeng J and Wei Q (2011). Support vector machine-based short-term wind power forecasting, 2011 IEEE/PES Power Systems Conference and Exposition. IEEE, 1-8.
  21. Zhao Y, Ye L, Li Z, Song X, Lang Y, and Su J (2016). A novel bidirectional mechanism based on time series model for wind power forecasting, Applied Energy, 177, 793-803. https://doi.org/10.1016/j.apenergy.2016.03.096