Journal of Wind Energy (풍력에너지저널)

Korea Wind Energy Association (한국풍력에너지학회)
권호 표지
DOI QR코드

DOI QR Code

Performance Evaluation of Stacking Models Based on Random Forest, XGBoost, and LGBM for Wind Power Forecasting

Random Forest, XGBoost, LGBM 조합형 Stacking 모델을 이용한 풍력 발전량 예측 성능 평가

  • Hui-Chan Kim ;
  • Dae-Young Kim ;
  • Bum-Suk Kim
  • 김희찬 (제주대학교, 대학원 풍력공학부) ;
  • 김대영 (제주대학교, 대학원 풍력공학부) ;
  • 김범석 (제주대학교, 대학원 풍력공학부)
  • Received : 2024.04.05
  • Accepted : 2024.08.30
  • Published : 2024.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

Wind power is highly variable due to the intermittent nature of wind. This can lead to power grid instability and decreased efficiency. Therefore, it is necessary to improve wind power prediction performance to minimize the negative impact on the power system. Recently, wind power prediction using machine learning has gained popularity, and ensemble models in machine learning have shown high prediction accuracy. RF, GB, XGB and LGBM are decision tree-based ensemble models and have high predictive performance in wind power, but these models have problems from over-fitting and strong dependence on certain variables. However, the stacking model can improve prediction performance by combining individual models and compensate for the shortcomings of each model. In this study, The MAE of RF, XGB and LGBM is 310.42 kWh, 217.07 kWh and 265.20 kWh, respectively, while the stacking model based on RF, XGB and LGBM is 202.33 kWh. Stacking models can improve prediction performance. Finally, it is expected to contribute to electricity supply and demand planning.

Keywords

Acknowledgement

본 연구는 한국동서발전 "풍력발전기 출력저하 및 누적피로하중 예측기술 개발"의 지원을 받아 수행한 연구 과제입니다. (과제번호 : C-2023-05). 그리고 풍력발전 데이터를 제공해 준 제주 에너지 공사에 감사드립니다.

References

  1. Mingli, Z,, Xu, H., Jianyuan, X. and Yun, T., "Study on application of the forecasting system for wind farm power," in Proceedings of the 2010 2nd International Conference on Signal Processing Systems, pp. 198-201
  2. Optis, M. and Perr-Sauer, J., "The importance of atmospheric turbulence and stability in machine-learning models of wind farm power production", Renewable and Sustainable Energy Reviews, Volume 112, Pages 27-41
  3. Leo, B., "Bagging predictors", Machine learning, 24, 123-140.
  4. Heinermann, J. and Kramer, O., "Machine learning ensembles for wind power prediction", Renewable Energy, Volume 89, Pages 671-679
  5. Demollia, H., Dokuzb, S. A., Ecemisb, A. and Murat Gokcekc, "Wind power forecasting based on daily wind speed data using machine learning algorithms.", Energy Conversion and Management, Vol. 198, pp. 111823
  6. Lee, J. H., Wang, W., Harrou, F. and Ying, Sun, "Wind power prediction using ensemble learning-based models.", IEEE Access, Vol. 8, pp. 61517-61527, 2020
  7. Wangjie, L. and Jia, L., "Wind Power Prediction Based on the Stacking Model of XGBoost and Random Forest." 2022 IEEE 11th Data Driven Control and Learning Systems Conference (DDCLS). IEEE, 2022.
  8. Jeong, H. S., Kim, Y. H. and Choi, H. W., "Characteristics of Wind Environment in Dongbok.Bukchon Wind Farm on Jeju", New & Renewable Energy, Vol 18, No. 1, pp.1-16
  9. Kim, D. Y., Kim, Y. H. and Kim, B. S., "Changes in wind turbine power characteristics and annual energy production due to atmospheric stability, turbulence intensity, and wind shear." Energy Volume 214, 119051
  10. Leo, B., "Random forests", Machine learning, 2001, Vol. 45, pp. 5~32.
  11. Singh, U.. and Rizwan, M., "SCADA system dataset exploration and machine learning based forecast for wind turbines.", Results in Engineering, Volume 16, December 2022
  12. Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q. and Liu, T. Y., "Lightgbm: A highly efficient gradient boosting decision tree." Advances in neural information processing systems 30, Long Beach, CA, USA. 4-9 DEC 2017
  13. Pavlyshenko, B., "Using stacking approaches for machine learning models." 2018 IEEE Second International Conference on Data Stream Mining & Processing (DSMP), Lviv, Ukraine, 21-25 August 2018
  14. Yang, L. and Shami, A., "On hyperparameter optimization of machine learning algorithms: Theory and practice", Neurocomputing, Volume 415, 20 November 2020, Pages 295-316