International Journal of Advanced Culture Technology

The International Promotion Agency of Culture Technology (국제문화기술진흥원)
권호 표지
DOI QR코드

DOI QR Code

Enhancing Wind Speed and Wind Power Forecasting Using Shape-Wise Feature Engineering: A Novel Approach for Improved Accuracy and Robustness

  • Mulomba Mukendi Christian (Dept. of Advanced Convergence, Handong Global Univ.) ;
  • Yun Seon Kim (School of Global Entrepreneurship and Information Communication Technology, Handong Global Univ.) ;
  • Hyebong Choi (School of Global Entrepreneurship and Information Communication Technology, Handong Global Univ.) ;
  • Jaeyoung Lee (School of Mechanical control and Engineering, Handong Global Univ.) ;
  • SongHee You (School of Spatial Environment System Engineering, Handong Global Univ.)
  • Received : 2023.10.01
  • Accepted : 2023.11.15
  • Published : 2023.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

Accurate prediction of wind speed and power is vital for enhancing the efficiency of wind energy systems. Numerous solutions have been implemented to date, demonstrating their potential to improve forecasting. Among these, deep learning is perceived as a revolutionary approach in the field. However, despite their effectiveness, the noise present in the collected data remains a significant challenge. This noise has the potential to diminish the performance of these algorithms, leading to inaccurate predictions. In response to this, this study explores a novel feature engineering approach. This approach involves altering the data input shape in both Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) and Autoregressive models for various forecasting horizons. The results reveal substantial enhancements in model resilience against noise resulting from step increases in data. The approach could achieve an impressive 83% accuracy in predicting unseen data up to the 24th steps. Furthermore, this method consistently provides high accuracy for short, mid, and long-term forecasts, outperforming the performance of individual models. These findings pave the way for further research on noise reduction strategies at different forecasting horizons through shape-wise feature engineering.

Keywords

References

  1. F. Porte-Agel, M. Bastankhah, and S. Shamsoddin, "Wind-Turbine and Wind-Farm Flows: A Review," Bound.-Layer Meteorol., vol. 174, no. 1, pp. 1-59, Jan. 2020, doi: 10.1007/s10546-019-00473-0.
  2. GWEC, "GWEC | GLOBAL WIND REPORT 2021," Global Wind Energy Council, 05 2021. [Online]. Available: https://gwec.net/wp-content/uploads/2021/03/GWEC-Global-Wind-Report-2021.pdf
  3. C. Gu and H. Li, "Review on Deep Learning Research and Applications in Wind and Wave Energy," Energies, vol. 15, no. 4, p. 1510, Feb. 2022, doi: 10.3390/en15041510.
  4. T. Hong, P. Pinson, Y. Wang, R. Weron, D. Yang, and H. Zareipour, "Energy Forecasting: A Review and Outlook," IEEE Open Access J. Power Energy, vol. 7, pp. 376-388, 2020, doi: 10.1109/OAJPE.2020.3029979.
  5. F. Fuentes-Penailillo, S. Ortega-Farias, C. Acevedo-Opazo, and D. Fonseca-Luengo, "Implementation of a Two-Source Model for Estimating the Spatial Variability of Olive Evapotranspiration Using Satellite Images and Ground-Based Climate Data," Water, vol. 10, no. 3, p. 339, Mar. 2018, doi: 10.3390/w10030339.
  6. Y. Zhu, C. Zhu, C. Song, Y. Li, X. Chen, and B. Yong, "Improvement of reliability and wind power generation based on wind turbine real-time condition assessment," Int. J. Electr. Power Energy Syst., vol. 113, pp. 344-354, Dec. 2019, doi: 10.1016/j.ijepes.2019.05.027.
  7. M. Ding, H. Zhou, H. Xie, M. Wu, Y. Nakanishi, and R. Yokoyama, "A gated recurrent unit neural networks based wind speed error correction model for short-term wind power forecasting," Neurocomputing, vol. 365, pp. 54-61, Nov. 2019, doi: 10.1016/j.neucom.2019.07.058.
  8. M. H. D. M. Ribeiro, R. G. Da Silva, S. R. Moreno, V. C. Mariani, and L. D. S. Coelho, "Efficient bootstrap stacking ensemble learning model applied to wind power generation forecasting," Int. J. Electr. Power Energy Syst., vol. 136, p. 107712, Mar. 2022, doi: 10.1016/j.ijepes.2021.107712.
  9. C.-M. Huang, S.-J. Chen, S.-P. Yang, and H.-J. Chen, "One-Day-Ahead Hourly Wind Power Forecasting Using Optimized Ensemble Prediction Methods," Energies, vol. 16, no. 6, p. 2688, Mar. 2023, doi: 10.3390/en16062688.
  10. A. F. Atiya, "Why does forecast combination work so well?," Int. J. Forecast., vol. 36, no. 1, pp. 197-200, Jan. 2020, doi: 10.1016/j.ijforecast.2019.03.010.
  11. Z. Cheng, L. Wang, and Y. Yang, "A Hybrid Feature Pyramid CNN-LSTM Model with Seasonal Inflection Month Correction for Medium- and Long-Term Power Load Forecasting," Energies, vol. 16, no. 7, p. 3081, Mar. 2023, doi: 10.3390/en16073081.
  12. M. Mathioudaki, D. Tsoukalas, M. Siavvas, and D. Kehagias, "Comparing Univariate and Multivariate Time Series Models for Technical Debt Forecasting," in Computational Science and Its Applications - ICCSA 2022 Workshops, vol. 13380, O. Gervasi, B. Murgante, S. Misra, A. M. A. C. Rocha, and C. Garau, Eds., in Lecture Notes in Computer Science, vol. 13380. , Cham: Springer International Publishing, 2022, pp. 62-78. doi: 10.1007/978-3-031-10542-5_5.
  13. N. H. Ismail, M. Du, D. Martinez, and Z. He, "Multivariate Multi-step Deep Learning Time Series Approach in Forecasting Parkinson's Disease Future Severity Progression," in Proceedings of the 10th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics, Niagara Falls NY USA: ACM, Sep. 2019, pp. 383-389. doi: 10.1145/3307339.3342185.
  14. D. Benvenuto, M. Giovanetti, L. Vassallo, S. Angeletti, and M. Ciccozzi, "Application of the ARIMA model on the COVID-2019 epidemic dataset," Data Brief, vol. 29, p. 105340, Apr. 2020, doi: 10.1016/j.dib.2020.105340.
  15. F. Li and H. Di, "Analysis of the Financing Structure of China's Listed New Energy Companies under the Goal of Peak CO2 Emissions and Carbon Neutrality," Energies, vol. 14, no. 18, p. 5636, Sep. 2021, doi: 10.3390/en14185636.
  16. M. Auger-Methe et al., "A guide to state-space modeling of ecological time series," Ecol. Monogr., vol. 91, no. 4, Nov. 2021, doi: 10.1002/ecm.1470.
  17. W. Mounter, C. Ogwumike, H. Dawood, and N. Dawood, "Machine Learning and Data Segmentation for Building Energy Use Prediction-A Comparative Study," Energies, vol. 14, no. 18, p. 5947, Sep. 2021, doi: 10.3390/en14185947.
  18. Namrye Son, "Wind power observation." figshare, p. 49095 Bytes, 2019. doi: 10.6084/M9.FIGSHARE.8330285.V3.
  19. Z. Wu, G. Luo, Z. Yang, Y. Guo, K. Li, and Y. Xue, "A comprehensive review on deep learning approaches in wind forecasting applications," CAAI Trans. Intell. Technol., vol. 7, no. 2, pp. 129-143, Jun. 2022, doi: 10.1049/cit2.12076.
  20. Shay Palachy Affek, "Detecting stationarity in time series data," https://towardsdatascience.com. [Online]. Available: https://towardsdatascience.com/detecting-stationarity-in-time-series-data-d29e0a21e638
  21. G. Petelin, G. Cenikj, and T. Eftimov, "Towards understanding the importance of time-series features in automated algorithm performance prediction," Expert Syst. Appl., vol. 213, p. 119023, Mar. 2023, doi: 10.1016/j.eswa.2022.119023.
  22. Y. Chen et al., "2-D regional short-term wind speed forecast based on CNN-LSTM deep learning model," Energy Convers. Manag., vol. 244, p. 114451, Sep. 2021, doi: 10.1016/j.enconman.2021.114451.
  23. C. Zhang, P. Chen, F. Jiang, J. Xie, and T. Yu, "Fault Diagnosis of Nuclear Power Plant Based on Sparrow Search Algorithm Optimized CNN-LSTM Neural Network," Energies, vol. 16, no. 6, p. 2934, Mar. 2023, doi: 10.3390/en16062934.
  24. C.-K. Ing, "Accumulated prediction errors, information criteria and optimal forecasting for autoregressive time series," Ann. Stat., vol. 35, no. 3, Jul. 2007, doi: 10.1214/009053606000001550.
  25. I. E. Livieris and P. Pintelas, "A novel multi-step forecasting strategy for enhancing deep learning models' performance," Neural Comput. Appl., vol. 34, no. 22, pp. 19453-19470, Nov. 2022, doi: 10.1007/s00521-022-07158-9.
  26. Nickalus Redell, "Direct Forecasting with Multiple Time Series," https://cran.r-project.org. [Online]. Available: https://cran.r-project.org/web/packages/forecastML/vignettes/grouped_forecast.html
  27. C. Emeksiz and M. Tan, "Multi-step wind speed forecasting and Hurst analysis using novel hybrid secondary decomposition approach," Energy, vol. 238, p. 121764, Jan. 2022, doi: 10.1016/j.energy.2021.121764.
  28. D. Chicco, M. J. Warrens, and G. Jurman, "The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation," PeerJ Comput. Sci., vol. 7, p. e623, Jul. 2021, doi: 10.7717/peerj-cs.623.
  29. A. Shrestha and A. Mahmood, "Review of Deep Learning Algorithms and Architectures," IEEE Access, vol. 7, pp. 53040-53065, 2019, doi: 10.1109/ACCESS.2019.2912200.
  30. Jason Brownlee, "How to Use Optimization Algorithms to Manually Fit Regression Models," Machine learning mastery. [Online]. Available: https://machinelearningmastery.com/optimize-regression-models/
  31. Department of CSE, VIIT(A), AP, India University, India, C. Sekhar, Department of CSE, VIIT(A), AP, India University, India, and P. S. Meghana, "A Study on Backpropagation in Artificial Neural Networks," Asia-Pac. J. Neural Netw. Its Appl., vol. 4, no. 1, pp. 21-28, Aug. 2020, doi: 10.21742/AJNNIA.2020.4.1.03.
  32. madrury, "Adding Noise to Regression Predictors is Ridge Regression," https://madrury.github.io/. [Online]. Available: https://madrury.github.io/jekyll/update/statistics/2017/08/12/noisy-regression.html
  33. Y. Yang et al., "A study on water quality prediction by a hybrid CNN-LSTM model with attention mechanism," Environ. Sci. Pollut. Res., vol. 28, no. 39, pp. 55129-55139, Oct. 2021, doi: 10.1007/s11356-021-14687-8.