Short-term wind power forecasting with responses transformation for taking into account wind turbine operational state D. Snegirev, A. Pazderin, V. Samoylenko and P. Bartolomey   2024/07/20

Abstract

The amount of electricity production provided by wind farms is increasing globally. In this regard, the accurate and reliable wind power forecasting for ensuring efficient operation of wind farms as part of energy systems is becoming crucial issue. The most rarely considered among the possible methods for prediction errors reducing is changing of the prediction responses and the individual wind turbines power aggregation method: forecasting at the turbine level, at the groups of turbines and at the wind farm level. There is no consensus yet on which approach is most effective. Also, the issue of taking into account the wind turbines operational state is not adequately covered in reported studies on the wind power forecasting. And the observations corresponding to the switched-off state of turbines are most likely considered as outliers. In this paper the effect of spatial smoothing within a wind farm is investigated and the evaluation of the response selection and individual turbines power aggregation approaches effectiveness in terms of the wind farm output prediction is performed. And finally, a new technique for taking into account the number of switched-off wind turbines based on response transformation for short-term wind power forecasting at the turbine groups and wind farm levels is proposed.

Authors and affiliations

D. Snegirev, A. Pazderin, V. Samoylenko and P. Bartolomey

Department of Automated Electrical Systems. Ural Federal University, Yekaterinburg (Russia)

Key words

Wind power forecast, aggregated forecast, data preprocessing, wind turbine operational state, response transformation, machine learning.

References

[1]  Q. T. Phan, Y. K Wu, Q. D. Phan. “An overview of data preprocessing for short-term wind power forecasting”, 7th International Conference on Applied System Innovation (ICASI), IEEE, (2021), pp.121-125.

[2]  Y. Zhao, L. Ye, W. Wang, H. Sun, Y. Ju, Y. Tang. “Data-driven correction approach to refine power curve of wind farm under wind curtailment”, IEEE Transactions on Sustainable Energy (2017, vol. 9.1, pp. 95-105.

[3]  H. Liu, C. Chen. “Data processing strategies in wind energy forecasting models and applications: A comprehensive review”, Applied Energy (2019), vol. 249, pp. 392-408.

[4]  H. Leng, X. Li, J. Zhu, H. Tang, Z.  Zhang, N. Ghadimi. “A new wind power prediction method based on ridgelet transforms, hybrid feature selection and closed-loop forecasting”, Advanced Engineering Informatics (2018), vol. 36, pp. 20-30.

[5]  Z. Liu, M. Hajiali, A. Torabi, B. Ahmadi, R. Simoes. “Novel forecasting model based on improved wavelet transform, informative feature selection, and hybrid support vector machine on wind power forecasting”, Journal of Ambient Intelligence and Humanized Computing (2018), vol. 9, pp. 1919-1931.

[6]  H. Huang, R. Jia, X. Shi, J. Liang, J. Dang. “Feature selection and hyper parameters optimization for short-term wind power forecast”, Applied Intelligence (2021), pp. 1-19.

[7]  S. Li, P. Wang, L. Goel. “Wind power forecasting using neural network ensembles with feature selection”, IEEE Trans. on sustainable energy (2015), vol. 6.4, pp. 1447-1456.

[8]  Y. Wang, R. Zou, F. Liu, L. Zhang, Q. Liu. “A review of wind speed and wind power forecasting with deep neural networks”, Applied Energy (2021), vol. 304, pp. 117766.

[9]  I. K. Bazionis, P. S. Georgilakis. “Review of deterministic and probabilistic wind power forecasting: Models, methods, and future research”, Electricity (2021), vol. 2.1, pp. 13-47.

[10] S. Hanifi, S. Lotfian, H. Zare-Behtash, A. Cammarano. “Offshore wind power forecasting—A new hyperparameter optimisation algorithm for deep learning models”, Energies (2022), vol. 15.19, pp. 6919.

[11]  J. Yan, Y. Liu, S. Han and M. Qiu. “Wind power grouping forecasts and its uncertainty analysis using optimized relevance vector machine”, Renewable and sustainable energy reviews (2013). Vol. 27, pp. 613-621.

[12]  N. Saleck, L. von Bremen. “Wind power forecast error smoothing within a wind farm”, Journal of Physics: Conference Series, IOP Publishing (2007), vol. 75:1, pp. 012051.

[13]  G. Yakoub, M. Sathyajith and L. Joao. “Direct and indirect short-term aggregated turbine-and farm-level wind power forecasts integrating several NWP sources”, Heliyon (2023) Vol. 9.11, pp. 1-14.

[14]  Y. Liu, X. Gao, J. Yan, S. Han, D. G. Infield. “Clustering methods of wind turbines and its application in short-term wind power forecasts”, J. of Ren. and Sust. Energy (2014), vol. 6.5.

[15]  F. Bilendo, A. Meyer, H.; Badihi, N. Lu, P. Cambron and B. Jiang. “Applications and Modeling Techniques of Wind Turbine Power Curve for Wind Farms—A Review”, Energies (2023), vol. 16:180, pp.1-38.

[16]  J. H. Friedman. “Greedy function approximation: a gradient boosting machine”, Annals of statistics (2001), pp. 1189-1232.

[17]  M. Feurer, F. Hutter. “Hyperparameter optimization”, Automated machine learning: Methods, systems, challenges (2019), pp. 3-33.

[18]  D. A. Snegirev, V. O. Samoylenko, A. V. Pazderin and P. I. Bartolomey. “The Selection of Machine Learning Model and Its Hyperparameters Using Bayesian Optimization for Short-Term Wind Power Forecasting”, Belarusian-Ural-Siberian Smart Energy Conference (BUSSEC) (2023), pp. 18-23.

[19]  The MathWorks, Inc. (2024). Statistics and Machine Learning Toolbox: Documentation (r2023a).