Abstract
The problem of wind energy production prediction has been one of the most prolific topics of study in the field of machine learning applied to the energy sector. Usually, these models receive data in tabular format. However, in this work we propose to solve the problem of predicting wind power like a spatio-temporal prediction problem as if it were an image or video analysis problem. On the one hand, energy production and the weather variables provided by Numerical Weather Prediction models (NWP) are time series, justifying the temporal treatment. On the other hand, NWP variables are provided in a regular grid format (in terms of latitude and longitude). Thus, the data are arranged as different meteorological variables in the shape of a grid, justifying the spatial treatment as if it were a low-resolution image, where the meteorological points are treated as pixels. For this reason, the goal of this article is to carry out an initial benchmark that compares the performance measured between different types of deep learning architectures that take advantage of these temporal and spatial features. The proposed architectures are CNN, LSTM, LSTM+CNN (Stacked), LSTM+CNN (Parallel), ConvLSTM and Vision Transformer.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Badrinath Krishna, V., Wadman, W., Kim, Y.: Nowcasting: accurate and precise short-term wind power prediction using hyperlocal wind forecasts (2018)
Baldi, P., Sadowski, P.J.: Understanding dropout. In: Burges, C., Bottou, L., Welling, M., Ghahramani, Z., Weinberger, K. (eds.) Advances in Neural Information Processing Systems, vol. 26. Curran Associates, Inc. (2013)
Chen, Y., Wang, Y., Dong, Z., Su, J., Han, Z., Zhou, D., Zhao, Y., Bao, Y.: 2-d regional short-term wind speed forecast based on CNN-LSTM deep learning model. Energy Conv. Manag. 244, 114451 (2021)
Clevert, D.A., Unterthiner, T., Hochreiter, S.: Fast and accurate deep network learning by exponential linear units (elus) (2015)
Díaz-Vico, D., Torres-Barrán, A., Omari, A., Dorronsoro, J.R.: Deep neural networks for wind and solar energy prediction. Neural Process. Lett. 46(3), 829–844 (2017)
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., Houlsby, N.: An image is worth 16\(\times \)16 words: transformers for image recognition at scale (2020)
Gao, M., Li, J., Hong, F., Long, D.: Short-term forecasting of power production in a large-scale photovoltaic plant based on LSTM. Appl. Sci. 9, 3192 (2019)
Hochreiter, S.: The vanishing gradient problem during learning recurrent neural nets and problem solutions. Int. J. Uncertain. Fuzziness Knowl.-Based Syst. 6, 107–116 (1998)
Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9, 1735–80 (1997)
Ioffe, S., Szegedy, C.: Batch normalization: accelerating deep network training by reducing internal covariate shift. In: Bach, F., Blei, D. (eds.) Proceedings of the 32nd International Conference on Machine Learning. Proceedings of Machine Learning Research, vol. 37, pp. 448–456. PMLR, Lille, France (2015)
Karim, F., Majumdar, S., Darabi, H., Chen, S.: LSTM fully convolutional networks for time series classification. IEEE Access 6, 1662–1669 (2018)
Lecun, Y., Haffner, P., Bengio, Y.: Object recognition with gradient-based learning (2000)
Lipton, Z.: A critical review of recurrent neural networks for sequence learning (2015)
Ren, J., Yu, Z., Gao, G., Yu, G., Yu, J.: A CNN-LSTM-lightgbm based short-term wind power prediction method based on attention mechanism. Energy Rep. (iCPE 2021-The 2nd International Conference on Power Engineering) 8, 437–443 (2022)
Schuster, M., Paliwal, K.: Bidirectional recurrent neural networks. IEEE Trans. Signal Process. 45, 2673–2681 (1997)
Shi, X., Chen, Z., Wang, H., Yeung, D.Y., Wong, W.k., Woo, W.c.: Convolutional lstm network: a machine learning approach for precipitation nowcasting. In: Cortes, C., Lawrence, N., Lee, D., Sugiyama, M., Garnett, R. (eds.) Advances in Neural Information Processing Systems, vol. 28. Curran Associates, Inc. (2015)
Staudemeyer, R., Morris, E.: Understanding LSTM—A tutorial into long short-term memory recurrent neural networks (2019)
Torres, J., Aguilar, R., Zúñiga, K.: Deep learning to predict the generation of a wind farm. J. Renew. Sustain. Energy 10 (2018)
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, L., Polosukhin, I.: Attention is all you need (2017)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
del Campo, R., Anguiano, E., Romero, Á., Dorronsoro, J.R. (2023). Deep Learning Applied to Wind Power Forecasting: A Spatio-Temporal Approach. In: Valenzuela, O., Rojas, F., Herrera, L.J., Pomares, H., Rojas, I. (eds) Theory and Applications of Time Series Analysis. ITISE 2022. Contributions to Statistics. Springer, Cham. https://doi.org/10.1007/978-3-031-40209-8_14
Download citation
DOI: https://doi.org/10.1007/978-3-031-40209-8_14
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-40208-1
Online ISBN: 978-3-031-40209-8
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)