Articles | Volume 16, issue 1
https://doi.org/10.5194/gmd-16-251-2023
© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/gmd-16-251-2023
© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Development and technical paper
| 
10 Jan 2023
Development and technical paper |
| 10 Jan 2023
| 10 Jan 2023
Adapting a deep convolutional RNN model with imbalanced regression loss for improved spatio-temporal forecasting of extreme wind speed events in the short to medium range
Daan R. Scheepens
CORRESPONDING AUTHOR
Research Group Data Mining and Machine Learning, Faculty of Computer Science, University of Vienna, Währingerstrasse 29, 1090 Vienna, Austria
Zentralanstalt für Meteorologie und Geodynamik (ZAMG), Hohe Warte 38, 1190 Vienna, Austria
Kateřina Hlaváčková-Schindler
CORRESPONDING AUTHOR
Research Group Data Mining and Machine Learning, Faculty of Computer Science, University of Vienna, Währingerstrasse 29, 1090 Vienna, Austria
Claudia Plant
Research Group Data Mining and Machine Learning, Faculty of Computer Science, University of Vienna, Währingerstrasse 29, 1090 Vienna, Austria
Related authors
No articles found.
Rainer Wöß, Katerina Hlavácková-Schindler, Irene Schicker, Petrina Papazek, and Claudia Plant
EGUsphere, https://doi.org/10.5194/egusphere-2024-3126, https://doi.org/10.5194/egusphere-2024-3126, 2024
This preprint is open for discussion and under review for Geoscientific Model Development (GMD).
Short summary
Short summary
HMMLVis is a causal inference, easy-to-use visualization software. It can be applied in any scientific discipline exploring time series and their relationships. The tool uses heterogeneous Granger causality. The tool is demonstrated on different types of applications related to meteorological events in a renewable energy, air pollution, and the EUMETNET postprocessing benchmark data. We believe HMMVis will serve climatologists or meteorologists as an interpretable causal visualization tool.
Cited articles
Alessandrini, S., Sperati, S., and Monache, L. D.:
Improving the Analog Ensemble Wind Speed Forecasts for Rare Events, Mon. Weather Rev., 147, 2677–2692, https://doi.org/10.1175/MWR-D-19-0006.1, 2019. a
Amato, F., Guignard, F., Robert, S., and Kanevski, M.:
A novel framework for spatio-temporal prediction of environmental data using deep learning, Sci. Rep.-UK, 10, 22243, https://doi.org/10.1038/s41598-020-79148-7, 2020. a
Ashkboos, S., Huang, L., Dryden, N., Ben-Nun, T., Dueben, P., Gianinazzi, L., Kummer, L., and Hoefler, T.:
ENS-10: A Dataset For Post-Processing Ensemble Weather Forecast, arXiv [cs.LG], https://doi.org/10.48550/ARXIV.2206.14786, 2022. a
Batista, G., Prati, R., and Monard, M.-C.:
A Study of the Behavior of Several Methods for Balancing machine Learning Training Data, SIGKDD Explorations, 6, 20–29, https://doi.org/10.1145/1007730.1007735, 2004. a
Burton, T., Sharpe, D., Jenkins, N., and Bossanyi, E.:
Reviewed Work: “Wind Energy Handbook”, Wind Engineering, 25, 197–199, http://www.jstor.org/stable/43749820 (last access: 2 January 2023), 2001. a
Cadenas, E., Rivera, W., Campos-Amezcua, R., and Heard, C.:
Wind Speed Prediction Using a Univariate ARIMA Model and a Multivariate NARX Model, Energies, 9, 109, https://doi.org/10.3390/en9020109, 2016. a
Chavas, D. R. and Dawson II, D. T.:
An Idealized Physical Model for the Severe Convective Storm Environmental Sounding, J. Atmos. Sci, 78, 653–670, https://doi.org/10.1175/JAS-D-20-0120.1, 2021. a
Chen, K. and Yu, J.:
Short-term wind speed prediction using an unscented Kalman filter based state-space support vector regression approach, Appl. Energ., 113, 690–705, https://doi.org/10.1016/j.apenergy.2013.08.025, 2014. a, b
Cuomo, S., di Cola, V. S., Giampaolo, F., Rozza, G., Raissi, M., and Piccialli, F.:
Scientific Machine Learning through Physics-Informed Neural Networks: Where we are and What's next, arXiv [cs.LG], https://doi.org/10.48550/ARXIV.2201.05624, 2022. a
Cutululis, N., Litong-Palima, M., and Sørensen, P.:
Offshore Wind Power Production in Critical Weather Conditions, in: Proceedings of EWEA 2012 – European Wind Energy Conference & Exhibition, European Wind Energy Association (EWEA), http://events.ewea.org/annual2012/ (last access: 2 January 2023), 2012. a
Darwish, A. S. and Al-Dabbagh, R.:
Wind energy state of the art: present and future technology advancements, Renew. Energy Environ. Sustain., 5, 7, https://doi.org/10.1051/rees/2020003, 2020. a
Deng, X., Li, W., Liu, X., Guo, Q., and Newsam, S.:
One-class remote sensing classification: one-class vs. binary classifiers, Int. J. Remote Sens., 39, 1890–1910, https://doi.org/10.1080/01431161.2017.1416697, 2018. a
Ding, D., Zhang, M., Pan, X., Yang, M., and He, X.:
Modeling Extreme Events in Time Series Prediction, in: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, KDD '19, Association for Computing Machinery, New York, NY, USA, https://doi.org/10.1145/3292500.3330896, 1114–1122, 2019. a
Friederichs, P., Wahl, S., and Buschow, S.:
Postprocessing for Extreme Events, Chapter 5 in: Statistical Postprocessing of Ensemble Forecasts, edited by: Vannitsem, S., Wilks, D. S., and Messner, J. W., Elsevier, 127–154, https://doi.org/10.1016/B978-0-12-812372-0.00005-4, 2018. a
Fyrippis, I., Axaopoulos, P. J., and Panayiotou, G.:
Wind energy potential assessment in Naxos Island, Greece, Appl. Energ., 87, 577–586, 2010. a
Gao, N., Xue, H., Shao, W., Zhao, S., Qin, K. K., Prabowo, A., Rahaman, M. S., and Salim, F. D.:
Generative Adversarial Networks for Spatio-Temporal Data: A Survey, ACM T. Intel. Syst. Tec., 13, 1–25, https://doi.org/10.1145/3474838, 2022. a
Ghosh, A., Sufian, A., Sultana, F., Chakrabarti, A., and De, D.:
Fundamental Concepts of Convolutional Neural Network, Springer, 519–567, https://doi.org/10.1007/978-3-030-32644-9_36, 2020. a
Goyal, S., Raghunathan, A., Jain, M., Simhadri, H. V., and Jain, P.:
DROCC: Deep Robust One-Class Classification, CoRR, arXiv [cs.LG], arXiv:2002.12718, 2020. a
Hassanaly, M., Perry, B. A., Mueller, M. E., and Yellapantula, S.:
Uniform-in-Phase-Space Data Selection with Iterative Normalizing Flows, arXiv [cs.LG], https://doi.org/10.48550/ARXIV.2112.15446, 2021. a
Hendrycks, D., Mazeika, M., and Dietterich, T.:
Deep Anomaly Detection with Outlier Exposure, in: International Conference on Learning Representations, https://openreview.net/forum?id=HyxCxhRcY7 (last access: 2 January 2023), 2019. a
Hersbach, H., Bell, B., Berrisford, P., Biavati, G., Horányi, A., Muñoz Sabater, J., Nicolas, J., Peubey, C., Radu, R., Rozum, I., Schepers, D. abd Simmons, A., Soci, C., Dee, D., and Thépaut, J.-N.:
ERA5 hourly data on pressure levels from 1979 to present, Copernicus Climate Change Service (C3S) Climate Data Store (CDS) [data set], https://doi.org/10.24381/cds.bd0915c6, 2018. a, b
Hogan, R. J. and Mason, I. B.:
Deterministic forecasts of binary events, chap. 3, John Wiley & Sons, Ltd, 31–59, https://doi.org/10.1002/9781119960003.ch3, 2012. a
Huang, C., Li, F., and Jin, Z.:
Maximum Power Point Tracking Strategy for Large-Scale Wind Generation Systems Considering Wind Turbine Dynamics, IEEE T. Ind. Electron., 62, 2530–2539, https://doi.org/10.1109/TIE.2015.2395384, 2015. a
IEA:
Global Energy Review 2021, https://www.iea.org/reports/global-energy-review-2021 (last access: 2 January 2023), 2021. a
Ji, Y., Zhi, X., Ji, L., Zhang, Y., Hao, C., and Peng, T.:
Deep-learning-based post-processing for probabilistic precipitation forecasting, Front. Earth Sci., 10, https://doi.org/10.3389/feart.2022.978041, 2022. a
Jung, J. and Broadwater, R. P.:
Current status and future advances for wind speed and power forecasting, Renew. Sust. Energ. Rev., 31, 762–777, https://doi.org/10.1016/j.rser.2013.12.054, 2014. a
Kavasseri, R. G. and Seetharaman, K.:
Day-ahead wind speed forecasting using f-ARIMA models, Renew. Energ., 34, 1388–1393, https://doi.org/10.1016/j.renene.2008.09.006, 2009. a
Lagerquist, R. and Ebert-Uphoff, I.:
Can we integrate spatial verification methods into neural-network loss functions for atmospheric science?, arXiv [cs.LG], https://doi.org/10.48550/ARXIV.2203.11141, 2022. a
Lei, M., Shiyan, L., Chuanwen, J., Hongling, L., and Yan, Z.:
A review on the forecasting of wind speed and generated power, Renew. Sust. Energ. Rev., 13, 915–920, https://doi.org/10.1016/j.rser.2008.02.002, 2009. a
Li, C., Xiao, Z., Xia, X., Zou, W., and Zhang, C.:
A hybrid model based on synchronous optimisation for multi-step short-term wind speed forecasting, Appl. Energ., 215, 131–144, https://doi.org/10.1016/j.apenergy.2018.01.094, 2018. a, b, c
Marndi, A., Patra, G. K., and Gouda, K. C.:
Short-term forecasting of wind speed using time division ensemble of hierarchical deep neural networks, Bulletin of Atmospheric Science and Technology, 1, 91–108, https://doi.org/10.1007/s42865-020-00009-2, 2020. a, b
Mohamad, M. A. and Sapsis, T. P.:
Sequential sampling strategy for extreme event statistics in nonlinear dynamical systems, P. Natl. Acad. Sci. USA, 115, 11138–11143, https://doi.org/10.1073/pnas.1813263115, 2018. a
Oliveira, M., Moniz, N., Torgo, L., and Santos Costa, V.:
Biased resampling strategies for imbalanced spatio-temporal forecasting, International Journal of Data Science and Analytics, 12, 205–228, https://doi.org/10.1007/s41060-021-00256-2, 2021. a
Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., Desmaison, A., Kopf, A., Yang, E., DeVito, Z., Raison, M., Tejani, A., Chilamkurthy, S., Steiner, B., Fang, L., Bai, J., and Chintala, S.:
PyTorch: An Imperative Style, High-Performance Deep Learning Library, in: Advances in Neural Information Processing Systems 32, Curran Associates, Inc., 8024–8035, http://papers.neurips.cc/paper/9015-pytorch-an-imperative-style-high-performance-deep-learning-library.pdf (last access: 2 January 2023), 2019. a
Pathak, J., Subramanian, S., Harrington, P., Raja, S., Chattopadhyay, A., Mardani, M., Kurth, T., Hall, D., Li, Z., Azizzadenesheli, K., Hassanzadeh, P., Kashinath, K., and Anandkumar, A.:
FourCastNet: A Global Data-driven High-resolution Weather Model using Adaptive Fourier Neural Operators, arXiv [preprint], https://doi.org/10.48550/ARXIV.2202.11214, 2022. a
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, E.:
Scikit-learn: Machine Learning in Python, J. Mach. Learn. Res., 12, 2825–2830, 2011. a
Petrović, V. and Bottasso, C. L.:
Wind turbine optimal control during storms, J. Phys. Conf. Ser., 524, 012052, https://doi.org/10.1088/1742-6596/524/1/012052, 2014. a
Phipps, K., Lerch, S., Andersson, M., Mikut, R., Hagenmeyer, V., and Ludwig, N.:
Evaluating ensemble post-processing for wind power forecasts, Wind Energy, 25, 1379–1405, https://doi.org/10.1002/we.2736, 2022. a
Racah, E., Beckham, C., Maharaj, T., Kahou, S. E., Prabhat, and Pal, C.:
Extreme Weather: A Large-Scale Climate Dataset for Semi-Supervised Detection, Localization, and Understanding of Extreme Weather Events, in: Proceedings of the 31st International Conference on Neural Information Processing Systems, NIPS'17, Curran Associates Inc., Red Hook, NY, USA, 3405–3416, https://doi.org/10.5555/3294996.3295099, 2017. a, b
Rao, A. R., Wang, Q., Wang, H., Khorasgani, H., and Gupta, C.:
Spatio-Temporal Functional Neural Networks, in: 2020 IEEE 7th International Conference on Data Science and Advanced Analytics (DSAA), 81–89, https://doi.org/10.1109/DSAA49011.2020.00020, 2020. a
Rädler, A., Groenemeijer, P., Pistotnik, G., Sausen, R., and Faust, E.:
Identification of favorable environments for thunderstorms in reanalysis data, Meteorol. Z., 26, 59–70, https://doi.org/10.1127/metz/2016/0754, 2015. a
Ruder, S.:
An overview of gradient descent optimization algorithms, arXiv [preprint], https://doi.org/10.48550/arxiv.1609.04747, 2017. a
Salcedo-Sanz, S., Pérez-Bellido, A., Ortiz-García, E., Portilla-Figueras, A., Prieto, L., and Correoso, F.:
Accurate Short-Term Wind Speed Prediction by Exploiting Diversity in Input Data using Banks of Artificial Neural Networks, Neurocomputing, 72, 1336–1341, https://doi.org/10.1016/j.neucom.2008.09.010, 2009. a
Scheepens, D.: dscheepens/Deep-RNN-for-extreme-wind-speed-prediction: v1.0.0-beta (v1.0.0-beta), Zenodo [code], https://doi.org/10.5281/zenodo.7369015, 2022. a
Schmidl, S., Wenig, P., and Papenbrock, T.:
Anomaly Detection in Time Series: A Comprehensive Evaluation, Proceedings of the VLDB Endowment (PVLDB), 15, 1779–1797, https://doi.org/10.14778/3538598.3538602, 2022. a
Schweri, L., Foucher, S., Tang, J., Azevedo, V. C., Günther, T., and Solenthaler, B.:
A Physics-Aware Neural Network Approach for Flow Data Reconstruction From Satellite Observations, Frontiers in Climate, 3, 23, https://doi.org/10.3389/fclim.2021.656505, 2021. a
Shi, X. and Yeung, D.-Y.:
Machine Learning for Spatiotemporal Sequence Forecasting: A Survey, arXiv [preprint], https://doi.org/10.48550/arXiv.1808.06865, 2018. a
Shi, X., Chen, Z., Wang, H., Yeung, D.-Y., Wong, W.-K., and Woo, W.-C.:
Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting, in: Advances in Neural Information Processing Systems, edited by: Cortes, C., Lawrence, N., Lee, D., Sugiyama, M., and Garnett, R., Curran Associates, Inc., 28, https://proceedings.neurips.cc/paper/2015/file/07563a3fe3bbe7e3ba84431ad9d055af-Paper.pdf (last access: 2 January 2023), 2015. a, b, c, d, e, f, g, h, i, j
Tsonevsky, I., Doswell, C. A., and Brooks, H. E.:
Early Warnings of Severe Convection Using the ECMWF Extreme Forecast Index, Weather Forecast., 33, 857–871, https://doi.org/10.1175/WAF-D-18-0030.1, 2018. a
Vannitsem, S., Bremnes, J. B., Demaeyer, J., Evans, G., Flowerdew, J., Hemri, S., Lerch, S., Roberts, N., Theis, S., Atencia, A., Ben Bouallègue, Z., Bhend, J., Dabernig, M., De Cruz, L., Hieta, L., Mestre, O., Moret, L., Odak Plenkovic, I., Schmeits, M., and Ylhäisi, J.:
Statistical Postprocessing for Weather Forecasts – Review, Challenges and Avenues in a Big Data World, B. Am. Meteorol. Soc., 102, 1–44, https://doi.org/10.1175/BAMS-D-19-0308.1, 2020.
a
Wang, J., Zong, Y., You, S., and Træholt, C.:
A review of Danish integrated multi-energy system flexibility options for high wind power penetration, Clean Energy, 1, 23–35, https://doi.org/10.1093/ce/zkx002, 2017. a
Wang, S., Cao, J., and Yu, P.:
Deep Learning for Spatio-Temporal Data Mining: A Survey, IEEE T. Knowl. Data En., 34, 3681–3700, https://doi.org/10.1109/TKDE.2020.3025580, 2020. a, b
Wang, Y., Gao, Z., Long, M., Wang, J., and Yu, P.:
PredRNN++: Towards A Resolution of the Deep-in-Time Dilemma in Spatiotemporal Predictive Learning, in: Proceedings of the 35th International Conference on Machine Learning, edited by: Dy, J. and Krause, A., PMLR, 80, 5123–5132, https://proceedings.mlr.press/v80/wang18b.html (last access: 2 January 2023), 2018. a
Williams, R., Ferro, C., and Kwasniok, F.:
A comparison of ensemble post-processing methods for extreme events, Q. J. Roy. Meteor. Soc., 140, 1112–1120, https://doi.org/10.1002/qj.2198, 2014. a
Wiser, R., Yang, Z., Hand, M., Hohmeyer, O., Infield, D., Jensen, P. H., Nikolaev, V., O'Malley, M., Sinden, G., and Zervos, A.:
Wind Energy, in: IPCC Special Report on Renewable Energy Sources and Climate Change Mitigation, edited by: Edenhofer, O., Pichs-Madruga, R., Sokona, Y., Seyboth, K., Matschoss, P., Kadner, S., Zwickel, T., Eickemeier, P., Hansen, G., Schlömer, S., and von Stechow, C., Cambridge University Press, ISBN 978-1-107-60710-1,
2011. a, b, c, d
Xie, A., Yang, H., Chen, J., Sheng, L., and Zhang, Q.:
A Short-Term Wind Speed Forecasting Model Based on a Multi-Variable Long Short-Term Memory Network, Atmosphere, 12, 651, https://doi.org/10.3390/atmos12050651, 2021. a, b
Yang, Y., Zha, K., Chen, Y.-C., Wang, H., and Katabi, D.:
Delving into Deep Imbalanced Regression, in: Proceedings of the 38th International Conference on Machine Learning, edited by: Meila, M. and Zhang, T., PMLR, 139, 11842–11851, https://proceedings.mlr.press/v139/yang21m.html (last access: 2 January 2023), 2021. a, b
Yeo, I.-K. and Johnson, R. A.:
A New Family of Power Transformations to Improve Normality or Symmetry, Biometrika, 87, 954–959, 2000. a
Short summary
The production of wind energy is increasing rapidly and relies heavily on atmospheric conditions. To ensure power grid stability, accurate predictions of wind speed are needed, especially in the short range and for extreme wind speed ranges. In this work, we demonstrate the forecasting skills of a data-driven deep learning model with model adaptations to suit higher wind speed ranges. The resulting model can be applied to other data and parameters, too, to improve nowcasting predictions.
The production of wind energy is increasing rapidly and relies heavily on atmospheric...
