Hybrid SDE-Neural Networks for Interpretable Wind Power Prediction Using SCADA Data
Abstract
1. Introduction
2. Methods and Formulation
2.1. Recurrent Neural Networks (RNN)
2.2. Long Short-Term Memory (LSTM)
2.3. Convolutional Neural Networks (CNN)—LSTM
2.4. SDE-Neural Models and Network Models
2.4.1. Definition of SDE
2.4.2. Residual Networks (ResNets) and ResNet as a Discretisation of ODEs
2.4.3. Euler–Maruyama Approximation for SDE
2.4.4. Lévy-Based Stochastic Differential Equations
2.4.5. SDE with Lévy Motion and Neural Networks
3. Case Study and Experiment
3.1. Dataset
3.2. Data Exploration
3.3. Data Cleaning and Preprocessing
3.4. Evaluation Index
4. Results and Discussion
4.1. RNN Performance
4.2. LSTM Performance
4.3. CNN-LSTM Performance
4.4. SDE-Alfa Stable
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
Abbreviations
SCADA | Supervisory Control and Data Acquisition Institute |
NNs | Neural Networks |
RNN | Recurrent Neural Networks |
LSTM | Long Short-Term Memory |
CNN | Convolutional Neural Networks |
SDEs | Stochastic Differential Equations |
References
- International Energy Agency. Renewables 2024. IEA Technol. Rep. 2024. Available online: https://www.iea.org/reports/renewables-2024 (accessed on 9 October 2024).
- Foley, A.M.; Leahy, P.G.; Marvuglia, A.; McKeogh, E.J. Current methods and advances in the orecasting of wind power generation. Renew. Energy 2012, 37, 1–8. [Google Scholar] [CrossRef]
- Jung, J.; Broadwater, R.P. Current status and future advances for wind speed and power forecasting. Renew. Sustain. Energy Rev. 2014, 31, 762–777. [Google Scholar] [CrossRef]
- Couto, A.; Estanqueiro, A. Wind power plants hybridised with solar power: A generation forecast perspective. J. Clean. Prod. 2023, 423, 138793. [Google Scholar] [CrossRef]
- Abbasi, A.; Abbasi, S.; Ansari, J.; Rahmani, E. Effect of plug-in electric vehicles demand on the renewable micro-grids. J. Intell. Fuzzy Syst. 2015, 29, 1957–1966. [Google Scholar] [CrossRef]
- Hill, D.C.; McMillan, D.; Bell, K.R.; Infield, D. Application of auto-regressive models to UK wind speed data for power system impact studies. IEEE Trans. Sustain. Energy 2011, 3, 134–141. [Google Scholar] [CrossRef]
- Dowell, J.; Pinson, P. Very-short-term probabilistic wind power forecasts by sparse vector autoregression. IEEE Trans. Smart Grid 2015, 7, 763–770. [Google Scholar] [CrossRef]
- Yuan, C.; Liu, S.; Fang, Z. Comparison of China’s primary energy consumption forecasting by using ARIMAd GM (1,1) models. Energy 2016, 100, 384–390. [Google Scholar] [CrossRef]
- Lei, M.; Shiyan, L.; Chuanwen, J.; Hongling, L.; Yan, Z. A review on wind speed forecasting and generated power. Renew. Sustain. Energy Rev. 2009, 13, 915–920. [Google Scholar] [CrossRef]
- Giebel, G.; Brownsword, R.; Kariniotakis, G.; Denhard, M.; Draxl, C. The State-of-the-Art in Short-Term Prediction of Wind Power: A Literature Overview. 2011. Available online: https://backend.orbit.dtu.dk/ws/portalfiles/portal/128933990/GGiebelEtAl_StateOfTheArtInShortTermPrediction_ANEMOSplus_2011.pdf (accessed on 9 October 2024).
- Hochreiter, S.; Schmidhuber, J. Long short-term memory. Neural Comput. 1997, 9, 1735–1780. [Google Scholar] [CrossRef]
- LeCun, Y.; Bengio, Y.; Hinton, G. Deep learning. Nature 2015, 521, 436–444. [Google Scholar] [CrossRef]
- Luke, K.A.A.; Evangeline, D.P.; Anandhakumar, P. Short-term wind power prediction using deep learning approaches. In Advances in Computers; Elsevier: Amsterdam, The Netherlands, 2024; pp. 111–139. [Google Scholar] [CrossRef]
- Shapi’i, A.; Herawan, T.; Ghani, M.K.A.; Alam, M.M.; Ghosh, S. A Machine Learning-Based Gradient Boosting Regression Approach for Wind Power Production Forecasting: A Step towards Smart Grid Environments. Energies 2021, 14, 5652. [Google Scholar] [CrossRef]
- Di Persio, L.; Fraccarolo, N. Energy Consumption Forecasts by Gradient Boosting Regression Trees. Mathematics 2023, 11, 1068. [Google Scholar] [CrossRef]
- Ning, S.A. A simple solution method for the blade element momentum equations with guaranteed convergence. Wind Energy 2014, 17, 1327–1345. [Google Scholar] [CrossRef]
- Li, G.; Shi, J. A comparative study on direct and indirect wind power prediction methods. J. Renew. Sustain. Energy 2011, 3, 043106. [Google Scholar] [CrossRef]
- Jia, X.; Benson, A. Neural Jump Stochastic Differential Equations. arXiv 2019, arXiv:1905.10403. [Google Scholar]
- Kong, L.; Sun, R.; Zha, H. SDE-Net: Equipping deep neural networks with stochastic differential equations. Proc. AAAI Conf. Artif. Intell. 2020, 34, 4462–4469. [Google Scholar]
- Chen, X.; Yang, H.; Karniadakis, G.E. CGNSDE: Coarse-Grained Neural Stochastic Differential Equations for Learning Multiscale Dynamics. arXiv 2024, arXiv:2402.06168. [Google Scholar]
- Huang, L.-J.; Majka, M.B.; Wang, J. Approximation of heavy-tailed distributions via stable-driven SDEs. arXiv 2020, arXiv:2007.02212. [Google Scholar] [CrossRef]
- Imkeller, P.; Pavlyukevich, I. First exit times of solutions of nonlinear stochastic differential equations driven by symmetric Lévy processes with alpha-stable components. arXiv 2004, arXiv:math/0409246. [Google Scholar]
- Liu, X. On the α-dependence of stochastic differential equations with Hölder drift driven by α-stable Lévy processes. J. Math. Anal. Appl. 2022, 506, 125642. [Google Scholar] [CrossRef]
- Mienye, I.D.; Swart, T.G.; Obaido, G. Recurrent Neural Networks: A Comprehensive Review of Architectures, Variants, and Applications. Information 2024, 15, 517. [Google Scholar] [CrossRef]
- Mohsen, S.; Elkaseer, A.; Scholz, S.G. Industry 4.0-Oriented Deep Learning Models for Human Activity Recognition. IEEE Access 2021, 9, 150508–150521. [Google Scholar] [CrossRef]
- Diniz, A.P.M.; Ciarelli, P.M.; Salles, E.O.T.; Coco, K.F. Use of deep neural networks for clogging detection in the Submerged Entry Nozzle of the continuous casting. Expert Syst. Appl. 2024, 238, 121963. [Google Scholar] [CrossRef]
- Kim, T.Y.; Cho, S.B. Predicting residential energy consumption using CNN-LSTM neural networks. Energy 2020, 182, 72–81. [Google Scholar] [CrossRef]
- Pham, V.H.; Nguyen, A.B.; Pham, V.H.; Nguyen, N.A.; Giang, T.P.T.; Nguyen, C.H.; Ngo, P.T.T.; Bui, D.T. One-dimensional deep learning driven geospatial analysis for flash flood susceptibility mapping: A case study in North Central Vietnam. Earth Sci. Inform. 2024, 17, 4419–4440. [Google Scholar] [CrossRef]
- Khan, S.; Mazhar, T.; Shahzad, T.; Waheed, W.; Waheed, A.; Saeed, M.M.; Hamam, H. Optimizing Deep Neural Network Architectures for Renewable Energy Forecasting. Discov. Sustain. 2024, 5, 394. [Google Scholar] [CrossRef]
- Iversen, E.B.; Morales, J.M.; Møller, J.K.; Madsen, H. Short-term probabilistic wind speed forecasting using stochastic differential equations. Int. J. Forecast. 2016, 32, 981–990. [Google Scholar] [CrossRef]
- Di Persio, L.; Fraccarolo, N.; Veronese, A. Wind energy production in Italy: A forecasting approach based on fractional Brownian motion and generative adversarial networks. Mathematics 2024, 12, 2105. [Google Scholar] [CrossRef]
- Yang, L.; Gao, T.; Lu, Y.; Duan, J.; Liu, T. Neural network stochastic differential equation models with applications to financial data forecasting. Appl. Math. Model. 2023, 115, 279–299. [Google Scholar] [CrossRef]
- He, K.; Zhang, X.; Ren, S.; Sun, J. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 27–30 June 2016; pp. 770–778. [Google Scholar]
- Hmidi, A.; Jihene, M. Deep Residual Network in Network. Comput. Intell. Neurosci. 2021, 2021, 6659083. [Google Scholar] [CrossRef] [PubMed]
- Lu, Y.; Zhong, A.; Li, Q.; Dong, B. Beyond finite layer neural networks: Bridging deep architectures and numerical differential equations. arXiv 2018, arXiv:1710.10121. [Google Scholar]
- Chen, R.T.Q.; Rubanova, Y.; Bettencourt, J.; Duvenaud, D.K. Neural ordinary differential equations. arXiv 2018, arXiv:1806.07366. [Google Scholar]
- Zenodo. [Dataset] Zenodo Record 5841834. 2024. Available online: https://zenodo.org/records/5841834 (accessed on 1 February 2022).
- Hammer, F.; Barber, S. Data imputation for SCADA data using Graph Neural Networks. J. Phys. Conf. Ser. 2025, 3025, 012014. [Google Scholar] [CrossRef]
- The Wind Power. Senvion MM92-2050 Wind Turbine. Available online: https://www.thewindpower.net/turbine_en_327_senvion_mm92-2050.php (accessed on 28 August 2024).
- Lobo, B.A.; Özçakmak, Ö.S.; Madsen, H.A.; Schaffarczyk, A.P.; Breuer, M.; Sørensen, N.N. On the laminar–turbulent transition mechanism on megawatt wind turbine blades operating in atmospheric flow. Wind. Energy Sci. 2023, 8, 303–326. [Google Scholar] [CrossRef]
- Long, H.; Xu, S.; Gu, W. An abnormal wind turbine data cleaning algorithm based on color space conversion and image feature detection. Appl. Energy 2022, 311, 118594. [Google Scholar] [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Ghadiri, M.; Persio, L.D. Hybrid SDE-Neural Networks for Interpretable Wind Power Prediction Using SCADA Data. Electricity 2025, 6, 48. https://doi.org/10.3390/electricity6030048
Ghadiri M, Persio LD. Hybrid SDE-Neural Networks for Interpretable Wind Power Prediction Using SCADA Data. Electricity. 2025; 6(3):48. https://doi.org/10.3390/electricity6030048
Chicago/Turabian StyleGhadiri, Mehrdad, and Luca Di Persio. 2025. "Hybrid SDE-Neural Networks for Interpretable Wind Power Prediction Using SCADA Data" Electricity 6, no. 3: 48. https://doi.org/10.3390/electricity6030048
APA StyleGhadiri, M., & Persio, L. D. (2025). Hybrid SDE-Neural Networks for Interpretable Wind Power Prediction Using SCADA Data. Electricity, 6(3), 48. https://doi.org/10.3390/electricity6030048