RNN-Based Time Series Analysis for Wind Turbine Energy Forecasting

Abstract

One significant source of renewable energy is wind power, which has the potential to generate sustainable energy. However, wind turbines have many challenges, such as high initial investment costs, the dynamic nature of wind speed, and the challenge of locating wind-efficient energy regions. Wind power predicting is crucial for effective planning of wind power generation, optimization of power generation, grid integration, and security of supply. Therefore, highly accurate forecasts ensure the efficient and sustainable operation of the wind energy sector and contribute to energy security, economic stability, and environmental sustainability. This study proposes a deep learning (DL) approach based on recurrent neural networks (RNNs) for long-term wind power forecasting utilizing climatic data. The input data that forms the basis of this study is obtained directly from a wind turbine system operating under real-world conditions. The proposed model in this study is based on a multilayer back-propagation neural network (RNN) architecture specifically designed to effectively handle complex data sets and time-dependent series. The architecture of the model is built on an RNN consisting of four separate layers, each with 50 hidden neurons, carefully structured to increase its capacity to capture complex features. To improve the robustness of the model and avoid overlearning, each RNN layer is followed by a dropout (regularizing) layer that randomly deactivates 20% of the neurons to enhance the generalization ability of the network. To finalize the prediction capability of the model, a linear function was chosen in the last layer to directly match the actual values. Evaluating the model performance metrics, the proposed architecture achieved a prediction accuracy of 91% R2 on the test dataset. The findings indicate that proposed method based on multilayer RNN can successfully capture the relationships between the sequential data of the wind turbine.

Keywords

• Recurrent Neural Networks
• Machine learning
• Wind power forecasting
• Regression.

Rüzgar Türbini Enerji Tahmini için RNN Tabanlı Zaman Serisi Analizi

Abstract

Önemli yenilenebilir enerjinin kaynaklarından biri, sürdürülebilir enerji üretme potansiyeline sahip olan rüzgar enerjisidir. Ancak rüzgâr türbinleri, yüksek ilk yatırım maliyetleri, rüzgâr hızının dinamik yapısı ve rüzgâr açısından verimli enerji bölgeleri bulma problemleri gibi birçok zorluğa sahiptir. Rüzgâr enerjisinin tahmin edilmesi, rüzgâr enerjisi üretiminin etkili bir şekilde planlanması, enerji üretiminin optimizasyonu, şebeke entegrasyonu ve arz güvenliği için çok önemlidir. Bu nedenle yüksek doğrulukta tahminler rüzgar enerjisi sektörünün verimli ve sürdürülebilir bir şekilde çalışmasını sağlar ve enerji güvenliğine, ekonomik istikrara ve çevresel sürdürülebilirliğe katkıda bulunur. Bu çalışma iklimsel verileri kullanarak uzun vadeli rüzgar enerjisi tahmini için tekrarlayan sinir ağlarına (RNN'ler) dayalı bir derin öğrenme (DL) yaklaşımı önermektedir. Bu çalışmanın temelini oluşturan girdi verileri, gerçek dünya koşullarında çalışan bir rüzgâr türbini sisteminden doğrudan elde edilmiştir. Çalışmada önerilen model, karmaşık veri setlerini ve zamana bağlı serileri etkili bir şekilde işlemek için özel olarak tasarlanmış çok katmanlı bir geri yayılım sinir ağı (RNN) mimarisine dayanmaktadır. Modelin mimarisi, karmaşık özellikleri yakalama kapasitesini artırmak için dikkatlice yapılandırılmış, her biri 50 gizli nörona sahip dört ayrı katmandan oluşan bir RNN üzerine inşa edilmiştir. Modelin sağlamlığını artırmak ve aşırı öğrenmeyi önlemek için her RNN katmanını, nöronların %20'sini rastgele devre dışı bırakan bir bırakma (düzenleyici) katmanı takip eder.

Keywords

• Tekrarlayan Sinir Ağları
• Makine öğrenmesi
• Rüzgar gücü tahmini
• Regresyon

References

1. [1] BAYRAM, A. B., & YAKUT, K. (2022). RENEWABLE ENERGY SCENARIO IN ELECTRICITY SYSTEM FOR ISPARTA PROVINCE THE YEAR 2030. International Journal of Engineering and Innovative Research, 4(3), 163-177. https://doi.org/10.47933/ijeir.1144163
2. [2] Bektaş, Y., & Karaca, H. (2022). Red deer algorithm based selective harmonic elimination for renewable energy application with unequal DC sources. Energy Reports, 8, 588-596.
3. [3] Sevim, D., Fidan, Ş., POLAT, S., & OKTAY, H. (2017). Experimental and articial neural network based studies on thermal conductivity of lightweight building materials. European Journal of Technique (EJT), 7(1), 33-41. Retrieved from https://dergipark.org.tr/en/pub/ejt/issue/34033/376667
4. [4] Saglam, M., Spataru, C., & Karaman, O. A. (2022). Electricity Demand Forecasting with Use of Artificial Intelligence: The Case of Gokceada Island. Energies, 15(16), 5950. https://doi.org/10.3390/en15165950
5. [5] Karakaya, H., Fidan, Ş., Şen, İ. E., & Gündoğdu, A. (2017). Batman ili fotovoltaik güneş enerjisi potansiyelinin analiz ve değerlendirmesi. Retrieved from https://earsiv.batman.edu.tr/xmlui/handle/20.500.12402/3941
6. [6] De Giorgi, M. G., Congedo, P. M., & Malvoni, M. (2014). Photovoltaic power forecasting using statistical methods: impact of weather data. IET Science, Measurement & Technology, 8(3), 90-97. https://doi.org/10.1049/iet-smt.2013.0135
7. [7] SANCAR, M. R., & BAYRAM, A. B. (2023). Modeling and Economic Analysis of Greenhouse Top Solar Power Plant with Pvsyst Software. International Journal of Engineering and Innovative Research, 5(1), 48-59. https://doi.org/10.47933/ijeir.1209362
8. [8] Hassan, A., Rehman, A. U., Shabbir, N., Hassan, S. R., Sadiq, M. T., & Arshad, J. (2019, February). Impact of inertial response for the variable speed wind turbine. In 2019 International Conference on Engineering and Emerging Technologies (ICEET) (pp. 1-6). IEEE. https://doi.org/10.1109/CEET1.2019.8711826.

9. [9] Fidan, Ş., & Çimen, H. (2021). Rüzgâr türbinlerinde tork ve kanat eğim açısı kontrolü. Batman Üniversitesi Yaşam Bilimleri Dergisi, 11(1), 12-26. Retrieved from https://dergipark.org.tr/en/pub/buyasambid/issue/63446/880791 [10] Fidan, Ş. (2010). Değişken hızlı-değişken kanat açılı rüzgar türbinlerinin tork ve kanat açısı kontrolü (Master's thesis, Fen Bilimleri Enstitüsü).
10. [11] Mahmoud, T., Dong, Z. Y., & Ma, J. (2018). An advanced approach for optimal wind power generation prediction intervals by using self-adaptive evolutionary extreme learning machine. Renewable energy, 126, 254-269. https://doi.org/10.1016/j.renene.2018.03.035
11. [12] Süzen, A. A., Duman, B., & Şen, B. (2020, June). Benchmark analysis of jetson tx2, jetson nano and raspberry pi using deep-cnn. In 2020 International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA) (pp. 1-5). IEEE. https://doi.org/10.1109/HORA49412.2020.9152915. [13] Foley, A. M., Leahy, P. G., Marvuglia, A., & McKeogh, E. J. (2012). Current methods and advances in forecasting of wind power generation. Renewable energy, 37(1), 1-8. https://doi.org/10.1016/j.renene.2011.05.033
12. [14] Zazoum, B. (2022). Solar photovoltaic power prediction using different machine learning methods. Energy Reports, 8, 19-25. https://doi.org/10.1016/j.egyr.2021.11.183
13. [15] Woon, W. L., Aung, Z., Kramer, O., & Madnick, S. (Eds.). (2017). Data Analytics for Renewable Energy Integration: Informing the Generation and Distribution of Renewable Energy: 5th ECML PKDD Workshop, DARE 2017, Skopje, Macedonia, September 22, 2017, Revised Selected Papers (Vol. 10691). Springer.
14. [16] Aydin, I., Celebi, S. B., Barmada, S., & Tucci, M. (2018). Fuzzy integral-based multi-sensor fusion for arc detection in the pantograph-catenary system. Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit, 232(1), 159-170. https://doi.org/10.1177/0954409716662090
15. [17] Kingsford, C., & Salzberg, S. L. (2008). What are decision trees?. Nature biotechnology, 26(9), 1011-1013. https://doi.org/10.1038/nbt0908-1011
16. [18] POLAT, S., FİDAN, Ş., & OKTAY, H. (2020). Hafif Yapı Malzemelerinin Isıl İletkenlik Özelliklerinin Yapay Sinir Ağları Kullanılarak Tahmin Edilmesi. Batman Üniversitesi Yaşam Bilimleri Dergisi, 10(1), 28-41. Retrieved from https://dergipark.org.tr/en/pub/buyasambid/issue/55551/643721
17. [19] Shabbir, N., AhmadiAhangar, R., Kütt, L., Iqbal, M. N., & Rosin, A. (2019, October). Forecasting short term wind energy generation using machine learning. In 2019 IEEE 60th International Scientific Conference on Power and Electrical Engineering of Riga Technical University (RTUCON) (pp. 1-4). IEEE. https://doi.org/10.1109/RTUCON48111.2019.8982365.
18. [20] Birecikli, B., Karaman, Ö. A., Çelebi, S. B., & Turgut, A. (2020). Failure load prediction of adhesively bonded GFRP composite joints using artificial neural networks. Journal of Mechanical Science and Technology, 34, 4631-4640. https://doi.org/10.1007/s12206-020-1021-7
19. [21] POLAT, S., FİDAN, Ş., & OKTAY, H. (2020). Hafif Yapı Malzemelerinin Isıl İletkenlik Özelliklerinin Yapay Sinir Ağları Kullanılarak Tahmin Edilmesi. Batman Üniversitesi Yaşam Bilimleri Dergisi, 10(1), 28-41. retrieved from https://dergipark.org.tr/en/pub/buyasambid/issue/55551/643721
20. [22] Kubilay, H. A. N., ÖZTÜRK, G., & ASLAN, A. (2023, June). Yapay Sinir Ağları Kullanarak Yüzey Pürüzlülüğü Tespiti. In International Conference on Pioneer and Innovative Studies (Vol. 1, pp. 487-492).
21. [23] Agarwal, K., & Vadhera, S. (2022, April). Short-term Wind Speed Prediction using ANN. In 2022 International Conference on Sustainable Computing and Data Communication Systems (ICSCDS) (pp. 496-501). IEEE. https://doi.org/10.1109/ICSCDS53736.2022.9760899.
22. [24] Mason, K., Duggan, J., & Howley, E. (2018). Forecasting energy demand, wind generation and carbon dioxide emissions in Ireland using evolutionary neural networks. Energy, 155, 705-720. https://doi.org/10.1016/j.energy.2018.04.192
23. [25] Fidan, Ş., Cebeci, M., & Gündoğdu, A. (2019). Extreme Learning Machine Based Control of Grid Side Inverter for Wind Turbines. Tehnički vjesnik, 26(5), 1492-1498. https://doi.org/10.17559/TV-20180730143757 [26] Mahmoud, T., Dong, Z. Y., & Ma, J. (2018). An advanced approach for optimal wind power generation prediction intervals by using self-adaptive evolutionary extreme learning machine. Renewable energy, 126, 254-269. https://doi.org/10.1016/j.renene.2018.03.035
24. [27] ÇELEBİ, S. B., & EMİROĞLU, B. G. (2023). Alzheimer Teşhisi için Derin Öğrenme Tabanlı Morfometrik Analiz. Journal of the Institute of Science and Technology, 13(3), 1454-1467. https://doi.org/10.21597/jist.1275669 [28] Süzen, A. A., & Şimşek, M. A. (2020). A novel approach to machine learning application to protection privacy data in healthcare: Federated learning. Namık Kemal Tıp Dergisi, 8(1), 22-30. https://doi.org/10.37696/nkmj.660762
25. [29] Çelebi, S. B., & Emiroğlu, B. G. (2023). Leveraging Deep Learning for Enhanced Detection of Alzheimer's Disease Through Morphometric Analysis of Brain Images. Traitement du Signal, 40(4). https://doi.org/10.18280/ts.400405 [30] ÇALIŞKAN, A. (2022). classification of tympanic membrane images based on VGG16 model. Kocaeli Journal of Science and Engineering, 5(1), 105-111. https://doi.org/10.34088/kojose.1081402 [31] KARAMAN, Ö. A., & BEKTAŞ, Y. (2023). Makine Öğrenmesi ve Optimizasyon Yöntemleri ile Uzun Dönem Elektrik Enerjisi Tahmini: Türkiye Örneği. Mühendislik Bilimleri ve Araştırmaları Dergisi, 5(2), 285-292. https://doi.org/10.46387/bjesr.1306577 [32] Yaman, O., & Tuncer, T. (2022). Exemplar pyramid deep feature extraction based cervical cancer image classification model using pap-smear images. Biomedical Signal Processing and Control, 73, 103428. https://doi.org/10.1016/j.bspc.2021.103428
26. [33] Çalışkan, A., Demirhan, S., & Tekin, R. (2022). Comparison of different machine learning methods for estimating compressive strength of mortars. Construction and Building Materials, 335, 127490. https://doi.org/10.1016/j.conbuildmat.2022.127490
27. [34] Çalışkan, A. (2023). Diagnosis of malaria disease by integrating chi-square feature selection algorithm with convolutional neural networks and autoencoder network. Transactions of the Institute of Measurement and Control, 45(5), 975-985. https://doi.org/10.1177/01423312221147335
28. [35] Medsker, L. R., & Jain, L. C. (2001). Recurrent neural networks. Design and Applications, 5(64-67), 2.
29. [36] Çelebi, S. B., & Emiroğlu, B. G. (2023). A novel deep dense block-based model for detecting Alzheimer’s Disease. Applied Sciences, 13(15), 8686. https://doi.org/10.3390/app13158686 [ 37] Aydın, İ., Yaman, O., Karaköse, M., & Çelebi, S. B. (2014, June). Particle swarm based arc detection on time series in pantograph-catenary system. In 2014 IEEE International Symposium on Innovations in Intelligent Systems and Applications (INISTA) Proceedings (pp. 344-349). IEEE. https://doi.org/10.1109/INISTA.2014.6873642
30. [38] Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT press.
31. [39] Süzen, A. A., & Çakıroğlu, M. A. (2019). Prediction of rebound in shotcrete using deep bi-directional LSTM. Computers and Concrete, An International Journal, 24(6), 555-560. https://doi.org/10.12989/cac.2019.24.6.555
32. [40] Rahman, A., Srikumar, V., & Smith, A. D. (2018). Predicting electricity consumption for commercial and residential buildings using deep recurrent neural networks. Applied energy, 212, 372-385. https://doi.org/10.1016/j.apenergy.2017.12.051
33. [41] Wang, J., Li, X., Li, J., Sun, Q., & Wang, H. (2022). NGCU: A new RNN model for time-series data prediction. Big Data Research, 27, 100296. https://doi.org/10.1016/j.bdr.2021.100296
34. [42] Mahmoud, T., Dong, Z. Y., & Ma, J. (2018). An advanced approach for optimal wind power generation prediction intervals by using self-adaptive evolutionary extreme learning machine. Renewable energy, 126, 254-269. https://doi.org/10.1016/j.renene.2018.03.035
35. [43] Shabbir, N., Kütt, L., Jawad, M., Amadiahanger, R., Iqbal, M. N., & Rosin, A. (2019, November). Wind energy forecasting using recurrent neural networks. In 2019 Big Data, Knowledge and Control Systems Engineering (BdKCSE) (pp. 1-5). IEEE. https://doi.org/10.1109/BdKCSE48644.2019.9010593
36. 44] Kaggle.com. Online. Available: https://www.kaggle.com/datasets/berkerisen/wind-turbine-scada-dataset,. Accessed: 30-Oct-2023.
37. [45] Keyhani, A. (2016). Design of smart power grid renewable energy systems. John Wiley & Sons.
38. [46] Xu, H., & Deng, Y. (2017). Dependent evidence combination based on shearman coefficient and pearson coefficient. IEEE Access, 6, 11634-11640. https://doi.org/10.1109/ACCESS.2017.2783320
39. [47] Patro, S. G. O. P. A. L., & Sahu, K. K. (2015). Normalization: A preprocessing stage. arXiv preprint arXiv:1503.06462. https://doi.org/10.48550/arXiv.1503.06462
40. [48] Rahman, M. M., Shakeri, M., Tiong, S. K., Khatun, F., Amin, N., Pasupuleti, J., & Hasan, M. K. (2021). Prospective methodologies in hybrid renewable energy systems for energy prediction using artificial neural networks. Sustainability, 13(4), 2393. https://doi.org/10.3390/su13042393
41. [49] Li, G., Wang, H., Zhang, S., Xin, J., & Liu, H. (2019). Recurrent neural networks based photovoltaic power forecasting approach. Energies, 12(13), 2538. https://doi.org/10.3390/en12132538
42. [50] Karaman, Ö. A. (2023). Prediction of Wind Power with Machine Learning Models. Applied Sciences, 13(20), 11455. https://doi.org/10.3390/app132011455
43. [51] Öztekin, A., & Erçelebi, E. (2016). An early split and skip algorithm for fast intra CU selection in HEVC. Journal of Real-Time Image Processing, 12, 273-283. https://doi.org/10.1007/s11554-015-0534-2