Geliştirilmiş EEMD-EWT Tabanlı Yapay Sinir Ağı Modeli Kullanarak Çok Adımlı Rüzgar Hızı Tahmini

Öz

Rüzgar hızı tahminlemesi rüzgar güç dönüşüm sistemleri için oldukça önemlidir. Bu çalışmada kısa vadeli rüzgar hızı tahminlemesi için hibrit bir ayrıklaştırma yöntemi önerilmiştir. Önerilen yöntemde Toplu ampirik mod ayrıştırma (Ensemble Empirical Mode Decomposition, EEMD) ve Ampirik dalgacık dönüşümü (Emprical wavelet transform, EWT) birlikte kullanılmıştır. İlk defa kullanılan bu kombinasyon sonucunda elde edilen ayrıklaştırılmış rüzgar hızı sinyalleri kısmi otokorelasyon fonksiyonu (Partial autocorrelation function, PACF) ile öznitelik çıkarma işlemine tabi tutulmuştur. Elde edilen öznitelikler, geri beslemeli sinir ağına (Back propagation neural networks, BPNN) uygulanmak suretiyle çok adımlı rüzgar hız tahminleme işlemi gerçekleştirilmiştir. Önerilen modelin birbirinden bağımsız teknikler kullanılarak yapılan tekil ayrıklaştırmaya göre çok daha doğru ve güvenilir sonuçlar verdiği tespit edilmiştir. Çalışmada kullanılan veriler Tokat Gaziosmanpaşa Üniversitesi Taşlıçiftlik Kampüsü içerisinde kurulan ölçüm istasyonundan toplanmıştır. Önerilen hibrit model, yüksek hassasiyetli rüzgar hızı tahminleri için güvenilir, güçlü ve etkili olduğu kadar veri madenciliği uygulamalarında da kolaylıkla kullanılabilir. Tahmin performansının genel tahmin doğruluğu yaygın olarak kullanılan üç genel hata değerlendirme endeksi olan determinasyon katsayısı (determination coefficient (R2), ortalama mutlak yüzde hata (mean absolute percent error (MAPE) ve ortalama karekök hata (root mean square error (RMSE)) ile gerçekleştirildi.

Anahtar Kelimeler

• Rüzgar hızı tahmini
• Hibrit model
• Topluluk ampirik mod ayrıştırma
• Ampirik dalgacık dönüşümü
• Geri beslemeli sinir ağı

Multi-Step Wind Speed Estimation Using Improved EEMD-EWT Based Artificial Neural Network Model

Abstract

Wind speed estimation is very important for wind power conversion systems. In this study, a hybrid discretization method is proposed for short-term wind speed estimation. In the proposed method, Ensemble Empirical Mode Decomposition (EEMD) and Empirical wavelet transform (EWT) are used together. Discretized wind speed signals obtained as a result of this combination used for the first time were subjected to feature extraction process with partial autocorrelation function (PACF). The multi-step wind speed estimation process has been carried out by applying the obtained features to the feedback neural network (Back propagation neural networks, BPNN). It has been determined that the proposed model gives much more accurate and reliable results than the singular discretization using independent techniques. The data used in the study were collected from the measurement station established in Tokat Gaziosmanpaşa University Taşlıçiftlik Campus. The proposed hybrid model is reliable, powerful and effective for high precision wind speed predictions, as well as easily used in data mining applications. The overall prediction accuracy of the prediction performance was achieved with the three commonly used general error rating indices: determination coefficient (R2), mean absolute percent error (MAPE) and root mean square error (RMSE).

Keywords

• Wind speed prediction
• Hybrid model
• Ensemble empirical mode decomposition
• Empirical wavelet transform
• Feedback neural network

Kaynakça

1. Aghajani, A., Kazemzadeh, R., & Ebrahimi, A. (2016). A novel hybrid approach for predicting wind farm power production based on wavelet transform, hybrid neural networks and imperialist competitive algorithm. Energy Conversion and Management, 121, 232-240.
2. Cadenas, E., & Rivera, W. (2009). Short term wind speed forecasting in La Venta, Oaxaca, México, using artificial neural networks. Renewable Energy, 34(1), 274-278.
3. Emeksiz, C., & Demir, G. (2018). An investigation of the effect of meteorological parameters on wind speed estimation using bagging algorithm. International Journal of Intelligent Systems and Applications in Engineering, 6(4), 311-321.
4. Gilles, J. (2013). Empirical wavelet transform. IEEE transactions on signal processing, 61(16), 3999-4010.
5. Guo, Z., Zhao, W., Lu, H., & Wang, J. (2012). Multi-step forecasting for wind speed using a modified EMD-based artificial neural network model. Renewable Energy, 37(1), 241-249.
6. He, Q., Wang, J., & Lu, H. (2018). A hybrid system for short-term wind speed forecasting. Applied Energy, 226, 756-771.
7. Hu, J., Wang, J., & Ma, K. (2015). A hybrid technique for short-term wind speed prediction. Energy, 81, 563-574. Huang, N. E., Shen, Z., Long, S. R., Wu, M. C., Shih, H. H., Zheng, Q., Yen, N. C., Tung, C. C., & Liu, H. H. (1998). The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis. Proceedings of the Royal Society of London. Series A: mathematical, physical and engineering sciences, 454(1971), 903-995.
8. Jiang, P., & Li, C. (2018). Research and application of an innovative combined model based on a modified optimization algorithm for wind speed forecasting. Measurement, 124, 395-412.

9. Liu, H., Mi, X., & Li, Y. (2018). Comparison of two new intelligent wind speed forecasting approaches based on wavelet packet decomposition, complete ensemble empirical mode decomposition with adaptive noise and artificial neural networks. Energy Conversion and Management, 155, 188-200.
10. Liu, H., Tian, H. Q, Liang, X. F., & Li, Y. F. (2015). Wind speed forecasting approach using secondary decomposition algorithm and Elman neural networks. Applied Energy, 157, 183-194.
11. Liu, H., Wu, H., & Li, Y. (2018). Smart wind speed forecasting using EWT decomposition, GWO evolutionary optimization, RELM learning and IEWT reconstruction. Energy Conversion and Management, 161, 266-283.
12. Ozbay, Y., & Karlik, B. (2002). A fast training back-propagation algorithm on windows. Proceedings of the Third International Symposium on Mathematical & Computational Applications,
13. Peng, T., Zhou, J., Zhang, C., & Zheng, Y. (2017). Multi-step ahead wind speed forecasting using a hybrid model based on two-stage decomposition technique and AdaBoost-extreme learning machine. Energy Conversion and Management, 153, 589-602.
14. Qian, Z., Pei, Y., Zareipour, H., & Chen, N. (2019). A review and discussion of decomposition-based hybrid models for wind energy forecasting applications. Applied Energy, 235, 939-953.
15. Ren, C., An, N., Wang, J., Li, L., Hu, B., & Shang, D. (2014). Optimal parameters selection for BP neural network based on particle swarm optimization: A case study of wind speed forecasting. Knowledge-Based Systems, 56, 226-239.
16. Santhosh, M., Venkaiah, C., & Kumar, D. V. (2018). Ensemble empirical mode decomposition based adaptive wavelet neural network method for wind speed prediction. Energy Conversion and Management, 168, 482-493.
17. Sun, N., Zhou, J., Chen, L., Jia, B., Tayyab, M., & Peng, T. (2018). An adaptive dynamic short-term wind speed forecasting model using secondary decomposition and an improved regularized extreme learning machine. Energy, 165, 939-957.
18. Şenkal, S. (2014). Rüzgar Hızı Tahmin Yöntemleri - Örnek Bir Uygulama Ondokuz Mayıs Üniversitesi. Ondokuz Mayıs Üniversitesi, Fen Bilimleri Enstitüsü, Yüksek Lisans Tezi.
19. Tan, M. (2020). İkincil ayrıştırma tekniği kullanarak yapay sinir ağı temelli çok adımlı rüzgar hızı tahmini. Tokat Gaziosmanpaşa Üniversitesi, Fen bilimleri Enstitüsü, Yüksek Lisans Tezi.
20. Tascikaraoglu, A., & Uzunoglu, M. (2014). A review of combined approaches for prediction of short-term wind speed and power. Renewable and Sustainable Energy Reviews, 34, 243-254.
21. Torres, M. E., Colominas, M. A., Schlotthauer, G., & Flandrin, P. (2011). A complete ensemble empirical mode decomposition with adaptive noise. 2011 IEEE international conference on acoustics, speech and signal processing (ICASSP).
22. Wang, H., & Zhao, W. (2009). Arima model estimated by particle swarm optimization algorithm for consumer price index forecasting. International Conference on Artificial Intelligence and Computational Intelligence.
23. Wang, S., Zhang, N., Wu, L., & Wang, Y. (2016). Wind speed forecasting based on the hybrid ensemble empirical mode decomposition and GA-BP neural network method. Renewable Energy, 94, 629-636.
24. Wu, Z., & Huang, N. E. (2009). Ensemble empirical mode decomposition: a noise-assisted data analysis method. Advances in adaptive data analysis, 1(01), 1-41.
25. Xiao, L., Wang, J., Dong, Y., & Wu, J. (2015). Combined forecasting models for wind energy forecasting: A case study in China. Renewable and Sustainable Energy Reviews, 44, 271-288.
26. Zhang, W., Qu, Z., Zhang, K., Mao, W., Ma, Y., & Fan, X. (2017). A combined model based on CEEMDAN and modified flower pollination algorithm for wind speed forecasting. Energy Conversion and Management, 136, 439-451.