Makine Öğrenmesi Algoritmalarını Kullanarak Rüzgar Enerjisi Üretimi Tahmini

Öz

Yenilenebilir enerji dünyada giderek popüler hale gelmektedir, çünkü güneş, jeotermal, rüzgar enerjisi gibi yenilenebilir kaynaklar temiz, tükenmez ve doğal kaynaklardır. Rüzgar enerjisi, yenilenebilir enerjinin en önemli kaynaklarından biridir ve elektrik üretiminde kilit rol oynamaktadır. Bu nedenle, rüzgar enerjisi üretiminin doğru tahmin edilmesi enerji ticareti, planlama, zamanlama kararları ve rüzgar enerjisi üretim stratejilerini dengeleme zorluklarıyla başa çıkmada çok önemlidir. Bu çalışma, tarihsel rüzgar enerjisi üretim verileri ve hava durumu tahmin raporlarını kullanarak yenilenebilir enerji sektöründeki gerçek yaşam sorununu, önümüzdeki 24 saat için saat başına rüzgar enerjisi üretim miktarını makine öğrenmesi (ML) teknikleri ile doğru bir şekilde tahmin edebilmek için bir model önermektedir. Önerilen yaklaşımda; ilk olarak, veri setini anlamlı kümeler halinde gruplamak için denetimsiz bir ML yöntemi (K-Means kümeleme algoritması) uygulanır; daha sonra, bu kümeler yeni öznitelik değerleri olarak kabul edilir ve veri setini büyütmek için eklenir; son olarak, tahmin için denetimli bir ML yöntemi (regresyon) gerçekleştirilir. Bu çalışma dokuz denetimli öğrenme algoritmasını karşılaştırmaktadır: K-En Yakın Komşu, Destek Vektör Regresyonu, Rastgele Orman, Ekstra Ağaçlar, Gradyan Artırma, Ridge Regresyon, En Küçük Mutlak Daralma ve Seçme Operatörü, Karar Ağacı, ve Konvolüsyonel Sinir Ağı. Bu çalışmanın amacı, rüzgar türbinlerinin gerçek dünya verileri üzerindeki farklı ML algoritmalarının başarısını araştırmak ve rüzgar enerjisi üretimi tahmini için en doğru nihai modeli seçmek üzere çeşitli makine öğrenmesi algoritmalarını karşılaştırmak için bir metodoloji önermektir.

Anahtar Kelimeler

• Makine öğrenmesi,Rüzgar enerjisi üretimi tahmini,Rüzgar enerjisi tahmini,Yenilenebilir enerji

Wind Power Generation Prediction Using Machine Learning Algorithms

Öz

Renewable energy becomes progressively popular in the world because renewable resources such as solar, geothermal, wind energy are clean, inexhaustible and come from natural sources. Wind energy is one of the most significant resources of renewable energy and it plays a key role in the generation of electricity. Thus, accurate wind power estimation is crucial to deal with the challenges to balance energy trading, planning, scheduling decisions and strategies of wind power generation. This study proposes a prediction model to solve a real-life problem in the renewable energy sector by accurately estimating the amount of wind energy production per hour in the next 24 hours by applying machine learning (ML) techniques using historical wind power generation data and weather forecasting reports. In the proposed approach, first, an unsupervised ML method (i.e., the K-Means clustering algorithm) is applied to group data into meaningful clusters; then, these clusters are accepted as new feature values and added to the dataset to enlarge it; finally, a supervised ML method (i.e., regression) is performed for prediction. This study compares nine supervised learning algorithms: K-Nearest Neighbors, Support Vector Regression, Random Forest, Extra Trees, Gradient Boosting, Ridge Regression, Least Absolute Shrinkage and Selection Operator, Decision Tree, and Convolutional Neural Network. The aim of this study is to investigate the success of different ML algorithms on real-world data of wind turbines and propose a methodology to benchmark various machine learning algorithms to choose the most accurate final model for wind power generation prediction.

Anahtar Kelimeler

• Machine learning,Wind power generation prediction,Wind power forecasting,Renewable energy

Teşekkür

We would firstly like to acknowledge Türker Murat for his support in providing and processing of wind power generation data as well as Bıçakcılar Çandarlı Elektrik Üretim A.Ş. for the sharing of wind power generation data used in the development of this study. We would also like to thank Gülşah Murat for introducing us to this company and for her help in acquiring data used in this work.

Kaynakça

1. Dolara, A., Gandelli, A., Grimaccia, F., Leva, S., Mussetta, M. 2017. Weather-based Machine Learning Technique for Day-Ahead Wind Power Forecasting. IEEE 6th International Conference on Renewable Energy Research and Applications (ICRERA), 5-8 November, San Diego, CA, USA, 206-209. DOI: 10.1109/ICRERA.2017.8191267
2. Zhang, J., Jiang, X., Chen, X., Li, X., Guo, D., Cui, L. 2019. Wind Power Generation Prediction Based on LSTM. 4th International Conference on Mathematics and Artificial Intelligence, 12-15 April, Chengdu, China, 85-89. DOI: 10.1145/3325730.3325735
3. Zhang, J., Yan, J., Infield, D., Yongqian, L., Lien, F. 2019. Short-term Forecasting and Uncertainty Analysis of Wind Turbine Power Based on Long Short-term Memory Network and Gaussian Mixture Model, Applied Energy, Volume. 241, p. 229-244. DOI: 10.1016/j.apenergy.2019.03.044
4. Cali, U., Sharma, V. 2019. Short-term Wind Power Forecasting Using Long-short Term Memory Based Recurrent Neural Network Model and Variable Selection, International Journal of Smart Grid and Clean Energy, Volume. 8, p. 103-110. DOI: 10.12720/sgce.8.2.103-110
5. Li, L., Zhao, X., Tseng, M., Tan, R. 2019. Short-term Wind Power Forecasting Based on Support Vector Machine with Improved Dragonfly Algorithm, Journal of Cleaner Production, Volume. 242: 118447. DOI: 10.1016/j.jclepro.2019.118447
6. Okumuş, İ., Dinler, A. 2016. Current Status of Wind Energy Forecasting and a Hybrid Method for Hourly Predictions, Energy Conversion and Management, Volume. 123, p. 362-371. DOI: 10.1016/j.enconman.2016.06.053
7. Hong, Y., Rioflorido, C.L.P. 2019. A Hybrid Deep Learning-based Neural Network for 24-h Ahead Wind Power Forecasting, Applied Energy, Volume. 250, p. 530-539. DOI: 10.1016/j.apenergy.2019.05.044
8. Ma, Y., Zhai, M. 2019. A Dual-Step Integrated Machine Learning Model for 24h-Ahead Wind Energy Generation Prediction Based on Actual Measurement Data and Environmental Factors, Applied Sciences, Volume. 9, p. 2125. DOI: 10.3390/app9102125

9. Salcedo-Sanz, S., Cornejo-Bueno, L., Prieto, L., Paredes, D., García-Herrera, R. 2018. Feature Selection in Machine Learning Prediction Systems for Renewable Energy Applications, Renewable and Sustainable Energy Reviews, Volume. 90, p. 728-741. DOI: 10.1016/j.rser.2018.04.008
10. Zhang, L., Wang, K., Lin, W., Geng, T., Lei, Z., Wang, Z. 2019. Wind Power Prediction Based On Improved Genetic Algorithm and Support Vector Machine, IOP Conference Series Earth and Environmental Science, Volume. 252:032052. DOI: 10.1088/1755-1315/252/3/032052
11. Demolli, H., Dokuz, A.S., Ecemiş, A., Gokcek, M. 2019. Wind Power Forecasting Based on Daily Wind Speed Data Using Machine Learning Algorithms, Energy Conversion and Management, Volume. 198: 111823. DOI: 10.1016/j.enconman.2019.111823
12. Kramer, O., Gieseke, F., Satzger, B. 2012. Wind Energy Prediction and Monitoring with Neural Computation, Neurocomputing, Volume. 109, p. 84-93. DOI: 10.1016/j.neucom.2012.07.029
13. Marvuglia, A., Messineo, A. 2012. Monitoring of Wind Farms' Power Curves Using Machine Learning Techniques, Applied Energy, Volume. 98, p. 574-583. DOI: 10.1016/j.apenergy.2012.04.037
14. Wang, H., Lei, Z., Zhang, X., Zhou, B., Peng, J. 2019. A Review of Deep Learning for Renewable Energy Forecasting, Energy Conversion and Management, Volume. 198:111799. DOI: 10.1016/j.enconman.2019.111799
15. Zhang, Y., Cao, G., Wang, B., Li, X. 2019. A Novel Ensemble Method for K-Nearest Neighbor, Pattern Recognition, Volume. 85, p. 13-25. DOI: 10.1016/j.patcog.2018.08.003
16. Vapnik, V., Golowich, S.E., Smola, A. 1997. Support Vector Method for Function Approximation, Regression Estimation, and Signal Processing. 9th International Conference on Neural Information Processing Systems, 2-5 December, Denver, CO, USA, 281-287.
17. Basak, D., Pal S., Patranabis, D.C. 2007. Support Vector Regression, Neural Information Processing Letters and Reviews, Volume. 11(10), p. 203-224.
18. Breiman, L. 2001. Random Forests, Machine Learning, Volume. 45, p. 5-32. DOI: 10.1023/A:1010950718922
19. Geurts, P., Ernst, D., Wehenkel, L. 2006. Extremely Randomized Trees, Machine Learning, Volume. 63, p. 3-42. DOI: 10.1007/s10994-006-6226-1
20. Zhang, C., Zhang, Y., Shi, X., Almpanidis, G., Fan, G., Shen, X. 2019. On Incremental Learning for Gradient Boosting Decision Trees, Neural Processing Letters, Volume. 50, Issue. 1, p. 957-987. DOI: 10.1007/s11063-019-09999-3
21. Ohishi, M., Yanagihara, H., Fujikoshi, Y. 2020. A Fast Algorithm for Optimizing Ridge Parameters in a Generalized Ridge Regression by Minimizing a Model Selection Criterion, Journal of Statistical Planning and Inference, Volume. 204, p. 187-205. DOI: 10.1016/j.jspi.2019.04.010
22. Kim, Y., Hao, J., Mallavarapu, T., Park, J., Kang, M. 2019. Hi-LASSO: High-Dimensional LASSO, IEEE Access, Volume. 7, p. 44562-44573. DOI: 10.1109/ACCESS.2019.2909071
23. Trabelsi, A., Elouedi, Z., Lefevre, E. 2019. Decision Tree Classifiers for Evidential Attribute Values and Class Labels, Fuzzy Sets and Systems, Volume. 366, p. 46-62. DOI: 10.1016/j.fss.2018.11.006
24. Patel, S. 2020. A Comprehensive Analysis of Convolutional Neural Network Models, International Journal of Advanced Science and Technology, Volume. 29, Issue. 4, p. 771-777.
25. Windfinder. https://www.windfinder.com (Access Date: 09.03.2020)