Wind Power Prediction Based on Polynomial Regression Method

Öz

Accurate prediction of wind energy output from site-specific wind velocity data is essential for evaluating the feasibility and energy yield of wind farm installations. This study presents a predictive framework for wind power assessment based on daily wind velocity datasets from eight Iraqi cities: Duhok, Mosul, Kirkuk, Baghdad, Najaf, Wasit, Qadisiyyah, and Basra. The proposed method employs polynomial regression (POR) as a nonlinear estimation technique to correlate daily wind speed data with turbine power output. POR is selected for its computational simplicity and adequacy in capturing the nonlinearity inherent in wind power curves. The predictive models are calibrated individually for each city using historical wind data and the manufacturer's specified power curve for the turbine. The results indicate that the POR-based predictions closely align with the actual turbine power curve across all sites, demonstrating low prediction error and strong curve-fitting behavior. Notably, Wasit, Qadisiyyah, and Basra exhibited the highest potential for wind energy generation, with annual predicted outputs exceeding 2000 kWh per turbine, indicating promising conditions for small-scale wind energy exploitation. In contrast, northern cities such as Duhok and Mosul yielded significantly lower outputs (<1000 kWh annually), suggesting limited economic viability under the studied configuration.

Anahtar Kelimeler

• Artificial intelligence
• Machine learning
• Wind power
• Wind velocity

Polinom Regresyon Yöntemine Dayalı Rüzgar Gücü Tahmini.

Öz

Rüzgar hızından elde edilen çıkış enerjisini tahmin etmek, potansiyel lokasyonlardaki rüzgar gücü karakteristiklerini araştırmak ve rüzgar çiftliklerine üretilen güç akılarının doğru bir şekilde değerlendirilmesi için gereklidir. Bu çalışma, öğrenme faktörü olarak rüzgar türbini parametrelerini ve Irak'taki 8 şehrin (Duhok, Musul, Kerkük, Bağdat, Necef, Vasit, Kadisiye ve Basra) günlük rüzgar hızı veri setini kullanarak rüzgar gücünü tahmin etmek ve değerlendirmek için önerilen bir yaklaşım sunmaktadır. Doğrusal olmayan tahmini modeller, rüzgar türbini modelinden (GhrepowerFD21-50_61.2kW_21.5) potansiyel rüzgar enerjisini çıkarmak için daha az karmaşık bir makine öğrenme yöntemi olarak polinom regresyon tekniği (POR) aracılığıyla kurulmuştur. Sonuçlara göre, tüm lokasyonlardaki rüzgar gücü tahmin değerlerinin dağılım noktaları, önerilen türbin modelinin güç eğrisi çıktısını yüksek doğruluk ve minimum hata ile takip edebilir. Ayrıca, Wasit, Qadisiyyah ve Basra'nın 3 şehri, her türbin ünitesi için 2000 Kilowatt'tan (kW) fazla olan en iyi toplam yıllık rüzgar gücü miktarlarına sahiptir ve 1000 kW'tan az olan Duhok ve Musul'u geride bırakmıştır. Ek olarak, POR tekniği, yıllık rüzgar enerjisini çıkarmada ve regresyon yöntemi için tahmin prosedürlerinde yararlıydı.

Anahtar Kelimeler

• Yapay zeka
• makine öğrenmesi
• rüzgar enerjisi
• rüzgar hızı.

Kaynakça

1. [1] Daily Loads. Retrieved March 10, 2023, from Ministry of Electricity: http://www.moelc.gov.iq
2. [2] Altai, Hisham Dawood Salman, Faisal Theyab Abed, Mohammed H. Lazim, and Haider TH Salim ALRikabi, “Analysis of the problems of electricity in Iraq and recommendations of methods of overcoming them, ” Periodicals of Engineering and Natural Sciences (PEN) 10, no. 1, 607-614, 2022.
3. [3] Bacci, A. (2017). Iraq Petroleum 2018—Natural Gas Must Be an Asset for Iraq.
4. [4] Mohammadi, Kasra, Omid Alavi, and Jon G. McGowan, “Use of Birnbaum-Saunders distribution for estimating wind speed and wind power probability distributions: A review,” Energy Conversion and Management, 143, 109-122, 2017.
5. [5] Song, Dongran, Yinggang Yang, Songyue Zheng, Xiaofei Deng, Jian Yang, Mei Su, Weiyi Tang, Xuebing Yang, Lingxiang Huang, and Young Hoon Joo. “New perspectives on maximum wind energy extraction of variable-speed wind turbines using previewed wind speeds,” Energy conversion and management, 206, 112496, 2020
6. [6] Wang, Jianzhou, Yuansheng Qian, Linyue Zhang, Kang Wang, and Haipeng Zhang, “A novel wind power forecasting system integrating time series refining, nonlinear multi-objective optimized deep learning and linear error correction,” Energy Conversion and Management, 299, 117818, 2024. https://doi.org/10.1016/j.enconman.2023.117818
7. [7] Alkesaiberi, Abdulelah, Fouzi Harrou, and Ying Sun, “Efficient wind power prediction using machine learning methods, A comparative study,” Energies 15, no. 7, 2327, 2022. https://doi.org/10.3390/en15072327
8. [8] Mi, Xiwei, Hui Liu, and Yanfei Li, “Wind speed prediction model using singular spectrum analysis, empirical mode decomposition and convolutional support vector machine,” Energy conversion and management 180, 196-205, 2019.

9. [9] Aldossary, Yasmeen, Nabil Hewahi, and Abdulla Alasaadi, “Wind Speed Forecasting Based on Data Decomposition and Deep Learning Models, A Case Study of a Wind Farm in Saudi Arabia,” arXiv preprint arXiv,2412.13356, 2024.
10. [10] Meng, Anbo, Shun Chen, Zuhong Ou, Weifeng Ding, Huaming Zhou, Jingmin Fan, and Hao Yin, “A hybrid deep learning architecture for wind power prediction based on bi-attention mechanism and crisscross optimization,” Energy 238, 121795, 2022. https://doi.org/10.1016/j.energy.2021.121795
11. [11] Wang, Jianing, Hongqiu Zhu, Yingjie Zhang, Fei Cheng, and Can Zhou, “A novel prediction model for wind power based on improved long short-term memory neural network,” Energy 265, 126283, 2023. https://doi.org/10.1016/j.energy.2022.126283
12. [12] Tarek, Zahraa, Mahmoud Y. Shams, Ahmed M. Elshewey, El-Sayed M. El-kenawy, Abdelhameed Ibrahim, Abdelaziz A. Abdelhamid, and Mohamed A. El-dosuky, “Wind Power Prediction Based on Machine Learning and Deep Learning Models,” Computers, Materials and Continua, 75, no. 1, 2023.
13. [13] Voyant, Cyril, Gilles Notton, Soteris Kalogirou, Marie-Laure Nivet, Christophe Paoli, Fabrice Motte, and Alexis Fouilloy, “Machine learning methods for solar radiation forecasting, A review,” Renewable energy 105, 569-582, 2017. https://doi.org/10.1016/j.renene.2016.12.095
14. [14] Al-Ansari, Nadhir, “Topography and climate of Iraq,” Journal of Earth Sciences and Geotechnical Engineering 11, no. 2, 1-13, 2021.
15. [15] Adeeb, H. Q., and Y. K. Al-Timimi, “GIS techniques for mapping of wind speed over Iraq,” Iraqi Journal of Agricultural Sciences 50, no. 6, 2019. https://doi.org/10.36103/ijas.v50i6.852
16. [16] Iraqi Meteorological Organization and Seismology (IMOS). http://www.meteoseism.gov.iq
17. [17] SWCC Small Wind Certification Council, “ICC-SWCC Summary Report SWCC-15-01-P,” Brea, CA,Small Wind Certification Council, 2014. http://smallwindcertification.org/wpcontent/uploads/2019/09/Summary-Report-SWCC-15- 01-2019-P.pdf.
18. [18] Ostertagova, Eva, “Modelling using polynomial regression,” Procedia engineering, 48, 500-506, 2012. https://doi.org/10.1016/j.proeng.2012.09.545
19. [19] Shi, Maolin, Weifei Hu, Muxi Li, Jian Zhang, Xueguan Song, and Wei Sun, “Ensemble regression based on polynomial regression-based decision tree and its application in the in-situ data of tunnel boring machine,” Mechanical Systems and Signal Processing, 188, 110022, 2023. https://doi.org/10.1016/j.ymssp.2022.110022
20. [20] Imran, Hamza, Nadia Moneem Al-Abdaly, Mohammed Hammodi Shamsa, Amjed Shatnawi, Majed Ibrahim, and Krzysztof Adam Ostrowski, “Development of prediction model to predict the compressive strength of eco-friendly concrete using multivariate polynomial regression combined with stepwise method,”Materials 15, no. 1, 317, 2022. https://doi.org/10.3390/ma15010317
21. [21] Alahmer, Ali, Hussein Alahmer, Ahmed Handam, and Hegazy Rezk, “Environmental assessment of a diesel engine fueled with various biodiesel blends, Polynomial regression and grey wolf optimization,” Sustainability, 14, no. 3, 1367, 2022. https://doi.org/10.3390/su14031367
22. [22] Saleh, Hussein H, “Estimating the amounts of the solar energy, A multiple linear regression approach,” In AIP Conference Proceedings, vol. 2591, no. 1. AIP Publishing, 2023. https://doi.org/10.1063/5.0125118