Mersin Bölgesi Rüzgar Hız Verilerinin Yapay Sinir Ağları ile Analizi ve Uygulanabilirliği

Yıl 2020, , 39 - 51, 01.06.2020

Mustafa Akkaya , Aytaç Gültekin , Kadir Sabancı , Selami Balcı , Hakan Sağlam

https://doi.org/10.17100/nevbiltek.691120

Cited By: 2

Öz

Bu çalışmada, Mersin-Mut bölgesinde
yenilenebilir enerji kaynakları ile bir evin ısıtma ve elektrik sistemlerine
enerji sağlanabilmesi için rüzgar hızı verileri analiz edilmiştir. Bölgedeki üç
yıllık rüzgar hızı verileri Türkiye Meteoroloji Genel Müdürlüğü’nden
alınmıştır. İncelenen bölge için 28 günlük rüzgar hızı verileri kullanılarak
yapay sinir ağları ile yıllık tahmin gerçekleştirilmiştir. Rüzgar verilerinin
bir kısmı yapay sinir ağının eğitimi için, bir kısmıda test işlemi için
kullanılmıştır. Yapay sinir ağı modelinde gizli katmandaki nöron sayıları
değiştirilerek en başarılı model elde edilmiştir. Gizli katmanda sekiz nöron
kullanılarak yapılan analizde, en düşük MAE ve RMSE hata değerleri
hesaplanmıştır. Nöron sayısı sekiz iken, MAE ve RMSE değerleri sırasıyla 0.4056
ve 0.5403 olarak elde edilmiştir. Ayrıca, bu bölge için rüzgar verilerinin WAsP
yazılımı ile analiz çalışmaları da gerçekleştirilmiştir. Böylece, analiz
çalışmalarına göre ortalama anlık rüzgar hızı belirlenmiştir.

Anahtar Kelimeler

Yenilenebilir Enerji Kaynakları, Rüzgar Enerjisi, YSA, RMSE, MAE

Teşekkür

Türkiye Cumhuriyeti Meteoroloji Genel Müdürlüğü

Analysis and Applicability of Mersin Region Wind Speed Data with Artificial Neural Networks

Öz

In this study, wind speed data were analyzed
in order to provide energy to the heating and electrical systems of a house
with renewable energy sources in Mersin-Mut region. Three-year wind speed data
is taken from the Turkey General Directorate of Meteorology in the region.
Annual estimation was made with artificial neural networks using 28-day wind
speed data for the studied area. Some of the wind data were used for training
of the neural network, and some were used for testing. In the artificial neural
network model, the most successful model was obtained by changing the number of
neurons in the hidden layer. In the analysis made using eight neurons in the
hidden layer, the lowest MAE and RMSE error values ​​were calculated. While the
number of neurons was eight, MAE and RMSE values ​​were obtained as 0.4056 and
0.5403, respectively. In addition, analysis of wind data with WAsP software has
been carried out for this region. Thus, the average instantaneous wind speed
was determined according to the analysis studies.

Anahtar Kelimeler

Renewable Energy Sources, wind energy, YSA, RMSE, MAE

Kaynakça

• [1] Lightning U., Gungor A., “Green Home and Applications in Turkey”, Tmmob chamber Of Mechanical Engineers 5th Solar Energy Systems Symposium and Exhibition Proceedings Book. Mersin. Ankara: Tmmob Chamber of Mechanical Engineers, (M / 2011/562). 66-67, 2011. [2] Akin M. and Balci S., ”The Electromagnetic Modeling and the Co-Simulation of a Direct Drive Axial Flux Permanent Magnet Synchronous Generator”, Journal of Energy Systems, 2020 4(2), DOI: 10.30521/jes.690997, 2020. [3] Yılmaz Ş., “Natıonal Renewable Energy Action Plan.” MMO, 2018. [4] “Republic of Turkey Ministry of Energy and Natural Resource” URL 1: http://www.enerji.gov.tr/En-Us/mainpage, Last Access: 09.11.2019 [5] “Energy Efficiency and Environment Department” URL 2: http://www.yegm.gov.tr/mycalculator/pages/33.Aspx, Last Access: 09.11.2019 [6] Ozay C. and Celiktas M.S., "Statistical Analysis of Wind Speed Using Two-Parameter Weibull Distribution in Alaçatı Region", Energy Conversion and Management , 121:49-54, 2016 [7] Lange B. and Højstrup J., "Evaluation of the Wind-Resource Estimation Program WasP for Offshore Applications", Journal of Wind Engineering and Industrial Aerodynamics, 89.3-4: 271-291, 2001. [8] Pop L., Zbyněk S. and David H., "A New Method for Estimating Maximum Wind Gust Speed with a Given Return Period and a High Areal Resolution", Journal of Wind Engineering and Industrial Aerodynamics, 158:51-60, 2016. [9] Katinas V., Giedrius G. and Mantas M., (2018)."An Investigation of Wind Power Density Distribution at Location with Low and High Wind Speeds Using Statistical Model", Applied Energy, 218:442-451. [10] Đurišić Ž. and Jovan M., "A Model for Vertical Wind Speed Data Extrapolation for Improving Wind Resource Assessment Using WaSP", Renewable Energy, 41: 407-411, 2012. [11] Baseer Mohammed A. et al., "Wind Power Characteristics of Seven Data Collection Sites in Jubail. Saudi Arabia Using Weibull Parameters", Renewable Energy, 102: 35-49, 2017. [12] Shahsavari A. and Morteza A.,"Potential of Solar Energy in Developing Countries for Reducing Energy-Related Emissions", Renewable and Sustainable Energy Reviews, 90: 275-291, 2018. [13] Elsheikh Ammar H. et al., "Modeling of Solar Energy Systems Using Artificial Neural Network: A Comprehensive Review", Solar Energy, 180: 622-639, 2019. [14] Ali A, Rodríguez S. and Sailor D.,"Transforming a Passive House a Net-Zero Energy House: A Case Study in The Pacific Northwest of the Us", Energy Conversion and Management, 172: 39-49, 2018. [15] Sabancı K., Balcı S. and Aslan M.F.,“Estimation of the Switching Losses in Dc-Dc Boost Converters by Various Machine Learning Methods”, Journal of Energy Systems, 4(1). 1-11. Doı: 10.30521/Jes.635582, 2020 [16] Balci S. and Helvaci O., “A Comparative Simulation on the Grounding Grid System of a Wind Turbine with FEA Software”, Journal of Energy Systems, 3(4), 148-157, DOI: 10.30521/jes.613724, 2019. [17] Republic of Turkey General Directorate of Meteorology, Ankara (Turkey). [18] Ataseven B., "Yapay Sinir Ağları ile Öngörü Modellemesi", Öneri Dergisi, 10.39: 101-115, 2013 [19] Bomin K. et al.,"Effect of Surfactant on Wetting Due to Fouling in Membrane Distillation Membrane: Application of Response Surface Methodology (RSM) and Artificial Neural Networks (ANN)", Korean Journal of Chemical Engineering, 37.1:1-10, 2020. [20] “Republic of Turkey Ministry of Agriculture and Forestry General Directorate of Meteorology” URL 3: https://www.mgm.gov.tr/files/genel/sss/ruzgaratlasi.pdf, Last Access: 09.10.2019 [21] Bowen A. J.and Niels G. Mortensen.,"Exploring the Limits Ff WAsP: The Wind Atlas Analysis and Application Program", Proceedings of the 1996 European Union Wind Energy Conference, Göteborg. Sweden, 1996. [22] Hande K. and Ercan E.,“Forecasting Study on the Comparative Performance of Back Propagation Neural Network Algorithms”, Animal Production, 56(1): 22-27, 2015. [23] Maroufpoor S., Ahmad Fakheri-Fard and Jalal S., "Study of the Spatial Distribution of Groundwater Quality Using Soft Computing and Geostatistical Models", Journal of Hydraulic Engineering, 25.2: 232-238, 2019. [24] Semih G., “Rüzgar Enerjisi Potansiyel Hesaplamasında Kullanılan Bilgisayar Programlarının Karşılaştırılması”, Diss. Enerji Enstitüsü, 2014. [25] Dinçer F., Rüstemli S., Yılmaz Ş.and Çıngı A., “Kilis İli için Farklı Yüksekliklerdeki Rüzgâr Potansiyelinin Belirlenmesi”, Bitlis Eren Üniversitesi Fen Bilimleri Dergisi, 6(1). 12-20, 2017. [26] Vardar A., “Wind Turbine Types and Determination of Energy to Be Obtained From Wind” URL 4: http://slideplayer.biz.tr/slide/2335624/., (Last Access: 11.11.2019), 2013.