Araştırma Makalesi
BibTex RIS Kaynak Göster

HAVA GÖRÜNTÜLERİNİN DERİN ÖĞRENMESİ KULLANILARAK DAĞINIK FOTOVOLTAİK PİLLERİN OTOMATİK HARİTALANMASI VE SEGMENTASYONUHAVA GÖRÜNTÜLERİNİN DERİN ÖĞRENMESİ KULLANILARAK DAĞINIK FOTOVOLTAİK PİLLERİN OTOMATİK HARİTALANMASI VE SEGMENTASYONU

Yıl 2025, Cilt: 28 Sayı: 2, 835 - 850, 03.06.2025

Sinan Altun

Öz

Güneş enerjisi sistemleri arasında fotovoltaik piller aracılığıyla elektrik enerjisi üretimi, dünya genelinde yaygın bir eğilim haline gelmiştir. Güneş enerjisinin sınırsız bir kaynak olarak değerlendirilmesi ve geleneksel enerji santrallerinin yüksek sera gazı emisyonlarının fotovoltaik pillerle elektrik üretiminde bir engel teşkil etmemesi, küresel ısınmanın tehdit oluşturduğu günümüzde bu yöntemi oldukça cazip kılmaktadır. Bu çalışmada, Türkiye'nin önde gelen üç metropolü olan İstanbul, Ankara ve İzmir'deki dağıtık fotovoltaik sistemler, hava görüntüleri aracılığıyla incelenmiştir. İnceleme süreci, derin öğrenme teknikleri kullanılarak gerçekleştirilmiştir. Bilgilerimize göre, Türkiye'de bu alanda gerçekleştirilen ilk araştırmadır. Ülkede fotovoltaik sistemlere dair hava görüntülerini içeren bir veri seti bulunmadığı için, test veri setinin oluşturulmasında Google Earth platformu kullanılmıştır. Çalışmanın amacı, dünya genelinde ilgi gören hava fotoğraflarını kullanarak Türkiye'de güneş enerjisi sistemleri pazarının büyüme potansiyelini araştırmaktır. Elde edilen sınıflandırma ve segmentasyon sonuçları başarılı olup, dünya genelindeki benzer hava görüntüleri ile Türkiye için güneş enerjisi sistemleri pazar analizi yapılabileceğini ortaya koymaktadır. Sınıflandırma skorları: AlexNet AUC skoru 0.9, GoogLeNet 0.87 ve Inception için 0.83

Anahtar Kelimeler

Fotovoltaik Hava görüntüleri Derin öğrenme Görüntü segmentasyonu

Kaynakça

  • A Garai, S Biswas, S Mandal. A theoretical justification of warping generation for dewarping using CNN. Pattern Recognit., 109 (2021), https://doi.org/10.1016/j.patcog.2020.107621.
  • B Rausch, K Mayer, M-L Arlt, G Gust, P Staudt, C Weinhardt, et al. An enriched automated PV registry: combining image recognition and 3D building data. ArXiv (2020).
  • BB Kausika, D Nijmeijer, I Reimerink, P Brouwer, V. Liem. GeoAI for detection of solar photovoltaic installations in the Netherlands. Energy AI, 6 (2021), https://doi.org/10.1016/j.egyai.2021.100111.
  • C Wilson, A Grubler, N Bento, S Healey, Stercke S de, C Zimm. Granular technologies to accelerate decarbonization. Science (1979), 368 (2020), pp. 36-39.
  • D Stowell, J Kelly, D Tanner, J Taylor, E Jones, J Geddes, et al. A harmonised, high-coverage, open dataset of solar photovoltaic installations in the UK. Sci Data, 7 (2020), pp. 1-15, https://doi.org/10.1038/s41597-020-00739-0.
  • D Wang, M Zhao, Z li, S Xu, X Wu, X Ma, X Liu. A survey of unmanned aerial vehicles and deep learning in precision agriculture. EUR J AGRON, 164 (2025), https://doi.org/10.1016/j.eja.2024.127477.
  • Deep Learning by Ian Goodfellow, Yoshua Bengio and Aaron Courville published by MIT Press, 2016.
  • F Creutzig, P Agoston, JC Goldschmidt, G Luderer, G Nemet, RC. Pietzcker. The underestimated potential of solar energy to mitigate climate change. Nat Energy, 2 (2017), https://doi.org/10.1038/nenergy.2017.140.
  • G Kasmi, L Dubus, P Blanc, Y-M. Saint-Drenan. Towards unsupervised assessment with open-source data of the accuracy of deep learning-based distributed PV mapping Workshop on Machine Learning for Earth Observation (MACLEAN), in Conjunction with the ECML/PKDD 2022 (2022). p. hal-03778289.
  • Google Earth , Hava görüntüleri, https:// https://earth.google.com/web (Date of acces: 16/06/2023).
  • IEA PVPS task 1, Masson G, Kaizuka I, Bosch E, Plaza C, Scognamiglio A, et al. Trends in photovoltaic applications — 2022. 2022.
  • IEA PVPS, Fechner H, Johnston W, Neubourg G, Masson G, Ahm P, et al. Data model for PV systems — Data model and data acquisition for PV registration schemes and grid connection evaluations — Best practice and recommendations. 2020.
  • JM Malof, B Li, B Huang, K Bradbury, A. Stretslov. Mapping solar array location, size, and capacity using deep learning and overhead imagery. ArXiv (2019), abs/1902.1.
  • K Bradbury, R Saboo, TL Johnson, JM Malof, A Devarajan, W Zhang, et al. Distributed solar photovoltaic array location and extent dataset for remote sensing object identification. Sci Data, 3 (2016), pp. 1-9, https://doi.org/10.1038/sdata.2016.106.
  • K Mayer, B Rausch, ML Arlt, G Gust, Z Wang, D Neumann, et al. 3D-PV-Locator: Large-scale detection of rooftop-mounted photovoltaic systems in 3D. Appl Energy, 310 (2022), Article 118469, https://doi.org/10.1016/j.apenergy.2021.118469.
  • K Mayer, Z Wang, ML Arlt, D Neumann, R. Rajagopal. DeepSolar for Germany: a deep learning framework for PV system mapping from aerial imagery. Proceedings of the 3rd International Conference on Smart Energy Systems and Technologies (SEST) (2020), pp. 11-16, https://doi.org/10.1109/SEST48500.2020.9203258.
  • Kasmi G, Saint-Drenan Y-M, Trebosc D, el Jolivet R, Leloux J, Sarr B, et al. A crowdsourced dataset of aerial images with annotated solar photovoltaic arrays and installation metadata, 2022, p. 1–12.
  • L Kruitwagen, KT Story, J Friedrich, L Byers, S Skillman, C. Hepburn. A global inventory of photovoltaic solar energy generating units. Nature, 598 (2021), pp. 604-610, https://doi.org/10.1038/s41586-021-03957-7.
  • L Liu , B Lin , Y Yang. Moving scene object tracking method based on deep convolutional neural network. Alex. Eng. J., 86 (2024).
  • M Balcı, A Alkan. Identification of wart treatment evaluation by using optimum ensemble based classification techniques. Biomed. Signal Process. Control., 95 (2024).
  • M Hajabdollahi, R Esfandiarpoor, E Sabeti, N Karimi, SMR Soroushmehr, S Samavi. Multiple abnormality detection for automatic medical image diagnosis using bifurcated convolutional neural network. Biomed., 57 (2020), https://doi.org/10.1016/j.bspc.2019.101792.
  • M Jaxa-Rozen, E. Trutnevyte. Sources of uncertainty in long-term global scenarios of solar photovoltaic technology. Nat Clim Chang, 11 (2021), pp. 266-273, https://doi.org/10.1038/s41558-021-00998-8.
  • M Victoria, N Haegel, IM Peters, R Sinton, A Jäger-Waldau, C del Cañizo, et al. Solar photovoltaics is ready to power a sustainable future. Joule, 5 (2021), pp. 1041-1056, https://doi.org/10.1016/j.joule.2021.03.005.
  • Q Paletta, G Terrén-Serrano, Y Nie, B li, J Bieker, W Zhang, L Dubus, S Dev, C Feng. Advances in solar forecasting: Computer vision with deep learning. ADV APPL ENERGY, 11 (2023), https://doi.org/10.1016/j.adapen.2023.100150.
  • QGis, https://qgis.org/en/site/(Date of acces: 10/06/2023).
  • S Ren, W Hu, K Bradbury, D Harrison-Atlas, L Malaguzzi Valeri, B Murray, et al. Automated extraction of energy systems information from remotely sensed data: a review and analysis. Appl Energy, 326 (2022), Article 119876, https://doi.org/10.1016/j.apenergy.2022.119876.
  • SA Almalki, S Alsubai, A Alqahtani, AA, Alenazi. Denoised encoder-based residual U-net for precise teeth image segmentation and damage prediction on panoramic radiographs. J. Dent., 137 (2023), https://doi.org/10.1016/j.jdent.2023.104651.
  • TEİAŞ (2019), Elektrik İstatistikleri, Aylık Elektrik İstatistikleri, https://enerji.gov.tr/eigm-yenilenebilir-enerji-kaynaklar-gunes.
  • W Hu, K Bradbury, JM Malof, B Li, B Huang, A Streltsov, et al. What you get is not always what you see—pitfalls in solar array assessment using overhead imagery. Appl Energy, 327 (2022), Article 120143, https://doi.org/10.1016/j.apenergy.2022.120143.
  • Weiss W, Spörk-Dür M. Solar heat worldwide — 2021 Edition. 2021.
  • X Deng, Q Liu, Y Deng, S Mahadevan. An improved method to construct basic probability assignment based on the confusion matrix for classification problem. Inf. Sci., 340-341 (2016). https://doi.org/10.1016/j.ins.2016.01.033.
  • Y Yuan, Z Li, W Tu, Y Zhu. Computed tomography image segmentation of irregular cerebral hemorrhage lesions based on improved U-Net. J. Radiat. Res. Appl. Sci., 16-3 (2023), https://doi.org/10.1016/j.jrras.2023.100638.
  • Z Wang, M-L Arlt, C Zanocco, A Majumdar, R. Rajagopal. DeepSolar++: Understanding residential solar adoption trajectories with computer vision and technology diffusion models. Joule (2022), pp. 1-15, https://doi.org/10.1016/j.joule.2022.09.011.
  • Z Xia, Y Li, R Chen, D Sengupta, X Guo, B Xiong, et al. Mapping the rapid development of photovoltaic power stations in northwestern China using remote sensing. Energy Rep, 8 (2022), pp. 4117-4127, https://doi.org/10.1016/j.egyr.2022.03.039.

AUTOMATIC MAPPING AND SEGMENTATION OF SCATTERED PHOTOVOLTAIC BATTERIES USING DEEP LEARNING OF AERIAL IMAGERY

Yıl 2025, Cilt: 28 Sayı: 2, 835 - 850, 03.06.2025

Sinan Altun

Öz

Among solar energy systems, the generation of electrical energy through photovoltaic cells has become a widespread trend worldwide. The fact that solar energy is considered as an unlimited resource and the high greenhouse gas emissions of conventional power plants do not constitute an obstacle in electricity generation with photovoltaic cells makes this method very attractive today, when global warming is a threat. In this study, distributed photovoltaic systems in three leading metropolises of Turkey, Istanbul, Ankara, and Izmir, are analysed through aerial imagery. The investigation process was carried out using deep learning techniques. To the best of our knowledge, this is the first research in this field in Turkey. Since there is no dataset containing aerial images of photovoltaic systems in the country, Google Earth platform was used to create the test dataset. The aim of the study is to investigate the growth potential of the solar energy systems market in Turkey by using aerial photographs, which have attracted worldwide attention. The classification and segmentation results obtained are successful and reveal that solar energy systems market analysis can be made for Turkey with similar aerial images around the world. Classification scores: AUC value is 0.9 for AlexNet, 0.87 for GoogLeNet, and 0.83 for Inception.

Anahtar Kelimeler

Photovoltaic Aerial imagery Deep learning Image segmentation

Kaynakça

  • A Garai, S Biswas, S Mandal. A theoretical justification of warping generation for dewarping using CNN. Pattern Recognit., 109 (2021), https://doi.org/10.1016/j.patcog.2020.107621.
  • B Rausch, K Mayer, M-L Arlt, G Gust, P Staudt, C Weinhardt, et al. An enriched automated PV registry: combining image recognition and 3D building data. ArXiv (2020).
  • BB Kausika, D Nijmeijer, I Reimerink, P Brouwer, V. Liem. GeoAI for detection of solar photovoltaic installations in the Netherlands. Energy AI, 6 (2021), https://doi.org/10.1016/j.egyai.2021.100111.
  • C Wilson, A Grubler, N Bento, S Healey, Stercke S de, C Zimm. Granular technologies to accelerate decarbonization. Science (1979), 368 (2020), pp. 36-39.
  • D Stowell, J Kelly, D Tanner, J Taylor, E Jones, J Geddes, et al. A harmonised, high-coverage, open dataset of solar photovoltaic installations in the UK. Sci Data, 7 (2020), pp. 1-15, https://doi.org/10.1038/s41597-020-00739-0.
  • D Wang, M Zhao, Z li, S Xu, X Wu, X Ma, X Liu. A survey of unmanned aerial vehicles and deep learning in precision agriculture. EUR J AGRON, 164 (2025), https://doi.org/10.1016/j.eja.2024.127477.
  • Deep Learning by Ian Goodfellow, Yoshua Bengio and Aaron Courville published by MIT Press, 2016.
  • F Creutzig, P Agoston, JC Goldschmidt, G Luderer, G Nemet, RC. Pietzcker. The underestimated potential of solar energy to mitigate climate change. Nat Energy, 2 (2017), https://doi.org/10.1038/nenergy.2017.140.
  • G Kasmi, L Dubus, P Blanc, Y-M. Saint-Drenan. Towards unsupervised assessment with open-source data of the accuracy of deep learning-based distributed PV mapping Workshop on Machine Learning for Earth Observation (MACLEAN), in Conjunction with the ECML/PKDD 2022 (2022). p. hal-03778289.
  • Google Earth , Hava görüntüleri, https:// https://earth.google.com/web (Date of acces: 16/06/2023).
  • IEA PVPS task 1, Masson G, Kaizuka I, Bosch E, Plaza C, Scognamiglio A, et al. Trends in photovoltaic applications — 2022. 2022.
  • IEA PVPS, Fechner H, Johnston W, Neubourg G, Masson G, Ahm P, et al. Data model for PV systems — Data model and data acquisition for PV registration schemes and grid connection evaluations — Best practice and recommendations. 2020.
  • JM Malof, B Li, B Huang, K Bradbury, A. Stretslov. Mapping solar array location, size, and capacity using deep learning and overhead imagery. ArXiv (2019), abs/1902.1.
  • K Bradbury, R Saboo, TL Johnson, JM Malof, A Devarajan, W Zhang, et al. Distributed solar photovoltaic array location and extent dataset for remote sensing object identification. Sci Data, 3 (2016), pp. 1-9, https://doi.org/10.1038/sdata.2016.106.
  • K Mayer, B Rausch, ML Arlt, G Gust, Z Wang, D Neumann, et al. 3D-PV-Locator: Large-scale detection of rooftop-mounted photovoltaic systems in 3D. Appl Energy, 310 (2022), Article 118469, https://doi.org/10.1016/j.apenergy.2021.118469.
  • K Mayer, Z Wang, ML Arlt, D Neumann, R. Rajagopal. DeepSolar for Germany: a deep learning framework for PV system mapping from aerial imagery. Proceedings of the 3rd International Conference on Smart Energy Systems and Technologies (SEST) (2020), pp. 11-16, https://doi.org/10.1109/SEST48500.2020.9203258.
  • Kasmi G, Saint-Drenan Y-M, Trebosc D, el Jolivet R, Leloux J, Sarr B, et al. A crowdsourced dataset of aerial images with annotated solar photovoltaic arrays and installation metadata, 2022, p. 1–12.
  • L Kruitwagen, KT Story, J Friedrich, L Byers, S Skillman, C. Hepburn. A global inventory of photovoltaic solar energy generating units. Nature, 598 (2021), pp. 604-610, https://doi.org/10.1038/s41586-021-03957-7.
  • L Liu , B Lin , Y Yang. Moving scene object tracking method based on deep convolutional neural network. Alex. Eng. J., 86 (2024).
  • M Balcı, A Alkan. Identification of wart treatment evaluation by using optimum ensemble based classification techniques. Biomed. Signal Process. Control., 95 (2024).
  • M Hajabdollahi, R Esfandiarpoor, E Sabeti, N Karimi, SMR Soroushmehr, S Samavi. Multiple abnormality detection for automatic medical image diagnosis using bifurcated convolutional neural network. Biomed., 57 (2020), https://doi.org/10.1016/j.bspc.2019.101792.
  • M Jaxa-Rozen, E. Trutnevyte. Sources of uncertainty in long-term global scenarios of solar photovoltaic technology. Nat Clim Chang, 11 (2021), pp. 266-273, https://doi.org/10.1038/s41558-021-00998-8.
  • M Victoria, N Haegel, IM Peters, R Sinton, A Jäger-Waldau, C del Cañizo, et al. Solar photovoltaics is ready to power a sustainable future. Joule, 5 (2021), pp. 1041-1056, https://doi.org/10.1016/j.joule.2021.03.005.
  • Q Paletta, G Terrén-Serrano, Y Nie, B li, J Bieker, W Zhang, L Dubus, S Dev, C Feng. Advances in solar forecasting: Computer vision with deep learning. ADV APPL ENERGY, 11 (2023), https://doi.org/10.1016/j.adapen.2023.100150.
  • QGis, https://qgis.org/en/site/(Date of acces: 10/06/2023).
  • S Ren, W Hu, K Bradbury, D Harrison-Atlas, L Malaguzzi Valeri, B Murray, et al. Automated extraction of energy systems information from remotely sensed data: a review and analysis. Appl Energy, 326 (2022), Article 119876, https://doi.org/10.1016/j.apenergy.2022.119876.
  • SA Almalki, S Alsubai, A Alqahtani, AA, Alenazi. Denoised encoder-based residual U-net for precise teeth image segmentation and damage prediction on panoramic radiographs. J. Dent., 137 (2023), https://doi.org/10.1016/j.jdent.2023.104651.
  • TEİAŞ (2019), Elektrik İstatistikleri, Aylık Elektrik İstatistikleri, https://enerji.gov.tr/eigm-yenilenebilir-enerji-kaynaklar-gunes.
  • W Hu, K Bradbury, JM Malof, B Li, B Huang, A Streltsov, et al. What you get is not always what you see—pitfalls in solar array assessment using overhead imagery. Appl Energy, 327 (2022), Article 120143, https://doi.org/10.1016/j.apenergy.2022.120143.
  • Weiss W, Spörk-Dür M. Solar heat worldwide — 2021 Edition. 2021.
  • X Deng, Q Liu, Y Deng, S Mahadevan. An improved method to construct basic probability assignment based on the confusion matrix for classification problem. Inf. Sci., 340-341 (2016). https://doi.org/10.1016/j.ins.2016.01.033.
  • Y Yuan, Z Li, W Tu, Y Zhu. Computed tomography image segmentation of irregular cerebral hemorrhage lesions based on improved U-Net. J. Radiat. Res. Appl. Sci., 16-3 (2023), https://doi.org/10.1016/j.jrras.2023.100638.
  • Z Wang, M-L Arlt, C Zanocco, A Majumdar, R. Rajagopal. DeepSolar++: Understanding residential solar adoption trajectories with computer vision and technology diffusion models. Joule (2022), pp. 1-15, https://doi.org/10.1016/j.joule.2022.09.011.
  • Z Xia, Y Li, R Chen, D Sengupta, X Guo, B Xiong, et al. Mapping the rapid development of photovoltaic power stations in northwestern China using remote sensing. Energy Rep, 8 (2022), pp. 4117-4127, https://doi.org/10.1016/j.egyr.2022.03.039.
Toplam 34 adet kaynakça vardır.

Ayrıntılar

Birincil Dil Türkçe
Konular Görüntü İşleme, Derin Öğrenme, Fotovoltaik Güç Sistemleri
Bölüm Elektrik Elektronik Mühendisliği
Yazarlar

Sinan Altun 0000-0002-2356-0460

Yayımlanma Tarihi 3 Haziran 2025
Gönderilme Tarihi 10 Ocak 2025
Kabul Tarihi 22 Nisan 2025
Yayımlandığı Sayı Yıl 2025Cilt: 28 Sayı: 2

Kaynak Göster

APA Altun, S. (2025). HAVA GÖRÜNTÜLERİNİN DERİN ÖĞRENMESİ KULLANILARAK DAĞINIK FOTOVOLTAİK PİLLERİN OTOMATİK HARİTALANMASI VE SEGMENTASYONUHAVA GÖRÜNTÜLERİNİN DERİN ÖĞRENMESİ KULLANILARAK DAĞINIK FOTOVOLTAİK PİLLERİN OTOMATİK HARİTALANMASI VE SEGMENTASYONU. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, 28(2), 835-850.
 Kapak Resmi İndir