Abstract

Solar energy is one of the most vital energy sources for human life. Although the production and installation of photovoltaic panels is simpler and more economical compared to other energy generation technologies, it can lead to significant cost increases if failures are not addressed in a timely manner. Therefore, early detection of faults and rapid response play a critical role in maintaining system performance and minimizing potential risks. Therefore, in this study, deep learning-based fault detection is performed using an unmanned aerial vehicle (UAV) with a thermal camera. YOLOV7 (You Look Only Once) algorithm, one of the current deep learning algorithms published in 2022, was used for automatic fault detection. In the study, the images taken with the thermal camera were processed and systematically analyzed with the YOLO algorithm model used in the training phase. In the training phase of the algorithm, a data set consisting of a total of 325 thermal images was used. The training consisted of 120 epochs and was completed in approximately 19 hours. As a result of the tests, it was observed that the developed system was able to detect faults with an accuracy of 0.91. This approach is considered as an important innovation in industrial maintenance processes.

References

[1] Hıguckı Y., Babasaki T., Failure detection of solar panels using thermographic images captured by drone, 2018 7th International Conference on Renewable Energy Research and Applications (ICRERA)

[2] Kaycı B., İHA Tarafından Elde Edilen Termal Görüntüler Kullanılarak Fotovoltaik Sistemde Derin Öğrenme Tabanlı Arıza Tespiti ve Teşhisi, 2021

[3] https://tr.trotec.com/uygulama-alanlari/termografi/ruezgar-tuerbinlerinde-termografi/ Erişim:16.09.2024

[4] Pierdicca, R., Malinverni, E. S., Piccinini, F., Paolanti, M., Felicetti, A., and Zingaretti, P.: Deep Convolutional Neural Network For Automatic Detection Of Damaged Photovoltaic Cells, Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-2, 893–900

[5] Wei S., Li X., Ding S., Yang Q., Yan W., Hotspots Infrared detection of photovoltaic modules based on Hough line transformation and Faster-RCNN approach, 2019 6th International Conference on Control, Decision and Information Technologies (CoDIT)

[6] Akram M. W., Li Guiqiang, Jin Y., Chen, X., Zhu C., and Ahmad A., "Automatic detection of photovoltaic module defects in infrared images with isolated and develop model transfer deep learning", Solar Energy, 198: 175– 186, (2020)

[7] Naveen Venkatesh S., and Sugumaran V., "Fault detection in aerial images of photovoltaic modules based on deep learning", IOP Conference Series: Materials Science and Engineering, 1012: 1–9, (2021)

[8] Huerta Herraiz Á., Pliego Marugán A., and García Márquez F. P., "photovoltaic plant condition monitoring using thermal images analysis by convolutional neural network-based structure", Renewable Energy, 153: 334– 348, (2020)

[9] Carletti V., Greco A., Saggese A., and Vento M., "An intelligent flying system for automatic detection of faults in photovoltaic plants", J. Ambient Intell. Humaniz. Comput., 11: 2027–2040, (2020)

[10] Altaee. F. (2018). A Study On The İmprovement Of Solar Pv Panel Efficiency And Power Output, Yayımlanmış Yüksek Lisans Tezi, Gaziantep Üniversitesi, Gaziantep.

[11] Kawamoto, H. ve Shibata, T. (2015). Electrostatic Cleaning System Forr Emoval Of Sand From Solar Panels, Journal of Electrostatics, 73, 65-70.

[12] Saidan, M., Albaali, A. G., Alasis, E. ve Kaldellis, J. K. (2016). Experimental Study On The Effect Of Dust Deposition On Solar Photovoltaic Panels In Desert Environment, Renewable Energy, 92, 499-5055.

[13] Mejia, F., Kleissl, J., &Bosch, J. L. (2014). The Effect Of Dust On Solar PhotovoltaicSystems. EnergyProcedia, 49, 2370-2376.

[14] Xie X., Wei X., Wang X., Guo X., Li J., and Cheng Z., "Photovoltaic panel anomaly detection system based on Unmanned Aerial Vehicle platform", IOP Conference Series: Materials Science and Engineering, 768: 1–7, (2020)

[15] Diaz J. J. V., Vlaminck M., Lefkaditis D., Vargas S. A. O., and Luong, H., "Solar panel detection within complex backgrounds using thermal images acquired by UAVs", Sensors, 20: 1–16, (2020).

[16] Koç Y., Hasarli Güneş Paneli Hücrelerinin Derin Öğrenme Yöntemleriyle Siniflandirilmasi, Amasya Üniversitesi, Fen Bilimleri Enstitüsü, Temmuz 2022

[17] Güvenç S., Derin Öğrenme İle Güneş Panellerindeki Kirliliğin Siniflandirilmasi, Amasya Üniversitesi Fen Bilimleri Enstitüsü, Ağustos 2022

[18] Bakır H., Çatı Tipi Güneş Enerji Sisteminde Kızılötesi Termal Teşhis ile Arıza Tespiti ve Önemi, OKU Journal of The Institute of Science and Technology, 6(1): 90-105, 2023

[19] Kıymaz Y.E., Güneş Enerjisi Santrallerinde Derin Öğrenme Kullanilarak Elektrik Üretimi Tahmininin Yapilmasi, Enerji Sistemleri Mühendisliği Anabilim Dalı, Aralık-2020

[20] Ersan S., YOLOv7 algoritması ile diş röntgenlerinde lezyon ve çürük tespiti, Dicle Üniversitesi, Fen Bilimleri Enstitüsü, Biyomedikal Mühendisliği Ana Bilim Dalı, 2024

[21] OGUNBOYE O.E., YOLOv7 kullanarak bir yol trafik algılamasını incelemek, Üsküdar Üniversitesi, Fen Bilimleri Enstitüsü, Elektrik-Elektronik Mühendisliği Ana Bilim Dalı, 2023

[22] Arı A., Karadağ B., Akıllı Mobil Cihazlarda YOLOv7 Modeli ile Nesne Tespiti, Yıl 2023, Cilt: 26 Sayı: 3, 1207 – 1214

[23] Yanılmaz S., Türkoğlu M., Aslan M., Güneş Enerjisi Santrallerinde YOLO Algoritmaları ile Hotspot Kusurlarının Tespiti, Yıl 2024, Cilt: 36 Sayı: 1, 121 - 132

[24] https://www.direnc.net/nvidia-jetson-nano-hazir-yapay-zeka-kiti-jnx30d?language=tr&h=78bd4f64&gad_source=4&gclid=CjwKCAjw59q2BhBOEiwAKc0ijRUThVaoCvf0IRCIUDUe8sKA9H2Iuv2xe9kGYEF5fvlWNmJM91A9choCfEYQAvD_BwE Erişim : 03.09.2024

[25] https://www.tiklasepete.com/uni-t-unit-uti721m-akilli-telefon-termal-goruntuleyici-kamera-19611?gad_source=1&gclid=CjwKCAjw_4S3BhAAEiwA_64YhgrrIMGIMHotayxuB3OJOKER59V89MwFfAtikZM1Wf4VKuA5OellWBoCehEQAvD_BwE Erişim : 03.09.2024

[26] https://www.djistoreturkiye.com/dji-mavic-3-thermal-dji-m3t?srsltid=AfmBOoqIB7dh-vwYibmsmZBCCgswuYfQYIkyJOPU5BEdgl9rQFa9c53B

[27] https://medium.com/machine-learning-t%C3%BCrkiye/ad%C4%B1m-ad%C4%B1m-makine-%C3%B6%C4%9Frenmesi-b%C3%B6l%C3%BCm-4-denetimli-%C3%B6%C4%9Frenme-ve-denetimsiz-%C3%B6%C4%9Frenme-aras%C4%B1ndaki-fark-4aa174983380 Erişim : 03.09.2024

[28] Gökalp Ö.M., Makine Öğrenmesi, * Gazi Üniversitesi, Gazi Bilişim Enstitüsü, 2022

[29] https://towardsdatascience.com/reinforcement-learning-fda8ff535bb6 Erişim : 03.09.2024

[30] Ülker E., İnik Ö., Derin Öğrenme ve Görüntü Analizinde Kullanılan Derin Öğrenme Modelleri, Year 2017 Volume: 6 Issue: 3

[31] https://novelis.io/news/yolov7-artificial-intelligence-for-real-time-object-detection-in-an-image/ Erişim : 03.09.2024

[32] Zhu W., Wang Q., Luo L., Zhang Y., Lu Q., Chang Yeh W., Liang J., CPAM: Cross Patch Attention Module for Complex Texture Tile Block Defect Detection, November 2022

[33] Orman E., Köse B., İzmir Bakırçay Üniversitesi Güneş Enerji Santralinin Elektrik Üretiminin Saatlik Tahmini, (ICADA’22) 2nd International Conference on Artificial Intelligence and Data Science, June 17-June 18, 2022