Home / Articles / 2026-yil, 3 / Article
GEOLOGICAL AND MINERALOGICAL SCIENCES

DEEP LEARNING-BASED AUTOMATIC DETECTION OF SOILING AND FAULTS ON PHOTOVOLTAIC PANELS

Shakhzod Bobokulov Shakhzod B.

Abstract

Dust accumulation on the surfaces of photovoltaic (PV) panels, as well as mechanical and electrical defects, are among the main factors contributing to reduced energy generation at solar power plants. This problem is particularly pronounced in arid and dusty regions such as Central Asia. Manual inspection of large PV arrays is labour-intensive, time-consuming and prone to errors. This paper proposes a lightweight and explainable deep learning system for automatically classifying PV panel surface conditions into six classes: clean, dusty, contaminated with bird droppings, electrically damaged, mechanically damaged and snow-covered. Transfer learning is applied using the MobileNetV3-Large neural network pretrained on the ImageNet dataset. To mitigate class imbalance, a class-weighted loss function and data augmentation techniques are employed. The Grad-CAM method is integrated to localise damaged regions. For each detected condition, the system estimates a potential power loss range based on scientific literature and presents the results to users through a FastAPI web application with an Uzbek-language interface.

Keywords

photovoltaic panel dust accumulation defect classification deep learning transfer learning MobileNetV3 Grad-CAM explainable artificial intelligence power loss.
Authors
Shakhzod Bobokulov
Karshi State University, Karshi
ORCID: 0000-0002-2180-2278

How To Cite

Journal Style
Shakhzod, B. DEEP LEARNING-BASED AUTOMATIC DETECTION OF SOILING AND FAULTS ON PHOTOVOLTAIC PANELS. Innovatsion texnologiyalar, 2026, 63(3), 115-123. https://doi.org/10.70769/2181-4732.ITJ.2026-3.12
TXT (current style) BibTeX RIS

References

  1. IRENA, “Renewable Capacity Statistics 2025,” Abu Dhabi, 2025.
  2. T. Sarver, A. Al- Qaraghuli, and L. L. Kazmerski, “A comprehensive review of the impact of dust on the use of solar energy: History, investigations, results, literature, and mitigation approaches,” Renewable and Sustainable Energy Reviews, vol. 22, pp. 698 – 733, 2013, doi: 10.1016/j.rser.2012.12.065.
  3. A. Sayyah, M. N. Horenstein, and M. K. Mazumder, “Energy yield loss caused by dust deposition on photovoltaic panels,” Solar Energy, vol. 107, pp. 576 – 604, 2014, doi: 10.1016/j.solener.2014.05.030.
  4. M. R. Maghami, H. Hizam, C. Gomes, M. A. Radzi, M. I. Rezadad, and S. Hajighorbani, “Power loss due to soiling on solar panel: A review,” Renewable and Sustainable Energy Reviews, vol. 59, pp. 1307 – 1316, 2016, doi: 10.1016/j.rser.2016.01.044.
  5. K. Ilse et al., “Techno -economic assessment of soiling losses and mitigation strategies for solar power generation,” Joule, vol. 3, no. 10, pp. 2303 – 2321, 2019, doi: 10.1016/j.joule.2019.08.019.
  6. D. S. Pillai and N. Rajasekar, “A comprehensive review on protection challenges and fault diagnosis in PV systems,” Renewable and Sustainable Energy Reviews, vol. 91, pp. 18 – 40, 2018, doi: 10.1016/j.rser.2018.03.082.
  7. A. Mellit and S. A. Kalogirou, “Artificial intelligence and internet of things to improve efficacy of diagnosis and remote sensing of solar photovoltaic systems,” Renewable and Sustainable Energy Reviews, vol. 143, p. 110889, 2021, doi: 10.1016/j.rser.2021.110889.
  8. Y. Zefri, A. ElKettani, I. Sebari, and S. Ait Lamallam, “Thermal infrared and visual inspection of photovoltaic installations by UAV photogrammetry: Application case Morocco,” Remote Sens. (Basel)., vol. 10, no. 12, p. 1874, 2018, doi: 10.3390/rs10121874.
  9. S. Mehta, A. P. Azad, S. A. Chemmengath, V. Raykar, and S. Kalyanaraman, “DeepSolarEye: Power loss prediction and weakly supervised soiling localization via fully convolutional networks for INNOVATSION TEXNOLOGIYALAR INNOVATIVE TECHNOLOGIES ИННОВАЦИОННЫЕ ТЕХ НОЛОГИИ 2026 - yil 3 (63) - son 2026 volume 63, number 3 Том 63 No 3, 202 6 ISSN 2181 - 4732 E - ISSN 2181 - 4 015 123 solar panels,” in Proceedings of the IEEE Winter Conference on Applications of Computer Vision (WACV), 2018, pp. 333 – 342. doi: 10.1109/WACV.2018.00043.
  10. M. Yang, J. Ji, and B. Guo, “Soiling quantification using an image -based method: Effects of imaging conditions,” IEEE J. Photovolt., vol. 10, no. 6, pp. 1780 – 1787, 2020, doi: 10.1109/JPHOTOV.2020.3018493.
  11. S. Fan, Y. Wang, S. Cao, B. Zhao, T. Sun, and P. Liu, “A deep residual neural network identification method for uneven dust accumulation on photovoltaic (PV) panels,” Energy, vol. 239, p. 122302, 2022, doi: 10.1016/j.energy.2021.122302.
  12. A. Et- Taleby, Y. Chaibi, M. Boussetta, and M. Benslimane, “Application of deep learning based detector YOLOv5 for soiling recognition in photovoltaic modules,” in Proceedings of the 3rd International Conference on Innovative Research in Applied Science, Engineering and Technology (IRASET), Mohammedia, Morocco, 2023, pp. 1 – 6. doi: 10.1109/IRASET57153.2023.10152990.
  13. M. W. Akram, G. Li, Y. Jin, X. Chen, C. Zhu, and A. Ahmad, “Automatic dete ction of photovoltaic module defects in infrared images with isolated and develop-model transfer deep learning,” Solar Energy, vol. 198, pp. 175 – 186, 2020, doi: 10.1016/j.solener.2020.01.055.
  14. Á. H. Herraiz, A. P. Marugán, and F. P. G. Márquez, “Photov oltaic plant condition monitoring using thermal images analysis by convolutional neural network- based structure,” Renew. Energy, vol. 153, pp. 334 – 348, 2020, doi: 10.1016/j.renene.2020.01.148.
  15. S. Deitsch et al., “Automatic classification of defective photovoltaic module cells in electroluminescence images,” Solar Energy, vol. 185, pp. 455 – 468, 2019, doi: 10.1016/j.solener.2019.02.067.
  16. J. Wang et al., “Deep -learning-based automatic detection of photovoltaic cell defects in electroluminescence image s,” Sensors, vol. 23, no. 1, p. 297, 2023, doi: 10.3390/s23010297.
  17. N. Ledmaoui, A. El Maghraoui, M. El Aroussi, and R. Saadane, “Enhanced fault detection in photovoltaic panels using CNN- based classification with PyQt5 implementation,” Sensors, vol. 24, no. 22, p. 7407, 2024, doi: 10.3390/s24227407.
  18. M. Buda, A. Maki, and M. A. Mazurowski, “A systematic study of the class imbalance problem in convolutional neural networks,” Neural Networks, vol. 106, pp. 249 – 259, 2018, doi: 10.1016/j.neunet.2018.07.011.
  19. Kaggle, “Solar Panel Images: Clean and Faulty.” [Online]. Available: https://www.kaggle.com/datasets
  20. J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, and L. Fei- Fei, “ImageNet: A large -scale hierarchical image database,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2009, pp. 248 – 255. doi: 10.1109/CVPR.2009.5206848.
  21. J. Yosinski, J. Clune, Y. Bengio, and H. Lipson, “How transferable are features in deep neural networks?,” in Advances in Neural Information Processing Systems (NeurIPS), 2014, pp. 3320 – 3328.
  22. M. Tan and Q. V Le, “EfficientNet: Rethinking model scaling for convolutional neural networks,” in Proceedings of the 36th International Conference on Machine Learning (ICML), PMLR, 2019, pp. 6105 – 6114.
  23. Y. Cui, M. Jia, T.- Y. Lin, Y. Song, and S. Belongie, “Class -balanced loss based on effective number of samples,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 9268 – 9277. doi: 10.1109/CVPR.2019.00949.
  24. R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, “Grad -CAM: Visual explanations from deep networks via gradient- based localization,” in Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2017, pp. 618 – 626. doi: 10.1109/ICCV.2017.74.
  25. T.- Y. Lin, P. Goyal, R. Girshick, K. He, and P. Dollár, “Focal loss for dense object detection,” in Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2017, pp. 2980 – 2988. doi: 10.1109/ICCV.2017.324.