A Deep Learning Approach for Fault Detection in Photovoltaic Systems Using MobileNetV3

Abstract

This study investigates the use of the MobileNetV3 deep learning architecture for fault detection in Photovoltaic (PV) systems. The research developed a model capable of classifying solar panels under six different conditions: clean, physically damaged, electrically damaged, snow covered, bird droppings covered, and dusty panels. Using a dataset obtained from Kaggle, pre-processed and divided into training (70%) and test (30%) sets, the MobileNetV3 model achieved a validation accuracy of 95%. Confusion matrix analysis showed high classification accuracy, in particular 100% accuracy for snow-covered and bird droppings-covered panels, with F1 scores as high as 98.73% for certain classes. Training and validation curves confirmed stable learning with low loss values. Compared to models such as EfficientB0 + SVM and InceptionV3-Net + U-Net, MobileNetV3 demonstrated competitive accuracy and computational efficiency, making it suitable for resource-constrained devices. This approach improves energy efficiency, reduces manual inspection, and promotes sustainable energy production. Future work will expand the dataset to include different climatic conditions and fault scenarios, improving the robustness and real-world applicability of the model.

Keywords

• MobileNetV3
• Photovoltaic Systems
• Fault Detection
• Deep Learning

References

1. Bishop, C. M. (2007). Pattern recognition and machine learning (Information Science and Statistics). Springer.
2. Boubaker, S., Kamel, S., Ghazouani, N., & Mellit, A. (2023). Assessment of machine and deep learning approaches for fault diagnosis in photovoltaic systems using infrared thermography. Remote Sensing, 15(6), 1686. https://doi.org/10.3390/rs15061686
3. Cao, H., Zhang, D., & Yi, S. (2023). Real-time machine learning-based fault detection, classification, and locating in large scale solar energy-based systems: Digital twin simulation. Solar Energy, 251, 77–85. https://doi.org/10.1016/j.solener.2022.12.042
4. Eurostat. (2022). Wind and water provide most renewable electricity; solar is the fastest-growing energy source. URL
5. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. The MIT Press.
6. Hong, Y.Y., & Pula, R. A. (2022). Methods of photovoltaic fault detection and classification: A review. Energy Reports, 8, 5898–5929. https://doi.org/10.1016/j.egyr.2022.04.043
7. Howard, A., Sandler, M., Chu, G., Chen, L.-C., Chen, B., Tan, M., Wang, W., Zhu, Y., Pang, R., Vasudevan, V., Le, Q. V., & Adam, H. (2019). Searching for MobileNetV3. In 2019 IEEE/CVF International Conference on Computer Vision (ICCV) (pp. 1314–1324). https://doi.org/10.48550/arXiv.1905.02244
8. Huang, J. M., Wai, R. J., & Gao, W. (2019). Newly-designed fault diagnostic method for solar photovoltaic generation system based on IV-Curve measurement. IEEE Access, 7, 70919–70932. https://doi.org/10.1109/ACCESS.2019.2919337

9. Johnson, J., Montoya, M., McCalmont, S., Katzir, G., Fuks, F., Earle, J., Fresquez, A., Gonzalez, S., & Granata, J. (2012). Differentiating series and parallel photovoltaic arc-faults. In 2012 38th IEEE Photovoltaic Specialists Conference (pp. 720–726). https://doi.org/10.1109/PVSC.2012.6317708
10. Kaggle. (n.d.). Fault detection using ResNet50. Retrieved from URL
11. Kumar, U., Mishra, S., & Dash, K. (2023). An IoT and semi-supervised learning-based sensorless technique for panel level solar photovoltaic array fault diagnosis. IEEE Transactions on Instrumentation and Measurement, 72, 1–12. https://doi.org/10.1109/TIM.2023.3287247
12. Mustafa, Z., Awad, A. S. A., Azzouz, M., & Azab, A. (2023). Fault identification for photovoltaic systems using a multi-output deep learning approach. Expert Systems with Applications, 211, 118551. https://doi.org/10.1016/j.eswa.2022.118551
13. Rudro, R. A. M., Nur, K., Sohan, M. F. A. Al, Mridha, M. F., Alfarhood, S., Safran, M., & Kanagarathinam, K. (2024). SPF-Net: Solar panel fault detection using U-Net based deep learning image classification. Energy Reports, 12, 1580–1594. https://doi.org/10.1016/J.EGYR.2024.07.044
14. Sabbaghpur, M., & Hejazi, M. A. (2016). The comprehensive study of electrical faults in PV arrays. Journal of Electrical and Computer Engineering, 2016, 8712960. https://doi.org/10.1155/2016/8712960
15. Sepúlveda-Oviedo, E. H., Travé-Massuyès, L., Subias, A., Pavlov, M., & Alonso, C. (2023). Fault diagnosis of photovoltaic systems using artificial intelligence: A bibliometric approach. Heliyon, 9(11), e21491. https://doi.org/10.1016/J.HELIYON.2023.E21491
16. Voutsinas, S., Mandourarakis, I., Koutroulis, E., Karolidis, D., Voyiatzis, I., & Samarakou, M. (2022). Control and communication for smart photovoltaic arrays. In 26th Pan-Hellenic Conference on Informatics (PCI 2022) (pp. 6). https://doi.org/10.1145/3575879.3575983
17. Voutsinas, S., Karolidis, D., Voyiatzis, I., & Samarakou, M. (2023). Development of a machine-learning-based method for early fault detection in photovoltaic systems. Journal of Engineering and Applied Sciences, 70(27). https://doi.org/10.1186/s44147-023-00200-0
18. Zhang, D., & Duranay, Z. B. (2023). Fault Detection in Solar Energy Systems: A Deep Learning Approach. Electronics 2023, Vol. 12, Page 4397, 12(21), 4397. https://doi.org/10.3390/ELECTRONICS12214397