https://doi.org/10.65770/BZZF2385
ABSTRACT
Solar panel faults can significantly reduce energy output and reliability in renewable energy systems. Traditional fault detection methods often rely on manual inspection or rule-based systems, which struggle with accuracy and scalability. This paper introduces a deep learning-based fault diagnosis system designed to automatically detect and classify solar panel anomalies using infrared (IR) images. The proposed StackNet architecture builds upon the ResNet-18 backbone, with targeted modifications to enhance feature extraction efficiency while maintaining computational lightness. The model begins with a large-kernel convolutional layer for effective low-level feature capture, followed by normalization, activation, and pooling to reduce spatial resolution early and accelerate training. StackNet strikes a balance between architectural depth and training efficiency, offering strong representational capacity with minimal increase in complexity. This makes it particularly suited for solar anomaly detection tasks where both performance and computational cost are critical. The model was trained and evaluated on IR images for both binary (fault and no-fault) and multi-class (six fault types) classification. Experimental results demonstrate high diagnostic performance on both binary and multi-classification tasks. These outcomes validate the model’s robustness across different imaging modalities and fault scenarios. This bridges the gap between research and field application, offering practical deployment potential for solar maintenance. In summary, this paper highlights the effectiveness of combining deep learning and attention mechanisms for reliable solar panel fault detection, contributing to more intelligent and automated renewable energy management.
References
- [1] B. Duranay, ‘Fault Detection in Solar Energy Systems: A Deep Learning Approach’, Electronics, vol. 12, no. 21, Art. no. 21, Jan. 2023, doi: 10.3390/electronics12214397.
- [2] K. H. Rabaia et al., ‘Environmental impacts of solar energy systems: A review’, Sci. Total Environ., vol. 754, p. 141989, Feb. 2021, doi: 10.1016/j.scitotenv.2020.141989.
- [3] H. Han, T. Rahim, and S. Y. Shin, ‘Detection of Faults in Solar Panels Using Deep Learning’, in 2021 International Conference on Electronics, Information, and Communication (ICEIC), Jan. 2021, pp. 1–4. doi: 10.1109/ICEIC51217.2021.9369744.
- [4] Al-Mashhadani et al., ‘Deep Learning Methods for Solar Fault Detection and Classification: A Review’, Inf. Sci. Lett., vol. 10, p. 323, May 2021, doi: 10.18576/isl/100213.
- [5] Jaybhaye, V. Sirvi, S. Srivastava, V. Loya, V. Gujarathi, and M. D. Jaybhaye, ‘Classification and Early Detection of Solar Panel Faults with Deep Neural Network Using Aerial and Electroluminescence Images’, J. Fail. Anal. Prev., vol. 24, no. 4, pp. 1746–1758, Aug. 2024, doi: 10.1007/s11668-024-01959-x.
- [6] Jaybhaye, O. Thakur, R. Yardi, V. Raut, and A. Raut, ‘Solar Panel Damage Detection and Localization of Thermal Images’, J. Fail. Anal. Prev., vol. 23, no. 5, pp. 1980–1990, Oct. 2023, doi: 10.1007/s11668-023-01747-z.
- [7] Waqar Akram et al., ‘Improved outdoor thermography and processing of infrared images for defect detection in PV modules’, Sol. Energy, vol. 190, pp. 549–560, Sep. 2019, doi: 10.1016/j.solener.2019.08.061.
- [8] Gallardo-Saavedra, L. Hernández-Callejo, M. del C. Alonso-García, J. Muñoz-Cruzado-Alba, and J. Ballestín-Fuertes, ‘Infrared Thermography for the Detection and Characterization of Photovoltaic Defects: Comparison between Illumination and Dark Conditions’, Sensors, vol. 20, no. 16, p. 4395, Aug. 2020, doi: 10.3390/s20164395.
- [9] K. Segbefia, N. Akhtar, and T. O. Sætre, ‘Defects and fault modes of field-aged photovoltaic modules in the Nordics’, Energy Rep., vol. 9, pp. 3104–3119, Dec. 2023,doi: 10.1016/j.egyr.2023.01.126.
- -Y. Hong and R. A. Pula, ‘Methods of photovoltaic fault detection and classification: A review’, Energy Rep., vol. 8, pp. 5898–5929, Nov. 2022, doi: 10.1016/j.egyr.2022.04.043.
- Bhatia, ‘10 Causes of Solar Panel Damage and How to Avoid Them’, Sustvest Blog. Accessed: Mar. 11, 2025. [Online]. Available: https://blog.sustvest.com/solar-panel-damage/
- Cubukcu and A. Akanalci, ‘Real-time inspection and determination methods of faults on photovoltaic power systems by thermal imaging in Turkey’, Renew. Energy, vol. 147, pp. 1231–1238, Mar. 2020, doi: 10.1016/j.renene.2019.09.075.
- Triki-Lahiani, A. Bennani-Ben Abdelghani, and I. Slama-Belkhodja, ‘Fault detection and monitoring systems for photovoltaic installations: A review’, Renew. Sustain. Energy Rev., vol. 82, pp. 2680–2692, Feb. 2018, doi: 10.1016/j.rser.2017.09.101.
- R. Madeti and S. N. Singh, ‘Modeling of PV system based on experimental data for fault detection using kNN method’, Sol. Energy, vol. 173, pp. 139–151, Oct. 2018, doi: 10.1016/j.solener.2018.07.038.
- S. Madani, A. Abbaspour, M. Beiraghi, P. Z. Dehkordi, and A. M. Ranjbar, ‘Islanding detection for PV and DFIG using decision tree and AdaBoost algorithm’, in 2012 3rd IEEE PES Innovative Smart Grid Technologies Europe (ISGT Europe), Oct. 2012, pp. 1–8. doi: 10.1109/ISGTEurope.2012.6465818.
- Dhibi et al., ‘Reduced Kernel Random Forest Technique for Fault Detection and Classification in Grid-Tied PV Systems’, IEEE J. Photovolt., vol. 10, no. 6, pp. 1864–1871, Nov. 2020, doi: 10.1109/JPHOTOV.2020.3011068.
Download all article in PDF
