An Explainable Image-Derived Illumination Non-Uniformity Index for Photovoltaic Attenuation Estimation

Document Type : Research Article

Authors
Mothiga Shivani J 1
Ashok Kumar B 2
Senthilrani S 2
1 Department of Electrical and Electronics Engineering, Thiagajarar College of Engineering, Tamil Nadu, India
2 Department of Electrical and Electronics Engineering, Thiagarajar College of Engineering, Tamil Nadu, India
10.22059/jser.2026.416342.1760
Abstract
Partial shading caused by dust accumulation, surface contaminants, and nearby obstructions reduces the incident irradiance on photovoltaic (PV) modules, leading to significant performance degradation. Conventional approaches for assessing shading effects rely on electrical measurements, irradiance sensors, or model-based calibration, increasing system complexity and limiting scalability. Although image-processing techniques have been widely investigated for PV inspection, most existing studies emphasize qualitative fault identification rather than quantitative attenuation assessment. This paper presents a non-intrusive image-processing framework that interprets PV panel images as two-dimensional illumination fields and introduces an Illumination Non-Uniformity Index (INUI) derived from physically interpretable spatial and frequency-domain descriptors. The proposed index integrates global intensity variation, spatial discontinuity, and low-frequency illumination distortion to characterize diverse shading patterns. An image-derived attenuation metric is subsequently established to quantify attenuation severity, demonstrating a monotonic relationship with illumination non-uniformity without requiring learning-based models, electrical measurements, or irradiance sensors. Experimental evaluation under diverse shading conditions confirms consistent monotonic behaviour, robustness against imaging variations, and computational efficiency suitable for large-scale visual inspection. The proposed framework provides an explainable, lightweight, and scalable methodology for quantitative image-based assessment of shading-induced attenuation in photovoltaic modules, offering a practical alternative for visual condition monitoring and supporting future intelligent PV inspection systems.
Keywords
Photovoltaic panels
Illumination non-uniformity
Image-based attenuation estimation
Partial shading
Frequency-domain image analysis
Non-intrusive monitoring
[1]      Abdelsattar, M., AbdelMoety, A., & Emad-Eldeen, A. (2025). ResNet-based image processing approach for precise detection of cracks in photovoltaic panels. Scientific Reports, 15, 24356. https://doi.org/10.1038/s41598-025-09101-z
[2]      Araji, M. T., & Waqas, A. (2025). Integrated deep learning and image processing method for modeling energy loss due to shadows in solar arrays. Solar Energy, 297, 113623. https://doi.org/10.1016/j.solener.2025.113623
[3]      Naeem, U., et al. (2025). Aerial imaging-based soiling detection system for solar photovoltaic panel cleanliness inspection. Sensors, 25(3), 738. https://doi.org/10.3390/s25030738
[4]      Ahmed, S., et al. (2025). Deep learning-based recognition and classification of soiled photovoltaic modules using HALCON software for solar cleaning robots. Sensors, 25(5), 1295. https://doi.org/10.3390/s25051295
[5]      Chen, L., et al. (2025). A detection model for dust deposition on photovoltaic (PV) panels based on light transmittance estimation. Energy, 322, 135284. https://doi.org/10.1016/j.energy.2025.135284
[6]      Özüpak, Y. (2025). Real-time detection of photovoltaic module faults using a hybrid machine learning model. Solar Energy, 302, 114014. https://doi.org/10.1016/j.solener.2025.114014
[7]      Agbogla, J., et al. (2025). Soiling estimation methods in solar photovoltaic systems: Review, challenges and future directions. Results in Engineering, 26, 104810. https://doi.org/10.1016/j.rineng.2025.104810
[8]      Tao, Y., Yu, T., & Yang, J. (2025). Photovoltaic array fault diagnosis and localization method based on modulated photocurrent and machine learning. Sensors, 25(1), 136. https://doi.org/10.3390/s25010136
[9]      Rudro, R. A. M., et al. (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
[10]   Amiri, A. F., Kichou, S., Oudira, H., Chouder, A., & Silvestre, S. (2024). Fault detection and diagnosis of a photovoltaic system based on deep learning using the combination of a convolutional neural network and bidirectional gated recurrent unit. Sustainability, 16(3), 1012. https://doi.org/10.3390/su16031012
[11]   Winkel, P., et al. (2024). Cell-resolved PV soiling measurement using drone images. Remote Sensing, 16(14), 2617. https://doi.org/10.3390/rs16142617
[12]   Evstatiev, B. I., et al. (2024). PV module soiling detection using visible spectrum imaging and machine learning. Energies, 17(20), 5238. https://doi.org/10.3390/en17205238
[13]   Setiawan, E. A., et al. (2024). Fast partial shading detection on PV modules for precise power loss ratio estimation using digital image processing. Journal of Electrical and Computer Engineering, 2024, 9385602. https://doi.org/10.1155/2024/9385602
[14]   Sarkar, S., et al. (2024). A model for effect of partial shading on PV panels with experimental validation. Energy Reports, 12, 3460–3469. https://doi.org/10.1016/j.egyr.2024.09.015
[15]   Lin, W.-T., Chang, C.-M., Huang, Y.-C., Wu, C.-C., & Kuo, C.-C. (2024). Fault diagnosis in solar array I–V curves using characteristic simulation and multi-input models. Applied Sciences, 14(13), 5417. https://doi.org/10.3390/app14135417
[16]   Puppala, H., Maganti, L. S., Peddinti, P. R. T., & Motapothula, M. R. (2024). Thermographic inspections of solar photovoltaic plants in India using unmanned aerial vehicles: Analysing the gap between theory and practice. Renewable Energy, 237(Part B), 121694. https://doi.org/10.1016/j.renene.2024.121694
[17]   Eesaar, H., et al. (2023). SEiPV-Net: An efficient deep learning framework for autonomous multi-defect segmentation in electroluminescence images of solar photovoltaic modules. Energies, 16(23), 7726. https://doi.org/10.3390/en16237726
[18]   Duranay, Z. B. (2023). Fault detection in solar energy systems: A deep learning approach. Electronics, 12(21), 4397. https://doi.org/10.3390/electronics12214397
[19]   Cavieres, R., Barraza, R., Estay, D., Bilbao, J., & Valdivia-Lefort, P. (2022). Automatic soiling and partial shading assessment on PV modules through RGB image analysis. Applied Energy, 306, 117964. https://doi.org/10.1016/j.apenergy.2021.117964
[20]   Bessa, J. G., et al. (2021). Monitoring photovoltaic soiling: Assessment, challenges, and perspectives of current and potential strategies. iScience, 24(3), 102165. https://doi.org/10.1016/j.isci.2021.102165
[21]   Deitsch, S., et al. (2021). Segmentation of photovoltaic module cells in uncalibrated electroluminescence images. Machine Vision and Applications, 32, 84. https://doi.org/10.1007/s00138-021-01191-9
[22]   Gao, W., & Wai, R.-J. (2020). A novel fault identification method for photovoltaic array via convolutional neural network and residual gated recurrent unit. IEEE Access, 8, 159493–159510. https://doi.org/10.1109/ACCESS.2020.3020296
[23]   Coello, M., & Boyle, L. (2019). Simple model for predicting time series soiling of photovoltaic panels. IEEE Journal of Photovoltaics, 9(5), 1382–1387. https://doi.org/10.1109/JPHOTOV.2019.2919628
[24]   Mellit, A., Tina, G. M., & Kalogirou, S. A. (2018). Fault detection and diagnosis methods for photovoltaic systems: A review. Renewable and Sustainable Energy Reviews, 91, 1–17. https://doi.org/10.1016/j.rser.2018.03.062
[25]   Madeti, S. R., & Singh, S. N. (2018). Modeling of PV system based on experimental data for fault detection using kNN method. Solar Energy, 173, 139–151. https://doi.org/10.1016/j.solener.2018.07.038
[26]   Silvestre, S., Chouder, A., & Karatepe, E. (2013). Automatic fault detection in grid-connected PV systems. Solar Energy, 94, 119–127. https://doi.org/10.1016/j.solener.2013.05.001
Journal of Solar Energy Research
Volume 11, Issue 3
Summer 2026
Pages 3252-3266