DOI: https://doi.org/10.61435/ijred.2026.62298
View PDF Download fulltext

Towards self-diagnostic solar farms: Leveraging EfficientNet and class activation mapping for predictive maintenance

1Institute of Engineering, HUTECH University, Ho Chi Minh City, Viet Nam

2Institute of Information Technology and Electricity, Ho Chi Minh City University of Transport, Ho Chi Minh City, Viet Nam

3Institute of Maritime, Ho Chi Minh City University of Transport, Ho Chi Minh City, Viet Nam

4 Institute of Mechanical Engineering, Ho Chi Minh City University of Transport, Ho Chi Minh City, Viet Nam

View all affiliations
Received: 28 Oct 2025; Revised: 24 Jan 2026; Accepted: 5 Feb 2026; Available online: 18 Feb 2026; Published: 1 Mar 2026.
Editor(s): H Hadiyanto
Open Access Copyright (c) 2026 The Author(s). Published by Centre of Biomass and Renewable Energy (CBIORE)
Creative Commons License This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Citation Format:
Abstract

The high rate of utility photovoltaic (PV) system development has increased the demand for stable, automated, and interpretable fault diagnostic systems that can be utilised in real-world environments. Solar farms with a large size are increasingly making conventional manual inspection methods impractical, and triggering the use of intelligent data-driven solutions. This paper presents a justifiable deep learning model for automated fault classification of solar panels based on the EfficientNet-B2 architecture combined with Gradient-weighted Class Activation Mapping (Grad-CAM). A six-class image dataset made of clean panels and five prevalent fault types is used. The two stages of transfer learning used to train the model include a warm-up phase and selective fine-tuning of upper network layers. Data augmentation is also performed extensively to make it more robust to changing illumination, viewing angles, and environmental noise. The experimental findings reveal consistent convergence and excellent generalization ability, and a high level of classification accuracy of all types of faults, as it achieved high classification accuracy, macro-averaged F1-scores exceeding 0.90 for most fault classes, and a macro-averaged ROC–AUC of approximately 0.981, highlighting the robustness and reliability of the proposed diagnostic model. The suggested structure will provide a scalable, interpretable, and realistic predictive maintenance of solar farms of the next generation with self-diagnostic capabilities.

Keywords: Photovoltaic fault diagnosis; EfficientNet-B2; Explainable artificial intelligence; Grad-CAM; Deep learning; Solar energy systems

Article Metrics:

Article Info
Section: Original Research Article
Language : EN
Recent articles
Identification and comparison of pyrolysis products of different biomasses from agro-industrial using TGA-FTIR and Py-GC/MS Towards self-diagnostic solar farms: Leveraging EfficientNet and class activation mapping for predictive maintenance Design and optimization of an energy storage system for off-grid rural communities More recent articles
Most cited articles
Empowering Distributed Solar PV Energy for Malaysian Rural Housing: Towards Energy Security and Equitability of Rural Communities Optimization of Lipid Extraction from Municipal Scum Sludge for Biodiesel Production Using Statistical Approach Potency of Solar Energy Applications in Indonesia Automatic and Online Detection of Rotor Fault State Economic feasibility of large scale PV water pumping applications utilizing real field data for a case study in Jordan More cited articles
  1. Adib, A. U. R., Islam, M., Abid, M. S., & Ahshan, R. (2025). A deep learning based framework for solar panel segmentation and fault classification enhanced with explainable AI. Solar Energy, 302, 114058. https://doi.org/10.1016/j.solener.2025.114058
  2. Afandi, A., Birowosuto, M. D., & Sembiring, K. C. (2022). Energy-yield Assessment Based on the Orientations and the Inclinations of the Solar Photovoltaic Rooftop Mounted in Jakarta, Indonesia. International Journal on Advanced Science, Engineering and Information Technology, 12(2 SE-Articles), 470–476. https://doi.org/10.18517/ijaseit.12.2.14812
  3. Afroz, P. (2024). Solar Panel Images Clean and Faulty Images. Kaggle
  4. Agarwala, N. (2024). Is hydrogen a decarbonizing fuel for maritime shipping? Maritime Technology and Research, 6(4), 271244. https://doi.org/10.33175/mtr.2024.271244
  5. Ahmed, S. F., Islam, N., Kumar, P. S., Hoang, A. T., Mofijur, M., Inayat, A., Shafiullah, G. M., Vo, D.-V. N., Badruddin, I. A., & Kamangar, S. (2023). Perovskite solar cells: Thermal and chemical stability improvement, and economic analysis. Materials Today Chemistry, 27, 101284. https://doi.org/10.1016/j.mtchem.2022.101284
  6. Al-lami, A., Török, A., Alatawneh, A., & Alrubaye, M. (2025). Future Energy Consumption and Economic Implications of Transport Policies: A Scenario-Based Analysis for 2030 and 2050. Energies, 18(12), 3012. https://doi.org/10.3390/en18123012
  7. Allen, M. R., Frame, D. J., Friedlingstein, P., Gillett, N. P., Grassi, G., Gregory, J. M., Hare, W., House, J., Huntingford, C., Jenkins, S., Jones, C. D., Knutti, R., Lowe, J. A., Matthews, H. D., Meinshausen, M., Meinshausen, N., Peters, G. P., Plattner, G.-K., Raper, S., … Zickfeld, K. (2025). Geological Net Zero and the need for disaggregated accounting for carbon sinks. Nature, 638(8050), 343–350. https://doi.org/10.1038/s41586-024-08326-8
  8. Alruwaili, M., & Mohamed, M. (2025). An Integrated Deep Learning Model with EfficientNet and ResNet for Accurate Multi-Class Skin Disease Classification. Diagnostics, 15(5), 551. https://doi.org/10.3390/diagnostics15050551
  9. Arun, M., Barik, D., Chandran, S. S. R., Govil, N., Sharma, P., Khan, T. M. Y., Baig, R. U., Bora, B. J., Medhi, B. J., & Kumar, R. (2024). Twisted helical Tape’s impact on heat transfer and friction in zinc oxide (ZnO) nanofluids for solar water heaters: biomedical insight. Case Studies in Thermal Engineering, 56, 104204
  10. Basit, M. A., Dilshad, S., Badar, R., & Sami ur Rehman, S. M. (2020). Limitations, challenges, and solution approaches in grid‐connected renewable energy systems. International Journal of Energy Research, 44(6), 4132–4162. https://doi.org/10.1002/er.5033
  11. Bendale, H., Aswar, H., Bamb, H., Desai, P., & Aher, C. N. (2023). Deep Learning for Solar Panel Maintenance: Detecting Faults and Improving Performance. 2023 14th International Conference on Computing Communication and Networking Technologies (ICCCNT), 1–5. https://doi.org/10.1109/ICCCNT56998.2023.10307465
  12. Chen, H., Su, L., Shu, R., Li, T., & Yin, F. (2024). EMB-YOLO: A Lightweight Object Detection Algorithm for Isolation Switch State Detection. Applied Sciences, 14(21), 9779. https://doi.org/10.3390/app14219779
  13. Gandhi, A. M., Shanmugan, S., Kumar, R., Elsheikh, A. H., Sharifpur, M., Bewoor, A. K., Bamisile, O., Hoang, A. T., & Ongar, B. (2022). SiO2/TiO2 nanolayer synergistically trigger thermal absorption inflammatory responses materials for performance improvement of stepped basin solar still natural distiller. Sustainable Energy Technologies and Assessments, 52, 101974. https://doi.org/10.1016/j.seta.2022.101974
  14. Guo, J., Chong, C. F., Abreu, P. H., Mao, C., Li, J., Lam, C.-T., & Ng, B. K. (2025). Reparameterization convolutional neural networks for handling imbalanced datasets in solar panel fault classification. Engineering Applications of Artificial Intelligence, 150, 110541. https://doi.org/10.1016/j.engappai.2025.110541
  15. Hoang, A. T., Pandey, A., Lichtfouse, E., Bui, V. G., Veza, I., Nguyen, H. L., & Nguyen, X. P. (2023). Green hydrogen economy: Prospects and policies in Vietnam. International Journal of Hydrogen Energy, 48(80), 31049–31062. https://doi.org/10.1016/j.ijhydene.2023.05.306
  16. Hoang, A. T., Pham, V. V., & Nguyen, X. P. (2021). Integrating renewable sources into energy system for smart city as a sagacious strategy towards clean and sustainable process. Journal of Cleaner Production, 305, 127161. https://doi.org/10.1016/j.jclepro.2021.127161
  17. Hoang, A. T., Sandro Nižetić, Olcer, A. I., Ong, H. C., Chen, W.-H., Chong, C. T., Thomas, S., Bandh, S. A., & Nguyen, X. P. (2021). Impacts of COVID-19 pandemic on the global energy system and the shift progress to renewable energy: Opportunities, challenges, and policy implications. Energy Policy, 154, 112322. https://doi.org/10.1016/j.enpol.2021.112322
  18. Hoang, V.-T., Hoang, V.-D., & Jo, K.-H. (2023). Rethinking Mobile Inverted Bottleneck Convolution for EfficientNet (pp. 435–445). https://doi.org/10.1007/978-3-031-19694-2_39
  19. Huynh, D. N. L., Bandh, S. A., Malla, F. A., Nguyen, X. P., Wani, S. A., Rodríguez-Castellón, E., Truong, T. N., Nguyen, D. K. P., Nguyen, H. P., & Hoang, A. T. (2025). Sustainable transformation to carbon neutrality: Obstacles and solutions. Energy & Environment. https://doi.org/10.1177/0958305X251354874
  20. Jabbar, A., Yuan, J., Idris, S., I.Khan, M., & Mahmood, T. (2025). Bayesian–Causal Reinforcement Learning for adaptive and interpretable solar energy policy design. Egyptian Informatics Journal, 32, 100853. https://doi.org/10.1016/j.eij.2025.100853
  21. Jin, T., & Liu, J. (2025). A text classification method by integrating mobile inverted residual bottleneck convolution networks and capsule networks with adaptive feature channels. Scientific Reports, 15(1), 855. https://doi.org/10.1038/s41598-025-85237-2
  22. Jing, D., Mo, H., Han, L., Yin, H., Li, L., Zhang, Y., Li, M., Pan, M., & Guo, L. (2025). 3M‐Net: Automatic Modulation Recognition Based on Multiscale Mobile Inverted Bottleneck Convolution and Manhattan Self‐Attention Network. International Journal of Communication Systems, 38(9). https://doi.org/10.1002/dac.70109
  23. Joshua, S. R., Park, S., & Kwon, K. (2024). Solar Panel Fault Detection: Applying Convolutional Neural Network for Advanced Fault Detection in Solar-Hydrogen System at University. 2024 IEEE 24th International Conference on Software Quality, Reliability, and Security Companion (QRS-C), 289–298. https://doi.org/10.1109/QRS-C63300.2024.00045
  24. Kassim, M., & Lazim, F. (2022). Adaptive photovoltaic solar module based on internet of things and web-based monitoring system. International Journal of Electrical and Computer Engineering (IJECE), 12(1), 924. https://doi.org/10.11591/ijece.v12i1.pp924-935
  25. Kharesaxena, A., Saxena, S., & Sudhakar, K. (2020). Solar energy policy of India: An overview. CSEE Journal of Power and Energy Systems. https://doi.org/10.17775/CSEEJPES.2020.03080
  26. Korkmaz, D., & Acikgoz, H. (2022). An efficient fault classification method in solar photovoltaic modules using transfer learning and multi-scale convolutional neural network. Engineering Applications of Artificial Intelligence, 113, 104959. https://doi.org/10.1016/j.engappai.2022.104959
  27. Lau, L.-S., Choong, Y.-O., Ching, S.-L., Wei, C.-Y., Senadjki, A., Choong, C.-K., & Seow, A.-N. (2022). Expert insights on Malaysia’s residential solar-energy policies: shortcomings and recommendations. Clean Energy, 6(4), 619–631. https://doi.org/10.1093/ce/zkac043
  28. Le, M., Le, D., & Ha Thi Vu, H. (2023). Thermal inspection of photovoltaic modules with deep convolutional neural networks on edge devices in AUV. Measurement, 218, 113135. https://doi.org/10.1016/j.measurement.2023.113135
  29. Le, T. T., Le, H. C., Paramasivam, P., & Chung, N. (2024). Artificial intelligence applications in solar energy. JOIV: International Journal on Informatics Visualization, 8(2), 826–844. https://doi.org/10.62527/joiv.8.2.2686
  30. Ling, H., Liu, M., & Fang, Y. (2024). Deep Edge-Based Fault Detection for Solar Panels. Sensors, 24(16), 5348. https://doi.org/10.3390/s24165348
  31. Maghraby, Y. R., Ibrahim, A. H., Tayel, A., Mohamed El-Said Azzazy, H., & Shoeib, T. (2025). Towards sustainability via recycling solar photovoltaic Panels, A review. Solar Energy, 285, 113085. https://doi.org/10.1016/j.solener.2024.113085
  32. Manimegalai, V., Oviya, B., Kargvel, S. U., Vilashini, P. L., Mohanapriya, V., & Elakya, A. (2025). Deep Learning for Solar Panel Fault Detection: Integrating GAN and ResNet Models. 2025 6th International Conference on Mobile Computing and Sustainable Informatics (ICMCSI), 1285–1291. https://doi.org/10.1109/ICMCSI64620.2025.10883431
  33. Manimegalai, V., Oviya, B., Mohanapriya, V., Kargvel, S. U., Tejuaswene, R., & Ravi Raaghav, M. P. (2025). Deep Learning Framework for Solar Panel Fault Detection: Combining GAN and EfficientNet. 2025 5th International Conference on Pervasive Computing and Social Networking (ICPCSN), 1872–1877. https://doi.org/10.1109/ICPCSN65854.2025.11034949
  34. Meng, W., Zhang, Q., Ma, S., Cai, M., Liu, D., Liu, Z., & Yang, J. (2022). A lightweight CNN and Transformer hybrid model for mental retardation screening among children from spontaneous speech. Computers in Biology and Medicine, 151, 106281. https://doi.org/10.1016/j.compbiomed.2022.106281
  35. Missoum, M., & Loukarfi, L. (2021). Investigation of a Solar Polygeneration System for a Multi-Storey Residential Building-Dynamic Simulation and Performance Analysis. International Journal of Renewable Energy Development, 10(3), 445–458. https://doi.org/10.14710/ijred.2021.34423
  36. Mohamed, M., Mahesh, T., Vinoth Kumar, V., & Guluwadi, S. (2024). Enhancing brain tumor detection in MRI images through explainable AI using Grad-CAM with Resnet 50. BMC Medical Imaging, 24(1), 107. https://doi.org/10.1186/s12880-024-01292-7
  37. Moreno, J., Campagnolo, L., Boitier, B., Nikas, A., Koasidis, K., Gambhir, A., Gonzalez-Eguino, M., Perdana, S., Van de Ven, D.-J., Chiodi, A., Delpiazzo, E., Doukas, H., Gargiulo, M., Herbst, A., Al-Dabbas, K., Alibaş, Ş., Neuner, F., Le Mouël, P., & Vielle, M. (2024). The impacts of decarbonization pathways on Sustainable Development Goals in the European Union. Communications Earth & Environment, 5(1), 136. https://doi.org/10.1038/s43247-024-01309-7
  38. Munusamy, A., Barik, D., Sharma, P., Medhi, B. J., & Bora, B. J. (2023). Performance analysis of parabolic type solar water heater by using copper-dimpled tube with aluminum coating. Environmental Science and Pollution Research, 31(53), 62376–62391. https://doi.org/10.1007/s11356-022-25071-5
  39. Nguyen, H. B., & Nguyen, V. L. (2023). A Study on the Efficiency of Solar Radiation Collectors Applying for Agricultural Products and Food Drying. International Journal on Advanced Science, Engineering and Information Technology, 13(2 SE-Articles), 564–571. https://doi.org/10.18517/ijaseit.13.2.18712
  40. Ogwumike, C., Akponeware, A., Oyewole, A., Dawood, H., Pinedo-Cuenca, R., Ling-Chin, J., Roskilly, A. P., & Dawood, N. (2024). Transitioning or tinkering at a net-zero economy? Introducing an assessment framework for industrial cluster decarbonisation in the United Kingdom. Energy Research & Social Science, 110, 103459. https://doi.org/10.1016/j.erss.2024.103459
  41. Pala, C., Bollem, P., & Neelima, N. (2024). Advanced Deep Learning Solutions for Automated Diagnosis of Solar Panel Issues. 2024 1st International Conference on Trends in Engineering Systems and Technologies (ICTEST), 1–5. https://doi.org/10.1109/ICTEST60614.2024.10576090
  42. Pallakonda, A., Raj, R. D. A., Yanamala, R. M. R., B., R. R., Kolisetty, H., Pedamallu, S. M., & K., K. P. (2025). Lightweight hierarchical spatial feature extraction and sequential modeling for PV fault detection using pyramid network and GRU for edge applications. Energy Conversion and Management: X, 28, 101293. https://doi.org/10.1016/j.ecmx.2025.101293
  43. Panos, B., Kleint, L., & Zbinden, J. (2023). Identifying preflare spectral features using explainable artificial intelligence. Astronomy & Astrophysics, 671, A73. https://doi.org/10.1051/0004-6361/202244835
  44. Pathak, S. P., Patil, D. S., & Patel, S. (2022). Solar panel hotspot localization and fault classification using deep learning approach. Procedia Computer Science, 204, 698–705. https://doi.org/10.1016/j.procs.2022.08.084
  45. Pathak, S. P., & Patil, S. A. (2023). Evaluation of Effect of Pre-Processing Techniques in Solar Panel Fault Detection. IEEE Access, 11, 72848–72860. https://doi.org/10.1109/ACCESS.2023.3293756
  46. Peña-Asensio, E., Trigo-Rodríguez, J. M., Grèbol-Tomàs, P., Regordosa-Avellana, D., & Rimola, A. (2023). Deep machine learning for meteor monitoring: Advances with transfer learning and gradient-weighted class activation mapping. Planetary and Space Science, 238, 105802. https://doi.org/10.1016/j.pss.2023.105802
  47. Pham, D. T., Vujanović, M., Luu, V. C., Paramasivam, P., Nguyen, V. N., Efremov, C., Sănduleac, M., Nguyen, X. P., Tran, V. D., & Hoang, A. T. (2025). Transparent artificial intelligence models using sequential minimal optimization and extreme Gradient Boosting for accurate performance prediction of solar farms. Case Studies in Thermal Engineering, 76, 107337. https://doi.org/10.1016/j.csite.2025.107337
  48. Pham, N. D. K., Dinh, G. H., Pham, H. T., Kozak, J., & Nguyen, H. P. (2023). Role of Green Logistics in the Construction of Sustainable Supply Chains. Polish Maritime Research, 30(3), 191–211. https://doi.org/10.2478/pomr-2023-0052
  49. Polymeropoulos, I., Bezyrgiannidis, S., Vrochidou, E., & Papakostas, G. A. (2024). Enhancing Solar Plant Efficiency: A Review of Vision-Based Monitoring and Fault Detection Techniques. Technologies, 12(10), 175. https://doi.org/10.3390/technologies12100175
  50. Rosadi, M. I., Hakim, L., & M. Faishol A. (2023). Classification of Coffee Leaf Diseases using the Convolutional Neural Network (CNN) EfficientNet Model. Conference Series, 4(1), 58–69. https://doi.org/10.34306/conferenceseries.v4i1.627
  51. 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
  52. Senthil, S. P., Kuma, P., Kumaragurubaran, T., & Chiranjeevi, V, R. (2024). A Novel Approach for Plant Diseases Detection and Identification Using EfficientNet-B2. 2024 International Conference on Recent Innovation in Smart and Sustainable Technology (ICRISST), 1–5. https://doi.org/10.1109/ICRISST59181.2024.10921816
  53. Sahu, B. K. (2016). Solar energy developments, policies and future prospectus in the state of Odisha, India. Renewable and Sustainable Energy Reviews, 61, 526–536. https://doi.org/10.1016/j.rser.2016.04.027
  54. Shamshirband, S., Rabczuk, T., & Chau, K.-W. (2019). A Survey of Deep Learning Techniques: Application in Wind and Solar Energy Resources. IEEE Access, 7, 164650–164666. https://doi.org/10.1109/ACCESS.2019.2951750
  55. Shi, Y., & Luo, W. (2018). Application of solar photovoltaic power generation system in maritime vessels and development of maritime tourism. Polish Maritime Research, 25, 176–181. https://doi.org/10.2478/pomr-2018-0090
  56. Usha, G. P., & Alex, J. S. R. (2024). Advanced grad-CAM extensions for interpretable aphasia speech keyword classification: Bridging the gap in impaired speech with XAI. Results in Engineering, 24, 103414. https://doi.org/10.1016/j.rineng.2024.103414
  57. van Zyl, C., Ye, X., & Naidoo, R. (2024). Harnessing eXplainable artificial intelligence for feature selection in time series energy forecasting: A comparative analysis of Grad-CAM and SHAP. Applied Energy, 353, 122079. https://doi.org/10.1016/j.apenergy.2023.122079
  58. Waseer, W. I., Baqir, M. A., Saqlain, M., Mughal, M. J., & Khan, S. (2025). Predictive modeling of MXene-based solar absorbers using a deep neural network. Journal of the Optical Society of America B, 42(4), 763. https://doi.org/10.1364/JOSAB.550317
  59. Wattana, B., & Aungyut, P. (2022). Impacts of Solar Electricity Generation on the Thai Electricity Industry. International Journal of Renewable Energy Development, 11(1), 157–163. https://doi.org/10.14710/ijred.2022.41059
  60. Wei, C.-C. (2019). Evaluation of Photovoltaic Power Generation by Using Deep Learning in Solar Panels Installed in Buildings. Energies, 12(18), 3564. https://doi.org/10.3390/en12183564
  61. Zhang, H., & Ogasawara, K. (2023). Grad-CAM-Based Explainable Artificial Intelligence Related to Medical Text Processing. Bioengineering, 10(9), 1070. https://doi.org/10.3390/bioengineering10091070
  62. Zhang, X., Xu, L., Li, Z., & Huang, X. (2024). Causal Attention Deep-learning Model for Solar Flare Forecasting. The Astrophysical Journal Supplement Series, 274(2), 38. https://doi.org/10.3847/1538-4365/ad7386

Last update:

No citation recorded.

Last update: 2026-03-20 14:22:46

No citation recorded.