Intelligent system for real-time detection and classification of solar panel defects

Lesia Dubchak, Yevgeniy Bodyanskiy, Oleg Savenko, Volodymyr Kochan, Anatoliy Sachenko

Abstract


The subject matter of the study in the article involves the process of detecting and classifying defects in solar panels in real-time using unmanned aerial vehicles (UAV) and artificial intelligence technologies. The goal of this study is to develop an intelligent system that combines an active monitoring methodology with Fuzzy BSB-based model for real-time detection and classification of solar panel defects. This will allow for the timely detection of defects and reduce the costs of repairing or replacing solar panels. The tasks to be solved are: to develop a method for active monitoring of the condition of panels based on laser scanning; to integrate algorithms for data processing and classification of defects in real time; to investigate the application of the Fuzzy BSB (Braine-State-in-the-Box) model to increase the stability of classification under conditions of noise and incomplete data. The methods used are: active laser scanning from UAVs, fuzzy neural network algorithms, the Fuzzy BSB associative memory model, as well as methods for analyzing images and feature vectors. The following results were obtained. A methodology for detecting defects at the transportation stage and during the operation of solar panels is proposed. A Fuzzy BSB model is proposed for classifying detected defects, which is capable of providing an accuracy of about 80% even under conditions of significant noise and class overlap. It is found that the system effectively distinguishes the main types of defects, in particular cracks, contamination, shading, and mechanical damage, demonstrating competitive advantages compared to traditional passive methods. Conclusions. The scientific novelty of the results obtained is as follows: 1) adapting the combination of associative memory and fuzzy logic in the Fuzzy BSB model to the classification of solar panel defects, which allows to increase the reliability of this classification in conditions of incomplete or noisy data; 2) the concept of integrating active laser scanning with intelligent analysis algorithms is proposed, which opens up prospects for creating flexible and adaptive systems for monitoring the condition of solar power plants.

Keywords


solar panels; defects; laser scanning; unmanned aerial vehicles; neural networks; intelligent system; Fuzzy BSB

Full Text:

PDF

References


Schulte, E., Scheller, F., Sloot, D., & Bruckner, T. A meta-analysis of residential PV adoption: the important role of perceived benefits, intentions and antecedents in solar energy acceptance. Energy Research & Social Science, 2022, vol. 84, article no. 102339. DOI: 10.1016/j.erss.2021.102339.

Lopushanska, M., & Ivanov, Ye. Klimatychni chynnyky ta yikhnya rolʹ u rozvytku sonyachnoyi enerhetyky u Lʹvivsʹkiy oblasti [Climate factors and their role in the solar energy development in the Lviv region]. Ecological Sciences, 2022, vol. 45, pp. 54–59. DOI: 10.32846/2306-9716/2022.eco.6-45.9. (In Ukrainian).

Sun, X., Silverman, T. J., Zhou, Z., Khan, M. R., Bermel, P., & Alam, M. A. Optics-Based Approach to Thermal Management of Photovoltaics: Selective-Spectral and Radiative Cooling. IEEE Journal of Photovoltaics, 2017, vol. 7, no. 2, pp. 566-574. DOI: 10.1109/JPHOTOV.2016.2646062.

Al Dhafari, L. S., Afzal, A., Al Abri, O. K., & Khan, A. Solar-Powered UAVs: A systematic Literature Review. Proc. 2nd Int. Conf. on Unmanned Vehicle Systems-Oman (UVS), Muscat, Oman, 2024, pp. 1-8. DOI: 10.1109/UVS59630.2024.10467158.

Radiuk, P., Rusyn, B., Melnychenko, O., Perzynski, T., Sachenko, A., Svystun, S., & Savenko, O. Criticality Assessment of Wind Turbine Defects via Multispectral UAV Fusion and Fuzzy Logic. Energies, 2025, vol. 18, article no. 4523. DOI: 10.3390/en18174523.

Hurin, A., Khudov, H., Kostyria , O., Maslenko, O. & Siadrystyi, S. Comparative analysis of spectral anomalies detection methods on images from on-board remote sensing systems. Advanced Information Systems, 2024, vol. 8, iss. 2, pp. 48–57. DOI: 10.20998/2522-9052.2024.2.06.

Tripathi, A., Gundebommu, S. L., Mishra, H., Shravani, C., Alwetaishi, M., Atamurotov, F., Benti, N. E., Sura, S., & Saleel C., A. Integration of Solar PV Panels in Electric Vehicle Charging Infrastructure: Benefits, Challenges, and Environmental Implications. Energy Science & Engineering, 2025, vol. 13, iss. 4, pp. 2135-2152. DOI: 10.1002/ese3.70014.

Venkatakrishnan, G. R., Rengaraj, R., Tamilselvi, S., Harshini, J., Sahoo, A., Saleel, C. A., Abbas, M., Cuce, E., Jazlyn, C., Shaik, S., Cuce, P. M., & Riffat, S. Detection, location, and diagnosis of different faults in large solar PV system—a review. International Journal of Low-Carbon Technologies, 2023, vol. 18, pp. 659–674. DOI: 10.1093/ijlct/ctad018.

Pathak, S. P., Patil, S., & Patel, S. Solar panel hotspot localization and fault classification using deep learning approach. Procedia Computer Science, 2022, vol. 204, pp. 698-705. DOI: 10.1016/j.procs.2022.08.084.

Abdallaha, F. S. M., Abdullaha, M. N., Musirinb, I., & Elshamyc, A. M.Intelligent solar panel monitoring system and shading detection using artificial neural networks. Energy Reports, 2023, vol. 9, sup. 10, pp. 324-334. DOI: 10.1016/j.egyr.2023.05.163.

El-Banby, G. M., Moawad, N. M., Abouzalm, B. A., Abouzaid, W. F., & Ramadan, E. A. Photovoltaic system fault detection techniques: a review. Neural Computing and Applications, 2023, vol. 35, pp. 24829-24842. DOI: 10.1007/s00521-023-09041-7.

Li, T., Hua, M., & Li, Q. Intelligent Inspection Method for Photovoltaic Modules Based on Image Processing and Deep Learning. Advances in Artificial Intelligence, Big Data and Algorithms. IOS Press, 2023, pp. 48-56, DOI: 10.3233/FAIA230791.

Ali Jallal, M., El Yassini, A., Chabaa, S., & Zeroual, A. Towards Smart Monitoring Systems: Fault Detection and Diagnosis-Based Artificial Intelligence Algorithms in Solar PV Power Plants. Journal of Nano- and Electronic Physics, 2022, vol. 14, no. 5, article no. 05004. DOI: 10.21272/jnep.14(5).05004.

Et-taleby, A., Chaibi, Y., Ayadi, N., Elkari, B., Benslimane, M., & Chalh, Z. Enhancing fault detection and classification in photovoltaic systems based on a hybrid approach using fuzzy logic algorithm and thermal image processing, Scientific African, 2025, vol. 28, article no. e02684. DOI: 10.1016/j.sciaf.2025.e02684.

Nassreddine, G., El Arid, A., Nassereddine, M., & Al Khatib, O. Fault Detection and Classification for Photovoltaic Panel System Using Machine Learning Techniques. Applied AI Letters, vol. 6, iss. 2, article no. e115. DOI: 10.1002/ail2.115.

Ghahremani, A., Adams, S. D., Norton, M., Khoo, S. Y., & Kouzani, A. Z. Advancements in AI-Driven detection and localisation of solar panel defects. Advanced Engineering Informatics, 2025, vol. 64, article no. 103104. DOI: 10.1016/j.aei.2024.103104.

Khudov, H., Makoveichuk, O., Khizhnyak, I., Varvarov, V., & Zots, F. Experimental studies of the image segmentation method quality from unmanned aerial vehicles based on the ant colony optimization algorithm under the influence of additive Gaussian noise. Advanced Information Systems, 2025, vol. 9, iss. 3, pp. 14-21. DOI: 10.20998/2522-9052.2025.3.02.

Jia, Y., Chen, G., & Zhao, L. Defect detection of photovoltaic modules based on improved VarifocalNet. Scientific Reports, 2024, vol. 14, article no. 15170. DOI: 10.1038/s41598-024-66234-3.

Dhimish, M., & Badran, G. Photovoltaic Hot-Spots Fault Detection Algorithm using Fuzzy Systems. IEEE Transactions on Device and Materials Reliability, 2019, vol. 19, no. 4, pp. 671-679. DOI: 10.1109/TDMR.2019.2944793.

Park, Y. Optimal and robust design of brain-state-in-a-box neural associative memories. Neural Networks, 2010, vol. 23, no. 2, pp. 210–218. DOI: 10.1016/j.neunet.2009.10.008.

Prabhakaran, S., Uthra, R. A., & Preetharoselyn, J. Deep Learning-Based Model for Defect Detection and Localization on Photovoltaic Panels. Computer Systems Science & Engineering, 2023, vol. 44, no. 3, pp. 2683-2700. DOI: 10.32604/csse.2023.028898.

Sevrani, F., & Abe, K. On the Synthesis of Brain-State-in-a-Box Neural Models with Application to Associative Memory. Neural Computation, 2000, vol. 12, no. 2, pp. 451–472. DOI: 10.1162/089976600300015871.

Bodyanskiy, Y. V., & Teslenko, N. A. Adaptive learning of fuzzy BSB and GBSB neural models. Cybernetics and Systems Analysis, 2006, vol. 42, pp. 786–794. DOI: 10.1007/s10559-006-0119-y.




DOI: https://doi.org/10.32620/reks.2025.3.08

Refbacks

  • There are currently no refbacks.