 |
|
|
Mohd Zulhamdy Ab Hamid Centre of Electrical Engineering Studies, Universiti Teknologi MARA, Cawangan Pulau Pinang, Permatang Pauh, Malaysia Kamarulazhar Daud Centre of Electrical Engineering Studies, Universiti Teknologi MARA, Cawangan Pulau Pinang, Permatang Pauh, Malaysia Zainal Hisham Che Soh Centre of Electrical Engineering Studies, Universiti Teknologi MARA, Cawangan Pulau Pinang, Permatang Pauh, Malaysia Muhammad Khusairi Osman Centre of Electrical Engineering Studies, Universiti Teknologi MARA, Cawangan Pulau Pinang, Permatang Pauh, Malaysia Iza Sazanita Isa Centre of Electrical Engineering Studies, Universiti Teknologi MARA, Cawangan Pulau Pinang, Permatang Pauh, Malaysia Mohd Shawal Jadin Faculty of Electrical and Electronics Engineering Technology, Universiti Malaysia Pahang Al-Sultan Abdullah, Pekan, Pahang, Malaysia |
|
| Abstract | |
|
The effective management of solar photovoltaic (PV) arrays is vital to maximising energy generation and ensuring long-term performance reliability. The presence of faults, or partial shading, can give rise to hotspots in PV arrays, making their detection critical for timely maintenance and avoiding further impacts on their performance. This paper describes a new approach to the hotspot detection issue in thermal images of solar PV arrays by leveraging deep learning. This study employs the YOLOv10 (You Only Look Once, version 10) model to deliver very high accuracy and speed in identifying and localising hotspots under defined regions of interest (ROIs). The proposed method begins with taking thermal images from multiple solar PV installations while capturing a range of operational conditions and fault types. These images are annotated with hotspot regions to create a robust training dataset. Given YOLO's capability for real-time object detection with high precision and mean average precision (mAP), the YOLOv10 model is then implemented to train on this dataset. Multiple changes were made to the YOLOv10 hyperparameters to optimise them for detecting hotspots in thermal images. The experimental scenario of this paper demonstrates that this approach indeed performs significantly better than standard image processing methods as well as prior deep learning models for detecting both hotspot accuracy as well as speed of processing the images. The YOLOv10 method demonstrated the highest available classification performance with a mAP at intersection over union (IoU) of 0.5 accuracy of 0.91 with inference time suitable for real-time applications. The sample results demonstrated the model's continuous ability to detect hotspots and provide location data under various conditions, such as crossing through different times of day and weather. The results of this study demonstrate that YOLOv10 can be used for improved detection of hotspots in the explicably thermal imagery of solar PV arrays. |
|
| Keyword: Deep Learning, Thermal Imaging, Hotspot Detection, Solar PV Arrays, YOLOv10 | |
|
DOI: 10.24191/esteem.v20iSeptember.1861.g1773 |
|
| References: | |
|
[1] P. Bharadwaj, K. Karnataki, and V. John, “Formation of Hotspots on Healthy PV Modules and Their Effect on Output Performance,” in 2018 IEEE 7th World Conference on Photovoltaic Energy Conversion (WCPEC) (A Joint Conference of 45th IEEE PVSC, 28th PVSEC & 34th EU PVSEC), pp. 0676–0680, Jun. 2018. Available: doi: 10.1109/PVSC.2018.8548126. [2] S. Deng et al., “Research on hot spot risk for high-efficiency solar module,” in Energy Procedia, Elsevier Ltd, pp. 77–86, 2017. Available: doi: 10.1016/j.egypro.2017.09.399. [3] Y. Zefri, A. Elkettani, I. Sebari, and S. A. Lamallam, “Thermal infrared and visual inspection of photovoltaic installations by uav photogrammetry—application case: Morocco,” Drones, vol. 2, no. 4, pp. 1–24, Dec. 2018. Available: doi: 10.3390/drones2040041. [4] A. Wang et al., YOLOv10: Real-Time End-to-End Object Detection, May 2024, [Online]. Available: http://arxiv.org/abs/2405.14458 [5] U. Pruthviraj, Y. Kashyap, E. Baxevanaki, and P. Kosmopoulos, “Solar Photovoltaic Hotspot Inspection Using Unmanned Aerial Vehicle Thermal Images at a Solar Field in South India,” Remote Sens (Basel), vol. 15, no. 7, Apr. 2023. Available: doi: 10.3390/rs15071914. [6] O. E. Ikejiofor, Y. E. Asuamah, H. O. Njoku, and S. O. Enibe, “Detection of Hotspots and Performance Deteriotations in PV Modules under Partial Shading Conditions Using Infrared Thermography †,” Engineering Proceedings, vol. 2, no. 1, 2020. Available: doi: 10.3390/ecsa-7-08201. [7] J. Starzy?ski, P. Zawadzki, and D. Hara?czyk, “Machine Learning in Solar Plants Inspection Automation,” Energies (Basel), vol. 15, no. 16, Aug. 2022. Available: doi: 10.3390/en15165966. [8] K. Nitturkar, S. Vitole, M. Jadhav, and V. G. Sridhar, “Solar Panel Fault Detection Using Machine Vision and Image Processing Technique,” in 2023 International Conference on Next Generation Electronics, NEleX 2023, Institute of Electrical and Electronics Engineers Inc., 2023. Available: doi: 10.1109/NEleX59773.2023.10421272. [9] Ihtyaz Kader Tasawar, Abyaz Kader Tanzeem, Md. Mosaddequr Rahman, Tahmid Ahmed, Mohaimenul Islam, and Shah Faiza Zarin, “Fault Diagnosis of PV Modules Using Deep CNN,” in 2022 International Conference on Energy and Power Engineering (ICEPE), pp. 1–6, 2022. Available: doi: 10.1109/ICEPE56629.2022.10044899. [10] M. Vlaminck, R. Heidbuchel, W. Philips, and H. Luong, “Region-Based CNN for Anomaly Detection in PV Power Plants Using Aerial Imagery,” Sensors, vol. 22, no. 3, Feb. 2022. Available: doi: 10.3390/s22031244. [11] F. Fan, Z. X. Na, C. Z. Zhang, H. R. Li, and C. L. Tong, “Hot Spot Detection of Photovoltaic Module Infrared Near-field Image based on Convolutional Neural Network,” in Journal of Physics: Conference Series, Institute of Physics, 2022. Available: doi: 10.1088/1742-6596/2310/1/012076. [12] H. Bak?r, F. A. Kuzhippallil, and A. Merabet, “Automatic detection of deteriorated photovoltaic modules using IRT images and deep learning (CNN, LSTM) strategies,” Eng Fail Anal, vol. 146, Apr. 2023. Available: doi: 10.1016/j.engfailanal.2023.107132. [13] A. Di Tommaso, A. Betti, G. Fontanelli, and B. Michelozzi, A Multi-Stage model based on YOLOv3 for defect detection in PV panels based on IR and Visible Imaging by Unmanned Aerial Vehicle, Nov. 2021, [Online]. Available: http://arxiv.org/abs/2111.11709 [14] T. Sun, H. Xing, S. Cao, Y. Zhang, S. Fan, and P. Liu, “A novel detection method for hot spots of photovoltaic (PV) panels using improved anchors and prediction heads of YOLOv5 network,” Energy Reports, vol. 8, pp. 1219–1229, Nov. 2022. Available: doi: 10.1016/j.egyr.2022.08.130. [15] T. Sun, H. Xing, S. Cao, Y. Zhang, S. Fan, and P. Liu, “A novel detection method for hot spots of photovoltaic (PV) panels using improved anchors and prediction heads of YOLOv5 network,” Energy Reports, vol. 8, pp. 1219–1229, Nov. 2022. Available: doi: 10.1016/j.egyr.2022.08.130. [16] J. Wang et al., “Deep-Learning-Based Automatic Detection of Photovoltaic Cell Defects in Electroluminescence Images,” Sensors, vol. 23, no. 1, Jan. 2023, doi: 10.3390/s23010297. [17] Y. Li, S. Chen, T. Rapakoulia, H. Kuwahara, K. Y. Yip, and X. Gao, “Deep learning identifies and quantifies recombination hotspot determinants,” Bioinformatics, vol. 38, no. 10, pp. 2683–2691, May 2022. Available: doi: 10.1093/bioinformatics/btac234. [18] Tzutalin, LabelImg, 2015, Free Software: MIT License. Accessed: Jul. 02, 2024. [Online]. Available: https://github.com/tzutalin/labelImg [19] C.-Y. Wang, I.-H. Yeh, and H.-Y. M. Liao, YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information, Feb. 2024, Accessed: Jun. 05, 2024. [Online]. Available: http://arxiv.org/abs/2402.13616 [20] Q. Zheng, J. Ma, M. Liu, Y. Liu, Y. Li, and G. Shi, “Lightweight Hot-Spot Fault Detection Model of Photovoltaic Panels in UAV Remote-Sensing Image,” Sensors, vol. 22, no. 12, Jun. 2022. Available: doi: 10.3390/s22124617. [21] S. Jumaboev, D. Jurakuziev, and M. Lee, “Photovoltaics Plant Fault Detection Using Deep Learning Techniques,” Remote Sens (Basel), vol. 14, no. 15, Aug. 2022. Available: doi: 10.3390/rs14153728. [22] L. Pratt, J. Mattheus, and R. Klein, “A benchmark dataset for defect detection and classification in electroluminescence images of PV modules using semantic segmentation”, Systems and Soft Computing, vol. 5, p. 200048, 2023. Available: , doi: 10.1016/j.sasc.2023.200048.
|
|
Deep learning-driven thermal imaging hotspot detection in solar photovoltaic arrays using YOLOv10
