Implementation Framework for Deep Learning Network in Hot Spot Identification of Photovoltaic Panels

Wende Wang; Lei Tang; Qujie Chan; Qiao Jiang; Jianfeng Li; Ming Lei

doi:10.54097/jdb7nr60

Authors

Wende Wang
Lei Tang
Qujie Chan
Qiao Jiang
Jianfeng Li
Ming Lei

DOI:

https://doi.org/10.54097/jdb7nr60

Keywords:

Photovoltaic Panels, Hot Spot Detection, Deep Learning, Real-time Monitoring, Performance Comparison

Abstract

The proposed framework for photovoltaic panel hot spot identification demonstrates excellent stability and accuracy in various scenarios. Experimental results show that the framework accurately identifies hot spots in clear, unobstructed, cloudy, and low-light conditions. The accuracy for clear, unobstructed conditions reaches 99.2%, with an F1 score of 99.0%. Even in complex scenarios (such as early morning low light), the accuracy remains at 94.6%. The recognition speed is 22-28ms per image, meeting real-time monitoring requirements. Compared with traditional methods (threshold segmentation and SVM), the accuracy improves by 15.8%-20.1%. Compared with deep learning models (YOLOv5 and Faster R-CNN), the accuracy is 2.2%-2.7% higher, the recognition speed is 5.1-16.9ms per image faster, and the model size and training time are superior, demonstrating the efficiency and practicality of the framework for hot spot identification.

Downloads

Download data is not yet available.

References

[1] Hao, S., Wu, Y., Ma, X., Li, T. & Wang, H. Photovoltaic hot spot detection based on knowledge distillation and attention mechanism. Optical Precision Engineering, Vol. 31(2023) No. 24, p. 3640-3650.

[2] Jiang, C., He, J., Lu, Q., Wang, J., Yin, Y. & Luo, Y. A photovoltaic hot spot detection method based on improved YOLOv5 algorithm. Computer and Digital Engineering, Vol. 51(2023) No. 10, p. 2277-2281.

[3] Sun, J., Wang, L., Ma, J. & Gao, W. Photovoltaic module fault detection based on improved YOLO v5s algorithm. Infrared Technology, Vol. 45(2023) No. 2, p. 202-208.

[4] Lv, Y., Zheng, Q., Qi, X., Fang, F. & Liu, J. Research on photovoltaic module fault identification based on infrared image based on improved EfficientNet. Chinese Journal of Scientific Instrumentation, Vol. 45(2024) No. 4, p. 175-184.

[5] Wu, J., Fan, P. & Wang, M. Multi-scale fusion CRC-RetinaNet photovoltaic panel shadow detection method. Laser & Optoelectronics Progress, Vol. 59(2022) No. 16, p. 1615009-1615009.

[6] Wang, X., Liu, B., Sun, K., Zhao, J. & Chen, L. Review of photovoltaic array fault diagnosis technology. Transactions of the Chinese Society of Electrotechnical Engineering, Vol. 39(2024) No. 20, p. 6526-6543.

[7] Zhao, L. & Wu, Y. Research progress of surface defect detection methods based on machine vision. Chinese Journal of Scientific Instrumentation, Vol. 43(2022) No. 1, p. 198-219.

[8] Liu, Y. & Wu, Y. Review of solar cell defect detection algorithms based on machine vision. Optics and Precision Engineering, Vol. 32(2024) No. 6, p. 868-900.

[9] Qi, B., Ren, S., Du, Y., Wang, M. & Qin, H. Research on power information acquisition system based on embedded platform. Electronic Measurement Technology, Vol. 45(2024) No. 22, p. 1-6.

[10] Fan, T., Sun, T. & Liu, H. Hot spot detection algorithm for photovoltaic modules based on attention mechanism. Journal of Beijing University of Aeronautics and Astronautics, Vol. 48(2022) No. 7, p. 1304-1313.