Photovoltaic Panel Defect Detection Algorithm based on Improved YOLOv8s

Ziqing Song; Yupeng Yao; Jingbin Yang

doi:10.54691/5fdg3j51

Authors

Ziqing Song
Yupeng Yao
Jingbin Yang

DOI:

https://doi.org/10.54691/5fdg3j51

Keywords:

Photovoltaic Panel Defect Detection; Yolov8; Dynamic Convolution; Attention Mechanism.

Abstract

Aiming at the problems of insufficient accuracy and large parameter amount of the existing YOLO model in defect detection, an improved YOLOv8s algorithm was proposed for photovoltaic panel defect detection.On the basis of retaining the backbone network structure, the full-dimensional dynamic convolutional ODConv is introduced, ODConv is combined with C2f, and the C2f_ODConv module is constructed to replace C2f of the neck small target layer, which effectively improves the feature extraction ability of small targets. At the same time, the TripletAttention mechanism is embedded to enhance the model's attention to subtle defects.Finally, we introduce the Focaler-IoU loss function. It integrates Focal Loss and dynamically weights difficult samples to improve bounding box regression accuracy.The results show that the mAP of the improved model is increased from 90.6% to 94.2%, and the calculation amount is reduced to 8.8GFLOPs, especially in the "cover" defect detection, the mAP is increased by 9.1%. Compared with other mainstream algorithms, the proposed method has advantages in accuracy and speed, and is suitable for automatic inspection and defect identification tasks of photovoltaic power stations.

Downloads

Download data is not yet available.

References

[1] YAN Fengyuan. Development of Practical Monitoring Circuit For Photovoltaic Modules[D]. Beijing: North China Electric Power University, 2021.

[2] Gielen D, Boshell F, Saygin D, et al. The Role of Renewable Energy in the Global Energy Transformation[J]. Energy Strategy Reviews, 2019, 24: 38-50.

[3] LI Bing, ZHAO Kuan, BAI Yunshan, et al. Multi-scale photovoltaic panel defect detection method incorporating attention mechanism[J]. Electric Power Science and Engineering, 2023, 39(8): pp. 1-10.(in chinses)

[4] Y. Guan, G. Wu, W. Huang, J. et al. Gray Level Co-occurrence Matrix-based Defect Detection Method for Photovoltaic Power Plant Panels[C]. 2023 International Conference on Computers, Information Processing and Advanced Education (CIPAE), 2023, 703-707.

[5] LIAO Kun. Image Recognition and Analysis of Dust Accumulation on Photovoltaic Panels Based on Deep Learning [D]. Northeast Electric Power University, 2020. DOI:10.27008/d.cnki.gdbdc.2020. 000092..

[6] LIU Xiaoxiong, ZHENG Qianying. Research on defect classification and localization algorithm for photovoltaic panels based on deep learning [J]. Optoelectronic Technology, 2024, 44(1): pp. 54-60.

[7] ZHU Chengjie, LIU Lele, ZHU Hongbo.** Multi-scale YOLOv8-MNS Photovoltaic Panel Defect Detection Algorithm [J/OL]. Journal of Chongqing Technology and Business University (Natural Science Edition), 1-9 [2024-11-29]. http://kns.cnki.net/kcms/detail/50.1155.N.20240604. 1406.002.html.

[8] WANG Tao, LI Yuansong, SHI Rui, et al. Photovoltaic cell defect detection algorithm based on improved YOLOv8n [J/OL]. Journal of Chongqing Technology and Business University (Natural Science Edition), 1-12 [2025-07-25]. http://kns.cnki.net/kcms/detail/50.1155.N.20240626.0951.002.html.

[9] ZHANG Zijun.** Research on Defect Detection Method of Solar Cell Silicon Modules Based on Faster RCNN [D]. Xiangtan University, 2024. DOI:10.27426/d.cnki.gxtdu.2024.002519.

[10] YANG Changchun, HE Xuanxuan, WANG Rui, et al.** Photovoltaic Panel Defect Detection Algorithm Based on Improved YOLOv8 [J/OL]. Electronic Measurement Technology, 1-13 [2024-11-29]. http://kns.cnki.net/kcms/detail/11.2175.TN.20241121.1746.066.html.

[11] Zhang Z, Liu S, Xiong J. MLCA-YOLO: improved yolov8 for solar cell defect detection[C]. 2024 6th International Conference on Internet of Things, Automation and Artificial Intelligence (IoTAAI), 2024, 24-27.

[12] GAO Hang, QI Yunsong. Improved YOLOv8n-based defect detection in solar panels [J/OL]. Infrared Technology, 1-7 [2025-07-25]. http://kns.cnki.net/kcms/detail/53.1053.TN.20240923. 0931.002. html.

[13] HUSSAIN M. YOLO-v1 to YOLO-v8, the rise of yolo and its complementary nature toward digital manufacturing and industrial defect detection[J]. Ma-chines, 2023, 11(7): 677.

[14] TANG Yu, MA Xinna, ZHENG Xuepeng, et al. Multi-source domain rolling bearing fault diagnosis based on ODConv [C]. In: Nonlinear Vibration Committee, Chinese Society of Vibration Engineering (Ed.), Abstracts of the 19th National Conference on Nonlinear Vibration & 16th National Conference on Nonlinear Dynamics and Motion Stability. Shijiazhuang: School of Information Science and Technology, Shijiazhuang Tiedao University; State Key Laboratory of Mechanical Behavior and System Safety of Traffic Engineering Structures, Shijiazhuang Tiedao University; 2023: p. 300. doi:10.26914/c.cnkihy.2023.112705.

[15] WENG Junhui, CHENG Le, HUANG Manli, et al. Small target detection in UAV aerial images based on CS-YOLOv5s[J]. Electronic Measurement Technology, 2024, 47(7): pp. 157-162. doi:10.19651/ j.cnki.emt.2315250.

[16] Vergura Silvano, Marino Francescomaria.Quantitative and computer-aided thermography-based diagnostics for PV devices: part I—framework[J]. IEEE Journal of Photovoltaics, 2017, 7(3): 822-827.