Automated Defects Classification of Photovoltaic Modules Using YOLOV7 Model

The widespread adoption of photovoltaic (PV) systems for renewable energy generation has highlighted the importance of maintaining optimal performance and reliability of PV modules. The classification of defects in PV modules is crucial for effective maintenance and improved system performance. Elec...

Full description

Saved in:
Bibliographic Details
Published in2023 Innovations in Power and Advanced Computing Technologies (i-PACT) pp. 1 - 5
Main Authors Ng, WeiLun, Khairuddin, Anis Salwa Mohd, Shah, Noraisyah Mohamed, Omar, Azimah, Ali, Mohd Syukri, Rahim, Nasrudin Abd
Format Conference Proceeding
LanguageEnglish
Published IEEE 08.12.2023
Subjects
Computational modeling
convolutional neural network
deep learning
forecasting
Inspection
machine learning
Photovoltaic (PV)
Photovoltaic systems
Renewable energy sources
System performance
Technological innovation
YOLO
Online AccessGet full text

Cover

More Information
Summary:The widespread adoption of photovoltaic (PV) systems for renewable energy generation has highlighted the importance of maintaining optimal performance and reliability of PV modules. The classification of defects in PV modules is crucial for effective maintenance and improved system performance. Electroluminescence (EL) imaging is one of the important techniques for the inspection of PV modules. This work aims to develop a deep learning-based classification model to categorize defects in photovoltaic modules. This work proposed a YOLOv7 model for automated classification of defects on PV modules. The experimental results demonstrated the proposed YOLOV7 model achieved mAP of 99% in classifying the PV defects into 3 classes (mode A, mode B and mode C).
DOI:10.1109/i-PACT58649.2023.10434429