Year: 2025
Type: article
Abstract: Safe and efficient operation of photovoltaic (PV) solar panels depends on early defect detection during manufacturing. ‘Bright spots’ on Electro-Luminescence (EL) images of Photovoltaic (PV) solar panels are critical defects, leading to excess energy production, short circuits, overheating, and potential fires. After extensive benchmarking against state-of-the-art methods, this paper proposes a robust approach for reliable bright spot detection based on image classification using novel features and synthetic bright spot EL images generated by generative adversarial networks (GANs). The proposed approach achieved an area under the precision-recall curve (AUPRC) of 0.9526 signifying its real-world applicability for reliable bright spot detection in PV solar panels. CCS CONCEPTS • Computing methodologies • Machine learning approaches.
Source: Discover Data
Authors: Rahul Nair, Anand Thirwani, Kedar Kulkarni
Institution: Reliance Industries (India)
Language: English
Subfield: Industrial and Manufacturing Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $1,390
OpenAlex ID: W4415964907