Defect Identification with Predictive Maintenance for Enhanced Photovoltaic System Performance | ||||
Delta University Scientific Journal | ||||
Volume 7, Issue 3, November 2024, Page 67-77 PDF (572.01 K) | ||||
Document Type: Original research papers | ||||
DOI: 10.21608/dusj.2024.433443 | ||||
![]() | ||||
Authors | ||||
Eman Ashraf ![]() ![]() | ||||
1communication and electronics, faculty of engineering, delta university for science and technology | ||||
2Faculty of Engineering, Delta University for Science and Technology, Gamasa, Egypt | ||||
3Mechatronics Engineering Program, Faculty of Engineering, Mansura University, Mansura, Egypt | ||||
4Mechanical Power Engineering Department, Tanta University, Tanta, Egypt | ||||
5dept. of Communications and Electronics Engineering Nile Higher Institute for Engineering and Technology | ||||
Abstract | ||||
This research paper addresses the critical challenge of ensuring the optimal performance and longevity of photovoltaic (PV) systems through defect detection and maintenance prediction. PV cells play a pivotal role in renewable energy generation, and their efficiency can be compromised by defects. This study presents a novel approach that combines Deep Learning (DL) techniques for defect detection and Machine Learning (ML) models for maintenance prediction. The proposed Convolutional Neural Network (CNN) architecture accurately detects defects in PV cell images, achieving high precision, recall, and F1-score. Furthermore, a separate maintenance prediction model, based on extracted features from the defect detection model, effectively predicts the maintenance requirements of PV cells. The integration of these models provides a comprehensive solution for identifying defects and anticipating maintenance needs in PV systems, contributing to the sustainability and reliability of renewable energy sources. The developed defect detection model achieves exceptional accuracy and precision in identifying defects, while the predictive maintenance model enhances the reliability and cost-effectiveness of maintenance operations. | ||||
Keywords | ||||
Photovoltaic; Maintenance; Defect Detection; Machine Learning; Deep Learning | ||||
Statistics Article View: 50 PDF Download: 29 |
||||