Lightning damage detection on solar panels using portable infrared camera and convolutional neural network for enhancing panel maintenance

As the demand for renewable energy sources increases, the vulnerability of solar panels to lightning strikes becomes a critical concern. This research explores the correlation between lightning-induced voltage fluctuations and the resultant damage intensity on solar panels. Monocrystalline and polycrystalline solar panels were used in this study. A systematic approach was adopted, investigating the correlation between lightning-induced voltage assessment using 30kV, 60kV and 90 kV impulse voltage with multi-stage Marx impulse generator and the damage intensity on these two types of solar panels. The utility of active infrared thermography in capturing lightning-induced damage was also explored. Portable active infrared thermography equipment, tCam- Mini with wireless streaming was employed to conduct this research. Additionally, Convolutional Neural Network (CNN) based image classification techniques were integrated to enhance the efficiency of damage assessment. The application of neural networks allowed for automated and precise identification of lightning-induced damage patterns, improving accuracy and speed of image classification for damaged and undamaged samples. The experimental design involves induced lightning strikes, exposing solar panels to controlled conditions, collecting infrared images and performing neural network-based image classifications. The findings contribute valuable insights into enhancing the resilience of solar panel systems against lightning strikes, ultimately advancing the reliability and sustainability of solar energy infrastructure. A new convolutional neural network model was developed to classify the images obtained from thermography with 90.21% accuracy for grayscale and 85% accuracy on thermal images. The application of this research lies in improving the protection and maintenance strategies for solar panels, ensuring more durable and efficient renewable energy systems.

Text FK 2024 58 - Declaration Form.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (940kB) Text FK 2024 58 -Full Text.pdf Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (3MB) Text FK 2024 58.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (4MB)

Actions (login required)

View Item