Recent Papers

The detection of faults in solar panels is essential for generating increased amounts of renewable green energy.Solar panels degrade over time due to physical damage, dust, or other faults. Numerous studies have been conducted to detect and monitor solar panel faults in real-time .To avoid these problems ,I present a fault detection model using Vision Transformers that make use of self-attention to capture the global and local dependencies in solar panel images. Even though deep learning models, especially convolutional neural networks, have been showing good prospects in this regard, recent developments in Vision Transformers may provide a different pathway toward increasing accuracy and robustness.The model is pre-trained on large-scale datasets and fine-tuned on solar panel images for the detection of various faults like physical damage, dust ,micro-cracks, dust deposition, and delamination. It is further empowered with multi-head self-attention layers, positional encoding, and layer normalization, followed by dense layers with LeakyReLU activation and batch normalization. Overall, the Vision Transformer-based approach outperforms traditional deep learning models by offering superior accuracy, precision, and robustness in detecting and classifying faults in solar panels. This method not only enhances the reliability of fault detection but also contributes to the optimization of solar energy systems, ultimately supporting the generation of more sustainable and efficient renewable energy.

Copyright © 2024 Jahnavi Allamsetty. This is an open access article distributed under the Creative Commons Attribution License.