Sign in Subscribe
Renewable energy

Advancing Renewable Energy Solutions with Deep Learning and Residual Network

Strategic Insights

— 1 min read

Research from South Valley University has proposed a novel mechanism for accurate cracking detection in photovoltaic (PV) panels using Electroluminescence (EL) images. The mechanism, based on Deep Learning (DL) and Residual Network (ResNet), was tested on a large PV power dataset composed of 2000 EL images collected from different polycrystalline and monocrystalline cells. The research demonstrates the application of advanced DL frameworks for early defect diagnosis, enhancing PV panel maintenance and bolstering the sustainability of solar systems.

Key Takeaways:

  • The proposed mechanism uses ResNet-based image processing to detect cracks in PV panels, achieving an F1-Score of 86.63% with ResNet34, 87.37% with ResNet50, and 88.89% with ResNet152.
  • The research focuses on the design of an efficient crack detection system trained on a large PV power dataset, composed of 2000 EL images collected from different polycrystalline and monocrystalline cells.
  • The dataset was split into training (70%), validating (20%), and testing (10%) subsets to guarantee the presence of many cell states in each subset.
  • The research has significant implications for the renewable energy sector, offering practical solutions for defect diagnosis and maintenance of PV panels.
  • The technology preserves the efficiency of solar modules and encourages clean energy solutions by accurately identifying PV panel faults.
  • The study lays a foundation for the further development of image-based defect detection methods in PV systems.

Statistics:

  • 2000 EL images were collected from different polycrystalline and monocrystalline cells.
  • The dataset was split into training (70%), validating (20%), and testing (10%) subsets.
  • The proposed mechanism achieved an F1-Score of 86.63% with ResNet34, 87.37% with ResNet50, and 88.89% with ResNet152.
  • 3 different ResNet architectures (ResNet34, ResNet50, and ResNet152) were tested and compared in terms of detection performance and computational performance.

Sources:

  • ResNet-based image processing approach for precise detection of cracks in photovoltaic panels. Scientific Reports, 2025;15(1):24356.
  • South Valley University, Montaser Abdelsattar, Electrical Engineering Department, Faculty of Engineering, South Valley University, Qena, 83523, Egypt.

Read more