Improving Solar Power Generation with InceptionV3 Dust Detection on the Solar Panel Energy Systems

As the demand for renewable energy increases, solar (PV) innovation has become a matter of concern. Diverse research proposals have been developed to derive the most significant benefit from the sun's rays, but dust gathering on solar panels and air pollution are two significant challenges. The installation of 40GW of residential solar panels and solar capacity connected to the grid is a significant accomplishment for the Indian government. Technology development to manage solar panel-based energy generation has been crucial for increasing reliability and decreasing costs. As a renewable energy source, solar panels produce zero pollution while generating electricity. When dust accumulates on the surface of the solar panel, the intensity of the light that falls on it is decreased. In turn, this reduces the amount of electricity the solar panel generates. Since solar panel cleaning is essential, continual surveillance and assessment are required to optimize these processes. This highlights the significance of using machine learning or deep learning to detect dirt and clean solar panels to enhance performance. In this research, we implemented dust detection of solar panels using the InceptionV3 model and obtained a remarkable 92.34 percent accuracy. The dataset was obtained from Kaggle and consisted of 1,200 images evenly divided between clean and dirty panels. Each image is 100 pixels by 100 pixels in size. The images have been assigned to the Clean and Dusty categories.