International Research Journal of Education and Technology

In response to the growing demand for sustainable energy solutions, this study presents a comprehensive framework aimed at improving the performance and longevity of solar panels through damage detection and real-time parameter monitoring. Leveraging machine learning techniques and convolutional neural networks (CNNs), the framework offers a robust solution for accurately identifying various types of damage, including cracks, scratches, and soiling, from images captured by drones or stationary cameras. Continuous model refinement ensures high efficiency and accuracy in the detection process, enhancing the system's reliability.

The proposed framework not only enables early detection of damage or performance degradation but also ensures real- time monitoring of operational parameters, thereby promoting the reliability, efficiency, and sustainability of solar power generation systems. By harnessing the power of machine learning and embedded systems, this approach contributes to the advancement of renewable energy technologies, addressing key challenges in the field of solar energy and paving the way for more efficient and reliable solar power generation.

Keywords – convolutional neural network(CNN), Machine learning(ML), real time.