|
چکیده
|
With its amazing developments, artificial intelligence (AI) has become a powerful tool in many fields, including the analysis and classification of data from solar systems, allowing for improved efficiency and performance. The use of renewable energy sources has become more important in recent years due to the depletion of fossil fuel reserves and the environmental effects of their consumption, and their adoption is growing quickly. However, a number of environmental conditions can affect solar power facilities, especially large-scale installations. A substantial decrease in power generation, equipment failure, and increased maintenance expenses can result from factors like dust buildup, bird droppings, physical and electrical damage, and snow cover. Therefore, maintaining optimal performance and prolonging the lifespan of these systems requires careful analysis and condition monitoring. Transformer-based Deep Neural Networks (DNNs) have become innovative and reliable tools for the classification and intricate analysis of data obtained from solar equipment in this regard. These models, which can process spatiotemporal data simultaneously, make it easier to find latent patterns and offer precise insights for the best possible decision-making. Increased energy efficiency, lower maintenance costs, and improved system reliability can all be substantial benefits of using transformers. Using a hybrid dataset gathered from Kaggle and online sources, Transformer-based DNNs were used in this study to classify solar system conditions with an accuracy of 98.82% across multiple classifications. These outcomes show how effective this method is at classifying and analyzing data, providing useful information for improving the operation of the solar system.
|