Determining the potential of rooftop solar energy systems in van province with machine learning supported satellite image analysis

Energy production from favorable natural resources is of great importance in line with sustainable development goals. In this context, the effective use of renewable energy sources plays a critical role in reducing environmental problems caused by fossil fuels. Solar energy stands out among renewable energy sources with its unlimited potential and environmentally friendly structure. However, in order to utilize solar energy efficiently, it is necessary to accurately identify potential regions. In recent years, innovative methods such as machine learning and satellite image analysis have become powerful tools for identifying solar energy potential. In this study, a new model is proposed and a comprehensive analysis method is presented to identify areas suitable for rooftop photovoltaic (PV) panel installation in Van province. Combining high-resolution satellite imagery and machine learning techniques, this approach analyzes the physical characteristics of rooftops, solar radiation potential and environmental factors in detail. Satellite imagery and geospatial datasets are integrated, and a comprehensive classification is applied on top of these data. Using Microsoft Building Footprints dataset and RGB histogram analysis, a total of 85,298 roof areas were evaluated with image processing methods and classified in terms of suitability with the help of Random Forest algorithm. In addition, using PVGIS, the potential installed capacity was calculated as 558.27 MW and the total annual energy production as 566928.80 MWh and it was revealed that Van province has a significant potential that can provide regional energy security and economic benefits. At the same time, based on the findings obtained, an annual CO2 reduction of 255117.95 tons was calculated. The study provides a strategic roadmap for a more effective utilization of the solar energy potential in Van province. It is a critical guide especially in terms of increasing energy independence and aligning local governments with sustainable energy policies. It is emphasized that these methods can be applied for potential analyses in other regions.