Determination of rooftop photovoltaic energy potential using satellite images

Today, the increase in the demand for energy with the development of technology requires policies to increase the use of renewable energy sources. Renewable energy sources are unlimited and clean sources and reduce foreign dependency in energy. Such factors form the basis of the demand for renewable energy sources. Factors such as low installation cost, ease of maintenance and variety of installation areas make photovoltaic systems prominent among renewable energy sources. In this context, it's important to determine the potential of the photovoltaic systems. In this thesis study, the PV energy capacity is determined for the roofs on the buildings by image processing method using satellite images. Satellite images of different zoom levels and sizes were obtained with latitude-longitude values used as geolocation information. The images obtained in the study evaluated with two methods: HSV color model method and K-Means algorithm filtering methods. In the first method, the hue, saturation and value values of the images are brought to optimum levels, and a suitable image is obtained for the application of separation by thresholding. The roof edges determined and drawn by taking the determined location as the center after thresholding. The PV capacity of the location tried to be estimated over the area within the determined edges. In the second method, after applying the K-Means filtering algorithm to the images, the roof area waa determined and the PV capacity tried to be determined. In order to evaluate the applicability of the methods, images from 5 different locations obtained and filters applied. The accuracy of the filtering methods compared with the results obtained with the PV*SOL simulation program and the efficiency of the methods tested. As a result, it is expected that the methods applied in the study will provide significant benefits to personal users in terms of effectively determining the rooftop photovoltaic capacity of their locations.