Work

Assessment of the Solar Potential of Buildings Based on Photogrammetric Data

Year: 2025

Type: article

Abstract: In recent years, a growing demand for alternative energy sources, including solar energy, has been observed. This article presents a methodology for assessing the solar potential of buildings using images from Unmanned Aerial Vehicles (UAVs) and point clouds from airborne LIDAR. The proposed method includes the following stages: DSM generation, extraction of building footprints, determination of roof parameters, map solar energy generation, removing of the areas that are not suitable for the installation solar systems, calculation of power per each building, conversion of solar irradiance into energy, and mapping the potential for solar power generation. This paper describes also the Detecting Photovoltaic Panels algorithm with the use of deep learning techniques. The proposed algorithm enabled assessing the efficiency of photovoltaic panels and comparing the results of maps of the solar potential of buildings, as well as identifying the areas that require optimization. The results of the analysis, which had been conducted in the test areas in the village and on the campus of the university, confirmed the usefulness of the above proposed methods. The analysis provides that the UAV image data enable generation of solar potential maps with higher accuracy (MAE = 8.5 MWh) than LIDAR data (MAE = 10.5 MWh).

Source: Energies

DOI: 10.3390/en18040868

Authors: Paulina Jaczewska, Hubert Sybilski, Marlena Tywonek

Institution: Military University of Technology in Warsaw

Language: English

Topic: Solar Radiation and Photovoltaics

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

SDGs: Affordable and clean energy, Sustainable cities and communities, Climate action, Industry, innovation and infrastructure

Open Access status: gold

Estimated APC paid: $3,137