Rapid development of semantic 3D city models for urban energy analysis based on free and open data sources and software

Abstract

Geospatial data, specifically semantic 3D building data, plays a crucial role in urban energy analysis as spatial calculations using 3D geometries usually form the basis for energy simulation and modelling needed for numerous smart cities applications. Additional information describing the building stock, such as building materials and energetic properties but also data not directly linked to urban morphology such as weather data, environmental data, vegetation or socio-demographic data sets are required for these applications. A major drawback in the widespread applicability of urban energy analysis is the lack of available data sets as well as the costly and lengthy labor-intensive process of generation of those data sets (e.g. 3D city models or LIDAR data). While recent years have seen an opening up of urban data sets through free and open data portals, web services, and APIs that are used for urban energy analysis, data standardization and varying data quality still raises big challenges. This research explores different methodologies for the generation and usage of semantic 3D city models based on free and open data sources and software. In this paper, we describe four different methodologies for the generation of semantic 3D city models from available open data (geospatial data portals, LIDAR data, Open Street Map data, and remote sensing) and the tools required to achieve the task. To evaluate the suitability of these open-data sets for smart cities applications, multiple energy models, such as an energy performance model and a vertical solar radiation tool, previously developed in EIFER, have been applied to evaluate the applicability of these generated city models.