Interpretation of Remote Sensing Data for Energy Performance Assessments of Existing Buildings

Abstract

Energy usage in buildings, particularly for heating, contributes significantly to ongoing global warming. The urgently required reduction of this contribution by countermeasures like energy system transformations energy retrofits of existing buildings is progressing unsatisfactorily slowly so far. This is also caused by a lack of knowledge about the conditions of the building stock. This dissertation investigates how automated interpretation approaches for remote sensing data can contribute to creating and/or improving building energy simulation models on different scales. On district level, a case study quarter is used to demonstrate how 3D building models can enhance the evaluation of aerial infrared thermography for deriving surface temperatures of façades and roofs. Additionally, TomoSAR satellite data are presented as a possible way to generate information about which attics in a quarter are heated. For application on single buildings, the methods developed for the dissertation and in the project it was associated with enable a fast, automatable, and hence inexpensive overall approach to estimate heat demands and whole building heat transfer coefficients pre- and post-retrofit from drone imagery. For this purpose, modelling software was developed, open-source tools for building energy simulation were adapted, and the approach was applied to a case study building for demonstration. The resulting model was able to reproduce the temperatures during an actual measurement campaign on the building with satisfying accuracy. Furthermore, this work comprises material investigations that connect permittivity and thermal parameters of common building materials. With the measured values, a method for identifying the interior structure of walls using microwave radar can additionally deliver insights about U-values.