Start your career with an exciting master thesis at the DLR Institute for Solar Research!

Minimizing CO2 emissions from heating existing buildings is a central task of the energy transition. Reliable information on the building stock is needed for the development of renovation strategies. However, the current state of the art lacks measurement techniques that are fast, cheap and automatable and thus attractive for building owners, managers and energy consultants.

In this master's thesis, you will develop a data-driven calibration procedure for the 5RC building model that is used to automate the optimization of retrofit measures. The 5RC model enables thermal simulation of buildings that can be linearly optimized, is based on less input data, and is significantly faster than conventional methods. However, for certain types of buildings there is a deviation in the calculated heat load profiles between the 5RC building model and conventional methods, which thus reduces the quality of the analysis of refurbishment strategies. The goal is to minimize these deviations by calibrating the 5RC building model with building databases for reference buildings such as Tabula using conventional simulation methods, such as TEASER.

Data-driven approaches will be independently researched, selected, and applied for calibration. In addition, there is an opportunity to develop a method to automatically calibrate the model for buildings that deviate from the database and apply it to a real building. Data-driven approaches and/or machine learning methods may be used for this purpose.

You will be closely integrated into our diverse and highly motivated team in the "Buildings and Neighborhoods" group at the Jülich site. Experienced colleagues will work closely with you and support you throughout the master thesis. The topic of the thesis will be coordinated with the responsible professor at your university. Theses at the Faculty of Mechanical Engineering at RWTH Aachen University are supervised directly by our institute directors.