Based on the Repository Site Selection Act (StandAG) Germany is currently searching for a repository site for high-level radioactive waste. Therefore investigations are conducted in deep geological formations (depth > 300m) which are expected to provide the best possible safety. The planning of a deep repository is based on engineering models, whose validity must be permanently verified by measurements of changes in state during the excavation of the underground cavities, the emplacement operation and the closure phase. As part of a self-learning process, the data obtained offers the possibility of recognizing undesirable developments and, if necessary, deriving consequences from them.
The overall goal of the research project is to develop a machine learning-based modeling method for high-level-radioactive-waste disposal systems, which enables an optimization process during the planning phase and a calibration process during the excavation phase in the area of geomechanical and geotechnical issues. This method is being developed as an example for the virtual demonstrator “Emplacement drift with backfill and sealing structure in rock salt”.
State: 07.06.2024
01.05.2023 - 30.04.2026
Funding code: 02E12102
2024
Paul, L.; Fiaz, U.; Stahlmann, J.:
SEMOTI - Entwicklung einer selbstlernenden Modellierungsmethodik für die Einlagerungsstrecke eines Tiefenlagers
In: Röhlig, K.-J. (Hrsg.). 4. Tage der Standortauswahl 18./19.04.2024 in Goslar. https://doi.org/10.21268/20240416-1, S.126-127, Clausthal-Zellerfeld, 2024.
Univ.-Prof. Dr.-Ing. Joachim Stahlmann
Tel.: +49 (0)531 391 62000
j.stahlmann(at)tu-braunschweig.de
Lennart Paul, M.Sc.
Tel.: +49 (0)531 391 62016
lennart.paul(at)tu-braunschweig.de
Umer Fiaz, M.Sc.
Tel.: +49 (0)531 391 62013
umer.fiaz(at)tu-braunschweig.de