|Digitally optimized sectioning and packing of activated concrete structures into Konrad-Containers when dismantling nuclear facilities
|Bundesministerium für Bildung und Forschung
|01.10.2022 – 30.09.2025
The decision to shut down all nuclear power plants in the Federal Republic of Germany is accompanied by the necessity to dismantle the decommissioned nuclear power plants properly. Since activated components are to be expected, especially in the area around the reactor, special conditions must be observed for the dismantling and final disposal of radioactive waste with negligible heat generation. In order to store this waste in the “Schacht Konrad” repository near Salzgitter, Germany, it is packed in containers. These filled containers must fulfil specified boundary conditions. The storage of the containers is associated with high costs; moreover, there is only limited space available in the final repository.
In order to make maximum use of the filling volume of a container, the dismantling and packaging is to be optimized and simulated in advance with the help of digital methods such as Building Information Modelling (BIM) and the use of game engines. This procedure will be tested using the example of a biological shield. This is first created as a BIM model and then enriched with information regarding the material and the specific activity of the isotopes. The BIM model is then imported into a game engine, where it is decomposed using the AI method reinforcement learning. This AI approach needs to be optimized, taking into account the framework conditions and maximizing the storage capacity of the containers.