5-9 September 2016
Prague Congress Centre
Europe/Prague timezone

P3.200 Improved Solid Decomposition Algorithms for the CAD-to-MC Conversion Tool McCad

7 Sep 2016, 11:00
1h 20m
Foyer 2A (2nd floor), 3A (3rd floor) (Prague Congress Centre)

Foyer 2A (2nd floor), 3A (3rd floor)

Prague Congress Centre

5. května 65, Prague, Czech Republic
Board: 200
Poster J. Power Plants Safety and Environment, Socio-Economics and Technology Transfer P3 Poster session

Speaker

Lei Lu (Neutronics and Nuclear Data Group)

Description

McCad is a geometry conversion tool developed at the Karlsruhe Institute of Technology (KIT) for the automatic conversion of CAD models into the constructive solid geometry (CSG) representation. The resulting geometry models can then be used in Monte Carlo (MC) particle transport simulations applied in design analyses of fusion reactors like the DEMO tokamak developed within the European Power Plant Physics and Technology (PPPT) programme. The conversion of such a CAD model necessitates to decompose complex solids into a collection of disjoint and simple convex solids. The decomposition algorithm implemented previously in McCad turned out to be not very efficient and robust when applied to large and complex geometry models such as the DEMO tokamak. Frequently, programme crashes are encountered, irregular and fragmentized solids are produced, and a lot of CPU time and memory are consumed.To overcome such difficulties, new decomposition algorithms and functions have been developed and implemented in McCad. These include a new splitting surface generation algorithm for the decomposition  of solids with curved surfaces, a new collision detecting algorithm using triangular facets, and a new splitting surface sorting algorithm based on feature recognition techniques. Furthermore, a new software architecture was introduced in McCad together with the new decomposition functions:  These are implemented with an independent kernel module and thus can be integrated into any other CAD platform.This paper describes in detail the new decomposition algorithms, their implementation in McCad and the verification on the example of the European DEMO tokamak. The results show that this advanced McCad version is more efficient and robust and provides more accurate and less complex conversion results. It is thus concluded that the new McCad version is well suited for the conversion of highly complex tokamak models such as the European DEMO or ITER.

Co-authors

Fischer Ulrich (Neutronics and Nuclear Data Group, Karlsruhe Institute of Technology, Eggenstein-Leopoldshafen, Germany) Lei Lu (Neutronics and Nuclear Data Group, Karlsruhe Institute of Technology, Eggenstein-Leopoldshafen, Germany) Qiu Yuefeng (Neutronics and Nuclear Data Group, Karlsruhe Institute of Technology, Eggenstein-Leopoldshafen, Germany)

Presentation Materials

There are no materials yet.