OLYMPE, a multi-physic platform for fusion magnet design: development status and first applications

Abstract In the framework of design studies for tokamak superconducting magnets a first step is usually required of pre-dimensioning that provides a rough magnets dimensioning, followed by detailed analyses. Such analyses aim at verifying that predefined design criteria are met in normal and in off-normal operation conditions, in order to ultimately progress step by step towards the most optimized configuration with respect to defined factors of merit of the tokamak. These studies involve several physics and thus several specific numerical codes modelling those physics. In order to handle these process analyses as efficiently and reliably as possible, the development of a multi-physic platform is underway at CEA. This platform, called OLYMPE, integrates different solvers under the supervision of a single actor for interfaces management. Furthermore, the availability of an integrated tool such as OLYMPE also aims at providing a possibility to easily run parametric studies in an automatic mode, to access to more efficient methods for confrontation with experimental data. In this paper, the structure of the platform is described, detailing the different codes and the coupling between the codes. The development status of these different codes included in OLYMPE workflow will also be exposed. Another part of the paper will present the application of the two most advanced OLYMPE modules (electromagnetic and thermal-thermohydraulic) on two projects. The first application is about the modelling of the experimental tests performed on the JT-60SA Toroidal Field coils tested in the Cold Test Facility. Parametric studies on some defined parameters are performed to fit the experimental data (e.g. impact on the thermal loads of the winding pack detachment from the casing, effect of pancakes friction factor variability…). The second application focuses on the CEA design proposal for EU-DEMO Toroidal Field (TF) magnets, where other parametric analyses will be provided.