Thermal surrogate modelling for the monitoring of actively cooled first wall components in ITER

During ITER operation, Diagnostic First Walls (DFWs) will be exposed to intense radiative and nuclear heat loads, making active cooling mandatory. This work presents the adaptation of a mesoscale Porous-Media Approach (PMA) to the thermal modelling of actively cooled DFWs, aiming to support their quasi-real-time thermal monitoring. In this approach, the DFW is represented as an equivalent homogeneous medium, while the effects of the cooling network are incorporated through closure terms in the volume-averaged enthalpy transport equation. These terms are derived from coolant properties and cooling layout geometry. The resulting surrogate model is integrated into an end-to-end framework within the ITER Tokamak System Monitor (TSM). The PMA formulation is implemented in the FEM-meshfree MkniX solver to compute the slice-wise steady-state temperature solutions and a signed distance function-based 2D-to-3D linear interpolation strategy reconstructs the full 3D temperature field while capturing smooth geometrical transitions between adjacent reference planes. The method is demonstrated on plasma-facing components with complex cooling networks and evaluated quantitatively on one Equatorial Port Plug 12 DFW, where the surrogate model reproduces plasma-facing surface hot spots, far from apertures, within maximum relative errors of about 7% compared to CFX results. The 2D steady-state solutions converge in O(10 s) while the 3D interpolation takes less than a second. This seminal study supports the potential use of the MkniX-PMA approach for quasi-real-time temperature monitoring of ITER actively cooled first wall components.