Non-gray radiation modeling of methanol pool fires using the RC-FSK method in FDS

This study presents a high-fidelity modeling framework for capturing non-gray radiative heat transfer in methanol pool fires. A global non-gray radiation model, based on the Rank Correlated Full Spectrum k -distribution (RC-FSK) method was implemented as a subroutine within the Fire Dynamics Simulator (FDS). The simulations were conducted in decoupled and coupled computational fluid dynamics configurations, comparing the performance of RC-FSK to gray approach from RadCal. In the decoupled configuration, RC-FSK predictions showed strong agreement with photon Monte Carlo line-by-line (PMC-LBL) benchmark, significantly improving the accuracy of gas-phase radiative source term, radiative emission and radiative heat flux estimates. In coupled simulations, a two-zone subgrid approach was used in conjunction to RC-FSK and RadCal to model turbulence–radiation interaction. Both models reproduced the main flow features; however, RC-FSK predicted higher emission values within the flame core, resulting in a more pronounced radiative cooling effect. It also yielded elevated absorption levels, particularly in absorption-dominant regions. RadCal was shown to under-predict both radiative emission and absorption, producing a compensating effect that leads to acceptable agreement in the global integrated quantities.