MaCFP4_Follow-up_Compact modeling of convective heat transfer

[halbulqini]

Hello,

In MaCFP4 workshop, the problem of simulating parallel panel fires was discussed. One of the problems was the divergence of the simulation results away from the experimental data when refining the mesh. An explanation for that made by the presenter was the inaccurate description of convective heat transfer because the correlations in heat transfer textbooks assumes constant fluid temperature near the wall/constant wall heat flux, and this is not the case for parallel panel fire scenarios.

Relating to that, I previously came across a couple of papers that addressed a similar problem, and proposed a more general compact modeling of convective heat transfer taking into account these variations.

The papers addressed the laminar flow case, but I think there may be something useful in these papers that could be built on by the FDS developers to better model the convective heat transfer.

Here are the links to the papers:
1- https://doi.org/10.1016/j.ijthermalsci.2026.110810
2- https://doi.org/10.1016/j.ijheatmasstransfer.2017.04.129

I hope this could help, and I am looking forward to hearing your opinion about this idea.

Thank you.

[mcgratta]

I took at look at these papers. There are many papers like this in the heat transfer literature, where analytical solutions to the boundary layer transport equations are applied to very specific scenarios, like laminar flow in a duct, for example. However, in FDS, we cannot expect the user to pick and choose amongst the dozens of approaches. We need something that will work reasonably well in the majority of cases. If I am misreading these papers, could you tell me how, specifically, we could implement this technique in FDS, keeping in mind that grid cells are typically centimeters to tens of centimeters in size, we have one average temperature in each grid cell, and some very rough idea of the level of turbulence in the boundary layer.