Thunderhead Engineering Forum

Public Forum => PyroSim => Topic started by: edwardomull on February 04, 2012, 03:11:53 pm

Title: Wnd Driven Fire
Post by: edwardomull on February 04, 2012, 03:11:53 pm
I am trying to model an actual Wind Driven Fire that occurred in Glasgow. This is part of my dissertation for my honours degree. The fire was so intense that fire-fighters could not access the apartment to extinguish the fire. After an hour during the decay phase access was gained and the remnants of the fire extinguished. I am modelling two scenarios, one with wind and one without so that a comparative analysis can be obtained.

The wind was 45mph and the model has the fire starting in the living room which has a three seater settee and two armchairs as the fuel load. Two bedrooms have beds and bedside cabinets. The model has been run for twenty minutes and I am trying to determine the conditions that were faced by fire crews.

I think the problem might be that the windows are failing before the flashover. The windows are set to fail at 200 degrees.

•         All windows fail @ ~78 seconds (I believe that this could be better modelled if each had its own separate sensor control mechanism.)
•         ‘Flashover’ in room at 380s with no wind. (peak 30MW, then unrealistic decay to 2.5MW within 3 minutes)
•         ‘Flashover’ in room at 658s with wind. The wind dissipates the hot layer around the apartment and increases turbulent cooling. So takes longer for hot gases to accumulate and other surfaces to heat up. (Peak of ~35MW decaying to 8MW within 60 seconds)

Modelling this WDF is proving to be difficult and any help or assistance would be greatly appreciated.