Advanced

Osäkerhet vid skalning av brandtekniska försök

Lilja, Oscar LU (2014) In LUTVDG/TVBB VBRM01 20142
Division of Fire Safety Engineering
Abstract
The aim of this report was to investigate the effects on temperature when scaling down an experiment within the area of fire safety engineering. This was done by performing a literature review and by comparing results from FDS 6 simulations in full scale with simulation results in scale 1:4.
To properly scale an experiment it is necessary to take the fire power as well as the heat losses from the model into account, and also making sure that the flow is turbulent as it would be in full scale. The magnitude of the error on temperature caused by scaling is influenced by how much you scale the model, the experimental set up and where you make the measurements. From the literature review and the simulations it was possible to notice some... (More)
The aim of this report was to investigate the effects on temperature when scaling down an experiment within the area of fire safety engineering. This was done by performing a literature review and by comparing results from FDS 6 simulations in full scale with simulation results in scale 1:4.
To properly scale an experiment it is necessary to take the fire power as well as the heat losses from the model into account, and also making sure that the flow is turbulent as it would be in full scale. The magnitude of the error on temperature caused by scaling is influenced by how much you scale the model, the experimental set up and where you make the measurements. From the literature review and the simulations it was possible to notice some patterns. A model of a compartment fire before flashover, scaled down by a factor of 2, seems to have an error of around 10 %. Scaled down by a factor of 4 the error lies between 10 % and 20 %. In an adjoining room the error may be larger, up to 30 %. In the lower, cooler part of the room the error is smaller and overall the temperatures in scaled down models seem to be underestimating the temperatures. (Less)
Please use this url to cite or link to this publication:
author
Lilja, Oscar LU
supervisor
organization
alternative title
Uncertainty caused by the scaling of fire experiments
course
VBRM01 20142
year
type
M2 - Bachelor Degree
subject
keywords
Scaling, Froude scaling, uncertainty, functional analysis, FDS, skalning, froude skalning, osäkerhet, funktionalanalys
publication/series
LUTVDG/TVBB
report number
5476
ISSN
1402-3504
language
Swedish
id
5046188
date added to LUP
2015-02-09 13:32:43
date last changed
2015-02-09 13:32:43
@misc{5046188,
  abstract     = {The aim of this report was to investigate the effects on temperature when scaling down an experiment within the area of fire safety engineering. This was done by performing a literature review and by comparing results from FDS 6 simulations in full scale with simulation results in scale 1:4.
To properly scale an experiment it is necessary to take the fire power as well as the heat losses from the model into account, and also making sure that the flow is turbulent as it would be in full scale. The magnitude of the error on temperature caused by scaling is influenced by how much you scale the model, the experimental set up and where you make the measurements. From the literature review and the simulations it was possible to notice some patterns. A model of a compartment fire before flashover, scaled down by a factor of 2, seems to have an error of around 10 %. Scaled down by a factor of 4 the error lies between 10 % and 20 %. In an adjoining room the error may be larger, up to 30 %. In the lower, cooler part of the room the error is smaller and overall the temperatures in scaled down models seem to be underestimating the temperatures.},
  author       = {Lilja, Oscar},
  issn         = {1402-3504},
  keyword      = {Scaling,Froude scaling,uncertainty,functional analysis,FDS,skalning,froude skalning,osäkerhet,funktionalanalys},
  language     = {swe},
  note         = {Student Paper},
  series       = {LUTVDG/TVBB},
  title        = {Osäkerhet vid skalning av brandtekniska försök},
  year         = {2014},
}