Upp í vindinn - 01.05.2009, Blaðsíða 38

Evaluating the fire safety performance
Case Population
Total
1 4200
2 3700
2* 3700
3 2500
3* 2500
Evacuationtime
t (min.)
evac ' '
2ndfloor P'floor
3:30°** 7:50°
3:30°** 7:10°
3:30°** 9:20°
3:30°** 5:20°
3:30°** 7:00°
3 vents
2nd floor P'floor
1.57 1.60
1.57 1.74
1.57 1.34
1.57 2.34
1.57 1.79
No vents
2ndf1oor P’floor
1.33 1.09
1.33 1.19
1.33 0.91
1.33 1.59
1.33 1.21
Criticaltimest „
crit
3 vents 5:30° 12:30°
Novents 4:40° 8:30°
(*) 1 exitblocked
(**) estimated time
Table 2: Population, evacuation times and safety factors for various configuration cases.
the evacuation routes. Therefore, engineers
should take a safety factor higher than 1. How
high depends mainly on how conservative
were the choices made by the engineer for the
simulation.
The evacuation is simulated for different
number of occupants (see Table 2) in zone
1 (see Figure 4), but with fixed number of
occupants in Zones 2 and 3 (100 and 600).The
corresponding evacuation times are compared
to the critical times defined earlier, and K ,
is calculated. As recommended by guidelines,
tests with one of the exit doors blocked are
also simulated (case 2* and 3*).
The results show that the safety is not sufficient
in the worst case (Ksafc = 0.91 < 1) for 3700
people if vents are not functioning and one
exit door closed (case 2*). For 2600 people, the
safety level is sufficient (K ifc > 1.2), even with
the 2 failures in the safety system.
Prescriptive based methods based on
recommendations from building regulations
[9] allow calculating the maximum number of
people as a function of the size and number of
emergency exits. Applying this method to the
skating hall gives a maximum close to 1800
people.
Conclusion
In this article, we have described the fire
safety investigation process for new or existing
Figure 4: Evacuation modeling in case 2: population at t= 0 on 1 st (left) and 2nd floor (right).
Figure 5 : Evacuation modeling in case 2: population at t= 3 min. on 1 st (left) and 2nd floor (right).
building, using both CFD and evacuation
models. As an example, the fire safety of the
skating hall in Reykjavik is tested in a very
different configuration than the intended use
was, with a rock concert hosting about 3 times
the normal population of the hall.
The fire and smoke simulation carried out
using the CFD software FDS gave the critical
time for evacuation, which was then compared
to results from an evacuation model built with
BuildingExodus, tested for various populations
and exit conditions. It appeared that the
evacuation time remained smaller than a
critical time based on visibility conditions,
showing that both thc smoke exhaustion
system in the roof and the emergency routes
created a sufficient level of safety in the
case safety devices work properly. However,
failures must be considered while assessing
the safety of a building, and authorities and
fire safety engineers should always make
conservative choices and carry out sensitivity
analysis to account for possible unexpected
problems. The performance based evacuation
simulation suggested that the level of safety
wasn’t sufficient for 3600 people (case 2) in
the likely case of 2 failures in the system. The
prescriptive method suggested a significantly
lower value close to 1800 people, but the safety
level for 2600 people (K fc <1.2; lVz minute
safety delay) was confirmed to be sufficient
considering the conservative choices made
during the simulations.
REFERENCES
[1] Karlsson B., Quintiere J.G. (2000):"Enclosure
Fire Dynamics", CRC Press.
[2] Drysdale D.D. (1992): "An Introduction to
Fire Dynamics", Wiley-lnterscience.
[3] SFPE Handbook of Fire Protection
Engineering, 2nd ed. (1995): National Fire
Protection Association.
[4] Fire Dynamic Simulator FDS and
Smokeview homepage: http://www.fire.nist.
gov/fds/
[5] BuildingExodus homepage, http://fseg.gre.
ac.uk/exodus/
[6] Bryan J.L. (1995):"Behavioral response to
fire and smoke", SFPE Handbook of Fire
Protection Engineering", 2nd ed., National
Fire Protection Association.
[7] PaulsJ.(1995):"Movementof people",SFPE >
Handbook of Fire Protection Engineering",
2nd ed„ National Fire Protection
Association.
[8] Nelson H.E, MacLennan H.A. (1995):
"Emergency movement", SFPE Handbook
of Fire Protection Engineering" 2nd ed.,
National Fire Protection Association.
[9] Byggingarreglugerð 441/1998.
38 I ... upp í vindinn