BRI2002: TWO LAYER ZONE SMOKE TRANSPORT MODEL

1
BRI2002TWO LAYER ZONE SMOKE TRANSPORT MODEL

• TANAKA Takeyoshi
• DPRI, Kyoto University
• YAMADA Shigeru
• Fujita Corporation

2
INTRODUCTION
CONTENTS
1 OUTLINE OF THE MODEL 1. 1 CONCEPTUAL MODEL
OF THE MODEL 1. 2 MATHEMATICAL DESCRIPTION OF
ZONE PHYSICS
1. 3 COMPONENT PROCESS MODELING 1. 3. 1
Combustion and Heat Release 1. 3. 2 Species
Generation Due to Combustion 1. 3. 3 Burning
Rate Of Gasified Fuel 1. 3. 4 Opening Flow
Rate 1. 3. 5 Fire Plume Flow Rate 1. 3. 6
Opening Jet Plume Flow Rate 1. 3. 7
Penetration Of Fire And Opening Jet Plumes Into
Layers 1. 3. 8 Thermal Radiation Heat
Transfer 1. 3. 9 Convective Heat Transfer
1. 3. 10 Thermal Conduction In Walls 1. 3. 11
Efficiency Of Mechanical Smoke Extraction
3
2 OUTLINE OF THE COMPUTER PROGRAM 2. 1
STRUCTURE OF THE PROGRAM 2. 2 MAIN PROGRAM
2. 3 DATA I/O SUBPROGRAMS 2. 4 COMPONENT
PHYSICS SUBPROGRAMS 2. 5 NUMERICS SUBPROGRAMS
3 USE OF THE PROGRAM 3. 1 EXECUTION OF THE
PROGRAM 3. 2 OUTPUT FILES 3. 3 DATA
INPUT FORMAT 3. 4 SAMPLE CALCULATIONS
4 LIMITATIONS AND FUTURE ISSUES 4.1
LIMITATION ON PHYSICS 4.2 PROBLEMS IN
APPLICATIONS
4

• It is no more than recent 15 years since that a
  variety of new smoke control methods, which are
  not prescribed in the building codes, has begun
  to be introduced into actual buildings in Japan.

• In performance-based smoke control designs, some
  engineering tool that can predict the smoke
  behavior with a reasonable accuracy under a
  prescribed design fire condition is indispensable.

• For this purpose, two layer zone models,
  particularly multi-story, multi-room smoke
  transport model BRI2 have been extensively used.

5

• On the other hand, more than 15 years have
  already passed since the first version of BRI2
  was made available from Building Center of Japan.

• The understanding has significantly progressed in
  several aspects of fire owing to the active
  research during this period.

• So it is thought to be appropriate to take the
  opportunity of reprinting the manual to revise
  the model taking into account the new research
  results and more convenience for users.

6
Mechanical smoke extraction
Mechanical air supply
Upper layer
Room i
Room j
Fire plume
Opening jet plume
Lower layer
Schematic of Two Layer zone Model
7
(a) any space in a building is filled with an
upper and a lower layers
(b) the upper and the lower layers are distinctly
divided by a horizontal boundary plane
(discontinuity)
(c) each of the layer is uniform with respect to
physical properties by virtue of vigorous mixing
(d) mass transfer across the boundary of a layer
occurs only through a fire plume, doorjets and
doorjet plumes
(g) radiation heat transfer between rooms is
neglected.
8
(e) heat transfer across a layer boundary occurs
by the radiation heat exchange among the layers
and the boundary surfaces the convective heat
transfer between a layer and the wall surface
contacting with the layer as well as that
associated with the mass transfer referred in (d)
(f) all the heat released by a fire source is
transported by the fire plume, in other word, the
flame radiation loss is neglected.
(g) radiation heat transfer between rooms is
neglected.

9

• Zone Conservation and State of Gas

10
Combustion product
Combustion
Residual char
Ts Residual char
Gasified fuel
Fuel at elevated temp
Tp Gasification (Pyrolysis)
H2O(v)
Gasification
Temp. rise
H2O
H2O(v)
100?
Original fuel
T0 Initial temp.
11
(No Transcript)
12
(No Transcript)
13
Opening Flow Rate
14
Fire Plume Flow Rate
15
Penetration of Fire and Opening Jet Plumes Into
Layers
16
Radiation Heat Transfer
17
Convective Heat Transfer
18
Efficiency of Mechanical Smoke Extraction
19
Heat Conduction in Wall
qout 0 BRI2002 does not chase the heat
transferred to an adjacent room through a wall
20
(No Transcript)
21
LIMITATIONS AND FUTURE ISSUES
(1) LIMITATION ON PHYSICS (1. 1) Entrainment of
Fire Plume The fire plume model integrated in
this model is based on the model by Zukoski et
al. The entrainment coefficient of which is
basically from measurements in a calm
environment, which can be attained only with
careful control of experimental conditions.
In more disturbed environment, which may be the
case in actual situation, the plume flow rate may
be significantly greater than the prediction. In
fact, the increase of the plume flow rate due to
HVAC or blow down effect by an inflow door jet is
reported by Zukoski et al. and Quintiere et al.
In BRI2002, such an effect on plume entrainment
is not considered since it was not clear to what
extent plumes from realistic burning items, a
chair for example, behave like the plume from
burner or pool fire sources.
22
(1. 2) Plume Penetration The penetration of a
plume, originated from fire source or a door jet,
through a layer discontinuity is simplistically
dealt with the critical temperature difference in
this model.
But it is suspected that not only temperature
difference but also plume momentum is involved in
this phenomenon. Further investigations will be
desired in this respect.
23
(1. 3) Air-tightness of Spaces The algebraic
equation for pressure condition in this model is
based on the premise that the pressure build up
in fire is relatively small so that it does not
affect the gas density. In addition, the room
pressures are taken relative to the pressure in
an outdoor space.
So a completely air-tight space is not allowed
but any system of spaces to which this model
apply must have some leak that connect the system
to outdoor space as the sink of mass flow. It
is not necessary, however, that every room in the
system has a leak to outdoor but is sufficient
that every room is connected with outdoor whether
directly of indirectly via other rooms.
24
(2) PROBLEMS IN APPLICATIONS
(2. 1) On Fire Source Conditions In this
model, scheduled burning rate or heat release
rate of fire source is specified as a given
condition, the manner of which is considered to
be appropriate for most of practical
applications. But since the fire spread itself
is not predicted, input of fire source conditions
is required to be realistic.
Particularly, due caution must be taken not to
input mass loss/heat release rate excessively
large compared with the size of room opening to
outdoor, or compared with the fire source area.
25
Both will limit the air supply for combustion so
result in excessive accumulation of gasified fuel
in the room to small ventilation rate in case of
the former and due to small entrainment rate in
case of the latter.
For fire sources in a real fire, a large
gasification rate without a sizable combustion in
a room is a contradiction. Furthermore, the
latent heat of gasification is automatically
subtracted proportionally to the mass loss rate
in the heat conservation equation in this model,
so the drop of the fire room temperature will be
caused if large mass loss rate continues without
enough heat release in the room.
26
There is no telling what the appropriate range
for fire conditions is but tentative
recommendations are
(a) maximum heat release rate lt 1,500AvH (kW)
(AvHopening factor) or maximum
mass loss rate lt 0.1 AvH (kg/s)
(b) maximum heat release rate per fire source
area lt1,000 (kW/m2)
27
(2. 2) Calculation Time Increment An adequate
calculation time increment will depend on how
fast the change of fire phenomena are, however, a
value about 0.5 -1.0 (sec) seems to take
care of most of the usual conditions, though
empirical.
28
(2. 3) On upper layer This model assumes that
upper layers exist at any time from the beginning
for the purpose of stability of calculation.
Whether an upper layer is a real smoke layer or
a pseudo layer should be judged taking into
account the temperature, the species
concentration etc.
In addition, a non-contaminated upper layer can
be developed by simple ventilation due to
temperature difference between rooms or between a
room and outdoor, or mechanical injection of air
of which temperature is higher than room
temperature.
29
(2. 4) Pseudo Rooms Like many other two layer
zone model, uniform temperature within a layer is
assumed in this model. In reality, this
assumption may not be appropriate for spaces
laterally very long or wide.
Such a room may be divided into an adequate
number of pseudo rooms for which uniformity
assumption can be insisted to hold, however, this
will be only possible by well knowledged user
since additional consideration is needed on
opening flow coefficient etc.