Numerical simulation of the spread of smoke in an atrium under fire scenario

https://doi.org/10.1016/j.buildenv.2008.01.014Get rights and content

Abstract

The Fire Dynamics Simulator code is used to investigate the smoke movement in an atrium under fire scenario. At first, by comparing with experimental data of the atrium fire under low and high heat release rates, reasonable model constants of Cs and Prt and appropriate grid system are determined for simulating smoke movement in the atrium, the simulation results are in good agreement with those experimental data. Then, the performance of different smoke exhaust methods in the atrium is studied. Smoke filling processes are investigated under different natural and enhanced smoke exhaust methods. Simulated results show that natural smoke exhaust method is preferred when the smoke exhaust vents are located at the ceiling of the atrium. On the other hand, when the smoke exhaust vents are located on the walls of the atrium, the higher positions of the smoke exhaust vents are preferred. In addition, the influence of the fire source locations on the smoke spreading process is presented in this paper, three kinds of fire source locations are studied, they are central fire, side wall fire and corner fire. Results indicate that the descending process of the smoke layer is the slowest when the fire source is at the corner of the atrium.

Introduction

An atrium within a building is a large open space created by an opening or a series of openings connecting two or more floors in a building. They are commonly found in hotels, commercial buildings, shopping malls and banks. It is generally known that smoke is often the main cause of deaths in an indoor fire. Protecting the occupants of a building during a fire is one of the primary objectives of the design of any fire protection system. As an atrium is an open design linking many floors, fire started on one floor could lead to smoke spreading to other parts of the building via the atrium. This is particularly important for an atrium joining large space, such as an indoor sports arena when many occupants are present.

Due to the size of the large buildings, it has been difficult to conduct experiments on their smoke exhaust system. Early investigations have been carried out theoretically and also with some reduced-scale experiments [1], [2]. When fire occurs in an atrium, flame and smoke spread vertically rather than horizontally to other parts of the building surrounding the atrium. The fire products are driven upwards within the interior of the atrium due to buoyancy. As smoke movement by the fire-induced flow is affected by many inter-related factors, it is difficult to depict the different features of fire spread in buildings using reduced-scale experiments. Recently, full-scaled experiments have been carried out on the spread of smoke within large-scaled burning facilities [3], [4], [5], [6], [7], [8]. These provided some understanding of the spread of smoke in large atria under realistic fire scenarios.

Numerical simulations play an important role in investigating the spread of smoke in large buildings. In general, there are two types of simulation method, the zone model and the field model. In the zone models, interior of the compartment is divided into a uniformly heated upward gas layer and a lower layer of cool gas of uniform properties. The conservation laws are then solved as a set of ordinary differential equations. The advantage of a zone model is its relative simplicity, allowing the inclusion of different phenomena in a given zone. Recently, numerical simulations of the spread of smoke in large buildings have been carried out using the zone model [9], [10], [11]. However, the codes developed were only suitable for building of specific shapes and sizes. Computational fluid dynamics models such as those developed by Chow [12] and Chow and Yin [13] do not have such limitations and are capable of describing the complex geometry of a building. The models solve the fundamental equations of fluid dynamics and heat transfer to provide detailed information on the fluid motions. Although the use of these models has been limited by their requirement of large computing power, modern computers have made them more accessible to fire safety designers. Sinclair [14] successfully adopted a CFD model to simulate fires in atria of different shapes and presented essential features of the spread of smoke. Cui and Chow [15] investigated three different types of atria commonly found in Hong Kong using the CFD model. Rho and Ryou [16] studied the spread of smoke for the three types of atria using both types of models. Chow also presented some numerical results [17] of balcony spill plumes in atria using the CFD model. More recently, Ding et al. [18] investigated numerically and experimentally the systems with both natural ventilation and smoke control.

When fire breaks out on the floor of an atrium, hot smoke plume rises to the ceiling by buoyancy force. When the smoke plume rises, relatively cool ambient air is entrained into the plume to reduce its temperature. However, this entrainment increases the mass flow rate of the contaminated air. When the smoke plume reaches the ceiling, momentum of the plume has an effect of a jet, spreading the smoke over the entire ceiling. Once the smoke covers the whole ceiling, the thickness of the smoke layer increases until either the entire atrium is filled with smoke or the rate of smoke entering the layer is balanced by the rate of exhaust. Due to this, smoke exhaust system is important in the design of fire safety in buildings with large space. The primary goal is to maintain an environment in which the impact of smoke and heat on occupants is reduced [19]. Ideally, thickness of the smoke layer is kept above all the occupants for a defined period of time to allow sufficient time for their evacuation.

In most buildings with large space, smoke is exhausted via the top of the atrium by means of either natural ventilation or extraction fans [20], [21], [22], [23], [24]. This paper focuses on the performance of different smoke exhaust methods in an experimental atrium (PolyU/USTC atrium), located at the State Key Laboratory of Fire Science of University of Science and Technology of China, which was built as a collaborative project between The Hong Kong Polytechnic University and the University of Science and Technology of China. While most previous investigations have been focused on fire scenarios where the fire source is located at the centre of the atrium, this paper simulates the spread of smoke with the fire source at different locations, using the Fire Dynamics Simulator (FDS) code [25], based on large eddy simulation (LES).

Section snippets

Mathematical model

In LES calculations, an approximate form of the Navier–Stokes equations as outlined by McGrattan et al. [26], appropriate for low Mach number applications, is used and the conservation equations to be solved numerically are

Continuity equation:ρt+u·ρ=-ρ·u,

Momentum equation:ut+u×ω+H=1ρ[(ρ-ρ)g+f+·τ],

Species equation:t(ρYl)+u·ρYl=-ρYl·u+·ρDYl+W˙l,

Divergence constraint:·u=γ-1γp0[q˙-·qr+·kT+·lcp,lTρDYl-1γ-1dp0dt],

Equation of state:p0(t)=ρTRlYlMl,where τ=μLES[2defu

Model verification

In order to validate the accuracy of the model, FDS is applied to simulate the fire experiment of Li et al. [31], which was carried out on the PolyU/USTC atrium. A schematic diagram of the atrium is shown in Fig. 1. The PolyU/USTC atrium is a full-scale burning facility, with an outside dimension of 27.6 m×18.1 m with a height of 30.6 m, and an inside dimension of 22.4 m×11.9 m and a height of 27 m, constructed for studying experimentally movement of atrium smoke. Dimensions of the two side doors are

Smoke exhaust systems

In this paper, the effects of different smoke exhaust systems on the spread of smoke in an atrium are simulated. To simulate the spread of smoke in a fire scenario, a 2 m×2 m oil burner with heat release rate of 1 MW per unit area is placed in the centre of the floor to form a symmetric fire plume.

Twelve different exhaust systems are simulated in this paper as outlined in Table 1. As shown in Fig. 1, there are 12 windows of 1.2 m×1.2 m each on the four side walls and eight rectangular natural smoke

Conclusions

The FDS code is used to simulate the smoke filling process in an atrium. Different grid systems and sensitivity of model constants are investigated and the values of Cs=0.2 and Prt=0.5 are chosen for modelling the spread of smoke in the atrium and found to be in good agreement with the experimental data.

Different natural and mechanical smoke exhaust systems are investigated. Results show that natural exhaust system is more favourable when the smoke exhaust vents are located on the roof of the

Acknowledgements

This work was partially supported by Ministry of Science & Technology (MOST) of China under project of National Key Basic Research Special Funds with Grant no. 2001CB409600, Beijing Municipal Commission for Science & Technology under Grant no. H023620020120, and the Research Committee of The Hong Kong Polytechnic University (Project Account Code A-PD99).

References (36)

  • J.A. Mike

    Smoke management in covered malls and atria, SFPE of Fire Protection of Engineering

    (1995)
  • L.M. Yuan et al.

    Theoretical analysis of hot smoke layer development in large space building fires

    Journal of Natural Disasters

    (1998)
  • F. You et al.

    One judgment on smoke layer interface of building fires with large space

    Fire Safety Science

    (2000)
  • R. Huo et al.

    Studies of smoke filling process in large space building

    Journal of Natural Disasters

    (2000)
  • W.K. Chow et al.

    Experimental studies on natural smoke filling in atria

    Journal of Fire Sciences

    (2000)
  • R. Huo et al.

    Studies of smoke filling process in large spaces

    Journal of Combustion Science and Technology

    (2001)
  • W.K. Chow et al.

    Plume equation for studying smoke-filling process in atria with a zone model

    Fire and Materials

    (1997)
  • W.K. Chow et al.

    Simulation on natural smoke filling in atrium with a balcony spill plume

    Journal of Fire Sciences

    (2001)
  • Cited by (57)

    • Artificial Intelligence tool for fire safety design (IFETool): Demonstration in large open spaces

      2022, Case Studies in Thermal Engineering
      Citation Excerpt :

      Fire size. The fire size or heat release rate (HRR) in the atrium, which significantly affects the total smoke generation, may vary according to practical experiences and experiments [48–52]. Seven different values of peak HRR (0.5, 1, 2, 3, 5, 8, and 10 MW) were considered with the same HRRPUA of 0.5 MW/m2.

    View all citing articles on Scopus
    View full text