By Maroun El-Khoury and Sharbel Haber, Dar Al-Handasah, Beirut, Lebanon; and A.K. Ahmed, Fluid Codes Ltd, Greenford, England

View the pdf of this article

The new shell-shaped concourse of Dubai International Airport

During the last two decades, building designs for hotels, shopping centers, hospitals, and airport terminals have undergone significant changes to accommodate the addition of atria, and walls and roofs constructed of glass. Accordingly, demands for improving the energy efficiency, thermal comfort, and safety of such buildings have significantly increased, challenging all parties involved, including the owners, insurance companies, and design engineers.

An internal section of Dubai International Airport showing the departure level (lowest level), where curtains are deployed, and other levels above it

Dar Al-Handasah, an internationally recognized engineering firm, is currently involved in the design and construction of a new concourse at Dubai International Airport, one of a handful of modern airports built recently. As part of the project, FLUENT has been used to design the ventilation system and validate the safety and thermal comfort within the concourse. Fluid Codes Ltd. provided the software, and supported Dar's staff during the investigation.

The concourse is designed predominantly for the new Airbus A380 aircraft. It consists of a superstructure of seven levels covered by a curved shell ceiling, and a substructure of four levels beneath the ground. A strong emphasis was put on safety during the design phase. In the event of a fire, an emergency smoke exhaust system will start to operate within one minute of detection. This system includes the deployment of emergency curtains, a popular means of creating zones within large indoor structures to control the spread of smoke. For any given structure, designers need to understand how smoke spreads in order to position curtains where they are most needed. For buildings containing atria, curtains are a prime smoke management strategy as they can efficiently isolate these large open spaces from the rest of the building.

A full scale 3D CFD model of the planned concourse (650m long, 100m wide and 42m high) was created, using a hex-core mesh of about 1.3 million cells. Fires initiated from different locations were considered, and one location - on the basement level near the sky-train, a glass elevator used to transport first-class passengers to the uppermost floor - was chosen for analysis. The fire was introduced as a 5MW heat source from a 10m² area. The fire power was allowed to grow for a period of 327 seconds, after which it remained steady. Transient simulations were carried out with and without curtains for a period of 15 minutes using a time step of one second. The simulations allowed the evolution of the smoke plume and the effect of the smoke extraction system to be examined. In particular, they were used to highlight the importance of using emergency curtains along the arrival and departure levels.

The smoke from the fire was represented as a scalar quantity introduced over the area of the fire. The smoke was convected with the airflow during the simulation and, hence, varied in concentration as a function of time and position throughout the concourse. Since the CFD analysis was done to gauge the effectiveness of alternative smoke management strategies, a key parameter studied was the visibility within the different levels of the concourse, particularly along the escape routes. For the corridor escape routes, the criterion applied was that the visibility should not fall below 10m.

Fire scenario after 15 minutes with emergency curtains in use; the fire originates on the basement level

Visibility contours across the concourse after 15 minutes for the fire scenario with activated emergency curtains

For the fire scenario with the emergency curtains activated, the CFD results showed that during the early stages of the fire, the smoke plume rises upwards and accumulates under the long shell ceiling. Accordingly, the main smoke exhaust fans planned for the top section of the shell ceiling would always be at the core of the smoke plume, and would necessitate an early response in activating the emergency smoke system. According to the smoke management criteria, the excess smoke after a time of 15 minutes did not cause the visibility to drop below 10 meters along the escape passages and corridors.

Visibility contours across the concourse after 15 minutes for the fire scenario without activating the emergency curtains; the smoke spreads more at the lower levels than when the curtains are in use

Fire scenario after 15 minutes without the emergency curtains activated; the fire originates on the basement level at left, and the smoke that puffs down through the sky train shaft is shown at right

For the fire scenario without the emergency curtains in use, the phenomenon of smoke plume accumulation along the shell ceiling is similar to that of the activated curtains scenario, but the plume is shorter in length, due to excessive spreading at lower levels. In particular, a significant amount of the smoke was found to fill both the arrival and departure levels, causing the visibility to drop to less than three meters in some of the escape route corridors. In addition, an unexpected phenomenon was observed. Smoke was found to puff down through the smoke-clear section of the sky train, reaching the basement level and becoming diluted with cold fresh air in that region. The time when puffing occurred coincided with the peak time of the most intensive smoke plume, when the arrival and departure levels were penetrated.

The use of CFD modeling has allowed Dar Al-Handasah to demonstrate successful ventilation designs for the new concourse, which will lead to a safer and more cost effective smoke management system.

FluentNEWS