Fluid dynamics leading the way to safer flights
By Bruce Flickinger
University researchers working to optimize airflow patterns in commercial airplane cabins have uncovered new ways of containing the spread of toxic and infectious agents in these environments, and have taken the first steps toward development of a contaminant alert and detection system that can pinpoint the source to an individual passenger.
The work is especially timely given heightened concerns about the use of chemical and biological agents by terrorists, and this summer’s news coverage of an airline passenger infected with a virulent strain of tuberculosis who traveled unimpeded on several airlines and put fellow passengers at risk of being exposed to the virus.
“Our simulation program can trace airborne contaminant release location to a single seat from our application in an aircraft cabin,” says Qingyan (Yan) Chen, PhD, principal director of the Air Transportation Center of Excellence for Airliner Cabin Environment Research and professor of mechanical engineering at Purdue University (West Lafayette, IN). “Zeroing in on the source of a contaminant is key for public safety because it’s those closest to the problem who are most likely to be affected.”
The system devised by Chen and doctoral student Tengfei Zhang uses four chemical sensors and a computer model they wrote called “inverse simulation.” The technique takes into account airline cabin variables that affect chemical dispersal patterns, such as airflow, velocity, and temperature, and the concentration of gases and particles suspended in the air. The model runs the pattern of dispersal backward to its source. “Essentially, the system tracks the contaminant backward as it travels from sensor to source,” Chen says.
Using a twin-aisle, four-row aircraft cabin mock-up, Chen and Zhang generated detailed airflow information and, using a tracer gas to simulate a contaminant, first developed a computer model of the fluid dynamics involved in how the tracer gas moved throughout the cabin. The mock cabin accommodated 28 “passengers” and included heaters on the seats to recreate the effect of body temperature on airflow, and tubes emitting gas to simulate passenger exhalation.
Chen found that the seating arrangement and ventilation system had far more effect on airflow and the direction a toxin may travel than the number of passengers onboard the plane. “Since the commercial airplane has a high occupation density and rate, the amount of air supplied to the aircraft cabin is high and constant,” he says. “Thus, the impact of the passengers is relatively small, because the air from the environmental control systems determines the airflow pattern in the cabin.” Because airborne contaminant transmission is subject to the airflow inside the cabin, “a good environmental control system is the best way to confine the contaminant transmission and also for better contaminant detection,” Chen says.
Commercial airliners typically employ mixing air distribution systems, which provide approximately 10 cfm of air per passenger, resulting in a complete cabin air exchange every two to three minutes. In these systems, high-velocity air from ceiling inlets curved toward the cabin walls on both sides of the cabin is flowed along the floor and mixed in the middle of the cabin. Because of this mixing effect, a virus could be spread throughout the cabin.
In a paper published earlier this year in Building and Environment, Chen and Zhang proposed a personalized air distribution system, in which air jets are directed to the breathing areas of individual passengers and then the air is driven upward to outlets on the ceiling by thermal plumes using a displacement ventilation system. They compared the two distribution systems and concluded the personalized distribution approach, among other advantages, “could effectively eliminate the risk of spreading infectious diseases in the cabin.” The inverse simulation algorithm grew out of the computational fluid dynamics model used in this work.
While early results are encouraging, several steps need to be taken to achieve practical implementation of the alert and detection system Chen envisions. “The program now takes a few weeks to complete a simulation, which impedes it from in situ application. The next step for us is to speed up the program simulation,” Chen says. One method the team is exploring involves the use of graphic processing units, where hundreds of microprocessors work simultaneously as opposed to a core central processing unit. “Our preliminary results show this method is quite promising. The numerical algorithms in our model should also be further developed to speed up the computing efficiency,” Chen says.
He notes, “While the computing cost is higher than other models available, the accuracy is much higher as well.” Most available sensor systems, for example, can only detect the presence but not the source of airborne contagions. And Chen says his technology could be used in buildings and other public transport vehicles, as well as in airplanes.
Another technical obstacle is that a computer model would need to be created for each type of plane. Sensors for various biological and chemical agents would also need to be installed throughout the cabin. But Chen says that for many contaminants, only one sensor is needed for every nine rows of seats if the unit is sensitive enough. Chen believes the inverse simulation program could be ready for use in commercial airplanes in a few years.
Ongoing work is planned in the design and development of optimal environmental control and sensor systems for aircraft. Biological sensors, in particular, are currently too slow to enable real-time detection of contagions. “It is essential to have sensors that can detect multiple chemical/biological agents in real time and with high precision,” Chen says.