How do you drop a 21-ton object from a plane? Very carefully-of course-but also virtually, if simulations from the Team for Ad vanced Flow Simulation and Mod eling (TAFSM) prove successful. For the past several years, the group of researchers from institutions including Rice University and the US Army Natick Soldier Center has been investigating the aerodynamics of airdrop systems, including the behavior of airborne parachutes, the effects of a plane's airflow on a paratrooper, and the consequences of a parachute crossing the wake of an aircraft.
The ultimate aim of this research, which is being conducted with a combination of TAFSM's own computational programs and commercial hardware and visualization software (www.mems.rice.edu/TAFSM), is to make airdrop systems more effective and less costly. In the case of multi-ton objects, if aircraft were able to accurately lower ar mored vehicles, food crates, and similarly weighty items from very high altitudes, they could avoid unnecessary exposure to enemy ground fire. But a more immediate goal for TAFSM personnel is to create simulations that have not previously been carried out in a laboratory and that will eventually complement costly and difficult field tests.
|Airflow streamlines surround a paratrooper who has jumped from a plane. The colored areas represent pressure distribution at one instant during descent. |
Simulating parachutes, planes, and the related airflow scenarios is a daunting task, explains TAFSM leader Tayfun Tezduyar, chairman of Mechanical Engineering and Materials Science at Rice University in Houston. "You can solve the equations that govern the flow of air around a moving aircraft, though it requires powerful visualization and computational tools to do so," he says. "Typi cally, we generate a mesh around an object when we compute the surrounding flow. But when a person jumps from the aircraft door, you also have to calculate the aerodynamic forces acting on the person. What's happening here is that the aircraft and the person are moving with respect to each other. So you need a [dynamic] mesh that includes both of the objects in motion and takes into account the aircraft-generated aerodynamic forces that are acting on the person."
The dynamics of the parachute itself are also complex. "It's basically a kind of fabric inflated with air," says Tezduyar. "So in addition to solving a set of equations governing the fluid mechanics, we need to solve another set that governs the deformations and motions of the parachute." In other words, as the parachute unfolds and inflates, its deformation must be computed, as must the aerodynamics of the flow field around it.
|To study the dynamics of airdrops, TAFSM created simulations of airflow around parachutes and paratroopers. |
The number of coupled equations needed to be solved in dual problems of fluid mechanics and structural analysis can run into the hundreds of millions for every time interval. Over the past several years (with the participation of undergraduates, in clud ing cadets from West Point), TAFSM has been developing the methodology, as well as the code, to carry out these computations, which are run on a 256-processor Cray T3E-1200 supercomputer at the Army HPC Research Center.
Key to making use of the large data sets that result from these equations is the ability to view them-no small order for a visualization program. TAFSM's parachute data is imported into EnSight visualization software from Computational Engineering In ter national (CEI; Minneapolis, MN), which runs on a dual-processor SGI Onyx2 Infinite Reality2 and a 20-processor SGI Onyx Real ity Engine II workstation. CEI, a spin-off of Cray Research, has worked with TAFSM for several years to provide visualizations of the group's numerous flow simulations.
"One of the difficulties people ran into early on," says CEI president Kent Mise gades, "is that they had no means of looking at results. The existing graphics tools in the mid-1980s were just not developed to handle large problems." In more recent times, the parachute simulations have posed a special challenge to the makers of EnSight. The adaptive mesh that TAFSM uses around the chutes actually increases in resolution in areas of change and deformity. "At every time step," says Misegades, "we have to assume that the mesh is not only de forming, but possibly changing its topology. That places some tough requirements on your visualization code."
TAFSM's paratrooper simulations add yet another ingredient to the computational mix: an aircraft's wake and its effect on descending parachutes and par a troopers. A typical military scenario, for example, involves multiple planes flying in formation, dropping a number of paratroopers from each plane in sequence. The first aircraft in the formation flies at a lower altitude than the second, the second flies lower than the third, and so on, ensuring that the earlier jumpers are not hit by the planes following behind. But the paratroopers are exposed to unsteady aerodynamic forces as they descend into the wakes of the preceding planes.
The TAFSM simulations demonstrate how the paratroopers' parachutes react when crossing a wake during multiple airdrops. Tezduyar explains, "You know, for example, that when a parachute is falling down, the air pressure under the parachute is higher than it is on top of it-otherwise the chute could collapse. If, however, a big rotational flow field came suddenly from the top and pushed the parachute down, we would like to be able to predict if the parachute would remain inflated."
Tezduyar looks forward to when TAFSM airdrop-related research can not only benefit the military but the rest of the world as well. Air traffic is one example. "Imagine a large aircraft landing at an airport, and a smaller one trying to land right behind it," he says. Though FAA rules exist regarding safe distances between two such aircraft (so that the smaller is not adversely affected by the unsteady wake of the larger), simulations could help confirm or adjust safety limits.
|The position and orientation of one paratrooper is shown at different instants during a simulation.|
And though Tezduyar is enthusiastic about the simulations being carried out by TAFSM, he is conservative about any immediate effect they may have on real-world operations-military or otherwise. He does point out that NASA Johnson Space Center is starting to make use of similar TAFSM research on large parafoil-type chutes and crew return vehicles. But, he cautions, "We are still developing the computational tools to demonstrate that studying these problems is possible. We are looking at a trend, a potential, rather than a definitive answer at this point."
In the meantime, while waiting for ever-faster processors, TAFSM continues to develop methodologies and software that will enable it to carry out reliable, meaningful airflow simulations that may one day change procedures for commercial as well as military aircraft and parachutists.
Ket Tool EnSight, CEI (www.ceintl.com)