Publications

In this paper the current status of an unsteady 3D parachute simulation code that is being developed at Sandia National Laboratories under the Department of Energy's Accelerated Strategic Computing Initiative (ASCI) is discussed. The Vortex Inflation PARachute code (VIPAR) that embodies this effort is being developed to perform complete numerical simulations of ribbon parachute deployment, inflation, and steady descent utilizing several thousand processors on one of the DOE "teraFLOP" computers. First generation working serial and parallel versions of the uncoupled fluids code that simulate unsteady 3D incompressible flows around bluff bodies with complex geometries have been developed. Preliminary results from the uncoupled fluids code along with the fluid-structure coupling strategy are presented herein.

A numerical flow model is developed to simulate two-dimensional fluid flow past immersed, elastically supported tube arrays. This work is motivated by the objective of predicting forces and motion associated with both deep-water drilling and production risers in the oil industry. This work has other engineering applications including simulation of flow past tubular heat exchangers or submarine-towed sensor arrays and the flow about parachute ribbons. In the present work, a vortex method is used for solving the unsteady flow field. This method demonstrates inherent advantages over more conventional grid-based computational fluid dynamics. The vortex method is non-iterative, does not require artificial viscosity for stability, displays minimal numerical diffusion, can easily treat moving boundaries, and allows a greatly reduced computational domain since vorticity occupies only a small fraction of the fluid volume. A gridless approach is used in the flow sufficiently distant from surfaces. A Lagrangian remap scheme is used near surfaces to calculate diffusion and convection of vorticity. A fast multipole technique is utilized for efficient calculation of velocity from the vorticity field. The ability of the method to correctly predict lift and drag forces on simple stationary geometries over a broad range of Reynolds numbers is presented.