Wing-body Aeroelasticity on iPSC/860 Computer Using ENSAERO Based on High Fidelity Equations

Objective: The objective is to compute accurate aeroelastic responses of wing-body configurations using high fidelity flow models based on the Euler/Navier-Stokes equations coupled with high fidelity structural models based on the plate/shell finite-elements.

Approach: The three-dimensional Euler/Navier-Stokes flow equations coupled with finite-element structural equations of motion are solved using a time accurate numerical integration scheme with aeroelastic shape conforming moving grid. A central-difference scheme based the finite-difference method is used to solve flow equations. The configuration is modeled using patched zonal structured grids. The structures of the body and wing are modeled using shell and plate finite elements, respectively. In general, computations are divided into fluids and structural domains and they are carried-out in parallel using intercube communications available on the Intel iPSC/860 computer. Within each domain, computations are done in parallel. Finite-difference fluids and finite-element structures grid are divided into several zones based on the topology and assigned to different processors. Information between fluids and structural domains are efficiently parsed using robust fluid-grid to structural-node type data interface.

Accomplishment: For the first time a state-of-the-art parallel computer is used for routine aeroelastic computations on wing-body configurations using high fidelity equations. Typical aeroelastic response computation are made for the Boeing 1807 HSCT (high speed civil transport) model using 32 and 8 processors for fluids and structures, respectively. The computer required was only about three times that required by the Cray C-90 single processor.

Significance: This first-of-its-kind work on a parallel computer is a major foundation to develop general purpose multidisciplinary codes using high fidelity equations for accurate computation of aeroelastic responses associated with transonic/vortical/separated complex flows.

Status/Plans: The capability of ENSAERO code will be further enhanced to more complex geometries that include moving control surfaces and empennage. This planned development will be implemented on the HPCC testbed 1 (IBM SP2).

Points of Contact:

Return to the Table of Contents

curator: Larry Picha